Scratching head about a linear PS for small Class A amp

Well, I think I need to re-approach this this with a lager power transformer (as suggested over on the Pass Labs forum where I posted originally) with 18v secondaries so I don't have to drop so much voltage. And a choke with much less dc resistance. PS filter caps. Oh and the bridge!

Other than that, I got most if it right!

This is the cost of the alternative I've used with success before.
Screen Shot 2023-08-04 at 4.30.01 PM.png


It just doesn't seem to be heavy enough. Not enough iron!.
 
FWIW: I don’t do class A (at all) but low power AB and D amplifiers with regulated linear PSUs with very good results. When using low power amplifiers with relatively low current demands things get a lot easier.

Your 3A continuously makes it a harder challenge. I don’t have any knowledge of Pass designs but the ACA needs 19V isn’t it? Anyway you want lower voltage drop and very low ripple voltage.
 
FWIW: I don’t do class A (at all) but low power AB and D amplifiers with regulated linear PSUs with very good results. When using low power amplifiers with relatively low current demands things get a lot easier.

Your 3A continuously makes it a harder challenge. I don’t have any knowledge of Pass designs but the ACA needs 19V isn’t it? Anyway you want lower voltage drop and very low ripple voltage.
The ACA originally came out with 19v, but the current version can take 24v. That is 24v at 1.5 amps continuously for about 8 watts per channel!

I have a 100 watt transformer, but sounds like I need 24v x 1.5A x 2 channels x 2, which is closer to 150 watts. Digikey has a 18v primary 180 watt Hammond for $80!!!!

Maybe I will buy a 35A rated Graetz bridge just to see what happens. But I still have the issue of the under rated choke.

We'll see.

Thanks,
Dan
 
Reality kicks in. When the new path has more drawbacks than benefits maybe one has chosen the non optimal path. Normally the followers follow the designers choices without any doubt so why rediscovering the obvious? Ah well, any choice is better than no choice 😉
 
Last edited:
  • Like
Reactions: lt_texan
Reality kicks in. When the new path has more drawbacks than benefits maybe one has chosen the non optimal path. Normally the followers follow the designers choices without any doubt so why rediscovering the obvious? Ah well, any choice is better than no choice 😉
Yes, it was an experiment to check out a CLC power supply filter. It has gone wrong, and there is really no reason to spend money to fix it.

But not able to completely let go, I have some heavier duty rectifier diodes in my parts bin (MBR735 7.5A), so I had to give those a try this morning. And that didn't even make a difference.

So now I can really throw in the towel.

I already have one of those Meanwell power supplies on the shelf, guess I need to spend $20 to finish the 2 ACA bridged monoblocks.

Thanks again,
Dan

IMG_2579.JPG


Screen Shot 2023-08-05 at 12.12.22 PM.png
 
MBR735 is also too tight with regard to maximum reverse voltage. Sorry.

Yes, in both cases the reverse voltage of the bridge diodes is too low and they will break down briefly on every cycle.
When the diode breaks down it conducts in reverse which lowers the voltage on the first filter cap and the rest of the output circuit.
The diode PRV needs to be a minimum of 2 x the peak unloaded transformer secondary voltage plus some headroom for AC line voltage fluctuation
In this case 2 x 22.1Vrms x 1.4 = 62V
Use 100V or greater diodes.
 
FWIW I also learnt to use twice the peak voltage when using Graetz bridges. No explanation given but probably to be safe in all possible conditions and scenarios. Diodes used to be so cheap overdimensioned ones were chosen anyway.

Let’s say that people that do like that never run into such issues 🙂
 
Last edited: