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
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. »

Without writing the whole chapter (One of my patents was for a heat sink design) there are several moving parts to consider.

Heat sink effectiveness is about coupling to ambient. Many very thin vanes have a higher surface area to mass ratio (good), but closely spaced fins can clog up with dust (bad). Long skinny fins experience temperature drop along their length (bad). If you are going for 'Bling" (looks) buy one of the artsy looking (and expensive) chip cooler heat sinks, Wayne linked to a source for more conventional heat sinks.

The old school sink designs are still in use because they work and don't break the bank.

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:The old school sink designs are still in use because they work and don't break the bank.

Yeah, you're right. And I did some modelling to see what it would look like when combining this type of heatsink with a typical amp case - and guess what: it looks totally ridiculous! Idea dropped. I also regret my use of the word "bling"; I'm actually much more of a minimalist, and much more "old school", than what that sounds like... But I do like coming up with unconventional solutions, and looking around at what can be re-purposed for other uses. In any case(!), it's far too early to be thinking about enclosures :) Thanks for your input though!
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 »

Oh, I see - so the headphone amp board is perfectly usable for a power amp as is?
Correct. There are no mods.
Just off-board power devices and jumping around the balanced input using the level control connector if the balanced input is not needed.

I mounted the emitter resistors standing off the board not only due to lead spacing but to give them and the board some "air."
Flying leads to the power devices is probably more reliable since the transistor to PC solder bond isn't thermally stressed.

If you do use a stepped attenuator go for the Goldpoint and build your own custom.
In the "Self" "shared gain" configuration you ideally want to use a make-before-break switch which the Goldpoint/Elmas are.
The heat is a feature, not a drawback! :)
Why don't you use a conventional heatsink and add bling by having it thermally pump a Lava Lamp?
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 »

mediatechnology wrote:Correct. There are no mods. Just off-board power devices and jumping around the balanced input using the level control connector if the balanced input is not needed.

Ok, thanks. That makes doing my own boards seem a little unnecessary... Hmmm. Re. balanced inputs; I don't see this as essential - my pro audio days are (likely) behind me, and yes, I do realise that means I'm probably in the wrong forum! My workshop soundcard does have balanced outputs though, so it's not completely without merit. While I no longer deal with long cable runs, the interference rejection could be valuable in an (electrically) noisy environment...
mediatechnology wrote:I mounted the emitter resistors standing off the board not only due to lead spacing but to give them and the board some "air." Flying leads to the power devices is probably more reliable since the transistor to PC solder bond isn't thermally stressed.
Yep, I understand that those resistors need some cooling, and will take that into account. Is it preferable that these are a non-inductive type? What power handling would be required for the 10W design? And re. the output devices - mounting them off-board would be the only option if I was to use the headphone amp boards, and even if I made my own layout I'd prefer them off-board, to make it easier to fit the amp into some off-the-shelf case and to allow for different transistor case types. The thermal stress is also a good point, and it would also move the main heat source away from the input stage etc. Would it be ok to use screw terminals for this, or is soldered connections preferable? How long do you think the leads can be before they start causing problems?
mediatechnology wrote:If you do use a stepped attenuator go for the Goldpoint and build your own custom. In the "Self" "shared gain" configuration you ideally want to use a make-before-break switch which the Goldpoint/Elmas are.
Thank you, that's a valuable tip, and exactly the kind of thing I would be likely to overlook. The prices even of unpopulated Goldpoint switches is shocking though; over $150 for a 2-pole model! Hopefully I'll be able to find something cheaper... Am I right in thinking that with a variable gain stage (as opposed to an attenuated input) I'd only need a two pole switch (or 2x one pole switches) since it would be switching the voltage to the gain stage rather than the signal itself? Should it still be a "shorting" type?
mediatechnology wrote:Why don't you use a conventional heatsink and add bling by having it thermally pump a Lava Lamp?
Hehe, that's a fun idea :) I've thought about water-cooling, but that would only make sense for much bigger amps - water is a great medium for dumping heat into, as high-end PC builders know... But having the heated water perform some kind of "work" is a novel idea! TBH though, I think something like the case below (311 x 260 x 70mm, or 12 1/4 x 10 1/4 x 2 3/4 inches, costs ~$50) is the way to go, and leave the crazy cooling ideas for later - I think I'll have enough on my hands just to get the amp itself working, and the money saved could be spent having the front-panel professionally engraved instead. Does it look to you as if the heatsinks on that case are substantial enough for this amp? Unsurprisingly the manufacturer(s) say nothing about the thermal conductivity...

