EEVblog
ฝัง
- เผยแพร่เมื่อ 12 ธ.ค. 2018
- A look at the dual bipolar transistor zener clamp circuit, and it's usefulness as a fast low leakage overload clamp.
The first part is lifted from video #1000 with 12 minutes of extra commentary.
Interview with Dave Taylor, designer of the famous Fluke 8060A multimeter:
theamphour.com/180-an-intervi...
Forum: www.eevblog.com/forum/blog/eev...
EEVblog Main Web Site: www.eevblog.com
The 2nd EEVblog Channel: / eevblog2
Support the EEVblog through Patreon!
/ eevblog
AliExpress Affiliate: s.click.aliexpress.com/e/c2LRpe8g
Buy anything through that link and Dave gets a commission at no cost to you.
Stuff I recommend:
kit.com/EEVblog/
Donate With Bitcoin & Other Crypto Currencies!
www.eevblog.com/crypto-currency/
T-Shirts: teespring.com/stores/eevblog
Likecoin - Coins for Likes: likecoin.pro/@eevblog/dil9/hcq3 - วิทยาศาสตร์และเทคโนโลยี
This got to be one of your best videos ever. I love the clarity of the explanation and the speed was just right. Awesome job Dave!
FINALLY, ELECTRONICS FROM AN ELECTRONICS CHANNEL
Or maybe I've just been watching too many debunking videos :(
That was a great addition to the original video! Thanks Dave!
Fantastic video, Dave! I needed the refresher on clamping circuits for work. Thank you a bunch!
Note: 2SC3326 is not a jellybean transistor. It is for MUTING and switching application. It is quite special with the fact that it has very high reverse current gain (they even tell you in the beginning of the datasheet). These transistors are quite rare. The high Vebo is coupled with the high reverse hFE due to the transistor construction: It's BE and BC junctions are much more symmetrical, unlike normal transistor, where you need very high gains and don't care about the reverse one. These transistors can be used to switch AC currents without introducing much distortion into the signal, where larger voltage swings are present.
Luckily the pricing is similar to jellybean parts. And they are available, current availability with Digikey is 117788!
Jellybean means it is widely available and low cost. Not that it's a boring or shit part. Cool info though
Sixta16 Thanks for the info about switching AC currents with this npn trasnsitor.... i was like wtf... - interesting, - have you got any advise for my H-Bridge. . im having issues switching over 430~ volts (under isnt too bad) , because the current is leaking through my darlington pair, c3998 transistors. , and biasing the base, despite the resistor too ground. ... i have solved it by using a opto isolator and 9volt batteries, instead of taking the base current from a darlington. - but its abit micky mouse. - any other solutions too switching on h bridge transistors ?. - the bottom ones particularly /// great video steve !
Thanks Dave. Its really good to learn something new about that humble BJT. I'm looking forward to designing it into a product.
Hey thank you so much for this channel; especially Fundamental Fridays! After years of being a professional programmer, I've started down the path of E.E! *Most* tutorials on youtube never go into enough depth & are always hum-drum (BORING) on a subject. After going through your series, even though sometimes years old, it seems timeless. Schmitt Trigger, Zener Diodes, Transistors, Opamps, Overload detectors etc. Not only informational, but fun at the same time. Here's a how ya doin' from Canada! Cheers
I first saw it on my crappy Vichy 99 DMM circuit. I couldn't understand it until I finally viewed your original EEVBlog vid many years ago.
Great Ep! Keep the circuit design vids flowin'!
Quite alright reusing that video clip. I didn't even remember it, so this was like seeing it for the first time, plus I always love fundamentals like that!
Cool stuff! Transistors are definitely full of surprises.
Thanks for useful stuff! This alternative to zener i never seen before.
I've never seen this before. Excellent!
Magnifficient! I wish I knew that trick when I was in desperate need for a really low leakage protection.
I've been repairing stuff since 1983 but my knowledge of how stuff really works is weak at best. I barely know how to use an O'scope. You're explanations of the engineering side of electronics are top notch. Maybe one day I'll understand what I'm repairing, lol.
Man, this is good!! Thank you !!!
This time last year we were counting Dave to 500K subscribers. He's travelling exponentially! Good job, Dave.
Linear actually!
