The work you are doing here is immeasurably important. I have NEVER felt like I understood audio electronics (even though I'm an audio DSP programmer) prior to your videos. If we want to have a future full of people who are continuing to build audio electronics, we need great teachers like you to make it accessible. Thank you so much.
Audio stuff is a bit strange because unlike the instrumentation sort of designs, so distortions are perfectly fine and others are a disaster. To make it even harder, I don't think there is any simple rule to tell one from the other. A little soft limiting sounds fine. Sometimes hard limiting is really really good. Other times hard limiting is really really awful. This makes for a lot of "listen and see" sort of tests.
@@MadScientist267 To be clear - my initial comment is not about people manipulating audio and not understanding it. It's about me, an audio DSP programmer, understanding very well how to manipulate audio in the digital domain, but not knowing how to achieve similar results in the analog electronics domain.
@@erich1394 I think you mean the analog and/or hardware domain not the "electronics domain." If you're an audio dsp programmer then you most likely work almost exclusively in the electronics domain.
@@davidsotomayor8713 Nope, I meant exactly what I said. I currently work in the digital domain and I'm interested in the analog electronics domain. DSP stands for Digital Signal Processing. Analog Electronics are literally "Electronics that are Analogous" in the case of audio, they are "Electronics that carry and process electrical signals, where those signals are analogous to audio signals" Hope that clarifies things.
Really appreciate the updates to the design here Mortiz. Grateful for you listening to the feedback. Also, I can’t understate how useful the animations are, (particularly for the LED VU meter section).
In the days of b&w analog TV, the subcircuit at 17:20 ('ac-couple-dc-restore') was known as a 'black level clamp' - it derived a DC reference for luminence from an unreferenced waveform. I have also seen it used to fix-up a poorly designed signalling system where lengthy PSU wires - used also as a signal reference - experienced significant voltage drop. The signal ground would rise-up as a result, offsetting the perceived signal level. AC-couple-DC-restore saved the day!
Moritz I am watching your videos only because your animations are pure genious. I love the pulsating LED while emitting small rays of light. So funny. I think if I was not already into electronics I would start right now after watching this video.
This! I have been researching forever but I've never picked up the soldering iron. Every time I watch one of Mortiz's videos I feel closer to actually doing it. I'm sure that when I have time I will build this compressor (or a derivation of it) as my first project.
Just to let everyone know that some of the best sounding mixes from the early 90's mastered on to cassette tape sounded amazing. But not because it was cassette tape, but because the cheap cassette recorders being used to do the mastering (DAT back then was only for the super wealthy), had little compressor chips inside that automatically kept the levels in check to prevent distortion from accidental overload and to raise the volume of the quiet bits. This lead to a super-lovely sound that I've never been able to recreate on any plug in or even hardware compressor. Also, another trick was using home-consumer dBx compressors that had link connections to join to other similar units, and that acted as what later came to be known as a sidechain. Running a breakbeat though them with an 808-style kickdrum at the link input resulted in some pretty unique sounding dub drums, in particular if you slowed the break down after sampling. Listen to Meat Beat Manifesto's remixes of the Shamen to see where I'm coming from. As for the cheapo cassette trick, highly likely Selected Ambient Works vol 1 by Aphex Twin mastered on shitty home cassette deck. That one I'm guessing, but 1st track has giveaway 'chewed tape' sound on 1st track and sounds very similar, as everything is lovely, warm and nicely compressed. Loving this series, Moritz. :)
Wrong. This sounds like *shit* and is the number one reason compressors should be eliminated today. They're not used correctly anymore, they're used as some sort of "effect". A compressor doing its job *correctly* has *no noticable effect* on the audio being processed. If you can *hear* it working, you're *doing it wrong* Period.
These are by far the best audio circuit explanations I have ever sen on the internet and I have been working in audio circuits for ten years, mostly self taught through the internet.
This is a very nice design but I will throw a few bricks are it for perhaps a Version-2.0 A few points based on my having made compression circuits in the very distant past. 1) The string of diodes making a current controlled attenuator can also be done with transistors. The advantages of transistors is that you can get them as semi matched parts in an array. 2) Your rectifier will have trouble with high frequency signals because it takes a bit of time for the opamp to recover from being jammed against the rail. There are rectifier circuits that avoid this. 3) Your VU meter is OK so long as temperatures are normal and transistors are close to matched. You are using Vbe of the transistor as a threshold. A clever trick is to use the forward voltage of the pilot light LED as a reference to compare with. If you use the right LED, you can match the temperature variation of the Vbe with the forward drop of the LED. 4) The "I am compressing" indicator could be better. If you used an emitter resistor on that part, you could get LED current that is more linear against voltage once you start having any voltage.
thanks for these! couple thoughts: 1) i like the simplicity of using diodes, also because they're so easy to get anywhere. are transistor array ICs widely available in your experience? 2) true, this is an issue. i think the compressor is biased towards low frequency signals on top of that. so i guess we should at least add a high pass before the rectifier. 3) could you give me simulator link for this one? not sure i follow! 4) sounds reasonable, but do we really want the LED to light up linearly? since we're indicating gain reduction, i.e. volume?
@@MoritzKlein0 1) digikey carries a selection 2) Other rectifier topologies are available that don't demand so much from the op-amp. 3) I don't have a link handy but maybe later 4) Point taken
It's amazing how much information that single threshold LED can convey. Glad Erica Synths talked you into adding that VU meter though... not only are additional LEDs awesome, it might be helpful for someone figuring compressors out (who doesn't immediately leap to TH-cam). The interaction of the threshold LED and the VU meter, coupled with their ears, should be everything they need to know to use this module. 👍
One possible noise source could be the diode in the full wave rectifier. I build this rectifier myself and in my tests I found that the 1N4148 introduced quite some noise into the positive part of the rectfier. Using a 1N4007 instead worked better. Altough I can't explain why, so it might be only a problem in my setup.
yeah, when i listened to the rectifier's output directly, i also noticed some added high frequency noise/distortion. figured that it would get killed by the peak detector, but maybe using a different type of diode would make sense!
Another problem could be the missing clamp diode (which unfortunately cannot be used in Moritz' brilliant rectifier variation) which causes the OpAmp to go to the negative supply voltage during positive inputs. When the input goes negative again, the OpAmp output takes some time to travel back to the zero volt line which cuts into the signal. This produces distortions which could be perceived as noise. The 1N4007 diode has a somewhat lower forward voltage which might be beneficial here.
Its actually a great starting point to make it for studio application Things to change. 1.More detailed VU output meter scaled in dB 2.Gain reductuon VU meter, analog or also in led array (i'd preffer analog) so i know if its actually makin' reduction -3dB for example. 3. Input and output socked on the back (to connect with patchbay and mixing console) 4. Bypass switch. 5. Im not sure about SIDE in input.. if also on back, or in front.. it could be on back, and then additional switch to use it or not. So this input could be also in patchbay 6. two paralell supply sockets to easly stack up multiple of these compressors next to each other. Im tryin to design PCB for this application. Also i've changed almost all components for SMD, to make it more compact. Also making potentiometers off-board allows to make stereo version of this compressor. Just two of these boards and stereo pots.
Hello Moritz! I design a tape saturator with nonlinear amplitude dependence low pass filter. And I was looking for a simple peak detector with full rectification. Your idea is exactly what I need. Great solutions.
Get a bigger tip for your soldering iron… or better yet, get an iron that takes bonded heater-tip cartridges (such as “T-15” style cartridges, or a TS-80 or TS-100 iron) and use the “chisel” or “hoof” style tips. The small tips like you were using don’t have enough heat capacity to effectively heat up the joint. Also, consider adding extra flux. I generally use a paste flux (I’ve tried MG Chemical and Chip Quik with good results and I’m going to try some Amtech with the next batch of boards), and clean it off afterwards with 99% isopropyl alcohol (it’s a “no-clean” flux so it won’t corrode the parts, but it’s kinda stickey and gets on my fingers if I leave it there).
You make it really easy to understand not only circuits, but one of the most unintuitive effects for people new to audio! Looking real forward to ordering some of your modules soon. The sound is great but I would change a couple things about the display - first, the input LED ladder would probably be more useful if it were after the input gain stage, so it displays what the compressor stage is actually processing. Second, the compression amount indicator is really nice, but I think you could double up the LED meter circuit but upside down, and put that next to the input LEDs. That way, the volume reduction can be directly compared to the input volume.
By-pass switch could be a good thing to have quick refference -before/after-, but the VU meter should still work in both cases (bypass -> shows oryginal signal, on-> shows after compression)
Sir, I randomly discovered your videos yesterday. It's a shame I didnt find them earlier beacuse I designed an anolog compressor earlier this year and your content is quite insightful. I appreaciate the quality of your work. If I may suggest that you trigger the scope view to render static waves, it vould be interesting to appreciate visually the effects of the circuit as well as the distortion. Sincerely
Your videos blow my mind how great they are thank you so much for doing this videos. I understand that you are developing a new product here that later on you most likely going to offer as module to buy. Question or better said request could you please offer this compressor as a DIY kit? I would love to build dual band compressor with frequency splitter in my own enclosure. As bass guitar player it is nightmare to get even sound with different playing styles on one compressor this is why dual band is the answer . The bass will be crushed even and the highes will be more moderate compressed. Our maybe you can offer such PCBs?
This is epic! Would love to see something related to design of USB audio interfaces, it could really help bridge a gap designing audio equipment for modern hybrid studio setups...
So cool. You seem to be designing lots of effects lately, so why not make some more! I would absolutely love to see you design some other nice effects like a vocoder, delay, phaser, etc. P.S. Thank you SO much for your videos, they helped me a lot in understanding basic audio circuit design, and the mixer video really helped me make a spring reverb module. Thank you so much.
