the advantage of active crosoovers are not hat they are way easier to design. But that they are way easier to implement and you can reuse them in the future if you ever retire your first build by just reconfiguring them, not needing to buy all new parts. And I'd like to share, you don't need to pay 100-200 for MiniDSP or MiniDSP HD to have active crossover, depending on your application there's free software like voicemeeter (that's actually a pay what you want software) or EqualizerAPO, that can help you with that depending on your setup. I personally have a pair of active bookshelfs and just bought a pair of subwoofers, I was desperate to find a way of connecting them all togheter witouth spending a lot of money, I considered MiniDSP but I didn't want to pay 100 for it. Well voicemeeter banana just saved me. What I did is I bought a second SMSL M2 DAC and used voicemeeter to send signal to both of them, then I configured a high pass to the first and a low pass to the second M2 DAC, now I have a crosoover at 70hz about 36db/octave. But I didn't test that yet as the dual subs and the second M2 didn't arrive yet. But yeah, very cheap, not quite easy to use but the price makes up for the trouble. And since voicemeeter allow you to setup macros I made 3 presets, one for movies, a second for music and a third to just send full-range to the monitors when it's late night and I don't want to disturb anybody. I'm pretty sure I'll need to make some fine adjustments when the subs and second DAC finally arrives.
Hi, im from indonesia. Mini dsp is $105 at PE, with shipping and tax its $161 at my door. Dont have budget for it. I go with $3 caps, and $6 inductors. With free xsim, im doing fine with passive.
Just as a generic passive crossover isn't plug & play the same applies for the Mini DSP. Once you factor the same things you need to take into account with a passive you can do the same with the Mini DSP. As you mentioned the Mini DSP takes phase alignment into consideration & gives you the tools to correct this, so with the Mini DSP active crossover it takes a couple tweaks on the laptop vs a complete redesign & build with a passive crossover.
exactly - in other words it took about 5 min to make a correction in the MiniDSP and would take several hours putting together and retesting a new passive unit....
Agreed. No generic crossover is going to fit any specific speaker without some tweaking. But with the Mini you can tweak in real time. You can do close field tweaking and then do room correction within minutes instead of hours. Saying that a good passive crossover is better than a bad active crossover is not fair. Of course it is and a good active crossover is way better than a bad passive crossover.
Agree, this video isn't a very fair assessment of the differences between an active and passive crossover. His poor performance on the active at first should have been compared to a generic off the shelf pre-built passive crossover. A good active crossover can do everything a passive can but it can also be re-tuned whenever you move your speakers to a new room.
@@rts100x5 - Agree 100%. To get a passive crossover design right, you really need an RTA analysis using white noise for each of the drivers. The crossover point usually will be where their rolloff curves intersect (target would be -3 dB with properly matched drivers). then design accordingly. Once you have that information, you can get preassembled passive crossovers at places like parts-express.com and some of their decent crossovers have user-selectable rolloff points. One other thing to consider (that I have not seen mentioned here yet) - passive crossovers depending on the filter type only provide a rolloff of around 6 dB, If you want to have a third or fourth degree filter configuration, for passive this is *very* expensive in terms of complexity, cost, AND attenuation (which must be made up with more amplifier power). What the big name speaker manufacturers do is settle with a 2nd degree filter but match it *perfectly* with the drivers they are including in the speaker design. Even with that, it is difficult to achieve a perfectly flat frequency response curve but any minor peaks or valleys can be corrected with small amounts of equalization. With active crossovers you can swap different driver types all day long and, with a few keystrokes, get the response curve exactly where you want it with an ultra precise 24 db rolloff. But you still have to put a lot of work in looking at the output graphs and (to do the *best* possible job) you want to complete this analysis and testing in the actual environment where the system is to be installed. And (this cannot be overstated) DO NOT OVER RELY ON HEAVY EQUALIZATION! By needing to do this indicates a major design error probably with the matching of the drivers and by overcorrecting using equalization you also run the risk of adding distortion *especially* with forcing the lower frequencies through the tweeters.
Active crossovers are much easier to design. They won't solve the issues of poor driver choice or bad enclosure design, but they're so much easier to work with. Try building a passive 24dB/oct 3-way crossover with subsonic filter. I did that in a couple of hours with active.
Looked a heck of a lot easier than digging through a box of crossover components and using that software to design a passive crossover. You even showed how to compensate for phase using DSP. All in all (having watched all the videos you've posted) I'm inclined to think that active is easier and more effective than passive.
Daniel Deverell you're not wrong. It is slightly easier. The title is a bit misleading I suppose. The point was that all the design work necessary (the actual hard part) is the exact same as passive. Many people view active as a button clicking excersize when it's far from that. Software is still necessary to design a proper active crossover and digging through a parts bin and doing some soldering is only 1% of the trouble. Fair comment though :)
Wouldn't it be possible to do a trial and error with an active crossover, measuring the response, then adjusting the delay, and measure again, until the smoothest response is obtained? Then adjust further with EQ. Not sure, but that might still be easier than a software design which would still need to be verified with measurements.
Greg Worrel I meant measuring software. But even design software would be the way I'd go. Trial and error is possible with either active or passive but I find I'm faster and easier using XO design software. Usually end up with better results too. But if someone wanted to do it by trial and error using measurements they could no problem. That's how I solved the need for correcting phase in this video, I didn't bother to use XO design software for a quick thing like that. It still takes a long time in the lab to do it by trial and error and then you need to listen then drag your amps and XO back out into the lab and back and forth. Not easier than passive IMO.
@@ImpulseAudioSpeakers The real problem is having that parts bin to dig through. Not everyone does. The MiniDSP can implement complex filters and equalizers and can be adjusted repeatedly. It can be reused later with a completely different set of drivers. Certainly, it's not fool proof and requires the same level of expertise, but if you mess up, it's a few mouse clicks to fix versus ordering a few hundred dollars of new components. Plus, with many of their devices, you can store multiple configurations and switch between them with a button press. Personally, I think it has lots off merit, especially for the beginner who wants to experiment and learn.
@@scotdixon777 I just feel that if someone actually knows how to design a XO properly they’re going to have the parts. If not, well then fine, the active XO has an advantage there. Not easier to design, easier to implement.
The conclusion at the end makes no sense. You make it out like a passive crossover is easier to design properly than it is to change settings in a program on the minidsp. It sounds a lot like the arguments of purists when it comes to tube amps vs solid state. You seem invested in passives being better. They are definitely convenient once designed but the minidsp can give you a better modeled response easily if it's not perfect and you can correct for the room after you get everything set up.
Keith Anderson sorry that wasn't my intention. I was just trying to show that active isn't simply picking a XO point and hooking it up. I get that sort of comment often. Certainly no purist type of commitment to passive here. I enjoy active and as you can see I own a minidsp. Active has many benefits. Hopefully that clarifies my position on the subject.
Keith Anderson By the way, I have a similar video about the pitfalls off passive XOers for the same reason. The two videos were merely meant to inform people who are new to the hobby that crossover design (passive or active) is more challenging than it can first appear. Hope you can get some enjoyment out of my other videos if this didn't sit well with you. Sorry again.
Impulse Audio the video is good. I loved the explanation of phase. I always knew it was a timing problem but never visualized what the actual issue was. I see the minidsp as a great tool to test drivers before making a passive crossover.
Tubes = epic win. Passive crossovers = epic win. How many DSPs will still be around in 50 years? I don't see a point in using disposable and fragile microchips in something that could last a lifetime.
@@westelaudio943 have you ever considered the fact that DSPs are only going to get cheaper and cheaper with time not to mention a miniDSP is only $100? Even if a $100 DSP only lasts you 10 years (and it should really last much more than that) that's incredible value. They are not going anywhere dude you sound absolutely crazy implying otherwise.
Now it all makes sense...This explains why the"sweet spot" for different speakers varies room to room. Your tutorial also calls out the difficulty in designing speakers AND X-overs. Thanks
Firstly. It’s important to understand the difference between the various types of crossovers. miniDSP is technically not an active crossover. It’s a dsp with multiple outputs, and not like the traditional active crossover that is analogue. I built my system with miniDSP but with separate AD and DA converters using the I2S in and outputs. The sound quality and flexibility is unbeatable. It costs a bit more but that was the best investment I ever made. Now I can build whatever speaker designs I want at no further cost
M D well, the dsp is in the active position (before the amplifier) rather than the passive position (doing the job after amplification), but sure ok it’s not a traditional active XO.
I think you make valuable points in this video with adding another channel into an active setup introducing potential phase issues, and you also explained how to use the tools to remedy those issues. The mini dsp is going to be much more versatile than a passive cross over and can be used in conjunction with passive. The beauty is in its modularity. The tools in that little box will yield much better results than passive crossovers alone. The time correction alone will allow you to achieve focus in irregular or odd shaped listening environments and you can make instantaneous changes on the fly that can't be accounted for with calculations.
I'm old school and installed a set of passive Morel's in my Ford. After listening to it for months, I bit the bullet and installed an Audison DSP, time aligned it and used an RTA to dial things in. OMG what a difference. I'd say that acitve takes even more expertise than passive because of the additional capabilities inherently included. With passive networks, you fight components to get things equalized and muck with time alignment the whole time. It's an endless battle. With DSP, the capabilities are infinite but you need to know what you are doing for sure.
Good video. Watching because I want a paramedic EQ for a passive B&W home subwoofer. From what you say, the mini DSP is perfect for this function. No need to worry about phase if I’m just knocking down a modal peak.
"phase shift"/time misalignment is primarily due to the physical offset of the focal point (axis) of the drivers...Were both divers in perfect time alignment (i.e. propagation of both wave fronts in perfect alignment), there would be roughly a summing of +3dB... assuming that both drivers were being driven with an identical signal/impulse (mass/acceleration rates of the respective drivers, notwithstanding). A relative "delay" of the signal applied to one of the drivers could bring both wave-fronts 'in phase"; additionally, a steeper slope, along with adjustment of High-cut and high-pass roll-off points could reduce some of the phase interference problems. Differences in mass/acceleration rate of respective drivers can be compensated with "torque" (power applied). Access to B&K FFT is usually handy.
Quick question. We are discussing phase cancellation and phase shifting however why would these necessarily cancel out? I can see if the frequency range is the *same* you would get cancellation. But the human ear doesn't hear at just a *single* frequency. I do not believe a 100 Hz tone (from the woofer, obviously) that is 180 degrees out of phase with a 2 kHz tone (from the tweeter) will result in cancellation. One of the obvious reasons for this is that the wavelengths are very different. Now, if you have *two* subwoofer signals (stereo configuration - same frequency range therefore same wavelength), older amplifier designs would have a phase reversal switch for the low frequencies. This was needed because the physical size of the listening room (and the resulting acoustics) may throw the subs out of phase with each other and, since they are operating in the *same* frequency range, that would definitely result in attenuation of the sound (perceived by the listener as "weak" or "muddy"). One thing that would be really cool is for you to actually get an RTA and watch what it does with a white noise signal in real time (before any equalization is applied). Many of the midrange and most of the very expensive receiver amplifiers out there with built in 7.1 surround all have auto equalization built in *plus* an included microphone (with input on the back of the unit) so that everything will be adjusted automatically when you set it up. I am a big fan of active crossovers because, when done right, the sound is amazing. Where these configurations have been more commonly found is automotive sound systems since you typically would be running 4.1 or 5.1 with just highs and lows (active crossover split) going to LF/RF/LR/RR and then the full range signal going to the center channel in the front (difficult to install since you have to do major surgery on the dashboard and center console). The sub is normally in the far rear and modern systems only have a single mono sub since subwoofer frequencies are highly omnidirectional so there is no need for stereo (you would not easily discern the direction from a cannon blast from the left vs. right - the entire vehicle shakes.... :) . The trend even for automotive today is to allow Bose, Sony, Celestion, JL Audio, Soundstream work their magic getting the passive crossover design absolutely perfect in their anechoic lab testing chambers. Known as "separates", they contain separate drivers for the satellites plus the passive crossover module that is mounted out of view. Basically let the OEM get everything matched perfectly for a typical car (or living room at home). This greatly simplifies the design (meaning potentially just one five channel amp for everything except center and a small amp for the center channel) versus 20 channels of amplification or more (and two days of adjusting crossovers and positioning working with an EQ and an RTA). Even the $20,000 sound systems installed in movie theatres today do not use active crossover designs. But it can be done if you want to put the time in. The question then becomes 'can I do it better than the engineers at Bose?'. I am *very* good at all this and have built several sets of speakers with passive crossover designs plus a couple for my cars in the past that used active crossovers. I do a good job. But Bose (and all the others) are better. But the joy in the journey is the journey itself. The first thing you do when you buy an old performance car is tinkering and upgrading. Because that is how we learn! :)
You realize I used a white noise signal to generate these plots? Nobody would say we only hear one frequency at a time. Take some time to figure out the graphs and you’ll see why the active XO was not as good.
The problem here is you are assuming an out of phase condition is the cause of the dip at the crossover point. That is a reasonable assumption as a likely cause, but not necessarily the case. You should be able to adjust the crossover point of each driver independently, and you may be able to eliminate the dip. The other thing you can do to quickly determine a phase cause for the dip is to reverse the polarity of one of the drivers and retest. Measuring phase accurately is a complicated issue, but being able to time delay the output with the miniDPS is very cool and the ultimate way to correct phase coherence issues.
Speaker Builder I was not assuming, it was an out of phase condition. I measured it. Phase is not hard to measure. And the solutions you suggest are also what I suggested. The point of the video was that a lot of people don’t understand they have to adjust the XO point or delay. They just assume an active XO box is magic. It’s not, it requires just as much speaker design as passive. You basically said as much in your comment. Did you miss the part of my video where I show actual expensive products that don’t allow you to change the XO points independently or adjust delay, etc. If people are actually manufacturing these things and people are buying them, then we know a lot of people don’t understand that active XOers require just as much design thought as passive.
I did my system with my 10 channel minidsp, without room correction and phase compensation because it’s too hard for me😅. But still the sound is phenomenal ❤
If I screw up an active crossover, it doesn’t cost me hundreds of dollars in parts and waiting on postage to change to change a setting. what about using an active crossover to aid in designing a passive crossover?
Sure that can be done to save money. I don’t know about hundreds of dollars. But yes active XOers are very good as a primary XO or an aid if you want a passive XO. It’s difficult to translate an active XO settings to passive parts though. So it’ll only tell you if the driver is happy with what ever XO point or slope etc. You’ll still need to come up with the passive design. And as the video says, it’ll be just as difficult to design the XO actively as it is passively so you may do some double work.
So if i'm quite new to speaker design, can i just get the 95$ mini dsp and tweek the hell out of it, when i don't have a ton of components (0 actually) to tweak a passive crossover design ? Because even when using a software to design it, parts are not always the same sot they will have variations 5-10%. So i might have to buy even more parts so i can tweak it even more. Imagine a 3 way system, with the mini dps you can just plug it in, and spend hours changing controls and taking countless measurements.
Quetzalcoalt this is a fair question. The answer is yes and no. First, dont worry about parts variation. Far more important is the placement and duty of the component. As for the minidsp or equivalent, I would suggest you at least buy a microphone and a couple large (50uF) protection capacitors also. That way you can do two things. You can tweak like crazy AND measure what you are getting. Our brains will adapt to bad sound and sometimes it is not until we hear the right thing that we realize what we liked actually sounded bad. We also convince ourselves something sounds good when it doesnt. Measurements help keep your brain in check and bring you back to reality. This isn't me suggesting you should like what the mic says. I've been accused of that before. Rather it is a way to visual what you are hearing. There are some small things that the mic can help you see like a narrow dip or peak that we can hear but would never know what it is or what to change. The second thing you can do is take driver measurements and put them into Xsim or some other program and then simulate the minidsp inputs. This allows you to "tweak" much faster so you can get in the ballpark quickly. It will also allow you to quickly see a few different approaches to the XO settings. This is the way I do active XOers and it saves me hours of tweaking and measurements. The protective caps will just make sure you dont blow a tweeter if you make a mistake or get turn on thump from your amps.
Have you designed a fully active crossover speaker? The miniDSP has 2 inputs and 4 outputs, so you could make a pair of bookshelf speakers with only one, or a 3/4 way using one for each tower. Might be fun selecting different amps.
Yes many times. It is fun. The mini dsp is a great product. I wouldn’t touch one of those other crossover boxes that don’t provide the proper flexibility though.
Ok so you said the Mini dsp was not a fix all but you proved it was. You were able to correct the phasing, then with the eq would be able to tweak the response curve. Also on your crossover screen on the mini dsp it looked like you had a gap between the points for the woofer and tweeter. For those that are not savvy on designing and building the absolute correct crossover be it passive or active it looks like the mini dsp is a winner.
Cyber Trucker Road Show I'd have to watch my video again to see what I actually said, but the point was to say that unless you know what you're doing with it, it won't be any easier than passive. As for the gap, take a look at the screen shot again. Each channel has low pass and high pass capability but usually one of those is muted. In this case it is muted. Both channels were set to the same frequency for this test. Having said that, often the XO frequencies are different (sometimes a gap, or overlap) when designing a proper crossover. I've never had a properly designed speaker use the same frequency on the transfer functions. The acoustics points are what need to match.