@@EEVblog Well...it's good its at least pointing upwards. While I have you, I found your 2004 PCB Design Tutorial document years ago when researching how to make PCBs. I had it before your global fame and didn't know who you were. The other day I pulled it up and saw "Dave L. Jones" on the cover and it finally clicked - that's Dave! If you haven't featured it in a video in a while, it would be good to point it out for new viewers. It's one stop shopping for PCB mojo.
"Naughty" transistor circuits can be very useful. I've made a nanosecond pulse generator once. (For transmission line length / velocity factor / integrity measurements.)
Tamás Fábián Could you elaborate, how it was done using this circuit?
@@sergeyivanov3453 look up Jim Williams pulse circuit.
The reverse biased base emitter junction is one way of generating noise.
This is actually a very very very true. Having a super low leakage current is great, but if you get a ton of noise back for it, your whole circuit might be unusable. Something that definitely had the be in addressed as well.
I guess not an issue if you're measuring some kind of high resistance and can average for years, but for circuits that have to perform at (almost) real-time that's a totally different story.
@@merlingallagher4484 I was wondering whether this could be used as a V Ref. The voltage looked pretty stable in the video. How bad d'you both think the noise will be compared to a buried Zener?
@@ChristianRThomas It is stable for this particular device under test, but as it is a loosely defined parameter, it may change a lot between devices and over temperature. If you are designing for production, it may be good until the manufacturer's batch changes.
@@ChristianRThomas . Good question, that's why it's very interesting to investigate.
@@andrewwhite1793 Yes, of course. It looks like it could be very device dependent, though I'd need to think a bit harder to get the theory to map onto this application. As a V Ref, I'd have a trim, so changing voltage doesn't especially matter to me. (Temperature might!). What is striking here is that the voltage is so much higher than what lies at the heart of a V Ref (so you don't have to amplify the noise as much) and that, in normal use, transistor noise levels are tiny by comparison to a voltage reference. Unless you have had reason to use them you'd be horrified by the noise levels. 10s or even > 100uV. Even expensive ones from people like Linear aren't great.
Very informative ,Thank you 👍👍👍👍
I learned about this when I was in high school (~1972)
It kills the DC current gain of the transistors due to McDonald effect, but works fine.
What is 'McDonald effect' ?
@@FarleyHillBilly I think he is referring to not ordering any food because you were not hungry but as soon as you see the food you ordered for someone else you become hungry. I think the analogy is that it wants to block all current until it triggers the breakdown then rather than just letting it buy at the higher voltage it drops to virtually no resistance as the base sees the voltage. Basically block everything then when you see it (voltage at the base) let it though without delay. Or its not hungry for voltage until it (the base) sees voltage.
To me its like the opposite of a high frequency switch mode power supply rather than supplying voltage it limits it by turning on and off very very fast. Someone will correct me if I'm wrong.
@@FarleyHillBilly It is the reduction of current gain by the formation of a surface leakage path between the forward biased base and emitter. The reverse biased base-emitter junction produces hot carrier that occupy traps at the Si-SiO2 interface. The trapped carriers then create a surface leakage path.
@@divewizard Thanks , I've often wondered why.
I have seen this only in one design (as far as I can remember), someone used it for protection on the HF input of an DIY frequency counter. The biggest advantage is very low off capacitance (compared to a zener).
I think it's quite common, in things like log-amps made on silicon, to make diodes this way (albeit not as a clamp circuit!) by tieing the base to the collector, since making transistors on silicon is easy. But that's a very cool circuit and I have no idea how you figure these things out, but thank you for sharing!
@@aphenioxPDWtechnology thanks for the reply, it's great to know what the actual name is called!
Horizontal output transistors such as the BU2525 we drove the base into reverse zener region to reduce the storage time to as little as possible.
brilliant vid
OUTSTANDING
We use a similar circuit in guitar amps to clip the signal for distortion. Usually mosfets, but sometimes bjts.
Oh yep hahahaha
What characteristic of a MOSFET do you use for clipping a guitar signal? Breakdown of the gate/source is usually a one way ticket to a dead transistor.
For how it's positioned it might be just the channel that pinches and let's some current pass to the source and reach the gate through there, if it works this way I doubt it has the same "high speed" characteristic of the BJT one
How fast does your circuit react, Dave? Speed related to ft ?
thank you,.. super lectures.