6:18 for the software compressor i made, i used an approximate hilbert transform to get two signals that should always be 90 degrees apart, then gets the magnitude of that and lowpasses it to remove some noise. i’m not sure it actually makes a big difference in getting a more accurate volume envelope but idk i like doing things differently and it’s not that complex.
the tldr is it’s like 8 total allpasses, and while their phases become very not flat, they have a pretty consistent 90 degree separation. it should be flat or mostly flat for a pure sine wave, since the 90 degree shifted version fills out the space in between
This is not meant as fault finding... Just to make sure that things went the way you intended. IIUC, you said the snappy switch was to take the pot out of line. However, the way you added the switch in before the pot makes the pot 100% in line. Meaning, its full 100K resistance is at work.
Im not a user of Euro-rack stuff, but wouldn't a bypass switch be nice to have here? So one can check what the compressor is actually doing to the signal etc.:)
I have a fringe idea...could you make a tube based, even harmonics variant in a new video? I like your compressor, a lot as is by the way. I am used to running a really hot "line level" signal through several different preamps at about 3v. My vintage Carver amp loves hot signals with some modifications. My 3rd preamp pretty much has the VU totally into the red with vintage Tungsol tubes 😅 For modern music, I ditch 1 preamp from the chain as it is so compressed already. I really think even order distortion, used carefully, sounds beautiful. Clipping...uhhhh not so much. Can you get that kind of sound quality from op-amps? If so, sign me up. BTW I am a sound person over an engineer, but I have made a few circuit boards over the years.
I got one more question. I just got a new headphone amp for my computer. I run an Audigy RX, into a DAC into a Geshelli Labs amp. GESHELLI LABS ARCHEL 2.5XL HEADPHONE AMP AND J2 DAC SPECIFICATIONS Archel 2.5XL DESIGN: 2-channel, solid-state headphone amplifier. POWER OUTPUT: 2 Watts Per Channel @ 32ohms. THD + N: < .000095% THD+N @ 2Vrms. SNR: > 124db SNR @ 2Vrms. INPUTS: 1 pair RCA, 1 pair XLR OUTPUTS: 1 pair RCA pre-out, Single 1/4” phono output. GAIN: 2 levels (Unity or 6x) MSRP: $269.98 (with beechwood case) OPTIONS: Several wood and aluminum color options for casework and plexiglass colors for front and back. DIMENSIONS: 4-7/8” Deep x 6-3/4” Wide x 3” Tall (Wood case, as tested.) It sounds like liquid sex even though it is solid state. 😲 I use it to drive Sony 7520s. Holy heck can it go loud.
This is awesome, you're awesome ! Will soon support you on Patreon for closer inspection of those sweet sweet designs ! I love modular but I don't like the teeny tiny knobs so I plan to adapt your designs for big buttons and big knobs (something like Dreadbox FX pedals). You have any opinion on that (downsides or other) ?
@@MoritzKlein0 Oh interesting! So are you just running the output from the compressor into an interface? How are you capturing other signals to visualize, like the peak detector line on the graph you show at th-cam.com/video/Wag-yTyAxPA/w-d-xo.html ?
Can i use this design for electric guitar use? I wanna diy a compressor but almost all guitar circuits use obsolete OTAs. Im liking this approach and how clearly you explained each part.
Ordered! You got me hooked to home soldering Moritz! The question now is should I replace the previous VCA with the new one or order another empty rack to start a new synth ? ;)
Again a very good video as usual. Small question if i may ask. Where should i connect the LED VU meter circuit? The same opamp output as the audio output or another point in the circuit? That only was not clear to me. Thanks in advance for your help. Kind regards
the erica synths manual for the kit should have the complete schematic: www.ericasynths.lv/media/COMPRESSOR_MANUAL.pdf but yeah, should’ve included that in the video. my bad!
Hi Moritz, My compressor is working like it should be. Thanks. I only have one small question about the Threshold knob. Its a 10K log. potentiometer connected to +12v via a 10k resistor. I notice that this knob only does work when almost fully clockwise. lets say the last 10% of its range Did you have this also? Any advice on how to adjust this? Thanks again and keep up making these great video's. Kind regards Maarten Louer
assuming that the guitar's signal peaks at 200 mV (just got that from a quick google search), you'd want to remove the voltage divider from the signal input completely, and bump the output gain by a factor of 2. like this: tinyurl.com/262pqnhf
@@MoritzKlein0 I tried this but I'm getting a lot of distortion whenever the compressor kicks in, any tips? I did have to add a voltage divider at the output to reduce the signal back to the 200mV range
This is awesome. Can this be translated to a 19” rack unit format and stereo? I’ve been interested on a DIY stereo compressor unit. Any thoughts or recommendations?
regarding the VU meter, could you add more LEDs to it by simply repeating the LED driver part more times? I was thinking of adding 3 and using different resistors between the current 47/120/220/330k ones.
Would substituting 4562 op-amp and/or a 4564 op-amp aid in reducing output noise? I assume the choices of which would depend on which packages handle the signal path. I also suspect that if that type of op-amp were used, 100pf ceramic caps should be placed between the supply rails next to the op-amps and ground.