Impulse Audio Never mind lol I think you were just referencing active xovers in general. The phase issue looked like you could have tweaked the points to fix the gap woofer @ idk 4k then tweet to 3.6k have some overlap. Great vid, I didn't mean any offences, my observation.
Cyber Trucker Road Show no problem. I don't take offence to comments on my channel. I like when people comment so everyone can learn. I didn't spend time adjusting the XO points because it was intended to show the pitfalls of arbitrarily choose Xhz to XO using one of these "magic boxes". But yes definitely if doing it properly the point, slope, delay, etc. all need to be adjusted to get it right.
I agree that you still need to be aware of phase, but as long as your DSP allows you to correct it, I'd say a DSP is way easier to work with then a passive XO. Downside is the fact that you need more amplifiers, obviously.
I got around many of the issues you point out by going with a 4-way active and choosing drivers that had good behavior about an octave beyond each crossover point. Large baffles with Vifa AG35, Vifa P13, Vifa P26, Eminence DVC-12, with 3200Hz, 500Hz, 100Hz. (this was about 1992?)
Hey you didnt point out the major attraction from the active xover, that by splitting the audio into separate bands BEFORE the power amps, each amp and each speaker unit now handle only their allocated band of sound, or in other words, each amp now outputs LESS power at a given SPL, coz it no l9nger has to amplofy the other band(s) of aidio that otherwise gets sent to the speaker which filtres it out oventually in a passive xover design. I find active xovering an advanced level of playground for audiophiles and engineers that give much much more tuning possibilities than simply matching one amp with one whole speaker. The audio perfectionist's l9ve here.
That is true, active has many great advantages over passive. This video was only to show that they require equal knowledge of how to design a speaker. Seems I didn't make that point very well though.
I don 't get it. You say you have a dip at the crossover frequency with a phase shift of 140°. Would it make sense to simply revert the tweeter phase polarity, thus getting a -40° phase shift and a little bump like you have with the passive crossover?
didierleclerc66 in this case inverting the polarity would have helped a lot. But it won’t always be that way. It could have been 90 degree out. Then what? And you still are out 40 degrees which is pretty crap in my books. Plus all the other issues of wonky response not related to phase. Generally, even if you knew in this specific case that flipping polarity would help a lot, it’s still a crap speaker, far far far worse than the passive XO. So unless you want to go through the proper design just like a passive XO, the active XO will suck. Therefore it’s really no easier than passive.
My bad OP . I wasn’t thinking when I stated the drivers are going to have different wavelengths , as at some point they will have the same frequency response .as they crossover . A duh moment
Generally the group delay of the woofer is longer than the tweeter due to its higher inertial mass. Therefore to synchronize them in time the digital delay should be applied to the tweeter This is still not good enough to get phase coherence. To do that they need to have the same geometric center of propagatoon, in other words they have to be coaxial. This problem is like throwing two rocks in a still pond and getting them to produce ripples as though they were a single rock. To achieve that they not only have to hit the water at the same time but also in the same place, otherwise you will still get phase interference patterns.
Great Video!!! My question is how do you test for alignment? is oscilloscope the only option..? is there software and calibrated microphones? this is intriguing and confusing all at once, LOL ...thanks
pallusubaru To find to acoustic offset I use a measurement microphone. You can either take measurements and then use software to determine the settings, or you can flip the tweeter polarity and add delay while taking measurements until the null is at the greatest and flip the polarity back. You're right that it is not simple. Unfortunately people think active crossovers are easy, which is why I made this video. Eventually people will hopefully understand that to get good sound quality from passive or active XOers a little work is required.
I have been following this channel a little as an obsessive compulsive/audiophile and DIY guy myself. I had some interest in the content but I have to say this video breaks the cycle for me. At 9:10 he says that you cant just plug and play with DSP equipment, but that is in fact exactly what you do! You plug it in and you play with it. Lets analyze this shall we. For $100.00 you can get the MiniDSP unit. That is about what it costs to buy the equipment to build your own passive two way filter and you likely will only afford a 2nd or 3rd order filter at that. With the $100.00 for the MiniDSP you get to choose between first through 4th order filtering, you can sweep the entire spectrum at will in a few moments where to filter at. You can time align and multi band PEQ all in one nice simple package. This guy proposes you get crossover software model a filter and hope it functions real world. With some sweet DSP you can real world tune and get it dialed better faster and I mean way faster. Yes working with a complex DSP device can be technical and a beginner my struggle but there will be no better way to learn than playing on a DSP. You can try anything and just reset it back to zero no additional money spent. This guy, if he want after designing and building a passive filter network must redesign and go buy more gear build it up and then test and he does not know if it will be better until after all the work. That is idiotic to me im sorry! Now to go further and say that DSP equipment is not for beginners well how much easier is it to buy design software model a component build it install it and then prove that it is infact the best design possible! DSP is going to be more user friendly than that. There is another aspect, environment effect. If you near field test and design you can get the system to be good or correct but then when you set up the system in a permanent environment the sound will be different than was ever measured during real testing. If you run a DSP system you can set up the rig in the location/environment desired, then place the measuring mic/ears in the listening position and tune away and you can get better results this way than all the seemingly sophisticated things this guy is doing. I am aware that eqing a system to a single location will usually create a mess in other locations in the potential listening environment but that is another topic for another time suffice to say I strongly disagree with this video content!
Impulse Audio, might I suggest you use a DSP system in your design process. If you like building as much of the system as you can (like I do) and want to use self built custom passive filters in your audio systems you should but you also should take advantage of all the tech available. Using DSP and your spectrum analyzing/snapshot equipment you can model a filter and then test it, you can make some changes and test it all in under 1hour you could model and real time observe 10 different crossover configurations.
You know I don't even know why am going into this, the PEQ function alone is worth the cost to have. If anyone is a true audiophile looking for the best most correct sound quality/fidelity they can afford than DSP is the only way to go. It is no more expensive for the level of improvement you get compared to any passive design. It is flexible adaptable it is the end all be all solution to DIY audio!!!! Period. I mean you can take a poor speaker system and digitally correct most of the problems in the bad design. Check this out. For years horn loading was the way to go for the efficiency of it. The sound quality issues with horns at the early days were tolerated because of how functional they were. Today most horn loading is done away with to try to maintain better fidelity of the sound but with DSP technology where it is at today, you can eq out the misbehavior of a horn and then benefit from their designed efficiency!!! Come on guys think about it you need to get on the DSP train or you will miss out sonically!!!
Good vid. I'm not sure that I agree with your assessment that active without delay are useless given that passives don't have delay and there are other benefits to active. Keep 'em coming!
Warren Smith Passive does have delay, its just harder to do. You can add poles to the network, which also steepens the slope so you have to be ok with that, or you can use a very complex ladder network. Most people use an extra pole or make physical changes to the baffle. Definitely need to understand whats going on with the phase to be able to get it correct. Active is easier if you have delay capability. The active XO boxes with a fixed slope and no delay make it impossible to solve the phase issue. Good point about active having other benefits though. They do have a lot going for them. Especially the ability to dial in delay rather than the challenges of getting a passive right.
You make some good points, but just change the title!!! Active crossovers won't fix stupid (implementation) so to speak, but they are indeed so much easier to implement and adjust if need be. Passive crossovers take labor. One has to fabricate the system and then they are stuck at a less than optimal arrangement. With active you just have to be knowledgeable enough to push the correct buttons. With active, there are many benefits, and near endless adjustability for different source material and/or preferences.
Anonymous DueToFascists I'm not sure I agree passive XOers are labor intense :o A few components soldered together isn't a bit deal. The case against active always seems to be the number of wires and amps required. So really they're just different but both require good design. I do agree a superior result can be had with active, but it is smaller than most active proponents seem to argue. And there are advantages to passive such as potability, simplicity, etc. that really depends on the application the user is going for.
Hi, im from indonesia. Love your video. Mini dsp is $105 at PE, with shipping and tax its $161 at my door. Dont have budget for it. I go with $3 caps, and $6 inductors. With free xsim im doing fine with passive
When you would like to replace one of drivers to another with different the highest or the lowest frequency, or with a different sensitivity, or change amplifiers, or to reduce a room main resonance frequency, you'll get to know how easy DSP is. A passive crossover make impossible to use current source amplifiers for mid and high range drivers. And etc. If you build entire system for a personal use, this imho have to be a DSP with an active crossover and galvanic isolated channels, and bunch of mono amplifiers that match bands and power.
I have done that and no, it’s no easier. Sure, it’s a few buttons to actually program it, but any of those changes requires measurements and knowledge just the same as passive. Absolutely no difference.
@@ImpulseAudioSpeakers Sure. Getting the numbers is the most experienced part of the process. This all could be similar to digital and silver photography. 'Our' numbers are like proper light and composition. However photographer can find the best idea way more fast with a digital camera imho. Then he/she can replicate this photo using Hassy. At the moment I saw only one $5k PS Audio product which is not a PA digital crossover. May be others exist (like hyperx modules for example). But this number will increase. And market will start offering loudspeakers with drivers that can be replaced/upgraded. This is not 100% true of course, but manufacturers would like to use this idea I think -)
I understand the point you are making here, but it does not align with the title of the video, or in fact some of the other things you are saying. True, you cannot just shoe-horn any type of crossover (active or not) into your system by picking a crossover frequency and calling it a day. There are other considerations. But throughout the video you systematically disproved the title; active crossovers ARE easier than passive. They just objectively are. You can design and tweak one to suit - and it will do a good job - within a matter of minutes. With a passive crossover, not only do you have to understand impedance (a thoroughly non-trivial requirement) but even in the simplest case you have to design your crossover in software (let's say that is equivalent to setting up your active crossover in its software, although it's almost certainly more complicated), then buy the required parts and wire them up. At this point, if something in your simulation was wrong or not accounted for, you have to start over. You can just change a parameter in the active crossover. Moreover, passive components are susceptible to overheating and electro-magnetic interference. They are categorically worse, but in some instances, necessary and convenient. I use them because they sit tidily inside the box, I don't need a DSP unit lying around, and I only need half the channel count on my amplifiers, and frankly I like the challenge. When I do live sound however, it is the opposite. The flexibility of an active system in that context outweighs the extra gear required.
Active is clearly the better choice; always! As mentioned in other comments it's much quicker and cheaper to design and then tweak an active, especially a DSP version, crossover. I'm a bit chagrined that see the obvious error of his "apples and apples" comparison between the passive and active. The passive was a 2nd order, 12dB/octave design. Barring other factors, in order for the response to sum to flat at crossover you must flip the polarity of either the tweeter or the woofer. A 2nd order crossover will give a +90 degree shift to one driver and a -90 degree shift to the other at the crossover point. Combined it results in 180 degrees out from each other. This is physics gentlemen, not my opinion and certainly not a marketing position. In order to get a flat response manufacturers always flip the polarity of one of the drivers. This gives the best frequency response at crossover but, destroys the integrity of the overall phase response and, therefore, it's performance in the time domain. It's an unavoidable and poor compromise. Using a 4th order, 24dB/octave gives you even more phase shift at the crossover frequency; +180 and -180 degrees but, as is apparent the phase "wraps around' so that both drivers are in phase again at the crossover frequency! It also reduces the area of the crossover where both drivers are working and that's a good thing too. BUT; in this video he made no mention of "correcting" the flipped phase that was, I'm sure, in the 12dB, passive design so that they were in phase for the 24dB active design. If fact, if you do your crossover in the active design correctly and optimally it will guaranty a dip in the response at crossover if the polarity is flipped on one of the drivers. In fact, the depth of the dip is a figure of merit for the overall performance of the drivers and the crossover. The deeper the dip at crossover when you flip on driver the better the summation when you flip it back to correct polarity! You can do a 24dB/crossover in a passive version but, it is a doubling of parts ( two caps and two inductors per driver). If you use quality components for good performance then it will absolutely cost more than an active like the miniDSP and that assumes you get all the passive parts dialed in initially and you don't have to buy extras to tweak the design. He also didn't mention that an active crossover is also the best because you will then be biamping the speaker which has it's own virtues and is the best way to go. This is not the same as the pointless exercise of bi-wiring which produces no measurable or double-blind positive results and is simply a con job from higher end manufacturers. Don't drink that Kool-Aid. The reason you don't see active crossovers in consumer equipment is because if the end user is supplying the amplifiers they then have the responsibility ensuring that the output level is correct between the high and low outputs based on the sensitively of the drivers. This requires a good measurement microphone, an RTA or some other analysis device/software and some method of getting the room "out of the measurement". This is far and away beyond the skill, test equipment available, and the time invested by pretty much anyone but, people involved with this hobby. Lonny J.
Obsfucation I don’t think you understood the video and you don’t understand how phase works. I was showing that active requires just as much design consideration as passive. Not that passive was better. I explained that to other commenters but you obviously didn’t read my responses. I’m tired of explaining it so maybe I’ll do another video. I even wonder if you watched the video and tried to learn something or just watched a little bit and jumped to conclusions. Did you watch the ending where I describe how phase and time alignment works?
But you can also do that with passive. The point of the video was most people don’t. They don’t even know to measure and see that the phase is wrong. That’s why it’s not a magic box or easier. Better, sure ya I’d say active is better.
mark clausson the point of the video was about not measuring and using active XOers blind. So what if it was 90 degrees out of phase and I inverted the tweeter, then what. And if I hadn’t measured how would I even know the dip was there? Yes the dip can easily be corrected, but for those who don’t go through the proper process of figuring out the slopes, delay, eq, etc. then active crossovers are no better than passive.
can you get perfect phase response with a passive crossover? this would save me allot of money for the atmos system i'm trying to build for my bedroom ultimate competitive gaming setup. reason i'm being so obsessed with phase is i think it's crucial for our brains ability to perceive a sound location with the least processing time.
alex kirk yes passive can be phase aligned. Active can make alignment a click of the button while passive requires phase to be considered wholistically.
thanks for reply :) is there any chance you could do a video on it, or link me to one if it already exists? i know it's probably no easy task but after hearing my friends co-axial dsp monitors i've been obsessed with the idea of perfect phase in loudspeakers.
A passive version? My video series about the Life S5 speaker ends up with a passive crossover in phase. Or do you mean perfect "linear" phase? That's a bit different and usually requires a lot of other compromises to work. I've never bothered with one of those.
i was talking about perfect linear phase but i'm guessing it's way easier just bi amping it with a mini dsp. do you have any idea where i should start reading as i can't fid anything on google. btw thanks for all videos, thanks to you i'm no longer scared of passive crossover design.
alex kirk hmm, I can't think of any resources that deal with that subject. There's probably a few AES papers. Minidsp might have some info as well. Or Dirac. It's not a subject many people are interested in. Seach FIR filter because FIR can correct phase where as IIR filters generate phase shift. That would be a good place to start.
So about phase misalignment, how do you respond to this: "The K231 is an all analog design, engineered with only the highest quality audiophile grade components. Users will have complete control over sound and be able to eliminate distortion with subwoofer, high and mid-range channels, which are phase aligned to within fractions of a degree." www.releasewire.com/press-releases/sublime-acoustics-k231-3-way-active-crossover-set-to-revolutionize-consumer-audio-market-with-affordable-professional-grade-sound-capabilities-791166.htm
What is the root cause of the phase difference in this example? I was always under the impression that with a passive crossover the components used created a slight phase shift so that when you run a 12db per octave slope you create 180 degree out of phase signal around the crossover point. I was also under the impression that this would not happen with a DSP. But, clearly, it has happened. Is this due to driver placement (i.e. the center of the woofer is further away than the center of the tweeter).
Justin Gardner there is a phase difference due to the time delay of the tweeter relative to the woofer. But you have also learned that dsp also introduces phase so that is good!! The acoustic response changes therefore the acoustic phase changes as well. This happens with active or passive crossovers. For every order or slope you get 45 degrees of phase change.
I'm using a Pioneer active xover with my Tannoy 10" Gold Mon. and my JBL 4320 (this one as a subwoofer) to listen to vinyls. This must be much better if I switch to the digital minidsp. But how do I use that minidsp unit for analogue audio? Impossible without further complication, right?
Stick with your Pioneer or pick up an Ashly or Marchand. Both sound very natural. Not DSP's but once you get the points figured out by testing, set it and leave it. You'll want to fabricate a BSC network between x-over and amp for the best sound. Do a Google search for a diagram. It will only cost you a few dollars in parts. Be prepared to be blown away by the sound quality.
What I really don't get with the practicalities of phase is that the SPL can measure fine (flat) but I believe there can still be phase issues which play havoc with the sound. Any thoughts welcomed, cheers Ryan.
MrFrobbo what is usually happening is a case like this is the FR is flat but phase is wrong, so the power response (a combination of all points off axis) ends up being very peaky in the midrange (or where ever the XO point is). It can sound very glaring and forward. Something won’t sound natural.
Input = RCA (unbalanced) > DAC > (filter) > ADC > RCA (unbalanced) = output? What about SPDIF In > (filter) > SPDIF Out (FWIW, Coax vs. Optical = Jitter)... analog amplifier has DAC inputs... vs. PWM amp? Fine Phase adjustment to compensate for mounted driver offset (mSec.... µSec.)
if I have a 3 way speaker and run it with a miniDSP does that also mean I need to run a 3 channels of amplification? Or is there a way to somehow run only one amp to power all 3 speakers with the miniDSP?
