Build this three transistor mini audio amplifier

[@ahmedalshalchi]

Thank you John for teaching new age engineers to use discrete analog topologies whatever possible than Integrated Circuits....

[@generessler6282]

Nice grass, John. You should diversify into lawn care tips. 😊. The amp discussion is great, too.

[@MrAlFuture]

I love this type of circuit level video. Thanks, John!

[@VIPINSAINI_20]

Love to see lm3886 and 50w+50w new Amplifier videos, please upload as soon as possible

[@unregistereduser2]

Nice little amp. I sketched it up in KiCAD, with a few minor mods to suit components I had on hand. I also implemented a little RC lowpass filter on the input, because I'm using it with PWM sources: github.com/prosper00/JAT-3t-Amplifier/blob/master/JAT%203t%20amplifier.pdf Eventually, I might make up a PCB gerber. If/when I do, I'll include it here: github.com/prosper00/JAT-3t-Amplifier

[@LeonMatthews]

This is so cool, thanks for posting this. I wish I had more thumbs-ups to give!

[@harveyellis6758]

Great initiative to help novices with the KiCad files & schematic. Use of 1N4007 (1000v diodes), rather than 1N4001 is a bit of overkill, unless this is all you have in your junkbox to use.

[@unregistereduser2]

@@harveyellis6758 4007's are what are in my drawer. They cost about the same as 4001's as I recall. That was my first ever crack at using kicad. I do still intend to make up some PCB's, though the gerbers I have up right now are kinda... garbage. the circuit is okay, but the layout is not great, and I should have used a ground pour on one of the layers. I'll take another crack at it someday. An SMT version might be fun too.

[@eugenepohjola258]

Howdy. Really nice. I like the bootstrapping a lot. That actually resolved problems of reaching calculated power I have had. I would like to go a few steps further. The base-emitter voltage will go down as the end transistors warm up. Therefore there is a tendency for the idle current to grow. The 0.47 ohm resistors may not be smaller. The voltage drop over the resistors must grow steeper than does the base-emitter voltage decrease warming up. Otherwise a thermal avalanche is to be anticipated. My experience is that the idle current cold is about 40-50 mA and after running warm it has increased to 200-300 mA. When the transistors cool down the idle falls back to the former value. Attaching the 1N4001 diodes to the same heat sink with the transistors speeds up the settling after running warm. The resistors prevent reaching max. power though. One may connect 6A diodes, P600J, in parallel with the resistors. They will provide bypassing for heavier currents. They will not interfere with the thermal stabilization. In high regards.

[@tiagomello]

I like your fluent Chinese! Impressive!

[@bob4analog]

Nice amp! Im gonna build this one. Love your demonstration and diagnostics with the scope. The scope is our friend.

[@robertneill3057]

Jaycar Electronics (Australian supplier) has something like this amplifier in Short Circuits Book 1. They also have a higher rated output discrete component amplifier in Short Circuits Book 2. This uses the BD139/BD140 transistors on the output.

[@draztiqmeshaz6226]

Oh yay, you got it! You're welcome. There's a few things I'm curious about if it piques your interest. As far as the subject of this video, I have to wonder why I'm dealing with so much complexity trying to fix my old Hafler, when clean amplification can happen so simply. Thanks for sharing your knowledge!

[@PileOfEmptyTapes]

Make no mistake, this circuit is the barest of bare minimums and about the kind of thing you might have found in a basic mid-70s portable radio. You might get half a watt @ 8 ohms out of it on 9 V with distortion levels of multiple %, which is not exactly setting the world on fire. I have not found a single Hafler amp rated less than 60 wpc, most 100+ (some of them substantially more powerful into lower impedances), at HiFi-worthy full power distortion and damping factor levels. That's hard to obtain with less than about a dozen transistors in the power amp.

[@JohnAudioTech]

More power and high lop gain for low distortion figures need more parts. My JAT501 amp uses 13 transistors. Some amps have over 20 per channel but that's diminished returns to me.