I have a feeling - you can also calculate it of course - that quite some noise is added by the resistors. Their values have been increased greatly and that might be the problem. But of course, opamp noise is also a thing.
maybe. i'm not too sure the TL07Xs are adding that much noise themselves. probably more effective to use smaller resistors (as @krakamak suggested), or add some filtering to the rectifier!
@@MoritzKlein0 Thanks for posting this video. I find it very interesting and educational, and I plan to rewatch it and the "part 1" video, and to watch a number of your older videos for the first time. I admit my experience is somewhat limited, but my subjective assessment of LM4562 op-amp performance is that it is consistent with the superior published specs, and LM833 and TL07x op-amps are noisier and sound more muddy.
@@MoritzKlein0 I agree, the resistors add a significant part of the noise. Reduce the resistors for non-critical parts to something like 1k to 10k. For example, the two noninverting amps at the output! As well as the rectifier and the inverting amp. And, of course, reducing the gain too much and amplifying it afterwards isn't ideal too. I think you should measure first how much distortion your compressor has, and decide based on that how much reduction is needed. I think less than 0.1% THD is still quite good.
Question from an absolutely inexperienced electronics student: how hard would it be to make this sidechainable? I'd think it would be a switch to switch the input to the rectifier from the signal input as it is now to a new sidechain input, but would it actually be that simple?
yup, check my original compressor video (th-cam.com/video/Wag-yTyAxPA/w-d-xo.htmlsi=HaUjOQuUGZm7vjWf) - the last part actually talks about side chaining!
I don't care about electronic music at all but I love electronics - which is why I found this video fascinating. I just saw a video about optic compressor. It looked really simple but didn't have attack or sustain dials. Wonder if it could be modified to include those. And maybe some day I'll build my own compressor based on this (and the former one) video for my guitar setup.
Great video. One thing strikes me, and it’s not related to electronics. You are saying “anyways” when you should be saying “anyway” or some other phrase with the same general thrust. “Anyways” is an error little boys make all the time, which means it is commonplace on Discord and other SM sites with lots of gamer activity. Common errors are still errors. Just saying. I love your work.
The work you are doing here is immeasurably important. I have NEVER felt like I understood audio electronics (even though I'm an audio DSP programmer) prior to your videos. If we want to have a future full of people who are continuing to build audio electronics, we need great teachers like you to make it accessible. Thank you so much.
Audio stuff is a bit strange because unlike the instrumentation sort of designs, so distortions are perfectly fine and others are a disaster. To make it even harder, I don't think there is any simple rule to tell one from the other. A little soft limiting sounds fine. Sometimes hard limiting is really really good. Other times hard limiting is really really awful.
This makes for a lot of "listen and see" sort of tests.
That's the problem with audio today. People manipulating it that have no idea what the hell any of it actually means.
@@MadScientist267 To be clear - my initial comment is not about people manipulating audio and not understanding it. It's about me, an audio DSP programmer, understanding very well how to manipulate audio in the digital domain, but not knowing how to achieve similar results in the analog electronics domain.
@@erich1394 I think you mean the analog and/or hardware domain not the "electronics domain." If you're an audio dsp programmer then you most likely work almost exclusively in the electronics domain.
@@davidsotomayor8713 Nope, I meant exactly what I said. I currently work in the digital domain and I'm interested in the analog electronics domain. DSP stands for Digital Signal Processing. Analog Electronics are literally "Electronics that are Analogous" in the case of audio, they are "Electronics that carry and process electrical signals, where those signals are analogous to audio signals" Hope that clarifies things.
Really appreciate the updates to the design here Mortiz. Grateful for you listening to the feedback.
Also, I can’t understate how useful the animations are, (particularly for the LED VU meter section).
glad to hear!
In the days of b&w analog TV, the subcircuit at 17:20 ('ac-couple-dc-restore') was known as a 'black level clamp' - it derived a DC reference for luminence from an unreferenced waveform. I have also seen it used to fix-up a poorly designed signalling system where lengthy PSU wires - used also as a signal reference - experienced significant voltage drop. The signal ground would rise-up as a result, offsetting the perceived signal level. AC-couple-DC-restore saved the day!
Moritz I am watching your videos only because your animations are pure genious. I love the pulsating LED while emitting small rays of light. So funny. I think if I was not already into electronics I would start right now after watching this video.
This! I have been researching forever but I've never picked up the soldering iron. Every time I watch one of Mortiz's videos I feel closer to actually doing it. I'm sure that when I have time I will build this compressor (or a derivation of it) as my first project.
