The (high-cap.) unregulated PSU for chipamps

Status
Not open for further replies.
mangrovejack said:
Originally Carlos mentioned using 100uF + 100nF at the pins of the positve and negative inputs into the chip, but now with this latest revision, he mentions 2200uF + 100nF at the pins. Now that's a substantial difference. My question is, would 1000uF (Kind of a compromise I guess between the two values) be ok?

1,000 to 2,200uf is fine.
Notice that this big jump from 100uf to 2,200uf is because these big caps are now snubberized.
 
Upupa Epops said:
Carlos says, Carlos uses, Carlos think, Carlos recomend ... Have you your own brain, guys, or are you only little kids ? 😀

Yes I do have my own brain, but what I don't have is time, money, resources and sufficient electronics knowledge to be able to experiment with the things that Carlos does.

I'm not in the electrical industry as my profession, and I've only recently started getting into building my own amplifiers etc for my own use.

Whilst I would love to have the knowledge of some of the guys here on this board, I don't, but am slowly learning.

When someone like Carlos (who obviously has a fair bit of knowledge in this area) (and others like Nuuk etc) decides to publicly share his experiences and knowledge with the rest of us, I am extremely greatful. The reason I ask him questions regarding some of the values comes back to my first sentence in that I don't have the necessary resources to try every value under the sun.

To be honest, Upupa Epops in a lot of the threads I've read (I read most threads on the Chip Amp forum, but only in lurker mode mostly), I notice that you come in, and either slam or insult someone on the board, but you rarely actually offer decent advice. Its obvious that you are very knowledgable in this area too, so instead of slamming people, how about you give some of your knowledge to people?
 
Dear friend, hifi scene is full of prejudices and DIY audio also. People are accustomed to think by pattern and sometimes is it lost fight, change thinking is hard thing. People like simply solutions and simply thruth, 'cos mostly haven't time, experiences and one's own will to find different solution. If you in this case can give him advice, you strike to the wall of noncomprehension 😎 and it is closed circle. Look around and you will see it everywhere, people are very conservative 😉 . Was some amp good in 69' ? It must be still good, think people, 'cos they haven't own experience and manytimes they had read, that is good. If you will come with somethig new, first question will be "... and will be it better, than this older one ? ". If you say " Yes ", in mostly cases they don't belive. Etc., it's all hard thing 😎 .
 
percy said:


So did you try this ?


Yes but with different values.

Did with:
18,000uF + 6,800uF + 4,700uF + 3,300uF + 1,000uF + 6x 100nF
for each rail.

Now I am working in a improved version 😀

But I can tell you that this PSU rock 😎
(punchy bass and lot of detail in high frequencys)

(put the caps like a downstair, from big capacitance to small capacitance)
 
I have tested with a STK chip and a Bipolar amp(modified Pioneer A-702R).
In both cases this PSU performed very well.

I am just waiting for my SSM2142 and Bridged LM3886 PCB.
Than I will test with the LM3886 chip.

In the end of this weak I hope to test a new version:

18,000uF + 100nF + 6,800uF + 100nF + 4,700nF + 100nF + 4x 1,000uF + 2x 100nF

Lets see how this one performs 😉




PS: I got better results than with the Snubber version 1.
( I Know, I must test the new version.... but I don't have time for all )
 
XELB said:
I have tested with a STK chip and a Bipolar amp(modified Pioneer A-702R).
In both cases this PSU performed very well.

That's what I suspected.:dodgy:

XELB said:
In the end of this weak I hope to test a new version:

18,000uF + 100nF + 6,800uF + 100nF + 4,700nF + 100nF + 4x 1,000uF + 2x 100nF

Lets see how this one performs 😉

What are you expecting from this bazooka? :clown:
Don't take me wrong, but just by looking at this it makes me laugh.😀 😀 😀

XELB said:
PS: I got better results than with the Snubber version 1.

With those amps.

XELB said:
( I Know, I must test the new version.... but I don't have time for all )

It's easy and fast to install a snubber, even if in a non-definitive way, and test.
Much faster than the compulsive "bypassing" you are doing there.
 
What are you expecting from this bazooka?

Have lower ESR than the snubber PSU and better performance in high frequencies :dead:




That's what I suspected

And so what ?
Did you thought that I would make a PSU to sound great with your amp ? 😛
I just care which one sounds better in my amp.
And as I told before, the first version I did sounded better than the snubber V1.0!



It's easy and fast to install a snubber, even if in a non-definitive way, and test.
Much faster than the compulsive "bypassing" you are doing there.
I have to buy the resistors... I don't have them at home.
I hope to buy them tomorow.

But will not test it with 4,700uF + 10,000uF + 10,000uF.
I don't like capacitors of big capacitance at front....
A 10,000uF capacitor just not performs as well as 10x 1,000uF at high frequencies!
 
XELB said:
Have lower ESR than the snubber PSU and better performance in high frequencies :dead:

It depends of the value of the snubber.

XELB said:
I just care which one sounds better in my amp.
And as I told before, the first version I did sounded better than the snubber V1.0!

Good for you.
But that snubber was tested with LM chips, so...

XELB said:
But will not test it with 4,700uF + 10,000uF + 10,000uF.
I don't like capacitors of big capacitance at front....
A 10,000uF capacitor just not performs as well as 10x 1,000uF at high frequencies!

Considering that 2x1,000uf per rail, bypassed with 100nf (and even with 10u -or not- on the chips) already start to muck up the sound on the LM chips, I don't see so much advantage in that.
You are missing a much better amp than what you have (with LM chips), with a very good PSU.😉

Try the LM3886 and you will smack that STK chip and the Pioneer with a hammer.😀
Take this advice, you are loosing your time with those things.
 
carlosfm said:



Try the LM3886 and you will smack that STK chip and the Pioneer with a hammer.😀
Take this advice, you are loosing your time with those things.


I know the value of the LM3886 chip 😉
A friend of mine have a Jeff Rowland with LM chips, and it plays very well.


Carlos, I am just waiting for my PCB's.
I have all the material, 4xLM3886, 2xSSM2142 for SE to Balanced signal, trafo, etc....

My Bridged LM3886 is based in your schematic 😛 But I have gone to ~20gain because the SSM2142 gives me ~6dB gain.


Hope to finish everything in a couple of weaks, if I don't burn out the house before that :dead: :clown:
 
PSUs per Channel; Cap values; Cap ratings; Snubber Values

Dear Carlos,

1. Embarassing question: are all the unregulated PSUs in this thread for one channel? Are the later ones (with higher capacitance for two channels?

2. One of the distributors over here has 15 000uF 35V Capacitors by Elna extremely cheap ($1.00 US each).

a) Is the voltage rating too low?
b) Should these be OK with your standard snubber values?

Regards,

George
 
Status
Not open for further replies.