Very nice. I had totally forgotten about this, but I used to use reverse-biased B-E junctions as voltage references in regulated power supplies 40 years ago. How about white-board episodes covering other "lost arts"? Two suggestions: simple BJT transistor oscillators, and running simple CMOS logic gates as linear amplifiers (RCA used to publish apps information on that).
Not hidden but in the same vane..., look up the Tayloe (yes with an e) quadrature demodulator/mixer.
Pretty cool.
You can easily get
Эпитаксиальные транзисторы можно использовать как стабилитрон при шунтировании затвор-истока у мосфета. Как раз переход база-эммитер. Из + это маленькая емкость
For muting applications is NOT a jelly bean application. The high reverse hFE is useful for that, and that probably correlates in some way to the high Vebo. Philips used to use power transistors for muting small signals, presumably because of (undocumented) similar characteristics.
Ah, the HP 1740A, my current scope.
It's quite good, but i keep my fingers crossed every Keysight scope month for a modern one
Great video
Isn't it nice? I have its bigger brother, the 1741A, which has a storage tube. I use that scope much more often than my digital one...
4:25 Cool HP oscilloscope, still quite capable and low noise.
Hi Dave, I tried this out with a bunch of transistors from my jellybean box, (noname BC547,BC548,BC549,BC550,2N2222,BC557) however I couldn't get anything less than 10V Vebo. Most are around 10.8V am I doing something wrong? I used a 2K series resistor and 12.3V input.
How fast is this? Would it be useful against ESD or is it too slow?
I'll look into this more, but I think it might be a neat thing to use on my clunky prototypes... might save that moment where I bump the +12v onto the I2C bus. (Not that I ever did that while sailing of course!). Plus, for a couple cents it might save the circuit from some odd pulse.
Hi, Dave!
That's old as the world. I saw such clamping circuits in very old soviet industrial equipment.
Well, if you dig into untraditional usage of common transistors, you must tell people about other uncommon usages!
Expamples are: BJT or MOSFET as varicap, BJTs or JFETs as tunnel or backward diode, BJTs as unijunction or avalanche transistor, BJTs as diac or triac, BJT as noise source... there are more of them, actually. I saw a lot of such circuits in some old soviet/russian books.
I guess it might be interesting for students and some hobbysts.
Electro FAN Isn't this circuit in fact two BJTs acting as a DIAC ?
@@johnfrancisdoe1563 diac: radiostorage.net/uploads/Image/schemes/misc/schemexis-26.png
triac: radiostorage.net/uploads/Image/schemes/misc/schemexis-27.png
Don't forget cheap BJT (BC junction) as tunnel diode
@@johnfrancisdoe1563 No. A DIAC is like a TRIAC - once it turns on, it stays on until the current is reduced below a certain value, regardless of the voltage.
Dave, can you quote approximate collector current in the upper transistor above and below the "zener knee"?
LOVE the VERY Aussie T-shirt...😂😂😂👍🤘
Tl431 fundamentals Friday!
can i use this clampling to embed leds in there and protect them from overcurrent ?
zing! fantastic.
Would a "Capacitance Multiplier" be an input protection (reverse polarity)?
Ran across this looking up what this transistor configuration does after seeing it in a schematic from early 1960's.
The testing,.... do i hear a study intern student! (Dave3.0)
beautiful
Just a stupid question. What will happen if you do not connect the Collectors. Just the Base together and Emitters as you used it? Will it also be very noisy?
Paul Badenhorst : This is actually what the O is for in the spec, Vebo is with the collector open circuit.
Hello Dave, at the of the text at 16:56 that write's "you can find a device that even the ??".. can you tell us what he write (or this text there is anyware uploaded?).
So it is basically a bi-directional transient voltage suppressor (TVS) diode , isn't it? I can understand why you would want this sort of circuit if you would like to reuse existing BJTs, which is definitely a big deal for mass manufacturer like fluke. But other than that, wouldn't it be more suitable to look for an appropriate TVS diode? They are especially designed and screened for transient response, compared to whatever information you have to fuse together from BJT datasheet.
Price?
No advertisement with this video.
I'm guessing that Vbeo varies widely with temperature?