You’ll need 3 channels, or you can use a passive XO for the tweeter and mid and then use the dsp to cross the woofer to the mid and tweeter combo. This is a really good technique because the passive XO protects the tweeter, the mid and tweeter need less amp power, and it saves 2 channels on a stereo system.
You can do things in seconds (like Harsch crossovers) that would be extremely difficult with passives. It is far faster, more flexible, more precise and trivial to optimize, plus you get the gigantic benefit of biamping. DSPs are vastly better IMO.
But all you needed to do is reverse the phase of the tweeter like most speaker companies do and use the active eq in the nuke to flatten everything out. This is how we do it in every recording studio. This is also how JBL profesional theater speaker systems are done 🤷🏻♂️
What if the phase had been 90 degrees out. What if the speaker didn’t require polarity flip. Just because I sorta made it work with measurement equipment and knowledge doesn’t make active any easier to design than passive. I needed the same knowledge and measurements.
I want to purchase this products for my powered subs. I read a lot of good reviews in forums. Seems like people can achieve phase correction after few short learning curve. What is your opinion products like this for powered subs? Thanks!
itoosh it's a great product for speakers and subs. The video wasnt meant to suggest its a bad product. Only show that it is just as much design work involved in active as it is in passive. Good luck with the subs.
So I have yet to hear about a passive crossover network that time aligns drivers which would be a phase compensation technique. The truth is that passive crossovers have a number of drawbacks over full DSP units. BTW The mini DSP is a pretty no-frills unit. There are far superior processors out there to that although they do work. The other thing that seems to be lax in speaker design is that so many designers only consider the FR plot on the page and don't actually listen to the driver. Many high-frequency drivers exhibit serious distortion issues in the area below 2Khz. It is more common than not to design speakers now to cross-over above 2K because of that. We took old JBL 2206 cones and using a waterproofing solution stiffened the cones which changed the response curve to one that allowed us to roll it over at 2200Hz. It was coupled to an old 3" B+C compression driver mounted to a CD horn. A little time shift on the low drivers (it was a 2-12 stage monitor) to compensate for the physical time offset plus adjust for the phase misalignment at the crossover point and all it needed was a few little eq nudges. I could have mixed a record on it. The thing was with the crossover set to 1600 Hz made it painful and unlistenable. 1600Hz was where the designer said the crossover should be set. Why didn't it work? Becuase just like almost all designers they only went by what the specs say. Specs don't plot distortion vs frequency. You have to listen for yourself and decide. "Well, that is what pa speakers sound like". NOT GOOD ONES!!!
Rob Chapman I don't know any speaker designers that only care about FR and don't listen. Hmm, that actually sounds absurd. Have you actually met someone like that? Check out ladder delay networks. They are passive and compensate for phase. The minidsp is extremely advanced compared to the bricks I was describing in the video. But even among the top contenders minidsp sells some very powerful units. Tweeter distortion is entirely dependent on the model of tweeter. I would be careful generalizing that tweeters distort below 2khz. It may be a rough approximation where most do, but hardly accurate for all. HD is one thing while woofer beaming and cone break up are also commonly an issue above 2khz. So pick your poison I guess.
Sadly, I have met a few and have actually been chastised for not succumbing to their sonic philosophy. I view it this way. If it doesn't sound good regardless of what angle I hear it from it is an unusable piece. I guess the thing there the guys who like fidelics don't usually gravitate to high SPL systems as traditionally they have been polar opposites. The thing is we have high SPL systems that now are actually producing sound with a great degree of fidelity and that means it is time for the old guard to get out of the way. I wasn't actually trying to generalize but the majority do. I guess in a perfect world we would need to use a 3-way system for this. I don't think I would trade a bit of beaming for distortion though. Everything in balance. It is, of course, a trade-off but more often than not narrowing the coverage angle a bit is a good thing. Besides that if you were to measure most stage monitors on the market you start to question who is designing these things anyway. They are never flat and often have crazy phase holes.
Gara Von Hoiwkenzoiber I did not know how bad the phase would be. Yes phase is easier to correct with active. But the same knowledge and work is required to get there. That was the point of the video. Not which one is easier to program. It’s not a magic box like some people treat it and think it is. Not many people even know what phase is but they think active just makes it ok.
Hi, im feom indonesia. Mini dsp is $105 at PE, with shipping and tax its $161 at my door. Dont have budget for it. I go with $3 caps, and $6 inductor. With free XSim, im doing fine with passive.
Sorry, something is confusing me... The wave length of the frequencies! You can not just add frequencies as you do. The tweeter can not produce so loud frequencies at 20, 30, 100Hz or so low and bass can't do it on 10000Hz or higher. Also, the only case where two waves can interpolate is when they are of the same frequency (or same wave length). And when the phases are 180 degree opposed they will cancel themselves. On the diagrams we can see that drop but that is not so significant when the music is in question because it is not the permanent drop in the music spectrum. The question is can anybody register that drop masked with the whole other frequencies? It is only seen in the lab when you sweep the frequencies. For me, the active crossover behind the preamplifier with amplifiers for each individual range is the top. Can you imagine how can you fine tweak 6-way system? Even with the cheap components.
Nice video. Like you said the title might be slightly misleading . I could be totally wrong but what you explained was about the acoustic phase difference which you are compensating with delay . But what about electrical phase difference . I know they are connected but it will be good if you can make a video about how they are tied together
jojiran i could possibly do a video about electrical phase sometime, but that isnt a strong point for me. They arent really connected to acoustic phase either. Im more knowledgeable in the acoustic realm but Ill try to cover that topic as I go. Thanks.
Thanks for your response . Ive been trying to get my head around this subject. Since I work in a studio setting , it is important to understand these concepts well in these times of evolving technology What you did in the video ( adjusting delay) will produce phase coherent sound i.e. sound waves hitting the human ear as if all those frequencies emnated from a single single source even though they may be physically separated The phase response you see in common design tools is the electrical phase . That has got to do more with how the amplifier sees the load. I was very confused about these concepts initially but now I think I have a better understanding .
jojiran yes, and also be careful that some of those design tools show the theoritical acoustic phase caused by the crossover. That is not useful information because those design tools assume perfect woofers and tweeter, which none exist.
I get your point that one cannot naively convert a passive design into an active or design a system from scratch without having at least minimum required knowledge regarding designing loudspeaker systems. you give a very good example in this video yourself. I think rookie DIYer having no clue about phase issues would try to eq that dip out in minidsp and arrive at a badly designed system that potentially measures quite flat but sounds just not right. also one thing to remember is that those who attempt to design active systems do it not only for the sake of driving each driver with its own amplifier but also for possibility to take full advantage of DSP power - driver time alignment, linear phase xo, driver linearization, steeper slopes.
David Gale thank you for understanding the point of the video. It was lost on many. Your right about the advantage of DSP. It is very powerful and if you know what you’re doing then it is unmatched by passive.
DJ Harry Trivedi My next video is kicking off a 4 part series of building and designing a desktop speaker from scratch. I'm going to look at how to design a passive XO and it should help you a lot. Should be starting in a week. The design part in two or three weeks. Keep an eye out for that.
Don't bother and keep saving your money for just a bit. Go proper or go home. Get a dsp and 4 stereo amps; or 3 stereo amps and a plate amp or 2 for the sub(s). Power your 4-way and you'll have audio that will rival any passive speaker. Think 12 inch sub, 6 to 8 inch mid-woofer (2 if you want), 2.5" to 4 inch dome upper mid (or cone), and a 7/8ths or 1 inch tweeter. You'll be able to experience anything your heart desires. Turn up the subs. Turn off the subs. Then listen to a 2-way or a 3-way setup for vocal clarity. Compare and contrast all you want, and be an audio GOD in total control. 16hz to 20khz and above will be entirely in your reach. I've never understood the appeal of building your own passive speakers. You're then stuck with a speaker that can't do much of anything (just like all the other crappy passive speakers on the market for the average idiot consumer). And chances are, your passive built setup won't compare to the hundreds of passive speakers already on the market. I guess you could add an equalizer with a passive setup, but that's fairly lame and doesn't address the flaws of passive speakers. Have fun and build an active system which will sound better, do more, and can be tweaked and played with according to whatever your heart desires.
There is definitely a place for passive crossovers. I am a fan of active crossovers but for my 4 Atmos speakers, I did not want to deal with 8 amp channels and active crossovers. So I learned how to do a passive design and have been very happy with the result. Xsim is very cool software and can be learned in a few hours. Although I think I spent at least a couple of days last winter playing around with it.
Fair enough. And your'e right. I was just trying to oversell the fun and advantages of an active setup. Too many overlook the option and many don't even know it exists.
You need active electronic crossovers that you can actually regulate continuously until you balance everything out , Rane and dbx work well. I don't waste time and money with passive anymore.
I was always under the impression that you could just change the phase on a mini DSP. I think anybody who would buy a mini DSP would probably also buy a measurement mic and download REW.
Justin Gardner you can easily change the phase via delay on a minidsp. In my experience, a lot of people do not buy a measurements mic and use REW let alone know how to use it. And when you consider how many people buy the fixed slope active boxes, it's scary. A lot of people think these are magic boxes. It's why I made the video.
There is no comparison to which is better (active) but you just need to have a very good grasp of digital/analog design principles as well as standard (discrete component) X-over approaches, to design, implement, and build a good active 3 way system; that also measures well. If it (active 3 way speaker design) was easy, all the Hi Fi fiddlers and speaker companies would be doing it.
Marcus dahlén all dsp crossovers have these same issues. Some are more automated and can apply eq, but can’t fix a hole like that. Apply EQ until flat isn’t always a good idea either. No matter the hardware, it’s best to apply best practices and get the XO dialed nicely.
@@ImpulseAudioSpeakers Do you think phase is such a big issue when you can cross 48db? I like to configure my systems with dsp because it's so flexible and also very exact. Today I want to cross my ib sub at 50 instead of 45, tomorrow I want to ad some midrange:) Think I could make so many drivers sound so much better by controlling them like this and also it's fun:) Thanks for the reply man, I'll tell you what my air motion Transformers sounds like free air mounted if it's nice ☺️
Marcus dahlén certainly dsp is more flexible and precise, but phase is an important aspect even with steep slopes. Subs are probably less prone to issues when making tweaks, but the higher frequencies should be checked with measurements. Easy to think it sounds great and miss a lot of information. Especially with suck outs like I showed in the video.
@@ImpulseAudioSpeakers So I just realized I need to analyze the frequency response and correct possible phase issues with the delay function. What is your opinion on just letting the frequency ranges overlap just a little bit to avoid the Gap in response?
@@ImpulseAudioSpeakers also do you have a suggestion for a good midrange driver 800 - 2000hz Did you try the Accuton mids? I'm feeling like putting a dome on top of the speaker Best regards 😊
Then, of course, you've got your extremist Klipsch fanatics using Xilica XP-4080 and 8080 DSP's on their Jubilee rigs with the awesome K-402 and K-510 horns (Roy Delgado's secret sauce). And for them it is so much better than anything passive. But that's very extreme and niche. But yeah, for that little pair of DIY 2-way bookshelves passive is just peachy.
Active is the only way to go. I haven't touched passive for years. I have a miniDSP 2x4, MiniDSP DDRC24, Dayton DSP-408, amplifiers with built in DSP, etc etc. It makes life so much easier and cheaper. Different speaker projects become much easier for newbies to get into without having to have a degree in electrical engineering and soldering. With a DSP you can make changes on the fly while listening to music rather than taking everything apart, buying different components, connecting components then listening..no way to A/B compare or tune the system quickly.
davefred3 yes it would be much better. But someone would need to know to reverse it, and it would do nothing if the phase were out by 90 degrees instead of the 170ish degrees it was out here.
If you talk to any speaker designer Or engineer which I design all of my speakers custom myself they would rather use an active crossover but it's just too expensive to sell to the general public I will never use a passive cross over On my system ever the The average passive crossover is garbage And active crossovers do not get hot at all we know what happens when Things get hot
@@ImpulseAudioSpeakers That's the easy part I am a carpenter by trade The hardest part which I will concede to is theTuning of the crossover Per individual driver I run A four-way stereo takes time to Get it right and we're not even gonna talk about room treatment And speaker placement whole different animal
Active is definitely the way to go. Its for sure easier. You can actively reconfigure and compare and contrast results on the fly. Your phase issues have more to do with driver size and center to center spacing, there are rules to speaker design that you would be wise to follow to eliminate these issues....a crossover point can expose your poor designs if thats the case but crossovers don't cause this off the bat. A 48db LR crossover is designed to be in phase, so if you still have phase issues its because bad choices, else where in the design.....If you know better you can do better =)
DR.NNOO I showed in this video how an LR4 XO isn’t automatically in phase. Active has just as many design challenges as passive. If you think otherwise you’ve missed the point.
@@ImpulseAudioSpeakers Thats the key, you don't design active crossovers, you buy them and apply them how ever you like , where as a passive isn't something that you can so easily reconfigure. That, in itself, makes them easier than a passive. Anyone who assumes that a crossover of any type is going to automatically make everything in phase, is lost, anyway. Still, with an active crossover you can try unlimited configurations in the same amount time it takes you to assemble one passive crossover limited to the amount of passive components you've got. If you dont get it, yet....make a video of you trying 3 crossover points with an active, then make another video documenting you making the same 3 crossovers from scratch and we'll let the people decide which route they want to go =)
DR.NNOO ok first yes you still have to measure each driver and use software to determine slopes, eq, etc. You don’t just buy and start fiddling. But even that aside, fine, active is slightly faster as pulling out the computer and booting up the software, plugging in the USB, programming it, hooking it back up to multi channel amps, than alligator clipping together some passive parts. Yes it’s SLIGHTLY faster. All the rest takes the same time energy knowledge and effort. I like active XOers. All you active fan boys just don’t get what I’m saying. Sorry I’m kind of exploding on you because this keeps coming up. Yes, active has many many merits. I use active XOers myself. BUT it’s NOT really any easier and if you think it is you’re doing it wrong. Saving 4 minutes on XO assembly over the course of a 6 month design process is not helpful.
@@ImpulseAudioSpeakers Slightly faster is a complete understatement, my active crossovers are built into my amps which are connected to my computer, theres no disconnecting or reconnecting, even my mindsp is ran from computer and straight to the amps, no disconnecting there either, if it takes you 6 months to design your crossover point its because you never understood loudspeaker design. You design the system around a preselected xover point not visa versa, and if you don't know how to do that....keep on, you'll get there =)
r. grambo not quite sure what you mean but if I flipped the polarity it would have looked very close to the result with delay. It would have been redundant.
What if it was 90 degrees out of phase? Would you still invert the tweeter? And if a person doesn’t know that it’s out of phase, how does an active XO help them?
Delay? your tweeter is 140deg out of phase (according to you)... you've tested it with a generic passive and now with active.... Did it occur to you to flip the electrical polarity on the tweeter, in order for it to be acoustically in phase with the woofer. The delay hack is a horrible suggestion, because now the woofer and tweeter going to be OUT of phase EVERYWHERE on the spectrum, by the amount of the delay. It's not a big deal because the x-order is 24db/octave, but it's a hack - the delay was never meant to be used like this.
pretol the delay was absolutely meant to be used like this. It’s puts the drivers exactly in phase where as simply flipping polarity would still be out by 40 degrees. Not sure where you got the idea the passive XO was generic. It was properly designed which is why it measured way better.
You're right, I meant generic active (which is the hack job you've done for this "test") vs "properly"-tuned passive (according to you). Which is NOT good guide for people that are building speakers at this level, and YET, here you are making profound conclusions. You'd have to be a moron to design an active XO with generic parameters. Tunability is THE WHOLE point of active XO's (presumably with a calib'ed mic), and here you've done almost none (except wiring your tweeter backwards, and compensating for it with delay).
With a proper set up for measurement using a loopback on an impulse test tone, one can very effectively time align the drivers using measurements-- here is the instructions from mini dsp www.minidsp.com/applications/dsp-basics/time-alignment with a link to instructions on using the loopback.
Trevor Jones yes or you can Check the time between impulse peaks in the measurements. Provided the person knows either of these methods. Good link thank you.
Chris Wilson phase and delay are exactly the same for all intents and purposes. I applied delay, which changes phase as I tried to demonstrate on my hand drawn figures. Sorry they were drawn poorly. One day I'll try and show this better.
My beef with passive is STILL money !! You can design a phenomenal passive unit - but the minute you start using high quality components = the cost goes up $$$ PER SPEAKER !!! The obvious limitation also applies to active = more amps and added multi channel Active units..... so either way it never gets away from cost vs quality.....
He misses the main advantage of an active crossover for the hobbyist, which is that you can make all these adjustments without needing replacement components, so you don't need to get it right before you order the parts.
I didn’t miss that part at all. The video was about the idea that you can just click X frequency and X slope and have a finished XO. The benefit you list is pretty minor in my opinion. The greatest benefit to active crossovers is the efficiency compared to lossy passive crossover components. That is huge a very big reason to use active. The title is are active crossovers easier. Not are they better.
