diyAudio (
-   Solid State (
-   -   Class A Sounds Mighty Sweet! (

jgray 5th March 2011 03:12 AM

Class A Sounds Mighty Sweet!
3 Attachment(s)
I thought I would share my first amp design and build project with the group here. I started out to build the JLH Class A, but it evolved into something a bit different. After completing the JLH I was not satisfied with the output offset drift. I also didn’t like the requirement for a regulated power supply, so I wound up switching to a differential input section with constant current sources for both the input and output sections. I stayed with the quasi-complementary output stage per the JLH design. My simulation showed very good results, although I struggled to get the transistor parameters right, so I have some question as to how good it really is. I have to say it sure sounds great to me, and my musician sons both said it sounded “Not Too Bad, ” so I call it the NTB:)

I switched the output transistors from 2N3005 to 2SC5242 for better linearity and easier mounting. The bias current for the output devices is 1A. The power supply runs at +/- 15V. At 8ohms, that works out to about 14W per channel. A more conservative rating on the power supply of 13V gives more like 10W per channel. One critical component was the “current splitter” which feeds the two output devices. I had to find a device that had a very low base-collector output capacitance and went with the NTE373 at only 14pF. This greatly improved the overall performance from the original device I had chosen. The rest of the transistors are pretty common components. Because I had two transformers on hand, I went with separate power supply boards for each channel, each with 8800uF of filtering. There is another 2000uF on each amplifier board for local decoupling. I used separate ground returns for that.

I drew the schematics and laid out the boards using ExpressPCB. I have been very happy with their software and service. I did not get the solder mask or silkscreening option.

The heat sinks were found on ebay. After about 4 hours with no load, the maximum temperature I could measure on the output chips was about 150F. I have access to an IR thermometer from my work which sure makes things easier. My transformers are each rated at 2A, but they run pretty hot at about 140F. I wasn’t expecting a heat problem there, and I may add a heat sink to each of them.

I have attached the schematics and an output waveform picture. I will post a few pictures of the build in a separate post. The squarewave output is at 20kHz, and +/-10V. The slew rate looks like about 20V/usec, although it is hard to tell from the picture. The amp is very quiet with no trace of hum at the speakers. It has an output offset of about 4mv.

I have learned a lot hanging around this forum so thanks to all of you. It sure has been fun working on this.:D


WuYit 5th March 2011 04:18 AM

very nice, except the scanty PS capacitor bank.

jgray 5th March 2011 05:14 AM

Hi WuYit, thanks for your comments. During the development and testing I did try adding significantly more capacitance to the power supply. As expected, it lowered the power supply ripple, but I couldn't measure any difference in the output waveform, so decided to keep my board space a bit smaller. I think the constant current sources, substantial decoupling on the amp board, and careful grounding design contribute to the excellent PSRR.

WuYit 5th March 2011 06:41 AM

you don`t measure, you hear the difference.

Mooly 5th March 2011 08:50 AM

It's very nicely built :)
You say this design "evolved" from the JLH.

Did you listen much to the original design as the single ended input stage plays a huge part in the distortion spectra produced by the amp and consequently also contributes to its very musical sound. It would be interesting to know if you noticed sonic differences between the two designs.

CopperTop 5th March 2011 10:48 AM

Very nice. Do you get much of a 'thump' at power-up?

jgray 5th March 2011 02:30 PM

Mooly and CopperTop, thanks for the complements :) I do not get any audible thump at power up. I have tried powering it up with the audio source both off and on. The music starts almost instantly. I have not looked at it with a scope though. Maybe I should...

Unfortunately, I did not spend much time listening critically to the single ended front end arrangement; now I kind of wish that I had :)


Bigun 5th March 2011 02:38 PM

I like it ! - from your description it sounds to me that you've done a very nice job of this.
I suspect folks would love to see a photo of the hardware and although there will be some 'armchair designer' suggestions for 'improvements' I think your design looks very good.

My 'armchair designer' suggestions would be to try a lower base-collector capacitor on the phase splitter to hear what impact it has on the sound, also to try adding resistors in parallel with the virtual ground capacitors (e.g. 1k5) and hear if it impacts the sound.

jam 5th March 2011 02:45 PM


Try removing the current mirror and increasing the degeneration on the differential................just a thought.


P.S. I also think the value of R6 might be way too high.

jgray 5th March 2011 02:50 PM

Bigun, thanks for the kind words. I posted the pictures of the amp in the sticky pictures thread. Here's a link post 1719

I appreciate your ideas for improvement. My list already has a few things on it related to improving the PCB layouts:)


All times are GMT. The time now is 04:41 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio