Discrete audio amplifier project PT10 circuit layout and VAS stage current

[@1959Berre]

It is comforting to see someone practicing the pre apocalyptic art of rendering a schematic on a sheet of graph paper with a sharp pencil without the aid of ruler.

[@diabolicalartificer]

This audio amp project series of video's is the best there is on Utube. A guide to amp design par excellence. Thanks, much appreciated. Have a good holiday.

[@questionmark9684]

Hi John, Yes, I would agree, it is a truly fantastic series that makes us learn an incredible amount. I'm glad to support you on Patreon as your time and expenses dedicated to your channel are precious. Cheers Mark

[@McTroyd]

Just burned through your audio amplifier playlist. Thanks for all the info. I especially appreciate the detail of going through the process (including errors)... makes for longer videos, but it's an important perspective that gets lost when everything is edited down to 5 minutes. Looking forward to the next!

[@tc-bladeofgrass6719]

Hey John I only discovered your channel a few days ago, I've binge watched all of your amp design videos. I absolutely can't wait for this to be available as a kit so I can make a pair of these. Keep up your fantastic work

[@jp-um2fr]

First and foremost John - Happy Christmas. Somewhere in my loft is a sterio pair of Mullard 20 W amps with their 3 valve pre amps. That was the initiation into my hobby of electronics somewhere around 1965. Rather the deep end perhaps but where I worked there was ample advice and help. 0.05% distortion at 20W into 15 Ohms (max 1% at 27W) and all other parameters to the highest standards of the time. It took 3 of us to get the thing in the loft along with the spares. The chassis is over 3 feet long and each amp is the opposite hand from the centre. When it was driving my 15" 15 Ohm Tannoys close to flat out the blue haze in the EL34's was hypnotic. One could see the power. There were burn marks in the lid of the cabinete above the EL's. It could be heard a mile away. Apart from a re cap it might well work, but I have no mentor to call on if it didn't. I have followed your amp design with interest but again I am concerned that I would be unable to correct my mistakes. I have tried a couple of little kits but they turned out awful. These little black things with legs are fussy blighters. It is interesting though to see what a transistor equivalent to 'the beast' consists of. I'm afraid modern standards of performance appall me but I know what your are building is of 'the old school'. Thank you for that. Happy New year John - Regards.

[@JohnAudioTech]

Happy Holidays JP! It would be great to get the old Mullard amps singing again. It may or may not need caps depending on the type used. Careful reforming of the electrolytic caps can bring them back if they haven't dried out. Most people would tell you just to replace them.

[@SuburbanDon]

Wow old school. I still have a 0.1 x 0.1 graph pad from the 80s.

[@paulpaulzadeh6172]

Hi John, great channel, very good as X-ray Tony

[@iblesbosuok]

5000 A/A! Bloody large amount of current gain! Cheers from Indonesia

[@HillsWorkbench]

Like the circuit, minimal but using modern high gain transistors to get there. I suspect those diodes on the outputs will cause a little distortion (or worse) though.

[@YoussefEl.]

I think i missed something, but how did you connect the differential stage's constant current source to the voltage amplification stage's constant current source (the two ccs circuit from part 9). Im not seeing how tying the bases together through resistors makes it work.. .. im wondering how did we get from the last video to integrating them here together?

[@Markoul11]

you must add in parallel to your 100uF d.c. decoupling capacitor in your FB resistor ladder and 100nF cap. The relative large inductance of your electrolytic 100 uF cap will mess up all your amplifier design.

[@johnsenchakinternetsecurit8935]

What Snickers this time around?

[@snaprollinpitts]

Hi John, I can make a prototype PCB for you, either way you will need all the files to send to a PCB manufacture. I don't know if you have any experience cutting your own boards. I have the boards, chemicals, drill press, and more than enough experience to "getter done". I use Diptrace, you can download it for free at diptrace.com let me know mike

[@akratooos]

I recently bought a PAM8403. Can i somehow connect the output to some mosfets, which are powered by an external 28V power supply, and in that way increase the power? PLEASE REPLY!

[@kennethsimmons5235]

Merry Christmas

[@JohnAudioTech]

I found this in the spam folder. Why YT thanks it is spam is beyond me. Anyway, Happy Holidays!

[@joohop]

would it be a lot of work too much to change thing's to the 2N3O55 To3 case

[@HillsWorkbench]

3055's have very poor gain unless you cherry-pick some prime ones. And you would likely have to parallel them up to stay in their safe operating area if he's going to push 100W at 4 Ohms.

[@joohop]

@@HillsWorkbench Mmmm Ok Thanks For The INPUT

[@glpilpi6209]

I would advise anyone attempting these builds to get a semiconductor reference book then you won't need to ask silly questions about which transistor is suitable for so and so that's in the circuit.

[@anindyamitra5091]

Did you test the overall circuit putted together? I am running into a problem with the current mirror circuit at the input stage. When I connect two resistors(1.8kΩ to be specific), as a substitute for the current mirror circuit, the amplifier runs fine, but as soon as I change the two resistors with the current mirror circuit (pnp-pnp current mirror, towards the +ve rail), the amplifier becomes "unreliable", i.e., sometimes it works fine but sometimes it keeps on latching up while playing sound through it and so sound is obtained at the output of the current stage(2×2sc5200s as the current stage transistors) plus a dc offset, which makes the transistors "warm". I don't know that to do now, I want to include the current mirror somehow, please try to help! Global feedback resistors R1=47kΩ, then I tried changing it to 10kΩ and R2=56Ω, later I changed R2 to 100Ω and after that, to 1kΩ, but none of them helped.