Appears you are mistaken here, as explained in this post: www.stereo.net.au/forums/topic/23800-active-vs-passive-crossovers-and-phase-shift/page/3/ For gosh sakes already... "only if you know how to use it" goes for most things in life, and in audio. Duh.
Luv this guy's vids but this beyond his scope (no disrespect), you can't hack DSP, it's graduate level math. For example you can auto compensate a speaker that would not be possible by analog/passive means (theoretically impossible). DSP essentially emulates the analog parts, thus u can create a "part" that does not exist or even physically not possible, like negative capacitance (I'm keeping this simple, it's the same tech used in military guidance sys for example). Beyond that you can compensate the speaker's impulse response (not practically doable in the analog world). Then there's efficiency, the reason you have power components is the passive X-over is it throws power out the window to compensate (utterly ancient way of doing things). good news is you have software that can do it all work for you now, no thinking involved, like cell phones leading everyone by the nose (yeah i'm from the DSP world, b4 it was a buzz word ;) ).
Greg D no thinking necessary? Clearly using dsp is also beyond your scope. The point of the video was to show newbs that just as much thought was necessary. Did you watch the video? Not sure you did. Please watch the whole thing and tell me if you still don’t get it. A newb cannot just hook up a dsp magic box and enter 2000hz LR4 and expect good results. That’s not going to sound good. And I showed it here. Clearly my video was not easy for people to understand. For that I apologize. Still not sure why, I thought it was simple. It is just as much design work as passive. Period. And BTW, I like active, maybe even more than passive. This is not a hate on active speakers video!!
Wow man, i was being polite, you're having a reaction. Yes i watched the entire vid, did you read my entire message? You do good stuff man but your video was like someone who's never flown a jet then trying to demo it without any flight experience of any kind, but again, no disrespect. Or are you saying you're very familiar with DSP? enough to say it's beyond someone's scope? You know convolution cold and it's edge affects? u know self adapting communication channel compensation filters so these audio steady state one's are are like childs play? Because if you are, then spend a little more time on the vid to do it right, if not then maybe don't post until it is right. So yes, there are programs now that will analyse the speakers then adaptively create the dual FIR/IIR filter coefficients to completely compensate it in every way, you can follow a procedure so one doesn't actually have to know what one is doing ( here's one for the laymen: www.bodziosoftware.com.au/). Verses say going knee deep into MATLAB or writing your own code and download your own coefficients to the DSP, but again, then you have to know math at a graduate level. Also that MiniDsp unit you have will not do impulse response correction and i don't want to get into why but it's a key disadvantage. And really man, every wonder why it's only "home guys" posting on the net, not guys from the Boeing lab or Tesla or the DOD or other pro's? because they keep it inhouse, so it's odd to me when an expert might chime in and say, be careful, you might unknowingly be posting miss information (which is rampant now). So one of my "DSP buddies" saw your response and said "why do you bother man?" and yeah.. he's right, so good luck man, this post was for others and try beware of what you don't know (btw, spare me the "male fragility complex" which is what your "clearly dsp is beyond your scope" retort is (i qualified mine), unless again, you're in the DSP industry as i inquired above, which means you really do have some other issue).
Greg D alright sorry, my comment was over the line. Although it was really no different than your comment. Saying you like my other videos first doesn’t change the comment. I still don’t understand what you disagree with in this video. The video is saying that you cannot take pre determined electrical filters and simply plug them in and bam good results. That’s all the video is saying. Do you disagree? By the way, professionals do post online. All the time. I’ve seen forums and facebooks posts from engineers with JBL, pioneer, etc. Not sure what a Boeing engineer has to do with audio. Those auto calculators are only so good. First, they still require measurements, which is partly what this video was saying. Second, they can’t fix things like driver spacing, Pilar response, distortion, etc. The speaker builder needs to make engineering decisions about this. Writing dsp math has nothing to do with the video. My comment about watching the video was not meant to be rude. I seriously ask did you watch all of it? Understand it? I admitted in my comment that clearly the video was not good because you’re about the 20th person to misunderstand. Do you think newbs can just buy a mini dsp and hook up two drivers and then program it to use an iir filter of what ever frequency and get good results? I doubt you do. Because as shown in the video you will very likely get bad results.
@@gregd6022 hi Greg, what dsp crossover can do impulse response correction? Is lack of impulse response correction a flaw across the whole minidsp line? Any other links to software to analyze drivers you would recommend. Think of going dsp to gets some Altec 604Es to work as well as they can.
@@ImpulseAudioSpeakers The only person saying you can plug-in a miniDSP and automatically get good results....IS YOU. Having an argument with people noone made.
I love your videos and learning so much from them, but this one is bad, you conflicting yourself few times, sure if you do the DSP wrong it will not going to be as good as Pasive done right, but with a little knowledge you can get much much better sound with the mini DSP then you ever hope to get then with a passive crossover, I know because I upgraded 2 professionally design loud speaker ( JBL and Dayton ), from Passive to active and they sound way better now, sadly outside of studio monitoring the world don't like active because people think they will improve the sound by buying a $3000 amp but they don't want to buy 2 or 3 amp and a DSP , if they make a system with 3 $100 amps and mini DSP they with get much better sound. but I know it is hard for speaker makers to sell this idea, Anyways love your video ,, David
David Mann I didn’t say active wasn’t better. It is. I said they’re equally difficult to design. Obviously that wasn’t clear in this video because a lot of people, yourself included, disagree with the video.
seriously i'm confused ... you're going on and on about the difficulty of crossover design, but then apparently your audience needs a slow and careful explanation of phase cancellation, which i think i understood when i was twelve. i would think many things would be difficult for aspiring speaker builders who didn't know this
you are not doing active good, phase is eliminated in 4th order crossover , i dont know why that DSP doesnt do that , its bad design , however i would like to experiment but i need more , at least 4th order , not 2end order , wahts with the miniDSP what it can do , can it simulate a lot of different crossover or fixed at bad design
Pavle Pavlovic the mini dsp is great. I’m afraid to say you just don’t understand what I was teaching in this video. 4th order doesn’t automatically fix phase. And that’s the problem. You need to understand what it is that makes a XO work whether passive or active. Try watching again and doing some reading. Even read some of the comments to see what I mean. And also try it in real life.
No , man i didnt come to argue , but ether you are wrong , or this paper i was reading is wrong , and i quote "Today, the de facto standard for professional audio active crossovers is the 4th-order Linkwitz-Riley (LR-4) design. Offering in-phase outputs and steep 24 dB/octave slopes" , Do you believe me now . ITs no problem you are not the first ive found doing stuff the wrong way , these are PHD topics , one cant know everything , but we learn and go on , maybe you can do another video , i dont know it is what it is , that miniDSP can be a good thing but it needs to do it better , im interested in it off course , but no one will tolerate phase issues , thats just not going to happen , ive found DSPs that do it all but like 500$, its too much
Pavle Pavlovic your understanding of the article is wrong. The outputs are in phase but that doesn’t mean the tweeter and woofer phase is 0. In this case the tweeter is behind the woofer. Telling me I’m doing it is wrong is argumentative. Just because you read something doesn’t mean you understand it. I actually demonstrated the folly I see so many people fall for. Either believe what I’ve shown and try to understand why, or keep reading papers without grasping it. As for the mini dsp, it can do it right. I show at the end of this video when the delay is set properly to deal with the phase issue it works really well. Plus it does PEQ and other things. So when someone knows what they’re doing it is very powerful, and costs much less than $500. Maybe try watching the video again and really concentrate on what I described with my hand diagram. It might help. I don’t know everything, but trust me, I know I’ve done this right.
Yes i am aware that it comes out on a wrong note ,but its just the truth sometimes it hurts , from my understanding phase problems are due to filtering but filter inverted phase it will get flipped once again , and return to its original , hence no phase issues , there are liner phase EQs as well as linear phase crossovers , no problem , as i understand you seam to think that those dont exist , im telling you they do im using both liner EQs in production and liner crossovers in monitoring , you can get some and measure , i dont know how else to tell you . Im not trusting one paper , i know it , im using it , ppl use it every day you just need to find liner ones and you will see it in your results
here you go analysis of Crossover topology with tests , there is some phase issues but nothing as you experience , more like 1% or around th-cam.com/video/aiJQED7-1b8/w-d-xo.html Dont knwo waht else to say , need i call Linkwitz here :)
no, it is not correct, a DSP and a activ crossover device are diffent things. A DSP has to calculate the signal this takes time, a activ Crossover uses the same parts as a passiv crossover and a op-amp, but before the amplifier.
Most people use the terms dsp and active XO interchangeably. Even more technically true is that a dsp and passive parts before the amp are both active crossovers. They actively change the signal delivered to the amplifier. A passive crossover passively changes the amplified signal.
You said "if they can't compensate for phase, by, you know... using delay...." but in real life, phase and delay are related but, completely different things.... you would never! use delay to make adjustments for phase...thats a basic understanding in loud speaker design....I find your legitimacy questionable at this point lol!
DR.NNOO ok if you don’t know to use delay to correct a phase issue and that calls my knowledge into question, you better just ignore my nonsense. I’m clearly an idiot who doesn’t know what he’s talking about lol. I mean, who would use time to correct a time function. I mean, that would be as crazy as using amplitude eq to fix an amplitude problem!! Phase and delay are directly linked. They are married. Check with ANY body who knows how to design XOers and delay will be one of the ways (not the only way) used to solve a phase misalignment. Go ahead and check on me.
DR.NNOO just in case my last two posts weren’t clear. I LIKE active XOers. They’re just very misunderstood. Unfortunately you are one of these people. I don’t actually want you to ignore what I have to say. I think I have something you can learn. I just don’t understand the “active or die” mentality so many people have. Active XOers are great. This video did not say they weren’t. I’m gonna chill out now. Thanks for watching.
@@ImpulseAudioSpeakers by adjusting delay you are adding group delay to the whole channel...if the group delay of both channels were already in sync, you don't adjust delay to get the xover band in phase, only to take away coherency of the whole signal, thats ass nine. Once again, there are rules to follow, regarding center to center spacing, driver size, and intended xover. You would also look at the measurements of phase transfer function of each driver to choose an area where the drivers can cross over well, phase wise, but thats done BEFORE you even purchase the drivers. You would even look at the group delay chart to see how well two drivers play together where desired xover....also before a purchase....you choose drivers that play well in the intended area of crossover, thats simple ...From there you may have to time align them using delay, and that may "fix" the phase, in your head, but the phase wasn't broken, you just needed to time align the drivers. If there are issues in phase after you time align, and you can't get the intended results, where they should have been, adjusting delay to unsyncronize almost the whole bandwidth to gain phase coherency with the cross band is ridiculous. Delay and phase are of the time domain, yet not exactly the same thing. Maybe you are stuck within the passive mindset. Nowaday active crossovers have parametric PHASE eq's, or automation to adjust phase and ONLY phase...its a separate parameter than delay, in design philosophy and in real life application... though they affect one another. Getting time delay insync will improve phase on a greater scale than adjust delay to optimize it within only the crossover bandwidth, I feel like I'm repeating myself. Voicing to flat and room correction improve phase as well but at a certain point, phase is what it is for the situation and cannot be improved. If you don't make the right choices leading up to implementing the predetermined xover...you are just fk'd lol. If you don't get it by now...nothing I can probably tell you.
Wow! This video is misleading at best. It would take an enormous amount of time and boxes of components to make the changes with a passive crossover that you can make in seconds with an active crossover like the miniDSP. And all the work is done before amplification. You are not placing a bunch of "stuff" between the amplifier and the drivers. AND, you are dedicating amplifiers to different parts of the frequency spectrum. The amps will have less to do and can be optimized for performance and sound quality....Solid State for bass and tubes for highs for example.....Lastly, the last speakers I built, the passive, high quality components, cost way more than a miniDPS.....And if I got the components wrong, the cash register just keeps ringing. Love your videos....This one?...Not so much! :-)
This is ridiculous. The whole point is that you can tweak the active DSP really easily. What you show is the result of an off the shelf 1.5k active like say a Marchand. The whole point of a DSP crossover is that revising it is trivial, while revising a passive (or an electrical active) requires a lot of resoldering and a big parts bin. That's where its less work - you can iterate really quickly - and with the minidsp you can do it without even stopping playback. Do you have to measure? Yes. Did you show how easy it is to try inverted? Nope, but its just a click. How about adding a delay? Yes you did the result was good and yet 'I didn't work on this for very long'. Bingo. There you have it - the minidsp *is* a panacea if you accept the resampling, and so is eg camilladsp (perhaps more so if done before any DAC). The summary is self contradictory - perhaps the answer is 'if you want to go active, go DSP' but you appear so keen to lump DSP with fixed function active.
Too much simplistic explanation. Bottom line is -active crossover has huge advantages over passive crossover an it makes more economical sense in today world. With a passive crossover you should be ready to pay insane amount money for one capacitor . You can buy few god quality DSP for that price. How many of those capacitors do you need for 2 way crossover in one speaker ? We even not touching sound reproduction quality of several amps for one speaker V.S. one amp and what passive crossover does to that amp. Please don't generalize such view points from personal point of view. Biggest contra argument to your statement would be to find recording studio with passive crossovers in they speakers. As soon as active crossovers was available for market passive speakers/monitors have disappear from recording studios inventory and it is bloody god reason why. Well you must know what you are doing with DSP ! Yes, I agree , it is not as simple to handle DPS at least as much I so on your video.th-cam.com/video/o2_gRAbnzyk/w-d-xo.html
@@ImpulseAudioSpeakers I feel like it's just adding unnecessary complexity. You can't use a normal 2-channel receiver and if the DSP, being a very fragile little computer, breaks you have to start all over again. Only things they are good for are active PA speakers where weight really matters or sub/sat systems with very low cut-off.
@@westelaudio943 Lol fragile. I can promise you that if you drop it from 1m it will survive but if you drop your 15kg avr from 1m it will not. Smaller electronics is usually more durable.
This video is nonsense. Way to push your agenda. The reality is that active crossovers are much easier since it gives you all the tools to fix any problems.
I don’t think you understood the video. Active is more powerful, but all the steps to create the XO are the same. They’re not any easier than passive. Still need to build, measure, model, listen, tweak, etc. Not sure how that’s even debatable. I don’t have an agenda. I can do both quite well.
@@ImpulseAudioSpeakers is it easier to make a new bigger box with a better port tuning or to just amplify the lower frequency that goes into the woofer? Can passive crossovers pull better performance from cheaper smaller design without reducing efficiency like active crossovers can?
@@BobFrostV active is better. I said it’s not easier. Of course active has many advantages, there’s no debate. The point of the video is that a newb can’t just hook up an active box and expect it to work if they can’t design a passive cross over. Did you watch the video? And no, you can’t just boost the bass if you build your box to small. Even with active you must rebuild the box, unless you like to clip amplifiers.
![ILLSLICK - KILLSHOT REMIX [FULL VERSION]](http://i.ytimg.com/vi/RIK8RrqIAzE/mqdefault.jpg)
Agree - it does need to be said! Many people don't realise that the level of expertise to design either is about the same.
Red Spade Audio Yaaaaa, someone understood the point of this video! Thank you!!
the advantage of active crosoovers are not hat they are way easier to design. But that they are way easier to implement and you can reuse them in the future if you ever retire your first build by just reconfiguring them, not needing to buy all new parts.
And I'd like to share, you don't need to pay 100-200 for MiniDSP or MiniDSP HD to have active crossover, depending on your application there's free software like voicemeeter (that's actually a pay what you want software) or EqualizerAPO, that can help you with that depending on your setup.
I personally have a pair of active bookshelfs and just bought a pair of subwoofers, I was desperate to find a way of connecting them all togheter witouth spending a lot of money, I considered MiniDSP but I didn't want to pay 100 for it. Well voicemeeter banana just saved me. What I did is I bought a second SMSL M2 DAC and used voicemeeter to send signal to both of them, then I configured a high pass to the first and a low pass to the second M2 DAC, now I have a crosoover at 70hz about 36db/octave. But I didn't test that yet as the dual subs and the second M2 didn't arrive yet. But yeah, very cheap, not quite easy to use but the price makes up for the trouble. And since voicemeeter allow you to setup macros I made 3 presets, one for movies, a second for music and a third to just send full-range to the monitors when it's late night and I don't want to disturb anybody.
I'm pretty sure I'll need to make some fine adjustments when the subs and second DAC finally arrives.
You seem have forgotten the most important part. Active done properly.
Hi, im from indonesia.
Mini dsp is $105 at PE, with shipping and tax its $161 at my door.
Dont have budget for it.
I go with $3 caps, and $6 inductors.
With free xsim, im doing fine with passive.
I'd have to disagree there..
Just as a generic passive crossover isn't plug & play the same applies for the Mini DSP. Once you factor the same things you need to take into account with a passive you can do the same with the Mini DSP. As you mentioned the Mini DSP takes phase alignment into consideration & gives you the tools to correct this, so with the Mini DSP active crossover it takes a couple tweaks on the laptop vs a complete redesign & build with a passive crossover.
exactly - in other words it took about 5 min to make a correction in the MiniDSP and would take several hours putting together and retesting a new passive unit....
