just double value of output capacitance
in future , you can calc it by yourself :
F=1/(2 x Pi x R x C)
Yep the ACA-heatsinks should have at least 0.6 c/w or 0.6 k/w if you prefer Lord Kelvin over Mr. Celsius.
That said I have a few questions. I'd perhaps like to put two ACA-boards on a single large heatsink in an upright configuration. Fins to be vertically oriented of course.
How far apart would the boards need to be?
Depending on the width of the heatsink a staggered arrangement could be feasible. Is there any minimum distance the transistors should have from the edge of the heatsink?
For aesthetic reasons I'd prefer a square shaped heatsink - given enough thermal capacity and distance the geometry shouldn't matter or am I overlooking something?
Isn't that the formula for calculating the rolloff frequency of the CR Filter at the output. Doubling the capacitance lowers the rolloff from 0,8Hz to 0,4Hz. So far so good but what does that mean for the output impedance?
Sorry if the question might be extremely stupid .... I'm still struggling a bit with the whole concept of impedance....
Merijn. I read all the posts in this blog from about page 300 to the current page. I may have missed something of course, but as far as I recall there was not a single determination or estimate of the maximum voltage a PSU can supply to the ACA without problems. Of course if you want to increase the voltage you will need to change some of the components, and you will need a larger set of heatsinks to deal with it.
Thanks OC11!
I guess for now I just stay at using a powerbrick, or digging out my old ZEN PSU, since it got 2x18V secondaries @ 120VA. With a little rewiring it gives me ~2x25V dc. Bringing it down to 24V from 25V might not even be neccesary...
btw, I started yesterday evening with the reading session.
Last edited:
impedance = impedanz = wechselstromwiderstand.
modern opamps like 2134 or sowhat can drive 600 ohms - do not worry about the 10k input impedance of the aca.
in the formula you have to insert the value of the cap at the output of the preamp (C=2uF) and the input impedance of the aca (R=10k)
(simplifikation: ignore the res to ground at the output of the preamp);
the description of the preamp only means: if you have an amp with 22k input impedance and 2uF at the output of the p88, you will get a roll off at 0,8Hz (-1dB)
Making it twice as big would mean it would ideally have one half the impedance. The series cap does add to the output impedance of the preamp. You would typically size the cap so that the impedance is very low at audio frequencies so that there are no practical consequences due to the impedance of the cap.
I have seen problems with very small output caps(.1uf) causing hum pickup in long cables due to the high output impedance, but it is not typically a problem.
Thanks mjf, my fellow Austrian!
So let's see if I get this right.
1.) Opamp output impedance is usually very low. So a 10k input impedance of the following stage should not pose a problem
2.) The 100k output resistor and the 10k ACA input impedance will be in parallel, meaning that the 100k output resistance will not really make a difference in the calculation. (100k in parallel with 10k = 9.09k)
3.) In the "original" configuration of my preamp, 100k would be in parallel with as little as 22k. Using the Formula for a CR Filter (2uF cap) that would result in a rolloff point of 4.4 Hz.
4) The configuration of my preamp with 10k ACA impedance would result in a rolloff 8.7 Hz. Doubling the cap value to 4uf will bring the rolloff frequency back to 4,4Hz.
Long story short, Zen Mod was right ....as always. I just needed a bit longer to wrap my head around it. Thanks fellas!
So let's see if I get this right.
1.) Opamp output impedance is usually very low. So a 10k input impedance of the following stage should not pose a problem
2.) The 100k output resistor and the 10k ACA input impedance will be in parallel, meaning that the 100k output resistance will not really make a difference in the calculation. (100k in parallel with 10k = 9.09k)
3.) In the "original" configuration of my preamp, 100k would be in parallel with as little as 22k. Using the Formula for a CR Filter (2uF cap) that would result in a rolloff point of 4.4 Hz.
4) The configuration of my preamp with 10k ACA impedance would result in a rolloff 8.7 Hz. Doubling the cap value to 4uf will bring the rolloff frequency back to 4,4Hz.
Long story short, Zen Mod was right ....as always. I just needed a bit longer to wrap my head around it. Thanks fellas!
It’s V1.6 as stated in my post. Apparently, I somehow accessed the V1.5 guide by accident and thought I would have to build the resistor jumpers to use the amp in monoblocks mode. I understand now that I merely have to flip the toggle switch up and plug the rca input and speaker cables in as depicted in the V1.6 guide, step 55. What a dummy! Thanks for your response.
I am not sure if this is correct or not, but you may be fixing the power transistors in a single board to the same cooling fin of the heatsink. It might be preferrable for heat dissipation to fix the power transistors to separate fins.
I was thinking of running 2 metal bars down the side or maybe doing like I have seen some other builders do, and use a metal side, and then cut a hole for the parts that need to attach to the heat sinks, and attach them directly to the heatsinks. I also saw some beautiful cases for almost $400 that I would love to have, but I need 3! I have not had a problem with eBay transactions so far Alliexpress I am not too sure of as i have heard that sometimes it takes a while to get your money returned, but that you do get it back.
Last edited:
Thanks for your advice. I'd rather tweak the preamp than the ACA as the preamp is still sitting on a prototype breadboard.
But just to be sure what you mean .... could you let me know the number of the resistors you replaced?
Best explanation location and purpose I’ve heard
Thank you MEPER that was the best explanation I’ve heard today using the amp camp amp circuit board resistor numbers location in order of purpose. The lightbulb just went on for me I had a aha moment .
Thank you MEPER that was the best explanation I’ve heard today using the amp camp amp circuit board resistor numbers location in order of purpose. The lightbulb just went on for me I had a aha moment .
My measurements are quite a bit off (for the faulty channel):
Q1 S 0ohm
D 5,23k
G 629 ohm
Q4 D 670 ohm
G 328 k
S 521 k
Q3 C 10,56k
B 1,29 M Ohm
E 1,29 M Ohm
Q2 G 11k
D 536 ohm
S 1,313 M Ohm
Q1 S 0 volt
D 0 volt
G 0,44 volt
Q4 D 24 volt
G 0,02 volt
S 0,45 volt
Q3 B 21,20 volt
C 24 volt
E 21,22 volt
Q2 G 24 volt
D 24 volt
S 21,23 volt
The left (working channel) is much more as the values in the checklist:
Q1 S 0ohm
D 5,45k
G 560 ohm
Q4 D 670 ohm
G 340 k
S 0,8 k
Q3 C 10,5k
B 6,12 k
E 5,2 k
Q2 G 10,65 k
D 529 ohm
S 5,50 k
Q1 S 0 volt
D 11,89 volt
G 4,50 volt
Q4 D 24 volt
G 4,32 volt
S 4,48 volt
Q3 B 12,40 volt
C 17,26 volt
E 11,94 volt
Q2 G 17,21 volt
D 24 volt
S 12,77 volt
The main reason for me to have higher input impedance was that I started using a passive preamp with ACA. I used a 10k attenuator. The output impedance of the source has to be small compared to the input impedance of the ACA. If not the amount of NFB will change a lot by volumen setting. I found out that passive preamp and ACA is not a perfect match so now I use active preamp but still keep R11 and R12 as it seems to work fine using higher values. I use 0.1% resistors for R11 and R12 to have exact same gain in both channels. When I just order a few resistors…..then why not....