Varies with temperature and time, but is a lot more stable and repeatable than Vbco, as that has both a very large wafer to wafer variation as well as from manufacturer to manufacturer as well. might be given as 100v in the datasheet, but in reality can vary from 150V to 400V from different wafers and suppliers, and also is temperature sensitive. Better to use the much tighter controlled diffusions of base and emitter, as those are the gain setting ones, plus the leakage current in the buried junctions are a lot lower as well. If you need higher simply stack them or use a bias supply in addition, though you then get an offset in clamping.
Hi, very useful information, what is the price for your differential probe, I live in Pakistan.
The reverse breakdown voltage of B-E junction can be interesting in germanium transistors. For some it goes as high as 30V, while for most it stays at ~10V.
Germanium junctions go into negative resistance if you pulse enough current thru them.
Germanium junctions are basically black magic. There is very little info on them on the internet.
So could this be scaled up with larger transistors to make a very high energy TVSS device? Can they be put in series to increase clamping voltage? I would like to build my own protection devices for solar energy systems. At my company we see solar components destroyed all the time by nearby lighting strikes. There are MOV based products available but they can be expensive and have very high clamping voltages.
This circuit is only useful for lower power clamping. Once you get into power electronics, purpose built devices will perform better. The BE junction does not scale very well for power.
Just starting my journey into electronics, but how does a varistor compare to this solution ? Varistors are available in many flavors and clamping voltages.
FPV Enthusiast : varistors are specified for this type of operation, they can be used to clamp voltages etc. Transistors have a quirky spec that does this clamping under certain conditions but aren’t specified particularly well for this function. For example , what is the effect of temperature, where is the voltage current curve, batch to batch tolerance of these properties, maximum power capability etc. This is an interesting spec but the fact that fluke uses it is because they need one specific property, the low leakage. MOV’s and zeners have higher leakage and this would affect the voltage readings poorly in a voltmeter. This is a use at your own peril design.
What was the advantage of two transistors over two zeners? Speed? Resistance after breakdown?
If he went over it in the video I apologize, I must have missed it.
In reverse order, yes, and yes, and also, bipolar transistors are much cheaper than zener diodes.
And leakage current. Were you high? :)
In fact you allready need an enhanced Vebo if you want to operate the standard astable multivibrator design at voltages above 5 V.
Can they be connected in series to give higher clamping voltage?
If 2 pairs are connected in series, would they clamp over 6.6Vx2=~13.2V ?
Yes, that works. He also mentioned that in the video, that Fluke uses them serial for higher voltage application or parallel for higher power dissipation application.
@@benjaminfacouchere2395 Thanks for the reply.
Used inside chips for years
Very clever! But reverse biasing the junction isn't supposed to be harmful to the device?
It ruins Hfe for some strange reason.
I wish this was properly modeled in LTSpice, I would like to be able to simulate and quantify this.
Falstad's online circuit simulator is enough for me to understand the concept. 😁
Thanks for the lecture, if you have tutorials how to make analog transmitter and receiver pls 🙏 teach us.
The fluke clamp! I'm just a dumb guy trying to learn something, but could this type thing be used in solar lightening arrestors? I don't know why they call it that, I haven't seen much that's can save you from lightening! I have always figured it protects from the static build up before of after near by or near strikes,
I want that T-shirt :D
why not just put two clamps in series to achieve higher breakdown voltage, rather than using a single clamp with 'special' transistors?
maybe because the actual V_EBO voltage varies quite a bit for each transistor? idk.
Because two clamps in series = 12V, which is not 25V. You'd need four clamps in series which is getting a bit ridiculous.
@@jaro6985 You could probably more easily get two 10V clamps to work in series, but I kinda think that you would be killing the reaction time of the whole thing a bit. Not by much but still measurably, it would all depend on the differences in capacitance of transistors so you get a best and worse scenario. But on a side note, you would probably have even better leakage characteristics.
In the application, the Base Emitter junction is not on at all. It only is reverse biased when the fault condition is reached. No noise is generated when the ohms circuit is working properly. It is used to protect a DMM from someone applying external voltage while the meter is in the ohms mode (diode check, continuity, etc). Remember, this circuit is only practical as a fault clamp. All speculation about using this circuit as a zener is just wrong. So NO noise is generated when the DMM is used properly. The reason to use this circuit over zeners is low leakage current. If the clamp leaks, it causes errors in high ohms ranges.
I am using BAT54S for that purpose. Am i on correct way ?