[@draztiqmeshaz6226]

@@JohnAudioTech wild. Baby steps bring up the signal, hey? Also, I've got some questions about that little Wuzhi too, but that depends on if it's worth your time. Lots of smc's and hard-to-trace traces

[@draztiqmeshaz6226]

Things I noticed: 3.5mm input is VERY noisy, with peaks starting around 100Hz (not 60 or 120), but when bt is connected, floor drops a lot. But since it doesn't have aptX, kinda worst of both worlds. I thought to either re-route the input jack to avoid the bt module or at least change the trace length so it doesn't resonate, or perhaps change the resistor values around pin 8 (from memory here) on the TI(is it though?) to drop the gain, but at least one of the sm resistors is just brown with no numbers :( Despite the noise they do put out plenty of juice, so I fabbed a large copper heat sink for the workshop unit (I bought 2) and hooked it up to some insensitive old boom box speakers. Getting a lot more work done now! What else? I emailed them, and they sent me a Pic of where to solder in a bt antenna. Not where I thought..

[@richardspence6543]

This is a great channel. Thanks for all the projects. I have a 12v supply and a 4 ohm, 30w speaker. Will this circuit work ok? Or perhaps it could be adjusted?

[@woodcoast5026]

You would get more than 562mw with a Bridge Tied Load configuration. The BTL is not given much attention by hobbyists and text books, people still use the LM386 that was designed in 1969. With a BTL chip an output capacitor is not required.

[@noelj62]

Simple yet interesting design.

[@aurthorthing7403]

It would be cool to turn this into a preamp or booster for bass or guitar. You should check out some boost pedal schematics and see if they have anything you like. The Electroharmonix LPB-1 is a good example of a basic boost pedal. I want to build a basic boost or preamp with BBE harmonizer chip used in the BBE Sonic Maximizer pedals. The Sonic Maximizer is kinda like a parametric EQ but different.

[@rich1051414]

It could work as a buffer perhaps, but there isn't enough gain with this as is to power speakers directly with only a guitar input.

[@aurthorthing7403]

@@rich1051414 I'm guessing you don't play the electric guitar and if you do then you are not familiar with pedalboard accessories.

[@aurthorthing7403]

@@rich1051414 you can place a EHX-LPB-1 boost in between the guitar and this circuit. In my case, I prefer passive bass guitar circuit with an active DI box. But you could get all the gain you want from an active pickup circuit. Some of them are called "amp killers."

[@mystcat3]

My headphones sound very low due to my cellphone jack output, it burned a bit after using another heaphones with the charger and this amplifier will help me a lot

[@Brickoliciousness]

This might be the right place to ask. If I wanted to build an audio amplifier for some old iMac speakers, what would the smallest useful amplifier board for that be? I can't find an answer to that anywhere. My current googling is why I ended here ;)

[@spark300c]

well JFETs are good for guitar amps. because of high impedance you just need one stage but it has low currant gain. also they great for clean channels for guitar amps because they soft clip when you over drive them.

[@Dandan-fd9vc]

I can't open the website you mentioned in the introduction. There seems to be restrictions on unknown access? Please check.

[@hippo-potamus]

Can you do a video discussing impedance rise and your opinion on how best to mitigate against it, ie would it be advisable to wire subwoofers to amplifier at lower ohms knowing that after accounting for the expected rise the overall power output of the amp would be maximized? Or would you simply recommend upgrading to a more powerful amplifier? Thank you. -Cheers mate

[@Theineluctable_SOME_CANT]

I plan to build a higher power version and run it on 12 V and also 18 V and 40 V. I want to see how far up in power and down in distortion this circuit will go. I think a Vbe multiplier instead of the diodes so tge circuit bias can be trimmed down and up. A darlington for the input transistor. Could be a great cheap clean power amp for an active loudspeaker with three bands and three power amps per cabinet.

[@PeterMilanovski]

I hear bootstrap mentioned a lot but admittedly I have no idea what problem it's meant to solve or where to look in the circuit to find the problem in the first place! I haven't found anyone that has made a good simple video on the topic yet. I'm surprised at the output level and quality of this little amplifier! It's got me thinking about what a single ended class A amplifier with paralleld output small signal transistors to current share would perform and sound like... Might just have to bust out the breadboard and put something together I think...

[@researchandbuild1751]

John also has a bootstrap video explaining how and wny bootstrapping works and how to do it

[@johnnyboy0g460]

Good day John. I just recently started watching your videos as I received som small two inch 4ohm 3w speakers from some scavenged things. So now I'm trying to build a small enclosure and a little amp to drive these from my phone. Thing is; I have no idea what I'm doing. I know the basics of small electronics but it really does end there. So I have a small collection of different IC's, capacitors, resistors and transistors; where can I figure out if any of these are useable to create a small amp? All I want for now is a proof of concept, but I can't find any info other than using IC's which were already made to be used for OP amps.