Image
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 »

Prototype2 wrote: The prices even of unpopulated Goldpoint switches is shocking though; over $150 for a 2-pole model! Hopefully I'll be able to find something cheaper...
Rotary switches with any significant numbers of steps are disappearing from suppliers fast, there are very few to choose from. Quality switches with lots of steps cost a lot. There are a number of cheap Russian switches with many steps that DIYers are using. Labor value to reload and replace a low quality switch that failed is a losing proposition over using the expensive quality switch in the first place, both take a good bit of time to load up with resistors.
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 »

emrr wrote:Quality switches with lots of steps cost a lot.
How about these 23-step Dale attenuators, with 4-poles and populated with high quality resistors, for only ~$50?
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 »

raf wrote:Those switches are most definitely not Dale brand, they are cheap Uraltone Russian switches
Yeah, I realised this after posting, and I do understand that there are quality issues with these switches. I'm interested in the option of a having a variable gain input stage rather than a variable resistance in the signal path (attenuator), but I confess I know next to nothing about (pre) amplifier design and cannot assess whether this is a viable option. If it is, then the demands and complexity of a gain selector switch ought to be much lower. I also don't need that many steps since whatever signal I'll be feeding this amp will come from a precisely variable source (either my function generator or my 24/192 sound card) - 10 or 12 steps should be plenty. The reason I'm interested in a stepped control is more to do with wanting to be able to reliably repeat a certain level of amplification rather than a belief that it will "sound better".
raf wrote:Those switches are worth $15... Well, I say worth $15 but I doubt it...
$15 plus the work of soldering all those resistors! I don't know about you but I'm not sure if I would have the patience to do that...
raf wrote:the cheap ones were way closer to target values than the Dale ones.
If I was to engage in the masochistic exercise of building my own stepped resistor ladder then I might as well hand match the resistors as well :)

One option (at least with a variable gain stage) would be to get a PCB mounted 2-pole (or two 1-pole) rotary switch of good quality and put the resistors on the PCB instead.
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. »

Prototype2 wrote:
mediatechnology wrote:Why don't you use a conventional heatsink and add bling by having it thermally pump a Lava Lamp?
Hehe, that's a fun idea :) I've thought about water-cooling, but that would only make sense for much bigger amps - water is a great medium for dumping heat into, as high-end PC builders know... But having the heated water perform some kind of "work" is a novel idea! TBH though, I think something like the case below (311 x 260 x 70mm, or 12 1/4 x 10 1/4 x 2 3/4 inches, costs ~$50) is the way to go, and leave the crazy cooling ideas for later - I think I'll have enough on my hands just to get the amp itself working, and the money saved could be spent having the front-panel professionally engraved instead. Does it look to you as if the heatsinks on that case are substantial enough for this amp? Unsurprisingly the manufacturer(s) say nothing about the thermal conductivity...
I've looked into this and water never ever makes sense inside an audio power amp. It adds a level of complexity and potential failure modes that are unnecessary and very undesirable in that context.

You get a similar mechanism, but with better reliability from using sealed heat pipes where the working fluid (not water) is reliably kept inside where it belongs. I think some of those exotic heat sinks you posted pictures of before used fluid based heat pipes to spread heat from the hot end to the fins.

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:I've looked into this and water never ever makes sense inside an audio power amp. It adds a level of complexity and potential failure modes that are unnecessary and very undesirable in that context.
Tell that to the people who put water cooling systems inside their computer cases, which often contain parts that are orders of magnitude more valuable than some home-built amplifier... At least with an amplifier the heat sources are located in one or two accessible places, which would make it far easier - and safer - to cool them with water than for a computer where the water by necessity has to pass repeatedly in close proximity to the circuitry, and needs to have multiple connections in places where a leak would be catastrophical. If you want really crazy solutions, take a look at this :) But this is all academic, and the whole cooling issue is a massive side-track - I'm sorry I even brought it up!
JR. wrote:You get a similar mechanism, but with better reliability from using sealed heat pipes where the working fluid (not water) is reliably kept inside where it belongs. I think some of those exotic heat sinks you posted pictures of before used fluid based heat pipes to spread heat from the hot end to the fins.
They do indeed; heatpipes have become a pretty standard feature in most computer cooling systems. But again, you've already convinced me that a plain old aluminium heatsink is the best option, and I'm not going to consider the more exotic options at this point.
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. »

That stuff is just silly... But the audio business teaches us build it and somebody will probably buy it if it's exotic looking enough. :D

The dedicated chip coolers are actually decent thermodynamic design, which I why I used one in my Peltier experiment. Of course air is a poor thermal conductor which is why they add fans.

JR
Cancel the "cancel culture", do not support mob hatred.
Post Reply