Amp Camp Amp - ACA

Jump to Latest
Attila I

Attila I

Member
Joined 2009
hope this link does it
IMG_4856.jpg
 
K

kiven chen

Member
Joined 2012
Loudthud said:
That looks like it would work. Find values for the resistors by experimentation.



Zener protection is possible. There is a schematic in the thread I posted a link to above.



The R1-R2 to R3-R4 ratio, as well as their absolute value, (and the transconductance of Q2) determine how Q2 assists Q1 in supplying current to the load. Those resistors also influence the open loop gain and the distortion profile. IMHO the R15 mod is the best way to go, but seemingly little changes to the ACA change many other things.
Click to expand...

I think the gain is determined by R12R14,
R1~R4 small resistance helps to reduce output impedance, however constant current power supply derived from the R1~R4 sample, cannot readily be reduced.
Therefore, I think that needs more electricity, reduction of R1~R4 resistance was a better approach. Sampling R8R15 pressure better voltage.
 
Loudthud

Loudthud

Member
Joined 2007
kiven chen said:
I think the gain is determined by R12R14,
R1~R4 small resistance helps to reduce output impedance, however constant current power supply derived from the R1~R4 sample, cannot readily be reduced.
Therefore, I think that needs more electricity, reduction of R1~R4 resistance was a better approach. Sampling R8R15 pressure better voltage.
Click to expand...

Those resistors also influence the open loop gain and the distortion profile.
Click to expand...

R8-R15 control the closed loop gain, a totally different thing.

As Nelson suggested in the ACA article, I connected a series of .1 ohm resistor to replace R1-R4 with a switch to select a tap and listened to the ACA with the feedback (R12) disconnected. That's how I know it changes the open loop gain. Furthermore I found that when R3-R4 is effectively 0.82 ohm and R1-R2 is effectively 0.15 ohm, (R15 needs to be adjusted to set the quiescent current) open loop distortion is greatly reduced, on the order of 1% at 1W with the feedback disconnected. I posted a graph of this earlier in this thread. This is what leads me to the conclusion that lowering R3-R4 would increase distortion because a smaller signal would be driving Q2's Gate through C4. I can't say it sounds any better, but measured distortion is lower.

Nelson wrote an excellent article on feedback and distortion. Link: http://www.firstwatt.com/pdf/art_dist_fdbk.pdf
 
K

kiven chen

Member
Joined 2012
Loudthud said:
R8-R15 control the closed loop gain, a totally different thing.

As Nelson suggested in the ACA article, I connected a series of .1 ohm resistor to replace R1-R4 with a switch to select a tap and listened to the ACA with the feedback (R12) disconnected. That's how I know it changes the open loop gain. Furthermore I found that when R3-R4 is effectively 0.82 ohm and R1-R2 is effectively 0.15 ohm, (R15 needs to be adjusted to set the quiescent current) open loop distortion is greatly reduced, on the order of 1% at 1W with the feedback disconnected. I posted a graph of this earlier in this thread. This is what leads me to the conclusion that lowering R3-R4 would increase distortion because a smaller signal would be driving Q2's Gate through C4. I can't say it sounds any better, but measured distortion is lower.

Nelson wrote an excellent article on feedback and distortion. Link: http://www.firstwatt.com/pdf/art_dist_fdbk.pdf
Click to expand...

Thank you so much
But I need not increase power
The power I needed was comparatively small.
And how to do it.
I do with the chassis cooling, no special radiators.
About 0.5A current 9V voltage is enough.
Please have a look at #1543.
Look forward to your advice.
 
M

mboxler

Member
Joined 2011
Hi

I did the resistor mod (2K), and fired up the amps. Voltage re-adjustment was minimal, if any. Does that sound reasonable?

However, I continue to have a "popping" issue (both amps, 10 to 15 seconds apart). This happens with a passive tube preamp, and last night hooked up directly to USB DAC (Windows 7). This doesn't happen with any other amp. Any suggestions?

Thanks, Mike
 
B

bcmbob

Member
Joined 2009
mboxler said:
...and last night hooked up directly to USB DAC (Windows 7). This doesn't happen with any other amp. Any suggestions?

Thanks, Mike
Click to expand...

Can your windows player bypass the audio drivers? JRiver has that capability. I get a similar occasional pop when using optical feed cause the toslink plug fits loose in it's motherboard socket. The pop is the recovery/rewrite from the data glitch. A better USB cable might help.
 
Loudthud

Loudthud

Member
Joined 2007
kiven chen said:
Thank you so much
But I need not increase power
The power I needed was comparatively small.
And how to do it.
I do with the chassis cooling, no special radiators.
About 0.5A current 9V voltage is enough.
Please have a look at #1543.
Look forward to your advice.
Click to expand...

The problem that I see in the schematic you posted in relpy 1542 is that the pot (P1) is connected (through a resistor) to the 9V supply. (I'm not sure if you want to leave the LED in the circuit.) This would make the current of Q1 adjustable, but not regulated. (See note 1 below) The current in Q1 would change when the supply voltage changed or as Q1 and Q4 change temperature. Even if P1 is connected to the Drain of Q1, the voltage at the Drain changes only slightly because of the low resistance of the inductor.

I don't know what the Gate voltage of the Q1 you are using would be and if it would bias correctly with Q4, but you can probably adjust R9 to make it work. It may help to substitute an enhancement mode MOSFET for Q4. The Gate voltage of Q4 needs to be positive for the circuit to work.

Note 1: In the ACA Q3 sets Q2's current. The pot P1 senses the voltage at the Drain of Q1 and tightly controls Q1's current to match what is flowing in Q2 and maintain about 10V at the Drain of Q1. If the supply voltage changes, The current in Q2 does not change and the Drain voltage of Q1 changes very little.
 
K

kiven chen

Member
Joined 2012
Loudthud said:
The problem that I see in the schematic you posted in relpy 1542 is that the pot (P1) is connected (through a resistor) to the 9V supply. (I'm not sure if you want to leave the LED in the circuit.) This would make the current of Q1 adjustable, but not regulated. (See note 1 below) The current in Q1 would change when the supply voltage changed or as Q1 and Q4 change temperature. Even if P1 is connected to the Drain of Q1, the voltage at the Drain changes only slightly because of the low resistance of the inductor.

I don't know what the Gate voltage of the Q1 you are using would be and if it would bias correctly with Q4, but you can probably adjust R9 to make it work. It may help to substitute an enhancement mode MOSFET for Q4. The Gate voltage of Q4 needs to be positive for the circuit to work.

Note 1: In the ACA Q3 sets Q2's current. The pot P1 senses the voltage at the Drain of Q1 and tightly controls Q1's current to match what is flowing in Q2 and maintain about 10V at the Drain of Q1. If the supply voltage changes, The current in Q2 does not change and the Drain voltage of Q1 changes very little.
Click to expand...

R9=330 now, forgot to change the parameters in the figure.
Q1 gate voltage 1.78V.
Because adding a load switch, inductive voltage drop is different than Q2, P1 connect LED on.
 
A

Alistair72

Member
Joined 2007
A New ACA is born

Hi All,

I built my kit over Xmas. Loving it.

What does the increased bias bring? I will need to purchase some resistors to make the mod, but right now I'm not sure it is worth it.

It's fronted by a SB3 into MF V-DAC with a pair of Frugal horns (Mark Audio CHPs). The ACA replaced a pop pulse 2020 which I thought was doing a good job, but there is more bass oomph and generally smoother sound.

Cheers for all the posts and build guides. It took about 6 hours over a few days. Would have been a pain without all the help.

Festive cheers,

Al
 
You must log in or register to reply here.
Top Bottom