[@danbuffington75]

Would it be useful as a preamplifier?

[@micomrkaic]

John, why didn’t you bias the output stage with a Vbe multiplier? It is simpler than matching diodes to output transistors.

[@davekazoroski6548]

In most commercially built amplifiers of this type, the biasing of the output transistors is done with one diode and a low value resistor in series with it to get the correct bias voltage. It is simple, reliable, and uses only two parts to do the job.

[@electronicssciencebysherry5050]

Thank u 4 yo good work I really appreciate scientist

[@DIY.elektroMrakorosi]

Can i use these Circuit as a preamp for Class A amplifier?

[@bassblaster505]

the yaunjing brand sells a metric crap ton of audio bits. from class A, AB D, cases crossovers all kinds of stuff. id check out there website im gonna use one of there cases and TPA3255 boards when i made a subwoofer amplifier.

[@larryshaver3568]

if i build a stereo amplifier[2 of these on one board] will 6 volts be enough?

[@johnnytacokleinschmidt515]

His example here on breadboard is stereo.

[@DAVIDGREGORYKERR]

2048 Bins is the max as anything over will slow the trace down.

[@ArdiNugrohocilacap]

How using in 4 ohm speaker?

[@cassvirgillo3395]

Hello John, Good little amp, it sounds good. Have you done any small amps using a j-fet in the input stage, like a runoffgroove.com Ruby amp, which uses the LM386? I have read where they have some good properties, from an RCA book I have. Stay well, C.

[@EngineeringEssentials]

Wow that one is great. Could you please do a video on MOSFET output stage amplifiers, and a comparison with Transistor output power stage audio amplifiers? . I actually tried to replace Transistors directly with MOSFETS but failed. Can you help me?

[@PileOfEmptyTapes]

Every MOSFET is a transistor but not every transistor is a MOSFET. ;) MOSFETs require a different biasing setup with a different temperature coefficient from BJTs, and are likely to require gate stopper resistors of a few hundred ohms. The required gate-source voltage of generally about 3.5-4 V is also likely to cause you a headache on low supplies, so topologies suitable for BJTs may just not work at all or give poor output then. Optimum current levels are also going to differ. So while various kinds of _Three-Legged Critters_ are basically interchangeable, in practice the devil's in the details. Maybe start with something simple like an emitter follower / source follower.

[@mystcat3]

Mosfet will produce a Digital (extremely distorted) signal

[@spark300c]

@@mystcat3 that only if they not bias correctly or you using to low voltage to turn them on.

[@bogdangrigore8736]

Can it work on 12v ?

[@mystcat3]

DiodeGoneWild has a cat, so Jhon's Grieco

[@sayantanmaiti2513]

if we use any high power bjt like tip31/32 pair instead of bc337/327, how much the output power will be???

[@Dazzwidd]

I would try a BD139 and BD140 output driven by a BC337 and see how that goes... play with the idle current

[@wallybaldemor8523]

Does it sounds better than lm386 amp? Thanks

[@jossmoerkerken5732]

That was great thank u

[@johnyang799]

What about a 4 transistor diamond buffer?

[@christopherkise]