Just to let everyone know that some of the best sounding mixes from the early 90's mastered on to cassette tape sounded amazing. But not because it was cassette tape, but because the cheap cassette recorders being used to do the mastering (DAT back then was only for the super wealthy), had little compressor chips inside that automatically kept the levels in check to prevent distortion from accidental overload and to raise the volume of the quiet bits. This lead to a super-lovely sound that I've never been able to recreate on any plug in or even hardware compressor. Also, another trick was using home-consumer dBx compressors that had link connections to join to other similar units, and that acted as what later came to be known as a sidechain. Running a breakbeat though them with an 808-style kickdrum at the link input resulted in some pretty unique sounding dub drums, in particular if you slowed the break down after sampling. Listen to Meat Beat Manifesto's remixes of the Shamen to see where I'm coming from. As for the cheapo cassette trick, highly likely Selected Ambient Works vol 1 by Aphex Twin mastered on shitty home cassette deck. That one I'm guessing, but 1st track has giveaway 'chewed tape' sound on 1st track and sounds very similar, as everything is lovely, warm and nicely compressed. Loving this series, Moritz. :)
Wrong. This sounds like *shit* and is the number one reason compressors should be eliminated today. They're not used correctly anymore, they're used as some sort of "effect".
A compressor doing its job *correctly* has *no noticable effect* on the audio being processed. If you can *hear* it working, you're *doing it wrong*
Period.
These are by far the best audio circuit explanations I have ever sen on the internet and I have been working in audio circuits for ten years, mostly self taught through the internet.
Awesome. I’ll take a stab at this one over the long holiday weekend.
Thanks again Mr Klein. You’re one of the diy heroes
Nice design! You have a real knack for walking through a circuit in a very understandable way. Thank you!
Thanks, Moritz, for embracing all the suggestions and coming up with a really useful and flexible design.
This is a very nice design but I will throw a few bricks are it for perhaps a Version-2.0
A few points based on my having made compression circuits in the very distant past.
1) The string of diodes making a current controlled attenuator can also be done with transistors. The advantages of transistors is that you can get them as semi matched parts in an array.
2) Your rectifier will have trouble with high frequency signals because it takes a bit of time for the opamp to recover from being jammed against the rail. There are rectifier circuits that avoid this.
3) Your VU meter is OK so long as temperatures are normal and transistors are close to matched. You are using Vbe of the transistor as a threshold. A clever trick is to use the forward voltage of the pilot light LED as a reference to compare with. If you use the right LED, you can match the temperature variation of the Vbe with the forward drop of the LED.
4) The "I am compressing" indicator could be better. If you used an emitter resistor on that part, you could get LED current that is more linear against voltage once you start having any voltage.
thanks for these! couple thoughts:
1) i like the simplicity of using diodes, also because they're so easy to get anywhere. are transistor array ICs widely available in your experience?
2) true, this is an issue. i think the compressor is biased towards low frequency signals on top of that. so i guess we should at least add a high pass before the rectifier.
3) could you give me simulator link for this one? not sure i follow!
4) sounds reasonable, but do we really want the LED to light up linearly? since we're indicating gain reduction, i.e. volume?
@@MoritzKlein0 1) digikey carries a selection
2) Other rectifier topologies are available that don't demand so much from the op-amp.
3) I don't have a link handy but maybe later
4) Point taken
It's amazing how much information that single threshold LED can convey. Glad Erica Synths talked you into adding that VU meter though... not only are additional LEDs awesome, it might be helpful for someone figuring compressors out (who doesn't immediately leap to TH-cam). The interaction of the threshold LED and the VU meter, coupled with their ears, should be everything they need to know to use this module. 👍
This is the gold standard in DIY videos 😍🧙♂
Excellent topic and brilliant presentation as well.
nice! I recently designed a 4-stage wavefolder using consecutive full-bridge rectifiers
sounds great, got a falstad link?
Awesome, thanks alot for this video. I'm working on something completely different and i needed a simple full wave rectifier. You just provided that 😊
great, glad you got something out of it!
One possible noise source could be the diode in the full wave rectifier. I build this rectifier myself and in my tests I found that the 1N4148 introduced quite some noise into the positive part of the rectfier. Using a 1N4007 instead worked better. Altough I can't explain why, so it might be only a problem in my setup.
yeah, when i listened to the rectifier's output directly, i also noticed some added high frequency noise/distortion. figured that it would get killed by the peak detector, but maybe using a different type of diode would make sense!
Another problem could be the missing clamp diode (which unfortunately cannot be used in Moritz' brilliant rectifier variation) which causes the OpAmp to go to the negative supply voltage during positive inputs. When the input goes negative again, the OpAmp output takes some time to travel back to the zero volt line which cuts into the signal. This produces distortions which could be perceived as noise. The 1N4007 diode has a somewhat lower forward voltage which might be beneficial here.
@@ChristianMuenker If that is the case, a schottky diode might be even better.
Wow impressive. Thank you.
I have a similar design to this I've been toying around with for a few years. It's a really cool compressor imo. I need to try your version.
Really nice. What an amazing series. On the final module front plate, note that it's "threshold", not "treshold"
oh no, totally didn't catch that! will tell ES about that one. thanks!
Its actually a great starting point to make it for studio application
Things to change.
1.More detailed VU output meter scaled in dB
2.Gain reductuon VU meter, analog or also in led array (i'd preffer analog) so i know if its actually makin' reduction -3dB for example.