BAT54S is a dual Schottky diode, and while you can definitely use it in a clamp circuit, but you generally have to provide voltages for it to clamp to. Many circuits use Schottky diodes to clamp to Vcc or Vdd on the positive side and ground on the negative side, which is a good solution when you're just trying to avoid overdriving the inputs to ICs. The advantages of using bipolar transistors as Dave does, are that a) you only need two very cheap components to clamp to about +/- 6V, b) they have a very sharp knee on their current/voltage curve, and c) they have very low leakage current until you reach the avalanche voltage. Schottky diodes have fairly high leakage current, which can be problematic in high impedance applications.
What about frequency responses?
Are these faster than a Transorb or regular Zenner?
As per my understanding, avalanche breakdown can do permanent damage to a semiconductor device if the current is not limited. What is special in this circuit that prevents BJTs from getting damaged permanently?
The damage cause by Vbe breakdown results in a loss of gain which is not an important parameter in this application.
The higher VEBO means the emitter is doped lighter, and will have a higher emitter resistance
That's what you'd think, but in the two high-Vbeotransistors he mentions, their main selling point (aside from high Vbeo) is low Ron - about 1 ohm max. So it might have more to do with diffusion depth than doping. That's just a wild guess, though.
It looks like on the schematics, those transistors are only put in the circuit in resistance, Diode test, and Continuity? How does that apply to the CATIII rating? Still cool to know, learn something new every day!
This is used in resistance, diode test, and continuity circuit to protect the DMM from someone applying voltage which would damage the A/D input and/or current source circuitry used in ohms measurements. The clamp voltage shunts current to ground until the PTC switches to high resistance. The meter cannot have a clamp when measuring voltage.
@@davetaylor5730 Thanks for clarifying that for me. At 5:49 Dave Jones says the transistors in this configuration is used for the 1000V CATIII, so it confused me.
9:20 - Wouldn't Vbeo matter with this particular device? Wouldn't it be the maximum voltage that the device is able to safely *mute* or *switch* ?
Yes it would. Muting (ac switching) BJT transistors need to be symmetrical both in terms of gain (and reverse gain) and the breakdown voltage.
In muting applications, Vbeo isn't the spec that's important, but Ron needs to be very low. Ron isn't even specified for most bipolar transistors, and I would guess that in the transistors Dave mentions, high Vbeo was a side-effect of designing for lowest Ron.
Sorry - that's R(on).
Hmm, do IGBTs happen to have any interesting uses? I mean they are great for high power high voltage inverters, tesla coils and such, but is there more to them?
They withstand a very high collector-emitter voltage. Much higher than any other transistor type.
It's used for switching high voltages like in inverters.
There isn't really more to them.
@@Momchil0 Well, I doubt that "there isn't really more to them", but the main point of most "alternate" applications of BJTs is to do something with cheap, cheap, cheap transistors that they weren't intended for. In the case of IGBTs, cheap, cheap, cheap isn't a thing.
Interesting but why not just use two zeners instead? Wouldn’t that give you more options for the voltage threshold?
It has to do with leakage current.
The difference from standard zeners is that this circuit has extremely low leakage. Zeners leak in the uA range.
BJTs reverse-biased like this tend to burn out very easily, a purpose-designed zener is much more rugged. And you can get TVS diodes which are effectively back-to-back zeners in one package, even more convenient. He's right, its a naughty circuit...
Dave, Why'd you use the old Analogue Scope and not one of your fancy ones?
Misplaced the Fancy ones? ;)
The analog one IS fancy
@@EEVblog I love the analog display. They need to get more love :)
CRT scopes DO get a lot of love, but mostly from people "of a certain age" who grew up with CRT oscilloscopes and still prefer them over parametrically superior digital ones. The HP 1740A is one of my favorites, but my go-to portable is an HP 1726A, which is higher bandwidth (275 MHz) and has a really cool differential timebase that lets you measure the time between any two events on a waveform.
@@BrightBlueJimi got a cool pocket one that has a touch screen. so kewl. by jyetech. DSO112A
@@pepe6666 Yeah. I almost bought one of those, but instead I got an old CRT scope on eBay for about the same amount, and I love it!
So... why don’t people use a normal Zener diode or ESD/TVD diode? Is there any special feature that this little circuit have?
Leakage is mentioned, but I suspect that recovery/response time and sharpness could also be of a possible asset.
Would this set up be used for intrinsic safety circuits? The Atex Ex i a ratings.
No, approvals agencies don’t play with dodgy designs. It needs to be rated for breakdown application, including the power dissipation and valid over temperature. It also needs to fail in a known, predictable way, short circuit in this case. Why would you choose this sort of approach in a safety related function?
@@klave8511 Just wondered if this was the basis of design, not the final design. My understanding is that one way of limiting voltage spikes in the event of a fault is to use a combination of zener diodes, resistors and fuses to limit the maximum voltage and current that can reach a hazardous area
Gordon Brown : sure, as the basis of a design it is valid to limit the voltage applied to a circuit by the use of a zener and a resistor. The zener must, in combination with a resistor, render the protected circuitry “non incendiary” - limit the voltage and current without heating up above the temperature class. You would use 2 zeners in parallel in case one fails and because they fail to a short circuit it is safe. They are rated for continuous operation so if a fault persists they won’t get hotter than the rating. The standard uses 1.5 x the power to be heating one zener during a fault (IIRC). So in all this, you have to use known, documented guaranteed specifications. Using parts in non standard ways is frowned upon because someone has to assume the liability for the safety. Agencies allow you to put their sign on the part because they assessed it as safe and they have to rigidly follow tried and tested rules. Imagine these to be a design applied to your car brakes, you want solid performance, not crafty designs when the time comes for safe operation. If you can convince an approvals agency that your inverted transistor voltage limiting design will always be reliable, at all operating temperatures and for every transistor you put in there for all the years it will be manufactured then they will allow you to use their name on the part. They get these requests all the time as new parts and new technologies develop but they are very slow to change the rules that they know work. Let’s just say that approval agencies say no when there is any doubt as to the safety, they say yes when all doubt has been removed. Have a look at safety barriers, very simple designs, they haven’t changed in decades. Only zeners and resistors in there to do the heavy lifting. Not just any resistor either, they fail open circuit.
@@klave8511 Awesome. Thanks for such a detailed reply! I'm just a newbie to it all so thanks, much appreciated.
How they measure ft?
Can you use two zeners in series to double the breakdown voltage?
Not always work as you wish. Try it for yourself. The voltage can be lower than double or not conducting at all.
@@johnyang799 I was thinking that maybe it won't conduct at all, since the voltage is applied at two separate junctions… I'll try it myself sometime in the future. ;-)
Don't know what John is on about. This is done often in for example tube amps to generate a fixed screen voltage of 200V
@@thulinp Not always work. Have to use specific ones to see if they work as they should.
Thanks for sharing😀👍
Great circuit Dave, Cheers! Maybe I could use this clipping circuit for a guitar distortion? - I might have to give that a crack :)
The harder the edges the worse the sound IMO. That's why Germanium clipping sounds so awesome. It's exactly the opposite, smooth and overdriven.... unless your going for a digi-deathmetal vibe
@@UpcycleElectronics You can round off the hard rate of change of voltage with a low pass filter. It would be kinder to your amp too. But it's a great idea. Having said that I'm told that reproducing real amp distortion is very difficult - but perhaps that's not what's wanted.
@@ChristianRThomas
You're right. It will still be rather linear though even with a filter.
At one point a couple years ago I took an old powered subwoofer and did a whole bunch of mods to the 4 stage 4558 based filter. I probably built half or more of the circuits on runoffgroove and a few other effects circuit sites to put them inline with that thing. I have a little Fender Reverb setup too, and have owned a half dozen amps over the last 21 years.
The subwoofer hack was just an attempt to learn more about op amp audio filters. One of the best classic overdriven sounds is ROG's Fetzer Valve preamp circuit IMO. That with a couple of germanium diodes like the 1N34A make an awesome classic sound. After reconfiguring the filter on the powered sub, and adding that preamp/ clipping diodes it sounds really good or so I've been told. I would play it more often but I need to put a spring reverb in it like the one in my main amp as I like that sound too much. I play acoustic far more than electric though.
...(tangent)... anyways... I put a bunch of stuff on a scope back when I was goofing around with that setup. If you look at what is happening to the signal with clipping and compare sounds, a lot of what makes tube amps sound the way they do is the nonlinear sound and lack of sharpness when they clip. Most transistor circuits, and silicon or schotky diode based circuits I tried looked like 'desert plateaus' on the scope. While the JFET type-a preamp and germanium diodes had a completely different taper and smooth "infinite" transition. That franken-amp-enstein really does sound good even though the integrated solid state amplification stage and driver were originally designed as a home theater powered subwoofer.