Agreed. No generic crossover is going to fit any specific speaker without some tweaking. But with the Mini you can tweak in real time. You can do close field tweaking and then do room correction within minutes instead of hours. Saying that a good passive crossover is better than a bad active crossover is not fair. Of course it is and a good active crossover is way better than a bad passive crossover.
Agree, this video isn't a very fair assessment of the differences between an active and passive crossover. His poor performance on the active at first should have been compared to a generic off the shelf pre-built passive crossover. A good active crossover can do everything a passive can but it can also be re-tuned whenever you move your speakers to a new room.
@@rts100x5 - Agree 100%. To get a passive crossover design right, you really need an RTA analysis using white noise for each of the drivers. The crossover point usually will be where their rolloff curves intersect (target would be -3 dB with properly matched drivers). then design accordingly. Once you have that information, you can get preassembled passive crossovers at places like parts-express.com and some of their decent crossovers have user-selectable rolloff points.
One other thing to consider (that I have not seen mentioned here yet) - passive crossovers depending on the filter type only provide a rolloff of around 6 dB, If you want to have a third or fourth degree filter configuration, for passive this is *very* expensive in terms of complexity, cost, AND attenuation (which must be made up with more amplifier power).
What the big name speaker manufacturers do is settle with a 2nd degree filter but match it *perfectly* with the drivers they are including in the speaker design. Even with that, it is difficult to achieve a perfectly flat frequency response curve but any minor peaks or valleys can be corrected with small amounts of equalization.
With active crossovers you can swap different driver types all day long and, with a few keystrokes, get the response curve exactly where you want it with an ultra precise 24 db rolloff. But you still have to put a lot of work in looking at the output graphs and (to do the *best* possible job) you want to complete this analysis and testing in the actual environment where the system is to be installed. And (this cannot be overstated) DO NOT OVER RELY ON HEAVY EQUALIZATION! By needing to do this indicates a major design error probably with the matching of the drivers and by overcorrecting using equalization you also run the risk of adding distortion *especially* with forcing the lower frequencies through the tweeters.
Active crossovers are much easier to design. They won't solve the issues of poor driver choice or bad enclosure design, but they're so much easier to work with. Try building a passive 24dB/oct 3-way crossover with subsonic filter. I did that in a couple of hours with active.
Bogdan Serban I think you missed the point of the video. It was about applying blanket crossover settings with a dsp box.
Looked a heck of a lot easier than digging through a box of crossover components and using that software to design a passive crossover. You even showed how to compensate for phase using DSP. All in all (having watched all the videos you've posted) I'm inclined to think that active is easier and more effective than passive.
Daniel Deverell you're not wrong. It is slightly easier. The title is a bit misleading I suppose. The point was that all the design work necessary (the actual hard part) is the exact same as passive. Many people view active as a button clicking excersize when it's far from that. Software is still necessary to design a proper active crossover and digging through a parts bin and doing some soldering is only 1% of the trouble. Fair comment though :)
Wouldn't it be possible to do a trial and error with an active crossover, measuring the response, then adjusting the delay, and measure again, until the smoothest response is obtained? Then adjust further with EQ. Not sure, but that might still be easier than a software design which would still need to be verified with measurements.
Greg Worrel I meant measuring software. But even design software would be the way I'd go. Trial and error is possible with either active or passive but I find I'm faster and easier using XO design software. Usually end up with better results too. But if someone wanted to do it by trial and error using measurements they could no problem. That's how I solved the need for correcting phase in this video, I didn't bother to use XO design software for a quick thing like that. It still takes a long time in the lab to do it by trial and error and then you need to listen then drag your amps and XO back out into the lab and back and forth. Not easier than passive IMO.
@@ImpulseAudioSpeakers The real problem is having that parts bin to dig through. Not everyone does. The MiniDSP can implement complex filters and equalizers and can be adjusted repeatedly. It can be reused later with a completely different set of drivers. Certainly, it's not fool proof and requires the same level of expertise, but if you mess up, it's a few mouse clicks to fix versus ordering a few hundred dollars of new components. Plus, with many of their devices, you can store multiple configurations and switch between them with a button press. Personally, I think it has lots off merit, especially for the beginner who wants to experiment and learn.
@@scotdixon777 I just feel that if someone actually knows how to design a XO properly they’re going to have the parts. If not, well then fine, the active XO has an advantage there. Not easier to design, easier to implement.
The conclusion at the end makes no sense. You make it out like a passive crossover is easier to design properly than it is to change settings in a program on the minidsp.
It sounds a lot like the arguments of purists when it comes to tube amps vs solid state. You seem invested in passives being better. They are definitely convenient once designed but the minidsp can give you a better modeled response easily if it's not perfect and you can correct for the room after you get everything set up.
Keith Anderson sorry that wasn't my intention. I was just trying to show that active isn't simply picking a XO point and hooking it up. I get that sort of comment often. Certainly no purist type of commitment to passive here. I enjoy active and as you can see I own a minidsp. Active has many benefits. Hopefully that clarifies my position on the subject.
Keith Anderson By the way, I have a similar video about the pitfalls off passive XOers for the same reason. The two videos were merely meant to inform people who are new to the hobby that crossover design (passive or active) is more challenging than it can first appear. Hope you can get some enjoyment out of my other videos if this didn't sit well with you. Sorry again.
Impulse Audio the video is good. I loved the explanation of phase. I always knew it was a timing problem but never visualized what the actual issue was.
I see the minidsp as a great tool to test drivers before making a passive crossover.
Tubes = epic win.
Passive crossovers = epic win.
How many DSPs will still be around in 50 years? I don't see a point in using disposable and fragile microchips in something that could last a lifetime.
@@westelaudio943 have you ever considered the fact that DSPs are only going to get cheaper and cheaper with time not to mention a miniDSP is only $100? Even if a $100 DSP only lasts you 10 years (and it should really last much more than that) that's incredible value. They are not going anywhere dude you sound absolutely crazy implying otherwise.
do a tutorial on how to use the mini dsp properly on varios scenarios....for example how to use it with subwoofers to get the lows smoother etc...
ziggy8757 Good idea. I will. I wont be able to right away but its going on the list! Thanks for the suggestion.
Now it all makes sense...This explains why the"sweet spot" for different speakers varies room to room. Your tutorial also calls out the difficulty in designing speakers AND X-overs. Thanks
Firstly. It’s important to understand the difference between the various types of crossovers.
miniDSP is technically not an active crossover. It’s a dsp with multiple outputs, and not like the traditional active crossover that is analogue.
I built my system with miniDSP but with separate AD and DA converters using the I2S in and outputs. The sound quality and flexibility is unbeatable. It costs a bit more but that was the best investment I ever made. Now I can build whatever speaker designs I want at no further cost
M D well, the dsp is in the active position (before the amplifier) rather than the passive position (doing the job after amplification), but sure ok it’s not a traditional active XO.
If the tweeter is late then make the woofer delayed
yup, you can adjust delay in milliseconds on Inuke DSP at least, I don't know about miniDSP tho
I think you make valuable points in this video with adding another channel into an active setup introducing potential phase issues, and you also explained how to use the tools to remedy those issues.
The mini dsp is going to be much more versatile than a passive cross over and can be used in conjunction with passive. The beauty is in its modularity.
The tools in that little box will yield much better results than passive crossovers alone. The time correction alone will allow you to achieve focus in irregular or odd shaped listening environments and you can make instantaneous changes on the fly that can't be accounted for with calculations.
I'm old school and installed a set of passive Morel's in my Ford. After listening to it for months, I bit the bullet and installed an Audison DSP, time aligned it and used an RTA to dial things in. OMG what a difference. I'd say that acitve takes even more expertise than passive because of the additional capabilities inherently included. With passive networks, you fight components to get things equalized and muck with time alignment the whole time. It's an endless battle. With DSP, the capabilities are infinite but you need to know what you are doing for sure.
Good video.
Watching because I want a paramedic EQ for a passive B&W home subwoofer. From what you say, the mini DSP is perfect for this function. No need to worry about phase if I’m just knocking down a modal peak.
Yes pretty much. Phase could be altered if you we around the XO point with your mains, but not any more than the room will cause you grief.
"phase shift"/time misalignment is primarily due to the physical offset of the focal point (axis) of the drivers...Were both divers in perfect time alignment (i.e. propagation of both wave fronts in perfect alignment), there would be roughly a summing of +3dB... assuming that both drivers were being driven with an identical signal/impulse (mass/acceleration rates of the respective drivers, notwithstanding). A relative "delay" of the signal applied to one of the drivers could bring both wave-fronts 'in phase"; additionally, a steeper slope, along with adjustment of High-cut and high-pass roll-off points could reduce some of the phase interference problems. Differences in mass/acceleration rate of respective drivers can be compensated with "torque" (power applied). Access to B&K FFT is usually handy.
Yes it's hard to delay the bass driver on a passive design.
Great videos , glad I recently found your channel . I’m 57 and still learning volumes
Quick question. We are discussing phase cancellation and phase shifting however why would these necessarily cancel out? I can see if the frequency range is the *same* you would get cancellation. But the human ear doesn't hear at just a *single* frequency. I do not believe a 100 Hz tone (from the woofer, obviously) that is 180 degrees out of phase with a 2 kHz tone (from the tweeter) will result in cancellation. One of the obvious reasons for this is that the wavelengths are very different.
Now, if you have *two* subwoofer signals (stereo configuration - same frequency range therefore same wavelength), older amplifier designs would have a phase reversal switch for the low frequencies. This was needed because the physical size of the listening room (and the resulting acoustics) may throw the subs out of phase with each other and, since they are operating in the *same* frequency range, that would definitely result in attenuation of the sound (perceived by the listener as "weak" or "muddy").
One thing that would be really cool is for you to actually get an RTA and watch what it does with a white noise signal in real time (before any equalization is applied). Many of the midrange and most of the very expensive receiver amplifiers out there with built in 7.1 surround all have auto equalization built in *plus* an included microphone (with input on the back of the unit) so that everything will be adjusted automatically when you set it up.
I am a big fan of active crossovers because, when done right, the sound is amazing. Where these configurations have been more commonly found is automotive sound systems since you typically would be running 4.1 or 5.1 with just highs and lows (active crossover split) going to LF/RF/LR/RR and then the full range signal going to the center channel in the front (difficult to install since you have to do major surgery on the dashboard and center console). The sub is normally in the far rear and modern systems only have a single mono sub since subwoofer frequencies are highly omnidirectional so there is no need for stereo (you would not easily discern the direction from a cannon blast from the left vs. right - the entire vehicle shakes.... :) .
The trend even for automotive today is to allow Bose, Sony, Celestion, JL Audio, Soundstream work their magic getting the passive crossover design absolutely perfect in their anechoic lab testing chambers. Known as "separates", they contain separate drivers for the satellites plus the passive crossover module that is mounted out of view. Basically let the OEM get everything matched perfectly for a typical car (or living room at home). This greatly simplifies the design (meaning potentially just one five channel amp for everything except center and a small amp for the center channel) versus 20 channels of amplification or more (and two days of adjusting crossovers and positioning working with an EQ and an RTA).
Even the $20,000 sound systems installed in movie theatres today do not use active crossover designs. But it can be done if you want to put the time in. The question then becomes 'can I do it better than the engineers at Bose?'. I am *very* good at all this and have built several sets of speakers with passive crossover designs plus a couple for my cars in the past that used active crossovers. I do a good job. But Bose (and all the others) are better. But the joy in the journey is the journey itself. The first thing you do when you buy an old performance car is tinkering and upgrading. Because that is how we learn! :)
You realize I used a white noise signal to generate these plots? Nobody would say we only hear one frequency at a time. Take some time to figure out the graphs and you’ll see why the active XO was not as good.
The problem here is you are assuming an out of phase condition is the cause of the dip at the crossover point. That is a reasonable assumption as a likely cause, but not necessarily the case. You should be able to adjust the crossover point of each driver independently, and you may be able to eliminate the dip. The other thing you can do to quickly determine a phase cause for the dip is to reverse the polarity of one of the drivers and retest. Measuring phase accurately is a complicated issue, but being able to time delay the output with the miniDPS is very cool and the ultimate way to correct phase coherence issues.
Speaker Builder I was not assuming, it was an out of phase condition. I measured it. Phase is not hard to measure. And the solutions you suggest are also what I suggested. The point of the video was that a lot of people don’t understand they have to adjust the XO point or delay. They just assume an active XO box is magic. It’s not, it requires just as much speaker design as passive. You basically said as much in your comment. Did you miss the part of my video where I show actual expensive products that don’t allow you to change the XO points independently or adjust delay, etc. If people are actually manufacturing these things and people are buying them, then we know a lot of people don’t understand that active XOers require just as much design thought as passive.
Something to consider with an active crossover is managing gain & headroom
adhanda2017 yes these little minidsps can clip. Especially if adding boost to demanding bands, like below 200hz.
I did my system with my 10 channel minidsp, without room correction and phase compensation because it’s too hard for me😅. But still the sound is phenomenal ❤
If I screw up an active crossover, it doesn’t cost me hundreds of dollars in parts and waiting on postage to change to change a setting. what about using an active crossover to aid in designing a passive crossover?
Sure that can be done to save money. I don’t know about hundreds of dollars. But yes active XOers are very good as a primary XO or an aid if you want a passive XO. It’s difficult to translate an active XO settings to passive parts though. So it’ll only tell you if the driver is happy with what ever XO point or slope etc. You’ll still need to come up with the passive design. And as the video says, it’ll be just as difficult to design the XO actively as it is passively so you may do some double work.
So if i'm quite new to speaker design, can i just get the 95$ mini dsp and tweek the hell out of it, when i don't have a ton of components (0 actually) to tweak a passive crossover design ? Because even when using a software to design it, parts are not always the same sot they will have variations 5-10%. So i might have to buy even more parts so i can tweak it even more. Imagine a 3 way system, with the mini dps you can just plug it in, and spend hours changing controls and taking countless measurements.
Quetzalcoalt this is a fair question. The answer is yes and no.
First, dont worry about parts variation. Far more important is the placement and duty of the component.
As for the minidsp or equivalent, I would suggest you at least buy a microphone and a couple large (50uF) protection capacitors also. That way you can do two things. You can tweak like crazy AND measure what you are getting. Our brains will adapt to bad sound and sometimes it is not until we hear the right thing that we realize what we liked actually sounded bad. We also convince ourselves something sounds good when it doesnt. Measurements help keep your brain in check and bring you back to reality. This isn't me suggesting you should like what the mic says. I've been accused of that before. Rather it is a way to visual what you are hearing. There are some small things that the mic can help you see like a narrow dip or peak that we can hear but would never know what it is or what to change.
The second thing you can do is take driver measurements and put them into Xsim or some other program and then simulate the minidsp inputs. This allows you to "tweak" much faster so you can get in the ballpark quickly. It will also allow you to quickly see a few different approaches to the XO settings. This is the way I do active XOers and it saves me hours of tweaking and measurements.
The protective caps will just make sure you dont blow a tweeter if you make a mistake or get turn on thump from your amps.
Then the 50µF caps are put before the tweeter only? What does the cap do?
Have you designed a fully active crossover speaker? The miniDSP has 2 inputs and 4 outputs, so you could make a pair of bookshelf speakers with only one, or a 3/4 way using one for each tower. Might be fun selecting different amps.
Yes many times. It is fun. The mini dsp is a great product. I wouldn’t touch one of those other crossover boxes that don’t provide the proper flexibility though.
Ok so you said the Mini dsp was not a fix all but you proved it was. You were able to correct the phasing, then with the eq would be able to tweak the response curve. Also on your crossover screen on the mini dsp it looked like you had a gap between the points for the woofer and tweeter. For those that are not savvy on designing and building the absolute correct crossover be it passive or active it looks like the mini dsp is a winner.
Cyber Trucker Road Show I'd have to watch my video again to see what I actually said, but the point was to say that unless you know what you're doing with it, it won't be any easier than passive.
As for the gap, take a look at the screen shot again. Each channel has low pass and high pass capability but usually one of those is muted. In this case it is muted. Both channels were set to the same frequency for this test. Having said that, often the XO frequencies are different (sometimes a gap, or overlap) when designing a proper crossover. I've never had a properly designed speaker use the same frequency on the transfer functions. The acoustics points are what need to match.
Impulse Audio Never mind lol I think you were just referencing active xovers in general. The phase issue looked like you could have tweaked the points to fix the gap woofer @ idk 4k then tweet to 3.6k have some overlap. Great vid, I didn't mean any offences, my observation.
Cyber Trucker Road Show no problem. I don't take offence to comments on my channel. I like when people comment so everyone can learn. I didn't spend time adjusting the XO points because it was intended to show the pitfalls of arbitrarily choose Xhz to XO using one of these "magic boxes". But yes definitely if doing it properly the point, slope, delay, etc. all need to be adjusted to get it right.
I agree that you still need to be aware of phase, but as long as your DSP allows you to correct it, I'd say a DSP is way easier to work with then a passive XO. Downside is the fact that you need more amplifiers, obviously.
Jan Barelds fair enough.
I got around many of the issues you point out by going with a 4-way active and choosing drivers that had good behavior about an octave beyond each crossover point. Large baffles with Vifa AG35, Vifa P13, Vifa P26, Eminence DVC-12, with 3200Hz, 500Hz, 100Hz. (this was about 1992?)