3. Input and output socked on the back (to connect with patchbay and mixing console)
4. Bypass switch.
5. Im not sure about SIDE in input.. if also on back, or in front.. it could be on back, and then additional switch to use it or not. So this input could be also in patchbay
6. two paralell supply sockets to easly stack up multiple of these compressors next to each other.
Im tryin to design PCB for this application. Also i've changed almost all components for SMD, to make it more compact.
Also making potentiometers off-board allows to make stereo version of this compressor. Just two of these boards and stereo pots.
Amazing video, love the clean hand drawn schematics and animations!
x2
Thanks for the great video!
such a great and simple module, thank you
Very cool. I love the animations. I wish I'd had them when I studied electronics so many years ago :)
Very informative fun & explanation for an old DIY guy like me
Thanks a lot 👍
Hello Moritz! I design a tape saturator with nonlinear amplitude dependence low pass filter. And I was looking for a simple peak detector with full rectification. Your idea is exactly what I need. Great solutions.
Instant like, watch later. Moritz, I love your show so much - you do a fantastic job here!! Danke tausendmal!
~
nicht dafür!
Brilliant extension of your previous video; and your style of explanation is simply wonderful! Thanks so much!
Get a bigger tip for your soldering iron… or better yet, get an iron that takes bonded heater-tip cartridges (such as “T-15” style cartridges, or a TS-80 or TS-100 iron) and use the “chisel” or “hoof” style tips. The small tips like you were using don’t have enough heat capacity to effectively heat up the joint.
Also, consider adding extra flux. I generally use a paste flux (I’ve tried MG Chemical and Chip Quik with good results and I’m going to try some Amtech with the next batch of boards), and clean it off afterwards with 99% isopropyl alcohol (it’s a “no-clean” flux so it won’t corrode the parts, but it’s kinda stickey and gets on my fingers if I leave it there).
sounds good, will give that a shot!
@17:08 It is called Clamper circuit. Vout = 2Vin -Vbe There is also an active from of clamper circuit using opamp that its output is 2Vin.
Bob Carver at Phase Linear used this type of VCA in their 4000 preamplifier back in the mid-1970s. Clever and worked fine I thought
You make it really easy to understand not only circuits, but one of the most unintuitive effects for people new to audio! Looking real forward to ordering some of your modules soon.
The sound is great but I would change a couple things about the display - first, the input LED ladder would probably be more useful if it were after the input gain stage, so it displays what the compressor stage is actually processing.
Second, the compression amount indicator is really nice, but I think you could double up the LED meter circuit but upside down, and put that next to the input LEDs. That way, the volume reduction can be directly compared to the input volume.
So underrated channel! Amazing work
Need to wait only few hours for prototype with cnc, even those cheap (
By-pass switch could be a good thing to have quick refference -before/after-, but the VU meter should still work in both cases (bypass -> shows oryginal signal, on-> shows after compression)
Great animations! so informative.
Sir, I randomly discovered your videos yesterday. It's a shame I didnt find them earlier beacuse I designed an anolog compressor earlier this year and your content is quite insightful. I appreaciate the quality of your work. If I may suggest that you trigger the scope view to render static waves, it vould be interesting to appreciate visually the effects of the circuit as well as the distortion. Sincerely
Thiabis so awesome, thankss for sharing it
Your videos blow my mind how great they are thank you so much for doing this videos.
I understand that you are developing a new product here that later on you most likely going to offer as module to buy. Question or better said request could you please offer this compressor as a DIY kit?
I would love to build dual band compressor with frequency splitter in my own enclosure.
As bass guitar player it is nightmare to get even sound with different playing styles on one compressor this is why dual band is the answer . The bass will be crushed even and the highes will be more moderate compressed.
Our maybe you can offer such PCBs?
Not sure I follow - what kind of DIY kit are you thinking?
This is epic! Would love to see something related to design of USB audio interfaces, it could really help bridge a gap designing audio equipment for modern hybrid studio setups...
Man, this is so cool! Really inspiring :)) Thank you!
happy to hear!
That's awesome!
FUUUULLLLL WAAAVEEEE RECTIFIIIIIIIIERRRR!!!
*eyebrows intensify*
is this a reference i'm not getting?
@@MoritzKlein0 search for electroboom in youtube… best electric engineer ever!
@@MoritzKlein0 Electroboom
Somebody needs a bit of agressive compression ...
So cool. You seem to be designing lots of effects lately, so why not make some more! I would absolutely love to see you design some other nice effects like a vocoder, delay, phaser, etc. P.S. Thank you SO much for your videos, they helped me a lot in understanding basic audio circuit design, and the mixer video really helped me make a spring reverb module. Thank you so much.
nicely done!
Best tutor on the planet!!!!
Gaaah this is so cool i love you bro
i love you too!
Would love to see a stereo spring reverb cirquit build :) great stuff btw keep it up!
