Amp Camp Amp - ACA

I think I am going to do a amp sound blind test like I did with full range drivers. Class D TPA31xx vs class D IRS2092 vs class A ACA vs class AB Dx vs class AB LM3886. All same speaker, same mic position, adjust for same SPL at mic and post blind-labeled sound clips. It might be fun...

@xrk971... this comparison will be helpful for someone like me. Waiting for your impressions on these amps.

Sent from my Nexus 5 using Tapatalk
 
Multi channel setup

Hi all
I've been looking for options to power my active horn speakers and looks like this amp may be a good way to go.
Thinking about mounting 6 boards in a single enclosure, can i just setup a standard power supply to power them all or can i get a big enough plug pack to power them all?

Another question about the power supply, can i increase the supply voltage for say the bass amps and lower the voltage to reduce the power for the mid and hf amps?

Appreciate your thoughts.

Dave
 
Founder of XSA-Labs
Joined 2012
Paid Member
Even with the 2.2k mod I can't get more than 3v out of the outputs when using PC soundcard as source. With a pro mixer console as preamp I can get it to 8.05v before clipping. Is there any other resistor changes that can increase gain? 14dB is frankly a paltry amount of gain. 20dB to 26dB would be ideal.

When I made the change from erroneous 330ohm to 330k ohm resistor - gain went up. Would going to 680k increase gain?
 
diyAudio Editor
Joined 2001
Paid Member
nzlowie:

You can make a standard, linear power supply , with a transformer in which case I'd think about the 24v output option which would use an 18v transformer at about 350 VA. For 6 channels it might well end up cheaper also. Of course there would be some interaction between the channels with one PSU but if you overspec the PSU some it should be fine..

Other opinions?
 
Last edited:
Well ~22V would be okay, if it meets that condition at startup. However whatever voltage swing the jfet sees at R9, it sees more then double the opposite swing across R17 so there will be some miller effect. My guess is the amp will still work, but keeping the voltage at the drain of the jfet more constant would be preferred. The F5 Turbo v3 also has an example.

LBHjdu, thank you again for taking the time to respond and for endeavoring to enlighten me, a nearly futile task.

I read Nelson's F5T article, and I was able to understand some of it. Specifically, most of the words with 5 letters or less ;).

However, I was able to look at the pretty drawings, and I noticed that his F5T V.2 schem, which also has a 32V supply, does not cascode the input devices and instead has a mere 1K drain resistor for the input jfet. Perhaps the F5 deals better with the drain voltage swing? BTW, in my simulation of the ACA on Steroids the voltage at the Q4 drain, with a 3K resistor is ~19V with a ~0.35V swing.

I tried adding a ZTX450 cascode modeled after the F5T V.3 schem and I can't get it to work in simulation. I get virtually no drain voltage (~140mV). I'll try fiddling with the R values to see If I can get satisfactory results... will report back if I do.

At this point I'm inclined to just go with a single drain resistor, since I've already made the board and have begun to populate it.
 
diyAudio Editor
Joined 2001
Paid Member
NP feels that a lot of gain isn't optimal, and since lots of people have preamps it makes sense for them. Making a decent preamp might be a very worthwhile project for you as a DIY audiophile.


Even with the 2.2k mod I can't get more than 3v out of the outputs when using PC soundcard as source. With a pro mixer console as preamp I can get it to 8.05v before clipping. Is there any other resistor changes that can increase gain? 14dB is frankly a paltry amount of gain. 20dB to 26dB would be ideal.

When I made the change from erroneous 330ohm to 330k ohm resistor - gain went up. Would going to 680k increase gain?
 
nzlowie:

You can make a standard, linear power supply , with a transformer in which case I'd think about the 24v output option which would use an 18v transformer at about 350 VA. For 6 channels it might well end up cheaper also. Of course there would be some interaction between the channels with one PSU but if you overspec the PSU some it should be fine..

Other opinions?
Thanks
Do i need to make any changes if i alter supply voltage?

Cheers

Sent from my SM-T800 using Tapatalk
 
Founder of XSA-Labs
Joined 2012
Paid Member
NP feels that a lot of gain isn't optimal, and since lots of people have preamps it makes sense for them. Making a decent preamp might be a very worthwhile project for you as a DIY audiophile.

I have a JLH class A headphone amp/pre-amp in the works. I did look at the B1 but that has no gain. What would be a Pass class A pre-amp with gain?
 
We use power bricks in the kit, and they are surprisingly un-noisy. Is that "Papa Pass" magic? Possibly, but he does have an Audio Precision tester so he can actually measure this stuff ;)


My biggest concern about the bricks you linked to is there seem to be a lot of complaints about them in the reviews..

I don't have any measurable issues with the bricks. I think they are a great way of keeping it simple.

The only issue I encountered was a poor connection with the jack which has already been fixed in the latest update from what I understand.
 
The 2.2K mod does a couple of things that are not documented here. The main one is that it increases the open loop gain which lowers the output impedance and lowers THD. I measured the open loop gain at over 30dB with the mod. Closed loop gain is about 14.5dB with the stock feedback resistors. If R12 is increased to 150K, the closed loop gain can be increased to about 20dB and THD increases somewhat but is still about half that of the original ACA. The 24V power supply has minimal impact on open loop gain and THD. 12.5V seemed like a good set point for output bias Voltage.
 
Founder of XSA-Labs
Joined 2012
Paid Member
The 2.2K mod does a couple of things that are not documented here. The main one is that it increases the open loop gain which lowers the output impedance and lowers THD. I measured the open loop gain at over 30dB with the mod. Closed loop gain is about 14.5dB with the stock feedback resistors. If R12 is increased to 150K, the closed loop gain can be increased to about 20dB and THD increases somewhat but is still about half that of the original ACA. The 24V power supply has minimal impact on open loop gain and THD. 12.5V seemed like a good set point for output bias Voltage.

Thanks for the tip - this is exactly the type of tweaking I was thinking of to get 20dB closed loop gain. With this mod and 24v, I probably need bigger heatsink. I can do 6 seconds touch test at this point and smell the heat.