aaaamaaaagaaaad there is a bird.. BIRD BIRD BIRD:...... A-well-a ev'rybody's heard about the bird B-b-b-bird, b-birdd's a word A-well, a bird, bird, bird, bird is a word A-well, a bird, bird, bird, well-a bird is a word A-well, a bird, bird, bird, b-bird's a word A-well, a bird, bird, bird, well-a bird is a word A-well, a bird, bird, b-bird is a word A-well, a bird, bird, bird, b-bird's a word A-well, a bird, bird, bird, well-a bird is a word A-well, a bird, bird, b-bird's a word A-well-a don't you know about the bird? Well, everybody knows that the bird is a word A-well-a-bird, bird, b-bird's a word, a-well-a A-well-a everybody's heard about the bird Bird, bird, bird, b-bird's a word A-well, a bird, bird, bird, b-bird's a word A-well, a bird, bird, bird, b-bird's a word A-well, a bird, bird, b-bird's a word A-well, a bird, bird, bird, b-bird's a word A-well, a bird, bird, bird, b-bird's a word A-well, a bird, bird, bird, b-bird's a word A-well, a bird, bird, bird, b-bird's a word A-well-a don't you know about the bird? Well, everybody's talking about the bird! A-well-a bird, bird, b-bird's the word A-well-a bird, surfing bird, brrrrrrrrpagh! Aaah, bap-a-pa-pa-pa-pa-pa-pa-pa-pa-pa-pa-pa-pa-pap Ma-ma-mow, pa-pa, ma-ma-mow, pa-pa Ma-ma-mow, pa-pa, ma-ma-mow, pa-pa Ma-ma-mow, pa-pa, ma-ma-mow, pa-pa Ma-ma-mow, pa-pa, ma-ma-mow, pa-pa Ma-ma-mow, pa-pa, ma-ma-mow, pa-pa Ma-ma-mow, pa-pa, ma-ma-mow, pa-pa Ma-ma-ma-ma-mow, pa-pa, ma-ma-mow, pa-pa Ma-ma-ma-ma-mow, pa-pa, ma-ma-mow, pa-pa Ma-ma-mow, pa-pa, ma-ma-mow, pa-pa Ma-ma-mow, pa-pa, ma-ma-mow, pa-pa Ma-ma-mow, ma-ma-mow, pa-pa Ma-ma-mow, ma-ma-mow, pa-pa Ma-ma-ma-ma, ma-ma-mow Ma-ma-ma-ma, ma-ma-mow Ma-ma-mow, pa-pa, ma-ma-mow, pa-pa Ma-ma-mow, pa-pa, ma-ma-mow A-well-a don't you know about the bird? Well, everybody knows that the bird is a word A-well, a bird, bird, b-bird's a word A-well-a mow, mow, pa-pa, ma-ma-mow, pa-pa Ma-ma-mow, ma-ma, mow, pa-pa Ma-ma-mow, ma-ma, mow, pa-pa Ma-ma-mow, ma-ma, mow, pa-pa

[@saanjayray]

Hey john can you design one class H board for us pls

[@AstrosElectronicsLab]

BIRDY!!!!

[@gkdresden]

You can improve the low frequency roll off very easily, because the input impedance of this circuit is unnecessarily low (around 1.5k). First of all, you can reduce the boot strap current (currently 4 mA) at least by a factor of 4, because you only need 0.4 mA to drive an output current of 100 mA (the BC327- / BC337-40 has a minimum current amplification of 250 at 100 mA). If you use 2.2k for the boot strap resistors you only need 5 µA to drive the input transistor. So you can increase the values of the three input / feedback resistors by a factor of 10 (200k, 22k, 47k).

[@PileOfEmptyTapes]

It's the output capacitor that's limiting bass response here - with somewhat hi-fi ambitions you want at least 1000 µF for an 8 ohm load, preferably 1500 µF or even 2200. The input cap is ample by comparison. 0.23 W into 8 ohms also implies a peak output current of ~240 mA, at which point beta would have drooped a fair bit more. That might be about 1 mA from the previous stage if you're lucky, several mA if you're not - and suddenly the 4 mA of VAS current seems anything but luxurious. In fact, it's probably the limiting factor to power output, especially at slightly higher supplies like 9 V. Guess why AB speaker power amplifiers normally tend to be more complex than this.

[@gkdresden]

@@PileOfEmptyTapes I agree with you. This design is in principle also useful for higher output powers by the use of a complementary darlington transistor pair and 4 1N4001 diodes as bias spreader. I prefer another, bit unusual design to make simple amplifiers by buffering of a low power op amp. Using my design it is also not necessary to adjust anything. ibb.co/xX8qpbT

[@gkdresden]

The control of the output buffer transistors by resistors in the power lines of the opamp is very unconventional but it works really good. The harmonic distortion is still very small, when the opamp's open loop gain is large enough. ibb.co/xX8qpbT The magnification shows, that the cross over notch for the current between the op amp and the buffer transistors is in the order of 2 µs and therefore not audible. ibb.co/SVv16dW

[@PileOfEmptyTapes]

@@gkdresden This old trick circuit is quite particular when it comes to the choice of opamp type and resistor values, so not exactly noob-friendly. Well, at least it wouldn't be as likely to blow up as the one with Darlingtons and diode bias. I would try something like an NJM4556A opamp using a complementary emitter follower (maybe BD139/140 or TIP31/32 or MJE181/171) and a bias spreader transistor (~BD139)... slap all of that onto a common heatsink with some silpads for simplicity. Should work pretty well, but not sure whether it's really worth it over a chip amplifier sadly (LM1875 or something in that vein).

[@pezinski01]

Haha Wongping

[@electronicssciencebysherry5050]

Can u make 4 me a beginners guide to amplifier designs