Hey you didnt point out the major attraction from the active xover, that by splitting the audio into separate bands BEFORE the power amps, each amp and each speaker unit now handle only their allocated band of sound, or in other words, each amp now outputs LESS power at a given SPL, coz it no l9nger has to amplofy the other band(s) of aidio that otherwise gets sent to the speaker which filtres it out oventually in a passive xover design.
I find active xovering an advanced level of playground for audiophiles and engineers that give much much more tuning possibilities than simply matching one amp with one whole speaker.
The audio perfectionist's l9ve here.
That is true, active has many great advantages over passive. This video was only to show that they require equal knowledge of how to design a speaker. Seems I didn't make that point very well though.
I don 't get it. You say you have a dip at the crossover frequency with a phase shift of 140°. Would it make sense to simply revert the tweeter phase polarity, thus getting a -40° phase shift and a little bump like you have with the passive crossover?
didierleclerc66 in this case inverting the polarity would have helped a lot. But it won’t always be that way. It could have been 90 degree out. Then what? And you still are out 40 degrees which is pretty crap in my books. Plus all the other issues of wonky response not related to phase. Generally, even if you knew in this specific case that flipping polarity would help a lot, it’s still a crap speaker, far far far worse than the passive XO. So unless you want to go through the proper design just like a passive XO, the active XO will suck. Therefore it’s really no easier than passive.
My bad OP . I wasn’t thinking when I stated the drivers are going to have different wavelengths , as at some point they will have the same frequency response .as they crossover . A duh moment
Generally the group delay of the woofer is longer than the tweeter due to its higher inertial mass. Therefore to synchronize them in time the digital delay should be applied to the tweeter
This is still not good enough to get phase coherence. To do that they need to have the same geometric center of propagatoon, in other words they have to be coaxial.
This problem is like throwing two rocks in a still pond and getting them to produce ripples as though they were a single rock. To achieve that they not only have to hit the water at the same time but also in the same place, otherwise you will still get phase interference patterns.
Mark Fischer in this case the delay was applied to the woofer because the tweeter acoustic center was behind the woofer.
Great Video!!! My question is how do you test for alignment? is oscilloscope the only option..? is there software and calibrated microphones? this is intriguing and confusing all at once, LOL ...thanks
pallusubaru To find to acoustic offset I use a measurement microphone. You can either take measurements and then use software to determine the settings, or you can flip the tweeter polarity and add delay while taking measurements until the null is at the greatest and flip the polarity back. You're right that it is not simple. Unfortunately people think active crossovers are easy, which is why I made this video. Eventually people will hopefully understand that to get good sound quality from passive or active XOers a little work is required.
I have been following this channel a little as an obsessive compulsive/audiophile and DIY guy myself. I had some interest in the content but I have to say this video breaks the cycle for me. At 9:10 he says that you cant just plug and play with DSP equipment, but that is in fact exactly what you do! You plug it in and you play with it. Lets analyze this shall we. For $100.00 you can get the MiniDSP unit. That is about what it costs to buy the equipment to build your own passive two way filter and you likely will only afford a 2nd or 3rd order filter at that. With the $100.00 for the MiniDSP you get to choose between first through 4th order filtering, you can sweep the entire spectrum at will in a few moments where to filter at. You can time align and multi band PEQ all in one nice simple package. This guy proposes you get crossover software model a filter and hope it functions real world. With some sweet DSP you can real world tune and get it dialed better faster and I mean way faster. Yes working with a complex DSP device can be technical and a beginner my struggle but there will be no better way to learn than playing on a DSP. You can try anything and just reset it back to zero no additional money spent. This guy, if he want after designing and building a passive filter network must redesign and go buy more gear build it up and then test and he does not know if it will be better until after all the work. That is idiotic to me im sorry! Now to go further and say that DSP equipment is not for beginners well how much easier is it to buy design software model a component build it install it and then prove that it is infact the best design possible! DSP is going to be more user friendly than that. There is another aspect, environment effect. If you near field test and design you can get the system to be good or correct but then when you set up the system in a permanent environment the sound will be different than was ever measured during real testing. If you run a DSP system you can set up the rig in the location/environment desired, then place the measuring mic/ears in the listening position and tune away and you can get better results this way than all the seemingly sophisticated things this guy is doing. I am aware that eqing a system to a single location will usually create a mess in other locations in the potential listening environment but that is another topic for another time suffice to say I strongly disagree with this video content!
Impulse Audio, might I suggest you use a DSP system in your design process. If you like building as much of the system as you can (like I do) and want to use self built custom passive filters in your audio systems you should but you also should take advantage of all the tech available. Using DSP and your spectrum analyzing/snapshot equipment you can model a filter and then test it, you can make some changes and test it all in under 1hour you could model and real time observe 10 different crossover configurations.
You know I don't even know why am going into this, the PEQ function alone is worth the cost to have. If anyone is a true audiophile looking for the best most correct sound quality/fidelity they can afford than DSP is the only way to go. It is no more expensive for the level of improvement you get compared to any passive design. It is flexible adaptable it is the end all be all solution to DIY audio!!!! Period. I mean you can take a poor speaker system and digitally correct most of the problems in the bad design. Check this out. For years horn loading was the way to go for the efficiency of it. The sound quality issues with horns at the early days were tolerated because of how functional they were. Today most horn loading is done away with to try to maintain better fidelity of the sound but with DSP technology where it is at today, you can eq out the misbehavior of a horn and then benefit from their designed efficiency!!! Come on guys think about it you need to get on the DSP train or you will miss out sonically!!!
Good vid. I'm not sure that I agree with your assessment that active without delay are useless given that passives don't have delay and there are other benefits to active. Keep 'em coming!
Warren Smith Passive does have delay, its just harder to do. You can add poles to the network, which also steepens the slope so you have to be ok with that, or you can use a very complex ladder network. Most people use an extra pole or make physical changes to the baffle. Definitely need to understand whats going on with the phase to be able to get it correct. Active is easier if you have delay capability. The active XO boxes with a fixed slope and no delay make it impossible to solve the phase issue.
Good point about active having other benefits though. They do have a lot going for them. Especially the ability to dial in delay rather than the challenges of getting a passive right.
You make some good points, but just change the title!!! Active crossovers won't fix stupid (implementation) so to speak, but they are indeed so much easier to implement and adjust if need be.
Passive crossovers take labor. One has to fabricate the system and then they are stuck at a less than optimal arrangement. With active you just have to be knowledgeable enough to push the correct buttons.
With active, there are many benefits, and near endless adjustability for different source material and/or preferences.
Anonymous DueToFascists I'm not sure I agree passive XOers are labor intense :o A few components soldered together isn't a bit deal. The case against active always seems to be the number of wires and amps required. So really they're just different but both require good design. I do agree a superior result can be had with active, but it is smaller than most active proponents seem to argue. And there are advantages to passive such as potability, simplicity, etc. that really depends on the application the user is going for.
Hi, im from indonesia. Love your video.
Mini dsp is $105 at PE, with shipping and tax its $161 at my door.
Dont have budget for it.
I go with $3 caps, and $6 inductors.
With free xsim im doing fine with passive
Is correct what you say.
When you would like to replace one of drivers to another with different the highest or the lowest frequency, or with a different sensitivity, or change amplifiers, or to reduce a room main resonance frequency, you'll get to know how easy DSP is. A passive crossover make impossible to use current source amplifiers for mid and high range drivers. And etc.
If you build entire system for a personal use, this imho have to be a DSP with an active crossover and galvanic isolated channels, and bunch of mono amplifiers that match bands and power.
I have done that and no, it’s no easier. Sure, it’s a few buttons to actually program it, but any of those changes requires measurements and knowledge just the same as passive. Absolutely no difference.
@@ImpulseAudioSpeakers Sure. Getting the numbers is the most experienced part of the process. This all could be similar to digital and silver photography. 'Our' numbers are like proper light and composition. However photographer can find the best idea way more fast with a digital camera imho. Then he/she can replicate this photo using Hassy.
At the moment I saw only one $5k PS Audio product which is not a PA digital crossover. May be others exist (like hyperx modules for example). But this number will increase. And market will start offering loudspeakers with drivers that can be replaced/upgraded. This is not 100% true of course, but manufacturers would like to use this idea I think -)
I understand the point you are making here, but it does not align with the title of the video, or in fact some of the other things you are saying. True, you cannot just shoe-horn any type of crossover (active or not) into your system by picking a crossover frequency and calling it a day. There are other considerations.
But throughout the video you systematically disproved the title; active crossovers ARE easier than passive. They just objectively are. You can design and tweak one to suit - and it will do a good job - within a matter of minutes. With a passive crossover, not only do you have to understand impedance (a thoroughly non-trivial requirement) but even in the simplest case you have to design your crossover in software (let's say that is equivalent to setting up your active crossover in its software, although it's almost certainly more complicated), then buy the required parts and wire them up.
At this point, if something in your simulation was wrong or not accounted for, you have to start over. You can just change a parameter in the active crossover. Moreover, passive components are susceptible to overheating and electro-magnetic interference.
They are categorically worse, but in some instances, necessary and convenient. I use them because they sit tidily inside the box, I don't need a DSP unit lying around, and I only need half the channel count on my amplifiers, and frankly I like the challenge. When I do live sound however, it is the opposite. The flexibility of an active system in that context outweighs the extra gear required.
Request you to please continue your channel it’s the best till date for DIyers
Yes I’m really hoping to do that, but this renovation is just taking way too long. Soon!
Active is clearly the better choice; always! As mentioned in other comments it's much quicker and cheaper to design and then tweak an active, especially a DSP version, crossover. I'm a bit chagrined that see the obvious error of his "apples and apples" comparison between the passive and active. The passive was a 2nd order, 12dB/octave design. Barring other factors, in order for the response to sum to flat at crossover you must flip the polarity of either the tweeter or the woofer. A 2nd order crossover will give a +90 degree shift to one driver and a -90 degree shift to the other at the crossover point. Combined it results in 180 degrees out from each other. This is physics gentlemen, not my opinion and certainly not a marketing position. In order to get a flat response manufacturers always flip the polarity of one of the drivers. This gives the best frequency response at crossover but, destroys the integrity of the overall phase response and, therefore, it's performance in the time domain. It's an unavoidable and poor compromise. Using a 4th order, 24dB/octave gives you even more phase shift at the crossover frequency; +180 and -180 degrees but, as is apparent the phase "wraps around' so that both drivers are in phase again at the crossover frequency! It also reduces the area of the crossover where both drivers are working and that's a good thing too. BUT; in this video he made no mention of "correcting" the flipped phase that was, I'm sure, in the 12dB, passive design so that they were in phase for the 24dB active design. If fact, if you do your crossover in the active design correctly and optimally it will guaranty a dip in the response at crossover if the polarity is flipped on one of the drivers. In fact, the depth of the dip is a figure of merit for the overall performance of the drivers and the crossover. The deeper the dip at crossover when you flip on driver the better the summation when you flip it back to correct polarity!
You can do a 24dB/crossover in a passive version but, it is a doubling of parts ( two caps and two inductors per driver). If you use quality components for good performance then it will absolutely cost more than an active like the miniDSP and that assumes you get all the passive parts dialed in initially and you don't have to buy extras to tweak the design.
He also didn't mention that an active crossover is also the best because you will then be biamping the speaker which has it's own virtues and is the best way to go. This is not the same as the pointless exercise of bi-wiring which produces no measurable or double-blind positive results and is simply a con job from higher end manufacturers. Don't drink that Kool-Aid.
The reason you don't see active crossovers in consumer equipment is because if the end user is supplying the amplifiers they then have the responsibility ensuring that the output level is correct between the high and low outputs based on the sensitively of the drivers. This requires a good measurement microphone, an RTA or some other analysis device/software and some method of getting the room "out of the measurement". This is far and away beyond the skill, test equipment available, and the time invested by pretty much anyone but, people involved with this hobby.
Lonny J.
Obsfucation I don’t think you understood the video and you don’t understand how phase works. I was showing that active requires just as much design consideration as passive. Not that passive was better. I explained that to other commenters but you obviously didn’t read my responses. I’m tired of explaining it so maybe I’ll do another video. I even wonder if you watched the video and tried to learn something or just watched a little bit and jumped to conclusions. Did you watch the ending where I describe how phase and time alignment works?
Obsfucation for the record, I like active better. I never said otherwise in this video.
you can correct phase problem in digital domain with group delay or inverting or so. obviously active solution is superior to passive.
But you can also do that with passive. The point of the video was most people don’t. They don’t even know to measure and see that the phase is wrong. That’s why it’s not a magic box or easier. Better, sure ya I’d say active is better.
Why not just invert tweeter with invert button and see what it looks like then?
mark clausson the point of the video was about not measuring and using active XOers blind. So what if it was 90 degrees out of phase and I inverted the tweeter, then what. And if I hadn’t measured how would I even know the dip was there? Yes the dip can easily be corrected, but for those who don’t go through the proper process of figuring out the slopes, delay, eq, etc. then active crossovers are no better than passive.
can you get perfect phase response with a passive crossover? this would save me allot of money for the atmos system i'm trying to build for my bedroom ultimate competitive gaming setup. reason i'm being so obsessed with phase is i think it's crucial for our brains ability to perceive a sound location with the least processing time.
alex kirk yes passive can be phase aligned. Active can make alignment a click of the button while passive requires phase to be considered wholistically.
thanks for reply :) is there any chance you could do a video on it, or link me to one if it already exists? i know it's probably no easy task but after hearing my friends co-axial dsp monitors i've been obsessed with the idea of perfect phase in loudspeakers.
A passive version? My video series about the Life S5 speaker ends up with a passive crossover in phase. Or do you mean perfect "linear" phase? That's a bit different and usually requires a lot of other compromises to work. I've never bothered with one of those.
i was talking about perfect linear phase but i'm guessing it's way easier just bi amping it with a mini dsp. do you have any idea where i should start reading as i can't fid anything on google. btw thanks for all videos, thanks to you i'm no longer scared of passive crossover design.
alex kirk hmm, I can't think of any resources that deal with that subject. There's probably a few AES papers. Minidsp might have some info as well. Or Dirac. It's not a subject many people are interested in. Seach FIR filter because FIR can correct phase where as IIR filters generate phase shift. That would be a good place to start.
So about phase misalignment, how do you respond to this:
"The K231 is an all analog design, engineered with only the highest quality audiophile grade components. Users will have complete control over sound and be able to eliminate distortion with subwoofer, high and mid-range channels, which are phase aligned to within fractions of a degree."
www.releasewire.com/press-releases/sublime-acoustics-k231-3-way-active-crossover-set-to-revolutionize-consumer-audio-market-with-affordable-professional-grade-sound-capabilities-791166.htm
What is the root cause of the phase difference in this example? I was always under the impression that with a passive crossover the components used created a slight phase shift so that when you run a 12db per octave slope you create 180 degree out of phase signal around the crossover point. I was also under the impression that this would not happen with a DSP. But, clearly, it has happened. Is this due to driver placement (i.e. the center of the woofer is further away than the center of the tweeter).
Justin Gardner there is a phase difference due to the time delay of the tweeter relative to the woofer. But you have also learned that dsp also introduces phase so that is good!! The acoustic response changes therefore the acoustic phase changes as well. This happens with active or passive crossovers. For every order or slope you get 45 degrees of phase change.
I'm using a Pioneer active xover with my Tannoy 10" Gold Mon. and my JBL 4320 (this one as a subwoofer) to listen to vinyls. This must be much better if I switch to the digital minidsp. But how do I use that minidsp unit for analogue audio? Impossible without further complication, right?
Stick with your Pioneer or pick up an Ashly or Marchand. Both sound very natural. Not DSP's but once you get the points figured out by testing, set it and leave it. You'll want to fabricate a BSC network between x-over and amp for the best sound. Do a Google search for a diagram. It will only cost you a few dollars in parts. Be prepared to be blown away by the sound quality.
What I really don't get with the practicalities of phase is that the SPL can measure fine (flat) but I believe there can still be phase issues which play havoc with the sound. Any thoughts welcomed, cheers Ryan.
MrFrobbo what is usually happening is a case like this is the FR is flat but phase is wrong, so the power response (a combination of all points off axis) ends up being very peaky in the midrange (or where ever the XO point is). It can sound very glaring and forward. Something won’t sound natural.
@@ImpulseAudioSpeakers Thanks Ryan really useful...think I'm getting it...slowly!
Input = RCA (unbalanced) > DAC > (filter) > ADC > RCA (unbalanced) = output?
What about SPDIF In > (filter) > SPDIF Out (FWIW, Coax vs. Optical = Jitter)... analog amplifier has DAC inputs... vs. PWM amp?
Fine Phase adjustment to compensate for mounted driver offset (mSec.... µSec.)
if I have a 3 way speaker and run it with a miniDSP does that also mean I need to run a 3 channels of amplification? Or is there a way to somehow run only one amp to power all 3 speakers with the miniDSP?
You’ll need 3 channels, or you can use a passive XO for the tweeter and mid and then use the dsp to cross the woofer to the mid and tweeter combo. This is a really good technique because the passive XO protects the tweeter, the mid and tweeter need less amp power, and it saves 2 channels on a stereo system.