[@anindyamitra5091]

I am using c1815 and a1015 transistors at the input stage and running the circuit off of ±30V, with 10,000uF filters(so decoupling is plenty, I think). I have also used two transistor CCS stages one at VAS and the other at input stage.

[@JohnAudioTech]

The current mirror adds a lot of gain to the input stage. The amplifier layout must be good to avoid problems. Check for oscillation with a scope if you have one.

[@anindyamitra5091]

@@JohnAudioTech , first of all thank you very much for your reply. Unfortunately, I can't afford an oscilloscope for myself(at this age and in this region, specially), so I have to do everything relying on a multimeter(it can measure LCR, frequency, and hFe etc.). Does the current mirror circuit help in reducing THD? I just want the THD to be not more than .01%, below 100w, I will drive 8Ω loads though I will try to build it to handle 4Ω loads without any issues(like blowing up the 5200s). Lastly, how much quescient current through the main transistors do you recommend? Because I had just figured out today that the 5200s are having a base to emitter voltage of about 0.38V in cold condition (in both of them), so almost no current flows through the main transistors. Might that be a reason for the latch up, in addition to what you said?

[@JohnAudioTech]

@@anindyamitra5091 The current mirror adds a lot of gain so the open loop gain is very high which allows negative feedback to do its job. The amp must have local supply decoupling using film or ceramic caps on the board or it can oscillate even with large filter caps. Bias the amp in the 35-50ma range. An oscilloscope is a critical tool in building these circuits. Even the cheap ebay scope kits are better than nothing.

[@anindyamitra5091]

@@JohnAudioTech , ok I'll try everything you have mentioned.

[@kier182]

hi john, can you try test the "yiroshi" power amplifier. I think it's a very interesting design. kier from Philippines. tnx

[@jimhough6233]

GO JOHN GO!

[@MrBrymstond]

It lives, It lives!

[@1959Berre]

"It's alive! It's alive!"

[@notahuman1]

Hey John, one out of the topic question. Would you be able to reply or make a video on this too??! I have a Sony CFS-B5S old cassette player. I removed the tape head and tried giving it an AUX connection without modifying anything on board.. L,R,G connection. The output is way too loud and distorted. Its a pretty simple board, a Preamp + PowerAmp (single board), Transformer and Radio board. Any suggestions on how to improve the sound quality?!

[@JohnAudioTech]

Do you mean you are putting signals in the wires that once connected the tape head? If so, the tape head puts out a very small signal of only microvolts and needs a lot of amplification. Plus, it has equalization which would change the frequency response. I'd say it is not a good idea to do this.

[@notahuman1]

@@JohnAudioTech Thanks. I did it thinking to connect ipod or a bluetooth receiver to the aux 3.5mm jack. You are right.. amplification is too high causing distortion. I have to keep input volume at the lowest still cant increase the output volume from the player much higher. Still for a small room, it does work with minimal volume. Let me know if we can tweak it anyway for better results. 😊

[@abhinilbose24]

Can not find bc139 and bc140. What us the manufacturer ?

[@JohnAudioTech]

These are a pretty common part from ON Semi or ST Micro.

[@abhinilbose24]

@@JohnAudioTech I am getting BD139 and BD140, I guess they are same as the BC139 and BC140.

[@JohnAudioTech]

Oh sorry, my mistake. They are BD, not BC. I had them labeled wrong on the schematic.

[@abhinilbose24]

@@JohnAudioTech Thank you for reply.

[@pangolinamplifier422]

John could you please put the node dots on the schematic to indicated electrical junction connections, then your wires that cross over another wire without the node dots, will indicated a lack of electrical connection. I mean this is the convention that the electrical world has used for decades. Your use of a broken wire to indicate a cross over wire is confusing. People don't realize that the schematic symbols are not just symbols they are electrically correct as far current flow. This is why a MOSFET has something that looks like a capacitor at the gate; because that is what it is, even though is composed of the electrode pad, oxide layer and metalized silicone underneath the oxide layer. Dig?

[@rich1051414]

Why would you fuse both rails when you could just fuse ground and cover both rails?

[@JohnAudioTech]

A lot of times when output transistors blow, they both go shorted which makes a rail to rail short. Having the fuse on the ground will not clear a rail to rail short.

[@rich1051414]

@@JohnAudioTech Oh I see, I didn't think about that.

[@jimhough6233]

My tablet has autocorrect that I can't disable. Sorry, it drives me nuts too.

[@jimhough6233]

Yesyesyes! I see other people designing custom boards and the tech away all that free$$$ copper flood fill. WWHHYY?

[@Arylic]

hi John ,just drop your channel .it's amazing . i'm wondering if you could try to check one of our product up2stream wireless audio receiver board ,which can bring multiroom ,airplay ,dlna streaming to any existing stereo system . Do not find a way to message ,so try to reach here .hope you don't mind .

[@abhinilbose24]

Where to buy original transistor, getting fakes online and offline.

[@Tbonyandsteak]

Haaa hAAA .......first