A Simple 10W Direct-Coupled Class-A Power Amplifier

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Post by mediatechnology »

You need a couple of LEDs on the PCB so it looks more testicle if the punter opens up the case :)

Even better if the case has a transparent window so these can be seen from the outside. Or a tube with just the filament powered up.
Good points.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Post by mediatechnology »

I did a quick FFT to measure the THD with an idle current of 1.2A.
Unlike the previous measurement the generator was an analog IG-18.

The green channel is generator monitor.
The red channel is 8W into 8 Ohms with an Iq of 1.2A.
8W is approx 1.5 dB below the clip point with 18V supplies.
The 1 kHz THD is about 0.019%.

(Note: There is a 26 dB pad at the analyzer input. The measurement is 24 bit, 96 kHz so the noise floor is elevated by the A/D and lack of meaningful shielding.)

Image
A Simple 10W Direct-Coupled Class-A Amp Using The DRV134.
FFT of output at 8W into 8 Ohms with an Iq of 1200mA.


This test was performed with the amp powered by two Meanwell 60W switchers.
emrr
Posts: 570
Joined: Sun Mar 02, 2008 10:27 pm
Location: NC, USA
Contact:

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Post by emrr »

mediatechnology wrote:
One of the things I keep realizing over and over again with flea-power Class-A is the complete absence of low-level grit.
Monitoring at whisper levels, about 100 mV p-p into the speaker, this amp sounds far cleaner than the higher-powered Class-AB amps I usually use.
Fading the monitors down to -70 from maximum peak level I don't hear rising low-level grime.

For whatever reason, the improved low-level HF detail seems to make things image more precisely at any monitoring level.



Yes.
Best,

Doug Williams
Electromagnetic Radiation Recorders
Prototype2
Posts: 16
Joined: Fri Oct 30, 2015 9:06 am
Location: United Kingdom

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Post by Prototype2 »

I registered here after finding this design, thanks for sharing it! I've been wanting to build a small, and simple, class A amplifier for some time and this looks like a good candidate. I particularly like that it is DC coupled and single-ended, while the overall simplicity means I have a reasonable chance of making it work - and understanding why it works :)

One thing is a concern though: I see that you've simply re-used the PCB layout from the headphone amp which forms the basis of the design, which is fine for a prototype but won't satisfy my neatness requirements... While it would be interesting to take on the task of designing a dedicated PCB I understand that this is not necessarily easy to do and that there are many considerations to take into account beyond making the right connections. Do you have any plans to produce a PCB layout for this amp, or if not, would you be able to assist by reviewing my own layouts? I have done some PCB designs before (using KiCAD/EESchema) but mainly for non-critical digital trinkets, LED drivers and more recently a battery charger. I fear an audiophile class A amplifier is a little beyond my abilities...
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Post by JR. »

Neatness is over rated for one-off DIY.

A proved design that works is always a good thing.

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Post by mediatechnology »

Hi Prototype2 - Thank you for joining us here and your interest in the dual Class-A 10W.

Are you going to use a linear supply or a switcher?
The switchers worked great...

I recently sold out of the DCAO_173 layout and now have the DCAO_174 which has the ability to use either THAT124X or THAT 12X6 InGenius inputs.
When I updated the layout I removed a bunch of unused resistor and TO-92 locations on the PC board to tidy it up.
It's basically a balanced input stage directly to the DRV134 pre-driver.
The size, spacings etc. are identical to the earlier one.

Image
Dual Class-A Headphone Amp Line Driver

Other than unused components that might not be needed for an unbalanced input, the only real difference in the TO-220 low power and TO-3P 10W version are the size of the emitter resistors.
Those resistors could be located off-board - the ideal place would be at the speaker terminal with individual leads to the emitters.

If I were going to lay out a board specifically for the dual Class-A 10W I'd put in some variable voltage gain.
I prototyped a "Self-style" shared gain volume control, balance trim, mono and optional Width circuit that I still have on the bench.
I can't recall the exact range but I think it was -inf to +15 dB. It may have been +20.
I was able to get pretty decent tracking and "feel" with linear controls.
I favored this approach for my own use since I felt it would be quieter than attenuate, then amplify to get the voltage swing needed to hit 8-10 Watts.

If someone wanted to use this as a simple power amp then I would just build in some fixed gain.

I'll re-visit that circuit and scan the sketch I did.

In any case, if you want to do a layout then go for it.

Having the output devices directly soldered to the board would be a big time saver.

FWIW I found this heatsink source while I was doing this a year or so ago. HeatsinkUSA http://www.heatsinkusa.com/
Prototype2
Posts: 16
Joined: Fri Oct 30, 2015 9:06 am
Location: United Kingdom

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Post by Prototype2 »

Many thanks for the fast and detailed reply!
mediatechnology wrote:Are you going to use a linear supply or a switcher? The switchers worked great...
Yes, I think initially at least this makes a whole lot of sense. If I get the amp working well and with the features I want I might build a regulated supply with a monster torioid - if only for the added weight and bling :)
mediatechnology wrote:Other than unused components that might not be needed for an unbalanced input, the only real difference in the TO-220 low power and TO-3P 10W version are the size of the emitter resistors.
Oh, I see - so the headphone amp board is perfectly usable for a power amp as is? I'm probably still leaning towards a new layout though, since I want to add some functionality - more on that below.
mediatechnology wrote:Those resistors could be located off-board - the ideal place would be at the speaker terminal with individual leads to the emitters.