You can do things in seconds (like Harsch crossovers) that would be extremely difficult with passives. It is far faster, more flexible, more precise and trivial to optimize, plus you get the gigantic benefit of biamping. DSPs are vastly
better IMO.
Peter Campbell did you watch the video?
But all you needed to do is reverse the phase of the tweeter like most speaker companies do and use the active eq in the nuke to flatten everything out. This is how we do it in every recording studio. This is also how JBL profesional theater speaker systems are done 🤷🏻♂️
What if the phase had been 90 degrees out. What if the speaker didn’t require polarity flip. Just because I sorta made it work with measurement equipment and knowledge doesn’t make active any easier to design than passive. I needed the same knowledge and measurements.
I want to purchase this products for my powered subs. I read a lot of good reviews in forums. Seems like people can achieve phase correction after few short learning curve.
What is your opinion products like this for powered subs?
Thanks!
itoosh it's a great product for speakers and subs. The video wasnt meant to suggest its a bad product. Only show that it is just as much design work involved in active as it is in passive. Good luck with the subs.
All clear.
Thanks!
So I have yet to hear about a passive crossover network that time aligns drivers which would be a phase compensation technique. The truth is that passive crossovers have a number of drawbacks over full DSP units. BTW The mini DSP is a pretty no-frills unit. There are far superior processors out there to that although they do work. The other thing that seems to be lax in speaker design is that so many designers only consider the FR plot on the page and don't actually listen to the driver. Many high-frequency drivers exhibit serious distortion issues in the area below 2Khz. It is more common than not to design speakers now to cross-over above 2K because of that. We took old JBL 2206 cones and using a waterproofing solution stiffened the cones which changed the response curve to one that allowed us to roll it over at 2200Hz. It was coupled to an old 3" B+C compression driver mounted to a CD horn. A little time shift on the low drivers (it was a 2-12 stage monitor) to compensate for the physical time offset plus adjust for the phase misalignment at the crossover point and all it needed was a few little eq nudges. I could have mixed a record on it. The thing was with the crossover set to 1600 Hz made it painful and unlistenable. 1600Hz was where the designer said the crossover should be set. Why didn't it work? Becuase just like almost all designers they only went by what the specs say. Specs don't plot distortion vs frequency. You have to listen for yourself and decide. "Well, that is what pa speakers sound like". NOT GOOD ONES!!!
Rob Chapman I don't know any speaker designers that only care about FR and don't listen. Hmm, that actually sounds absurd. Have you actually met someone like that?
Check out ladder delay networks. They are passive and compensate for phase.
The minidsp is extremely advanced compared to the bricks I was describing in the video. But even among the top contenders minidsp sells some very powerful units.
Tweeter distortion is entirely dependent on the model of tweeter. I would be careful generalizing that tweeters distort below 2khz. It may be a rough approximation where most do, but hardly accurate for all. HD is one thing while woofer beaming and cone break up are also commonly an issue above 2khz. So pick your poison I guess.
Sadly, I have met a few and have actually been chastised for not succumbing to their sonic philosophy. I view it this way. If it doesn't sound good regardless of what angle I hear it from it is an unusable piece. I guess the thing there the guys who like fidelics don't usually gravitate to high SPL systems as traditionally they have been polar opposites. The thing is we have high SPL systems that now are actually producing sound with a great degree of fidelity and that means it is time for the old guard to get out of the way.
I wasn't actually trying to generalize but the majority do. I guess in a perfect world we would need to use a 3-way system for this. I don't think I would trade a bit of beaming for distortion though. Everything in balance. It is, of course, a trade-off but more often than not narrowing the coverage angle a bit is a good thing. Besides that if you were to measure most stage monitors on the market you start to question who is designing these things anyway. They are never flat and often have crazy phase holes.
Wow.
"look! The phase is wrong! I knew it would be bad! Don't use active crossovers guys"
later: *presses a button to correct phase*
Gara Von Hoiwkenzoiber I did not know how bad the phase would be. Yes phase is easier to correct with active. But the same knowledge and work is required to get there. That was the point of the video. Not which one is easier to program. It’s not a magic box like some people treat it and think it is. Not many people even know what phase is but they think active just makes it ok.
Hi, im feom indonesia.
Mini dsp is $105 at PE, with shipping and tax its $161 at my door.
Dont have budget for it.
I go with $3 caps, and $6 inductor.
With free XSim, im doing fine with passive.
Sorry, something is confusing me... The wave length of the frequencies! You can not just add frequencies as you do. The tweeter can not produce so loud frequencies at 20, 30, 100Hz or so low and bass can't do it on 10000Hz or higher. Also, the only case where two waves can interpolate is when they are of the same frequency (or same wave length). And when the phases are 180 degree opposed they will cancel themselves. On the diagrams we can see that drop but that is not so significant when the music is in question because it is not the permanent drop in the music spectrum. The question is can anybody register that drop masked with the whole other frequencies? It is only seen in the lab when you sweep the frequencies.
For me, the active crossover behind the preamplifier with amplifiers for each individual range is the top. Can you imagine how can you fine tweak 6-way system? Even with the cheap components.
Nice video. Like you said the title might be slightly misleading . I could be totally wrong but what you explained was about the acoustic phase difference which you are compensating with delay . But what about electrical phase difference . I know they are connected but it will be good if you can make a video about how they are tied together
jojiran i could possibly do a video about electrical phase sometime, but that isnt a strong point for me. They arent really connected to acoustic phase either. Im more knowledgeable in the acoustic realm but Ill try to cover that topic as I go. Thanks.
Thanks for your response . Ive been trying to get my head around this subject. Since I work in a studio setting , it is important to understand these concepts well in these times of evolving technology
What you did in the video ( adjusting delay) will produce phase coherent sound i.e. sound waves hitting the human ear as if all those frequencies emnated from a single single source even though they may be physically separated
The phase response you see in common design tools is the electrical phase . That has got to do more with how the amplifier sees the load. I was very confused about these concepts initially but now I think I have a better understanding .
jojiran yes, and also be careful that some of those design tools show the theoritical acoustic phase caused by the crossover. That is not useful information because those design tools assume perfect woofers and tweeter, which none exist.
OK I will look out. Currently I use VituixCAD and Ive been playing around with the Behringer DCX2496
I get your point that one cannot naively convert a passive design into an active or design a system from scratch without having at least minimum required knowledge regarding designing loudspeaker systems. you give a very good example in this video yourself. I think rookie DIYer having no clue about phase issues would try to eq that dip out in minidsp and arrive at a badly designed system that potentially measures quite flat but sounds just not right.
also one thing to remember is that those who attempt to design active systems do it not only for the sake of driving each driver with its own amplifier but also for possibility to take full advantage of DSP power - driver time alignment, linear phase xo, driver linearization, steeper slopes.
David Gale thank you for understanding the point of the video. It was lost on many. Your right about the advantage of DSP. It is very powerful and if you know what you’re doing then it is unmatched by passive.
can you teach me how to build passive cross over
DJ Harry Trivedi My next video is kicking off a 4 part series of building and designing a desktop speaker from scratch. I'm going to look at how to design a passive XO and it should help you a lot. Should be starting in a week. The design part in two or three weeks. Keep an eye out for that.
Don't bother and keep saving your money for just a bit. Go proper or go home.
Get a dsp and 4 stereo amps; or 3 stereo amps and a plate amp or 2 for the sub(s). Power your 4-way and you'll have audio that will rival any passive speaker.
Think 12 inch sub, 6 to 8 inch mid-woofer (2 if you want), 2.5" to 4 inch dome upper mid (or cone), and a 7/8ths or 1 inch tweeter.
You'll be able to experience anything your heart desires. Turn up the subs. Turn off the subs. Then listen to a 2-way or a 3-way setup for vocal clarity. Compare and contrast all you want, and be an audio GOD in total control. 16hz to 20khz and above will be entirely in your reach.
I've never understood the appeal of building your own passive speakers. You're then stuck with a speaker that can't do much of anything (just like all the other crappy passive speakers on the market for the average idiot consumer). And chances are, your passive built setup won't compare to the hundreds of passive speakers already on the market. I guess you could add an equalizer with a passive setup, but that's fairly lame and doesn't address the flaws of passive speakers.
Have fun and build an active system which will sound better, do more, and can be tweaked and played with according to whatever your heart desires.
There is definitely a place for passive crossovers. I am a fan of active crossovers but for my 4 Atmos speakers, I did not want to deal with 8 amp channels and active crossovers. So I learned how to do a passive design and have been very happy with the result. Xsim is very cool software and can be learned in a few hours. Although I think I spent at least a couple of days last winter playing around with it.
Fair enough. And your'e right.
I was just trying to oversell the fun and advantages of an active setup.
Too many overlook the option and many don't even know it exists.
ya but how to learn please help me im 📺 tv technician and dj too
You need active electronic crossovers that you can actually regulate continuously until you balance everything out , Rane and dbx work well. I don't waste time and money with passive anymore.
I don’t think you understood the point. The point was unless you can measure and understand speaker design, active is not easier.
I was always under the impression that you could just change the phase on a mini DSP. I think anybody who would buy a mini DSP would probably also buy a measurement mic and download REW.
Justin Gardner you can easily change the phase via delay on a minidsp. In my experience, a lot of people do not buy a measurements mic and use REW let alone know how to use it. And when you consider how many people buy the fixed slope active boxes, it's scary. A lot of people think these are magic boxes. It's why I made the video.
There is no comparison to which is better (active) but you just need to have a very good grasp of digital/analog design principles as well as standard (discrete component) X-over approaches, to design, implement, and build a good active 3 way system; that also measures well.
If it (active 3 way speaker design) was easy, all the Hi Fi fiddlers and speaker companies would be doing it.
hello fortunately I found your channel which I find very good content, I subscribed to not miss your videos
Thank you so much for clarifying.
Could you try out Behringer DCX2496 Ultradrive Pro
Or maybe another DSP you can use with a mic and automatically make a linear sound
Marcus dahlén all dsp crossovers have these same issues. Some are more automated and can apply eq, but can’t fix a hole like that. Apply EQ until flat isn’t always a good idea either.
No matter the hardware, it’s best to apply best practices and get the XO dialed nicely.
@@ImpulseAudioSpeakers
Do you think phase is such a big issue when you can cross 48db?
I like to configure my systems with dsp because it's so flexible and also very exact.
Today I want to cross my ib sub at 50 instead of 45, tomorrow I want to ad some midrange:)
Think I could make so many drivers sound so much better by controlling them like this and also it's fun:)
Thanks for the reply man, I'll tell you what my air motion Transformers sounds like free air mounted if it's nice ☺️
Marcus dahlén certainly dsp is more flexible and precise, but phase is an important aspect even with steep slopes. Subs are probably less prone to issues when making tweaks, but the higher frequencies should be checked with measurements. Easy to think it sounds great and miss a lot of information. Especially with suck outs like I showed in the video.
@@ImpulseAudioSpeakers
So I just realized I need to analyze the frequency response and correct possible phase issues with the delay function.
What is your opinion on just letting the frequency ranges overlap just a little bit to avoid the Gap in response?
@@ImpulseAudioSpeakers also do you have a suggestion for a good midrange driver 800 - 2000hz
Did you try the Accuton mids?
I'm feeling like putting a dome on top of the speaker
Best regards 😊
Then, of course, you've got your extremist Klipsch fanatics using Xilica XP-4080 and 8080 DSP's on their Jubilee rigs with the awesome K-402 and K-510 horns (Roy Delgado's secret sauce). And for them it is so much better than anything passive. But that's very extreme and niche. But yeah, for that little pair of DIY 2-way bookshelves passive is just peachy.
Active is the only way to go. I haven't touched passive for years. I have a miniDSP 2x4, MiniDSP DDRC24, Dayton DSP-408, amplifiers with built in DSP, etc etc. It makes life so much easier and cheaper. Different speaker projects become much easier for newbies to get into without having to have a degree in electrical engineering and soldering. With a DSP you can make changes on the fly while listening to music rather than taking everything apart, buying different components, connecting components then listening..no way to A/B compare or tune the system quickly.
I would like to see the response with the tweeter polarity reversed. I know it wouldn't be perfect, but shouldn't it be much better?
davefred3 yes it would be much better. But someone would need to know to reverse it, and it would do nothing if the phase were out by 90 degrees instead of the 170ish degrees it was out here.
If you talk to any speaker designer Or engineer which I design all of my speakers custom myself they would rather use an active crossover but it's just too expensive to sell to the general public I will never use a passive cross over On my system ever the The average passive crossover is garbage And active crossovers do not get hot at all we know what happens when Things get hot
Mustafa Akbar ok. Hope you know how to design them properly.
@@ImpulseAudioSpeakers That's the easy part I am a carpenter by trade The hardest part which I will concede to is theTuning of the crossover Per individual driver I run A four-way stereo takes time to Get it right and we're not even gonna talk about room treatment And speaker placement whole different animal
Active is definitely the way to go. Its for sure easier. You can actively reconfigure and compare and contrast results on the fly. Your phase issues have more to do with driver size and center to center spacing, there are rules to speaker design that you would be wise to follow to eliminate these issues....a crossover point can expose your poor designs if thats the case but crossovers don't cause this off the bat. A 48db LR crossover is designed to be in phase, so if you still have phase issues its because bad choices, else where in the design.....If you know better you can do better =)
DR.NNOO I showed in this video how an LR4 XO isn’t automatically in phase. Active has just as many design challenges as passive. If you think otherwise you’ve missed the point.
@@ImpulseAudioSpeakers Thats the key, you don't design active crossovers, you buy them and apply them how ever you like , where as a passive isn't something that you can so easily reconfigure. That, in itself, makes them easier than a passive. Anyone who assumes that a crossover of any type is going to automatically make everything in phase, is lost, anyway. Still, with an active crossover you can try unlimited configurations in the same amount time it takes you to assemble one passive crossover limited to the amount of passive components you've got. If you dont get it, yet....make a video of you trying 3 crossover points with an active, then make another video documenting you making the same 3 crossovers from scratch and we'll let the people decide which route they want to go =)
DR.NNOO ok first yes you still have to measure each driver and use software to determine slopes, eq, etc. You don’t just buy and start fiddling. But even that aside, fine, active is slightly faster as pulling out the computer and booting up the software, plugging in the USB, programming it, hooking it back up to multi channel amps, than alligator clipping together some passive parts. Yes it’s SLIGHTLY faster. All the rest takes the same time energy knowledge and effort.
I like active XOers. All you active fan boys just don’t get what I’m saying. Sorry I’m kind of exploding on you because this keeps coming up. Yes, active has many many merits. I use active XOers myself. BUT it’s NOT really any easier and if you think it is you’re doing it wrong. Saving 4 minutes on XO assembly over the course of a 6 month design process is not helpful.
@@ImpulseAudioSpeakers Slightly faster is a complete understatement, my active crossovers are built into my amps which are connected to my computer, theres no disconnecting or reconnecting, even my mindsp is ran from computer and straight to the amps, no disconnecting there either, if it takes you 6 months to design your crossover point its because you never understood loudspeaker design. You design the system around a preselected xover point not visa versa, and if you don't know how to do that....keep on, you'll get there =)
DR.NNOO ya I’m an idiot. I forgot you just pick a number and force it to work. Ok man.
that was a nice null. im not surprised you chose not to display the inverted result. it wouldnt have served the content very well.. ; )
r. grambo not quite sure what you mean but if I flipped the polarity it would have looked very close to the result with delay. It would have been redundant.
Out of phase invert the tweeter ?
What if it was 90 degrees out of phase? Would you still invert the tweeter? And if a person doesn’t know that it’s out of phase, how does an active XO help them?
Does minidsp introduce latency?
Probably a little. Not enough that I’ve ever noticed. But I haven’t games with it.
Not much. I play at games and I haven't noticed any latency
Delay? your tweeter is 140deg out of phase (according to you)... you've tested it with a generic passive and now with active.... Did it occur to you to flip the electrical polarity on the tweeter, in order for it to be acoustically in phase with the woofer. The delay hack is a horrible suggestion, because now the woofer and tweeter going to be OUT of phase EVERYWHERE on the spectrum, by the amount of the delay. It's not a big deal because the x-order is 24db/octave, but it's a hack - the delay was never meant to be used like this.
pretol the delay was absolutely meant to be used like this. It’s puts the drivers exactly in phase where as simply flipping polarity would still be out by 40 degrees. Not sure where you got the idea the passive XO was generic. It was properly designed which is why it measured way better.
You're right, I meant generic active (which is the hack job you've done for this "test") vs "properly"-tuned passive (according to you). Which is NOT good guide for people that are building speakers at this level, and YET, here you are making profound conclusions. You'd have to be a moron to design an active XO with generic parameters. Tunability is THE WHOLE point of active XO's (presumably with a calib'ed mic), and here you've done almost none (except wiring your tweeter backwards, and compensating for it with delay).
With a proper set up for measurement using a loopback on an impulse test tone, one can very effectively time align the drivers using measurements-- here is the instructions from mini dsp www.minidsp.com/applications/dsp-basics/time-alignment with a link to instructions on using the loopback.
Trevor Jones yes or you can Check the time between impulse peaks in the measurements. Provided the person knows either of these methods. Good link thank you.
Great video thanks for sharing
Phase isn't the same thing as delay though, so which did you apply?