You probably already know this or have tried it, but anyone interested in distortion and/or overdrive really should use an oscilloscope to quantify/qualify the sound/signal. At least for me, it made a big difference seeing the dynamic difference first hand.
The thing I really want to play with next is winding my own pickups to try to filter through all the fluff I've read and heard, but that's way down the list of rabbit holes I'm currently navigating.
-Jake
@@UpcycleElectronics Lucky I read the original full reply on my email! :) I have only ever thought about effects circuits - never actually done one. Proper overdrive distortion at low volume seems to be what everyone would like. I'd use a better op amp than a 4558, though. Mind you, if you only want to pay a few cents each then the choices are pretty limited; 5532, 4558, TLO72, LF353. The TLO72 has more guts than the other two BJTs but it's pretty ragged. My "musical" op amp is the LM6172, which are about a quid each. If you get a copy of Lancaster's Active Filter Cookbook (not easy to find cheaply but you can get pdf versions of it) then you'll be 8/10ths of the way to knowing all of the practical side of designing filters. (There is a little bit more needed for high precision applications using multiple feedback circuits but that's getting quite specialist and you won't see them in most loudspeaker/amp applications.)
I can help you with winding pickups, though. This is really quite easy. First of all it boils down to the inductance and the load it's playing into - giving you your first order rolloff. The tone control turns that into a second order, and that will change shape depending on the cap value and the load. You can model this easily in LTspice. Most people seem to have gone for a 1 Meg load and if that's what you choose then you'll have the same scope as everyone else. You can squeeze out more top end with a lower inductance or you can have it peak (say around 2k) and then have the tone control take the peak down. Personally I would send the signal into the highest impedance I could manage (using a second op amp to feedback to below the load resistor) and then do all the shaping with filters on the pre-amp. On the practical side Alnico or neo magnets will give you less distortion than ceramic (which is why the new Stratocasters sound overly smooth), and you'll get more distortion with scattered windings (but never the same pickup twice). Wires.co.uk sell some old wires for pickups and you can pick up a hand winder on ebay for £20.
Dave ,we need fundamentals friday .every weak.much more about fpga
I guess the Fluke engineers did this before TVS diodes were commonplace? My experience of reverse-biasing base-emitter junctions is instantly burnt-out transistors!
Can a transistor be made with 2 Zeners like your equivalent circuit?
No. It's only an approximation.
@@FarleyHillBilly Thanks.
This entire article is worth reading, but this one page covers your question. www.allaboutcircuits.com/textbook/semiconductors/chpt-4/meter-check-transistor-bjt/
I thought those first 6 minutes felt very familiar lol
I think you told abt this while reverse engineering a fluke multimeter??
Maybe as a distortion circuit??
Lol - "shin din electronics". I think I remember you doing a vid on the nightmare, even when buying from big name suppliers. Seems like you wanted a Motorola transistor and it was a crap-shoot as to what you got
121gw is that 1.21 gw??? Back to the future!...
What if use two pairs in series ?
think leakage
Why use this instead of two zeners? Not sure what benefit of this different arrangement gives.
Dave addresses that issue a few times in this video, watch the video again and pay attention.
some people make relaxation oscillator , using this property.
Clamp means? Without that knowledge I'm lost in this video
Clamp means to limit the voltage across a part. Any extra voltage that tries to exceed the breakdown voltage gets dumped down the zener as current, so as long as the part is fuctioning it won't exceed the breakdown voltage.
@@kouji71 thanks
another interesting term to look for is a crowbar circuit. Also a type of clamp ;)
It works in real life. I am wondering if will it works in a PSPice/LTSpice/etc simulation? Simulations are for many times erroneous.
It's not a problem with the simulation, but with the model. I've seen bipolar transistor SPICE models PUBLISHED BY SEMICONDUCTOR MANUFACTURERS that were just plain awful, resulting in "huh??" results from simulations that disagreed strongly with bench tests. The bottom line is, if the base-emitter avalanche is correctly modeled, the simulation of this circuit will give the correct results.
Regards inventing the wheel?.
Reminds me of a "diac trigger"
Francis X. Clampazzo: Should I use THE CLAMPS??