I might use housed resistors mounted to the outside of the case, again for the bling! (Yes, the bling factor is important!)
mediatechnology wrote:If I were going to lay out a board specifically for the dual Class-A 10W I'd put in some variable voltage gain. I prototyped a "Self-style" shared gain volume control, balance trim, mono and optional Width circuit that I still have on the bench. I can't recall the exact range but I think it was -inf to +15 dB. It may have been +20. I was able to get pretty decent tracking and "feel" with linear controls. I favored this approach for my own use since I felt it would be quieter than attenuate, then amplify to get the voltage swing needed to hit 8-10 Watts.

This is the most essential thing I'd like to see added, though I'm thinking of using a stepped attenuator (DIY) rather than a linear pot - not because I think it will "sound better" but because I want repeatable gain - and here I have to expand a bit:

The reason I got interested in class A amplifiers was because I got interested in measuring loudspeaker driver performance. I started looking for amplifiers with very low distortion and high linearity, with capacitance free outputs, and soon realised class A was the way to go, preferably single-ended. I got excited by the relative simplicity of most class A amps I looked at and realised that building one would not only be a nice way to get my own test & measurement amplifier, but that it would be a great little project in many other ways. I actually like the fact that they idle hot and and need big heatsinks and I look forward to (eventually) working on the case design as much as the electronics. The heat is a feature, not a drawback! :) But if I'm going to be able to reliably compare individual speaker drivers I need to be able to set the same output level repeatably, so stepped attenuators actually make sense here (as opposed in a "they sound better" sense) and is the perfect excuse to add some extra bling (again).
mediatechnology wrote:I'll re-visit that circuit and scan the sketch I did.

Thank you, that would be great!
mediatechnology wrote:In any case, if you want to do a layout then go for it.
I'll have a stab at it, if you can get me a sketch with the variable gain stage included. Apart from locating the resistors off-board, and (maybe) connecting the power transistors directly to the PCB, are there any other layout pointers you can give me? Like where things need to be kept well apart, which connections need extra copper, and whether it's worth having a ground plane?
mediatechnology wrote:Having the output devices directly soldered to the board would be a big time saver.
I'm temped to keep them off-board to make the case design more flexible, but I'll bear this in mind.
mediatechnology wrote:FWIW I found this heatsink source while I was doing this a year or so ago. HeatsinkUSA http://www.heatsinkusa.com/

Those are great prices, thanks for the tip! I'm based in the UK though, and I already have some plans for how to keep this thing cool, which probably won't involve conventional heatsinks... More on that later!
Prototype2
Posts: 16
Joined: Fri Oct 30, 2015 9:06 am
Location: United Kingdom

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Post by Prototype2 »

When I said I wouldn't necessarily use a conventional heatsink for cooling, this is the sort of thing I had in mind... Crazy idea?
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Post by JR. »

Forced air (fan) based chip coolers generally need a fan to be effective. For only several watts I'd lean toward a passive heat sink. I used a fan and chip cooler for my failed Peltier experiment and as I recall they are good for a few tens of watts so adequate for a 10W amp if you don't mind the noise.

JR
Cancel the "cancel culture", do not support mob hatred.
Prototype2
Posts: 16
Joined: Fri Oct 30, 2015 9:06 am
Location: United Kingdom

Re: A Simple 10W Direct-Coupled Class-A Power Amplifier

Post by Prototype2 »

JR. wrote:Forced air (fan) based chip coolers generally need a fan to be effective. For only several watts I'd lean toward a passive heat sink.

That's a very good point - I was wondering if the thin fins (and narrow air gaps) would be effective without a fan. On the other hand, even with a slow fan, and inside a warm case, these heatsinks are able to dissipate around 100W - the question is how much the efficiency is reduced in the absence of forced air. Could it be as much as 90%? Maybe. These odd looking things have been designed to be used without any forced air though:

Image

They consist entirely of some kind of thin heatpipes, and take up rather a lot of space (and money!), but the smaller version can dissipate "up to 80W" and the larger one an unspecified but likely larger number. The manufacturer specifically claims that these can be convection cooled alone. Read more here.

I can't say I like the look of them, and they're obviously far too large for this amp, but it shows you what is possible. The type I was thinking of testing looks like this:

Image

That's a "Be Quiet Shadow Rock TopFlow" (where do they get those names from) which can dissipate 160W with a fan. DxWxH measures 171 x 137 x 110mm (or 6 3/4 x 5 2/5 x 4 1/3 inches) and the flat part of the baseplate is about 50 x 35mm (or 2 x 1 2/5 inches). It costs about $50 which is lot of money for just the heatsink, but I think it could look fantastic :)
Post Reply