Chris Wilson phase and delay are exactly the same for all intents and purposes. I applied delay, which changes phase as I tried to demonstrate on my hand drawn figures. Sorry they were drawn poorly. One day I'll try and show this better.
great video, must be watched till the end!!! :)
Good video I have the MINI DSP 2x4 device with the 2x4 advanced PLUGGIN for my subwoofer applications.
Thank you for explaining this issue, I hadn't thought about phasing issues with the drivers. Very informative!
My beef with passive is STILL money !! You can design a phenomenal passive unit - but the minute you start using high quality components = the cost goes up $$$ PER SPEAKER !!! The obvious limitation also applies to active = more amps and added multi channel Active units..... so either way it never gets away from cost vs quality.....
He misses the main advantage of an active crossover for the hobbyist, which is that you can make all these adjustments without needing replacement components, so you don't need to get it right before you order the parts.
I didn’t miss that part at all. The video was about the idea that you can just click X frequency and X slope and have a finished XO. The benefit you list is pretty minor in my opinion. The greatest benefit to active crossovers is the efficiency compared to lossy passive crossover components. That is huge a very big reason to use active. The title is are active crossovers easier. Not are they better.
Right is better than wrong!
presented passive has a phase problem seen at 600-800 Hz and 3-5 KHz and active has the same. Development is a process.
So Like One day I tried to make a Crossover for my Full range speaker... And like... My life got turned upside down and I had to Move to Bel Air!
Yes phase is important in all design, one can screw it up with both passive and active.
Appears you are mistaken here, as explained in this post:
www.stereo.net.au/forums/topic/23800-active-vs-passive-crossovers-and-phase-shift/page/3/
For gosh sakes already...
"only if you know how to use it" goes for most things in life, and in audio. Duh.
Luv this guy's vids but this beyond his scope (no disrespect), you can't hack DSP, it's graduate level math. For example you can auto compensate a speaker that would not be possible by analog/passive means (theoretically impossible). DSP essentially emulates the analog parts, thus u can create a "part" that does not exist or even physically not possible, like negative capacitance (I'm keeping this simple, it's the same tech used in military guidance sys for example). Beyond that you can compensate the speaker's impulse response (not practically doable in the analog world). Then there's efficiency, the reason you have power components is the passive X-over is it throws power out the window to compensate (utterly ancient way of doing things). good news is you have software that can do it all work for you now, no thinking involved, like cell phones leading everyone by the nose (yeah i'm from the DSP world, b4 it was a buzz word ;) ).
Greg D no thinking necessary? Clearly using dsp is also beyond your scope. The point of the video was to show newbs that just as much thought was necessary. Did you watch the video? Not sure you did. Please watch the whole thing and tell me if you still don’t get it. A newb cannot just hook up a dsp magic box and enter 2000hz LR4 and expect good results. That’s not going to sound good. And I showed it here.
Clearly my video was not easy for people to understand. For that I apologize. Still not sure why, I thought it was simple. It is just as much design work as passive. Period.
And BTW, I like active, maybe even more than passive. This is not a hate on active speakers video!!
Wow man, i was being polite, you're having a reaction. Yes i watched the entire vid, did you read my entire message? You do good stuff man but your video was like someone who's never flown a jet then trying to demo it without any flight experience of any kind, but again, no disrespect. Or are you saying you're very familiar with DSP? enough to say it's beyond someone's scope? You know convolution cold and it's edge affects? u know self adapting communication channel compensation filters so these audio steady state one's are are like childs play? Because if you are, then spend a little more time on the vid to do it right, if not then maybe don't post until it is right. So yes, there are programs now that will analyse the speakers then adaptively create the dual FIR/IIR filter coefficients to completely compensate it in every way, you can follow a procedure so one doesn't actually have to know what one is doing ( here's one for the laymen: www.bodziosoftware.com.au/). Verses say going knee deep into MATLAB or writing your own code and download your own coefficients to the DSP, but again, then you have to know math at a graduate level. Also that MiniDsp unit you have will not do impulse response correction and i don't want to get into why but it's a key disadvantage. And really man, every wonder why it's only "home guys" posting on the net, not guys from the Boeing lab or Tesla or the DOD or other pro's? because they keep it inhouse, so it's odd to me when an expert might chime in and say, be careful, you might unknowingly be posting miss information (which is rampant now). So one of my "DSP buddies" saw your response and said "why do you bother man?" and yeah.. he's right, so good luck man, this post was for others and try beware of what you don't know (btw, spare me the "male fragility complex" which is what your "clearly dsp is beyond your scope" retort is (i qualified mine), unless again, you're in the DSP industry as i inquired above, which means you really do have some other issue).
Greg D alright sorry, my comment was over the line. Although it was really no different than your comment. Saying you like my other videos first doesn’t change the comment.
I still don’t understand what you disagree with in this video. The video is saying that you cannot take pre determined electrical filters and simply plug them in and bam good results. That’s all the video is saying. Do you disagree?
By the way, professionals do post online. All the time. I’ve seen forums and facebooks posts from engineers with JBL, pioneer, etc. Not sure what a Boeing engineer has to do with audio.
Those auto calculators are only so good. First, they still require measurements, which is partly what this video was saying. Second, they can’t fix things like driver spacing, Pilar response, distortion, etc. The speaker builder needs to make engineering decisions about this. Writing dsp math has nothing to do with the video. My comment about watching the video was not meant to be rude. I seriously ask did you watch all of it? Understand it? I admitted in my comment that clearly the video was not good because you’re about the 20th person to misunderstand. Do you think newbs can just buy a mini dsp and hook up two drivers and then program it to use an iir filter of what ever frequency and get good results? I doubt you do. Because as shown in the video you will very likely get bad results.
@@gregd6022 hi Greg, what dsp crossover can do impulse response correction? Is lack of impulse response correction a flaw across the whole minidsp line? Any other links to software to analyze drivers you would recommend. Think of going dsp to gets some Altec 604Es to work as well as they can.
@@ImpulseAudioSpeakers The only person saying you can plug-in a miniDSP and automatically get good results....IS YOU. Having an argument with people noone made.
I love your videos and learning so much from them, but this one is bad, you conflicting yourself few times, sure if you do the DSP wrong it will not going to be as good as Pasive done right, but with a little knowledge you can get much much better sound with the mini DSP then you ever hope to get then with a passive crossover, I know because I upgraded
2 professionally design loud speaker ( JBL and Dayton ), from Passive to active and they sound way better now, sadly outside of studio monitoring the world don't like active because people think they will improve the sound by buying a $3000 amp but they don't want to buy 2 or 3 amp and a DSP , if they make a system with 3 $100 amps and mini DSP they with get much better sound. but I know it is hard for speaker makers to sell this idea, Anyways love your video ,, David
David Mann I didn’t say active wasn’t better. It is. I said they’re equally difficult to design. Obviously that wasn’t clear in this video because a lot of people, yourself included, disagree with the video.
I'm used av ressiver Yamaha z in DSP crossover end 7 chanel
seriously i'm confused ... you're going on and on about the difficulty of crossover design, but then apparently your audience needs a slow and careful explanation of phase cancellation, which i think i understood when i was twelve. i would think many things would be difficult for aspiring speaker builders who didn't know this
It was meant to be an example of why newbs can’t just plug in an active XO and expect good results. Sorry.
you are not doing active good, phase is eliminated in 4th order crossover , i dont know why that DSP doesnt do that , its bad design , however i would like to experiment but i need more , at least 4th order , not 2end order , wahts with the miniDSP what it can do , can it simulate a lot of different crossover or fixed at bad design
Pavle Pavlovic the mini dsp is great. I’m afraid to say you just don’t understand what I was teaching in this video. 4th order doesn’t automatically fix phase. And that’s the problem. You need to understand what it is that makes a XO work whether passive or active.
Try watching again and doing some reading. Even read some of the comments to see what I mean. And also try it in real life.
No , man i didnt come to argue , but ether you are wrong , or this paper i was reading is wrong , and i quote "Today, the de facto standard for professional audio active crossovers is the 4th-order Linkwitz-Riley (LR-4) design. Offering in-phase outputs and steep 24 dB/octave slopes" , Do you believe me now . ITs no problem you are not the first ive found doing stuff the wrong way , these are PHD topics , one cant know everything , but we learn and go on , maybe you can do another video , i dont know it is what it is , that miniDSP can be a good thing but it needs to do it better , im interested in it off course , but no one will tolerate phase issues , thats just not going to happen , ive found DSPs that do it all but like 500$, its too much
Pavle Pavlovic your understanding of the article is wrong. The outputs are in phase but that doesn’t mean the tweeter and woofer phase is 0. In this case the tweeter is behind the woofer. Telling me I’m doing it is wrong is argumentative. Just because you read something doesn’t mean you understand it. I actually demonstrated the folly I see so many people fall for. Either believe what I’ve shown and try to understand why, or keep reading papers without grasping it.
As for the mini dsp, it can do it right. I show at the end of this video when the delay is set properly to deal with the phase issue it works really well. Plus it does PEQ and other things. So when someone knows what they’re doing it is very powerful, and costs much less than $500.
Maybe try watching the video again and really concentrate on what I described with my hand diagram. It might help. I don’t know everything, but trust me, I know I’ve done this right.
Yes i am aware that it comes out on a wrong note ,but its just the truth sometimes it hurts , from my understanding phase problems are due to filtering but filter inverted phase it will get flipped once again , and return to its original , hence no phase issues , there are liner phase EQs as well as linear phase crossovers , no problem , as i understand you seam to think that those dont exist , im telling you they do im using both liner EQs in production and liner crossovers in monitoring , you can get some and measure , i dont know how else to tell you . Im not trusting one paper , i know it , im using it , ppl use it every day you just need to find liner ones and you will see it in your results
here you go analysis of Crossover topology with tests , there is some phase issues but nothing as you experience , more like 1% or around th-cam.com/video/aiJQED7-1b8/w-d-xo.html Dont knwo waht else to say , need i call Linkwitz here :)
no, it is not correct, a DSP and a activ crossover device are diffent things. A DSP has to calculate the signal this takes time, a activ Crossover uses the same parts as a passiv crossover and a op-amp, but before the amplifier.
Most people use the terms dsp and active XO interchangeably. Even more technically true is that a dsp and passive parts before the amp are both active crossovers. They actively change the signal delivered to the amplifier. A passive crossover passively changes the amplified signal.
fUCKING BRUTAl! Thaks Ryan!
You said "if they can't compensate for phase, by, you know... using delay...." but in real life, phase and delay are related but, completely different things.... you would never! use delay to make adjustments for phase...thats a basic understanding in loud speaker design....I find your legitimacy questionable at this point lol!
DR.NNOO ok if you don’t know to use delay to correct a phase issue and that calls my knowledge into question, you better just ignore my nonsense. I’m clearly an idiot who doesn’t know what he’s talking about lol. I mean, who would use time to correct a time function. I mean, that would be as crazy as using amplitude eq to fix an amplitude problem!!
Phase and delay are directly linked. They are married. Check with ANY body who knows how to design XOers and delay will be one of the ways (not the only way) used to solve a phase misalignment. Go ahead and check on me.
DR.NNOO just in case my last two posts weren’t clear. I LIKE active XOers. They’re just very misunderstood. Unfortunately you are one of these people. I don’t actually want you to ignore what I have to say. I think I have something you can learn. I just don’t understand the “active or die” mentality so many people have. Active XOers are great. This video did not say they weren’t. I’m gonna chill out now. Thanks for watching.
@@ImpulseAudioSpeakers by adjusting delay you are adding group delay to the whole channel...if the group delay of both channels were already in sync, you don't adjust delay to get the xover band in phase, only to take away coherency of the whole signal, thats ass nine. Once again, there are rules to follow, regarding center to center spacing, driver size, and intended xover. You would also look at the measurements of phase transfer function of each driver to choose an area where the drivers can cross over well, phase wise, but thats done BEFORE you even purchase the drivers. You would even look at the group delay chart to see how well two drivers play together where desired xover....also before a purchase....you choose drivers that play well in the intended area of crossover, thats simple ...From there you may have to time align them using delay, and that may "fix" the phase, in your head, but the phase wasn't broken, you just needed to time align the drivers. If there are issues in phase after you time align, and you can't get the intended results, where they should have been, adjusting delay to unsyncronize almost the whole bandwidth to gain phase coherency with the cross band is ridiculous. Delay and phase are of the time domain, yet not exactly the same thing. Maybe you are stuck within the passive mindset. Nowaday active crossovers have parametric PHASE eq's, or automation to adjust phase and ONLY phase...its a separate parameter than delay, in design philosophy and in real life application... though they affect one another. Getting time delay insync will improve phase on a greater scale than adjust delay to optimize it within only the crossover bandwidth, I feel like I'm repeating myself. Voicing to flat and room correction improve phase as well but at a certain point, phase is what it is for the situation and cannot be improved. If you don't make the right choices leading up to implementing the predetermined xover...you are just fk'd lol. If you don't get it by now...nothing I can probably tell you.
I want one so bad. Find the best sound and then eq it. ( ˘ ³˘)♥
Wow! This video is misleading at best. It would take an enormous amount of time and boxes of components to make the changes with a passive crossover that you can make in seconds with an active crossover like the miniDSP. And all the work is done before amplification. You are not placing a bunch of "stuff" between the amplifier and the drivers. AND, you are dedicating amplifiers to different parts of the frequency spectrum. The amps will have less to do and can be optimized for performance and sound quality....Solid State for bass and tubes for highs for example.....Lastly, the last speakers I built, the passive, high quality components, cost way more than a miniDPS.....And if I got the components wrong, the cash register just keeps ringing. Love your videos....This one?...Not so much! :-)
Not sure you got the point of the video.
Exactly why they are not more popular.
This is ridiculous. The whole point is that you can tweak the active DSP really easily. What you show is the result of an off the shelf 1.5k active like say a Marchand. The whole point of a DSP crossover is that revising it is trivial, while revising a passive (or an electrical active) requires a lot of resoldering and a big parts bin. That's where its less work - you can iterate really quickly - and with the minidsp you can do it without even stopping playback. Do you have to measure? Yes. Did you show how easy it is to try inverted? Nope, but its just a click. How about adding a delay? Yes you did the result was good and yet 'I didn't work on this for very long'. Bingo. There you have it - the minidsp *is* a panacea if you accept the resampling, and so is eg camilladsp (perhaps more so if done before any DAC). The summary is self contradictory - perhaps the answer is 'if you want to go active, go DSP' but you appear so keen to lump DSP with fixed function active.
Need to address your bad lighting!
Ya I know. I think I got it better in my later videos.
Too much simplistic explanation. Bottom line is -active crossover has huge advantages over passive crossover an it makes more economical sense in today world. With a passive crossover you should be ready to pay insane amount money for one capacitor . You can buy few god quality DSP for that price. How many of those capacitors do you need for 2 way crossover in one speaker ? We even not touching sound reproduction quality of several amps for one speaker V.S. one amp and what passive crossover does to that amp. Please don't generalize such view points from personal point of view. Biggest contra argument to your statement would be to find recording studio with passive crossovers in they speakers. As soon as active crossovers was available for market passive speakers/monitors have disappear from recording studios inventory and it is bloody god reason why. Well you must know what you are doing with DSP ! Yes, I agree , it is not as simple to handle DPS at least as much I so on your video.th-cam.com/video/o2_gRAbnzyk/w-d-xo.html
Gin 8330 I think you missed the point of the video. Sorry I wasn’t clear.
Never liked DSP.
WESTEL Audio really? I don’t have a problem with dsp, it’s powerful that is for sure. So I use it where appropriate.
@@ImpulseAudioSpeakers
I feel like it's just adding unnecessary complexity. You can't use a normal 2-channel receiver and if the DSP, being a very fragile little computer, breaks you have to start all over again. Only things they are good for are active PA speakers where weight really matters or sub/sat systems with very low cut-off.
@@westelaudio943 Lol fragile. I can promise you that if you drop it from 1m it will survive but if you drop your 15kg avr from 1m it will not. Smaller electronics is usually more durable.
This video is nonsense. Way to push your agenda. The reality is that active crossovers are much easier since it gives you all the tools to fix any problems.
I don’t think you understood the video. Active is more powerful, but all the steps to create the XO are the same. They’re not any easier than passive. Still need to build, measure, model, listen, tweak, etc. Not sure how that’s even debatable. I don’t have an agenda. I can do both quite well.
@@ImpulseAudioSpeakers is it easier to make a new bigger box with a better port tuning or to just amplify the lower frequency that goes into the woofer? Can passive crossovers pull better performance from cheaper smaller design without reducing efficiency like active crossovers can?
@@ImpulseAudioSpeakers Is it easier to run REW and then modify and rebuild the speaker or to just push some settings?
@@BobFrostV active is better. I said it’s not easier. Of course active has many advantages, there’s no debate. The point of the video is that a newb can’t just hook up an active box and expect it to work if they can’t design a passive cross over.
Did you watch the video?
And no, you can’t just boost the bass if you build your box to small. Even with active you must rebuild the box, unless you like to clip amplifiers.
You really don't know what you're talking about and should quit dispensing advise.
william fitzpatrick haha. Well, you really should watch the whole video or go learn something.