6:18 for the software compressor i made, i used an approximate hilbert transform to get two signals that should always be 90 degrees apart, then gets the magnitude of that and lowpasses it to remove some noise. i’m not sure it actually makes a big difference in getting a more accurate volume envelope but idk i like doing things differently and it’s not that complex.
the tldr is it’s like 8 total allpasses, and while their phases become very not flat, they have a pretty consistent 90 degree separation. it should be flat or mostly flat for a pure sine wave, since the 90 degree shifted version fills out the space in between
This is not meant as fault finding... Just to make sure that things went the way you intended. IIUC, you said the snappy switch was to take the pot out of line. However, the way you added the switch in before the pot makes the pot 100% in line. Meaning, its full 100K resistance is at work.
Would be amazing if you shared a DIY preamp and EQ design as well! Then we could build our own mixers!
what would be cool is a voltage controlled gain. an amplifier that can boost the signal strength and reduce it depending on the control voltage.
Im not a user of Euro-rack stuff, but wouldn't a bypass switch be nice to have here? So one can check what the compressor is actually doing to the signal etc.:)
we originally planned to have a bypass switch in there, but it was cut in favor of the snappy attack switch. maybe in a future revision!
I have a fringe idea...could you make a tube based, even harmonics variant in a new video? I like your compressor, a lot as is by the way. I am used to running a really hot "line level" signal through several different preamps at about 3v. My vintage Carver amp loves hot signals with some modifications. My 3rd preamp pretty much has the VU totally into the red with vintage Tungsol tubes 😅 For modern music, I ditch 1 preamp from the chain as it is so compressed already. I really think even order distortion, used carefully, sounds beautiful. Clipping...uhhhh not so much. Can you get that kind of sound quality from op-amps? If so, sign me up.
BTW I am a sound person over an engineer, but I have made a few circuit boards over the years.
oh very good question. i have to do some research on that one and get back to you!
@@MoritzKlein0 ❤
I got one more question. I just got a new headphone amp for my computer. I run an Audigy RX, into a DAC into a Geshelli Labs amp.
GESHELLI LABS ARCHEL 2.5XL HEADPHONE AMP AND J2 DAC SPECIFICATIONS
Archel 2.5XL
DESIGN:
2-channel, solid-state headphone amplifier.
POWER OUTPUT:
2 Watts Per Channel @ 32ohms.
THD + N:
< .000095% THD+N @ 2Vrms.
SNR:
> 124db SNR @ 2Vrms.
INPUTS:
1 pair RCA, 1 pair XLR
OUTPUTS:
1 pair RCA pre-out, Single 1/4” phono output.
GAIN:
2 levels (Unity or 6x)
MSRP:
$269.98 (with beechwood case)
OPTIONS:
Several wood and aluminum color options for casework and plexiglass colors for front and back.
DIMENSIONS:
4-7/8” Deep x 6-3/4” Wide x 3” Tall (Wood case, as tested.)
It sounds like liquid sex even though it is solid state. 😲 I use it to drive Sony 7520s. Holy heck can it go loud.
Fantastic video. Well done...
@19:35 the reason you need more temperature is because either your tip is not clean or your solder is bad ( either old or not good quality solder)
This is awesome, you're awesome ! Will soon support you on Patreon for closer inspection of those sweet sweet designs !
I love modular but I don't like the teeny tiny knobs so I plan to adapt your designs for big buttons and big knobs (something like Dreadbox FX pedals). You have any opinion on that (downsides or other) ?
What do you use to animte the drawings
Compressor with 1 LED : 💵
Compressor with more LEDs💸💸💸💸💸💸💸💸
THANK U
Hey! Brilliant as always. Maybe a silly question, but what copringht license applies to your designs?
good question! as far as i'm concerned, my circuits are fully open source - so you can use them for anything you want. credit is appreciated, though!
@@MoritzKlein0 Thank you for your reply. Brilliant.
Any idea when the printed version of all the edu manuals will come out?
unfortunately not. i‘ll talk to erica again to see if we can speed up the process.
I thought the distortion was pretty pleasant sound!
How to further improve this whole circuit to deal with Stereo?
Hey Moritz, thanks so much for these videos! Could you tell me what oscilloscope you're using in these projects? Thanks!
i‘m using a free VST called Wave Observer!
@@MoritzKlein0 Oh interesting! So are you just running the output from the compressor into an interface? How are you capturing other signals to visualize, like the peak detector line on the graph you show at th-cam.com/video/Wag-yTyAxPA/w-d-xo.html ?
Can i use this design for electric guitar use? I wanna diy a compressor but almost all guitar circuits use obsolete OTAs. Im liking this approach and how clearly you explained each part.
Is there a way to view the full schematic? Fantastic video man thank you, very informative
if it had a by-pass switch, it will be perfectly pass into my home setup
let's see if we can work it in for a future revision!
Ordered! You got me hooked to home soldering Moritz! The question now is should I replace the previous VCA with the new one or order another empty rack to start a new synth ? ;)
Would be cool if LEDs were different colors; maybe green, yellow ranging to red at the high end.
Again a very good video as usual. Small question if i may ask. Where should i connect the LED VU meter circuit? The same opamp output as the audio output or another point in the circuit? That only was not clear to me. Thanks in advance for your help. Kind regards
the erica synths manual for the kit should have the complete schematic: www.ericasynths.lv/media/COMPRESSOR_MANUAL.pdf
but yeah, should’ve included that in the video. my bad!
Hi Moritz. Thanks😊
Hi Moritz, My compressor is working like it should be. Thanks. I only have one small question about the Threshold knob. Its a 10K log. potentiometer connected to +12v via a 10k resistor. I notice that this knob only does work when almost fully clockwise. lets say the last 10% of its range Did you have this also? Any advice on how to adjust this? Thanks again and keep up making these great video's. Kind regards Maarten Louer
I would be like to see an upgrade to VCO
Lin FM, soft Sync, hard Sync,
What would you change if input is instrument level (guitar)
assuming that the guitar's signal peaks at 200 mV (just got that from a quick google search), you'd want to remove the voltage divider from the signal input completely, and bump the output gain by a factor of 2. like this: tinyurl.com/262pqnhf
@@MoritzKlein0 I tried this but I'm getting a lot of distortion whenever the compressor kicks in, any tips? I did have to add a voltage divider at the output to reduce the signal back to the 200mV range
This is awesome. Can this be translated to a 19” rack unit format and stereo? I’ve been interested on a DIY stereo compressor unit. Any thoughts or recommendations?
How might you generate animated videos about circuitry?
Can you do a video on a snare or hihat module?
hihat/cymbal is next on my list!
regarding the VU meter, could you add more LEDs to it by simply repeating the LED driver part more times? I was thinking of adding 3 and using different resistors between the current 47/120/220/330k ones.
Would substituting 4562 op-amp and/or a 4564 op-amp aid in reducing output noise? I assume the choices of which would depend on which packages handle the signal path. I also suspect that if that type of op-amp were used, 100pf ceramic caps should be placed between the supply rails next to the op-amps and ground.
I have a feeling - you can also calculate it of course - that quite some noise is added by the resistors. Their values have been increased greatly and that might be the problem. But of course, opamp noise is also a thing.
maybe. i'm not too sure the TL07Xs are adding that much noise themselves. probably more effective to use smaller resistors (as @krakamak suggested), or add some filtering to the rectifier!
@@MoritzKlein0 Thanks for posting this video. I find it very interesting and educational, and I plan to rewatch it and the "part 1" video, and to watch a number of your older videos for the first time.
I admit my experience is somewhat limited, but my subjective assessment of LM4562 op-amp performance is that it is consistent with the superior published specs, and LM833 and TL07x op-amps are noisier and sound more muddy.
@@MoritzKlein0 I agree, the resistors add a significant part of the noise. Reduce the resistors for non-critical parts to something like 1k to 10k. For example, the two noninverting amps at the output! As well as the rectifier and the inverting amp.
And, of course, reducing the gain too much and amplifying it afterwards isn't ideal too.
I think you should measure first how much distortion your compressor has, and decide based on that how much reduction is needed. I think less than 0.1% THD is still quite good.
Can we please get the Eagle, KiCad, EasyEDA, or what else you use -Files to download?
❤
Would this work as Automatic Volume control, for say youtube or TV channels?
Just what I was looking for! Is it possible to replace the TL series ICs with NE5532? I'm gonna start building this.
yeah, that should work just fine!
@@MoritzKlein0 Thanks 🙏. Is the circuit diagram for the final version available?
@@anandha12 yup, here: www.ericasynths.lv/media/COMPRESSOR_MANUAL.pdf
@@MoritzKlein0 Wow, thanks 👍.
Question from an absolutely inexperienced electronics student: how hard would it be to make this sidechainable? I'd think it would be a switch to switch the input to the rectifier from the signal input as it is now to a new sidechain input, but would it actually be that simple?
yup, check my original compressor video (th-cam.com/video/Wag-yTyAxPA/w-d-xo.htmlsi=HaUjOQuUGZm7vjWf) - the last part actually talks about side chaining!
@@MoritzKlein0 sure will, thanks! Great stuff
Couldn't you go with a row of led's for gain reduction?
Awesome. But isn't it spelled Threshold? Or is that some kind of modular-synth-insider gag?
I don't care about electronic music at all but I love electronics - which is why I found this video fascinating. I just saw a video about optic compressor. It looked really simple but didn't have attack or sustain dials. Wonder if it could be modified to include those. And maybe some day I'll build my own compressor based on this (and the former one) video for my guitar setup.
what Voltage does this run at? is it 9 or 12?
???...?!?!? The OpAmps I am using are speced for 1A o.0 The messily 20mA of an LED are nothing for them...??
i thought the schematic had 2 switches
Video Idea: 20w Class AB Amplifier
oh my god I am so glad I dont have an LED symbol tattood on my hand
Who needs analog audio compressor nowadays
Great video. One thing strikes me, and it’s not related to electronics. You are saying “anyways” when you should be saying “anyway” or some other phrase with the same general thrust. “Anyways” is an error little boys make all the time, which means it is commonplace on Discord and other SM sites with lots of gamer activity. Common errors are still errors. Just saying. I love your work.
Why the hell would you bother? This project would just be abused like every other compressor out there.