The (high-cap.) unregulated PSU for chipamps

[XELB]

I made a few tests with my unregulated PSU.

I could not notice de differences between snubber (1 Ohm + 100nF) and just 100nF near the chip.
I think I have to test the new version....... 0,47 Ohm resistor. 🙂

 

[peranders]

I could not notice de differences between snubber (1 Ohm + 100nF) and just 100nF near the chip.

I doubt that anyone else can..... in a blind test but it matter if you feel it is a good thing.

 

[carlosfm]

I made a few tests with my unregulated PSU.
I could not notice de differences between snubber (1 Ohm + 100nF) and just 100nF near the chip.

I never recommended those values on the chip.
Btw, what electrolythic caps are you using on the chips?

 

[carlosfm]

Still driving him nuts?

Carlos, I still can't understand why you have the C7, C8 becuase you form a resonance circuit with those and C3-C6.

Don't worry, I like my music with all the harmonics and resonances.😎
I like to hear the sound of a string of a (good) acoustical guitar resonating for a long time, like it does in real life.

 

[XELB]

I never recommended those values on the chip.
Btw, what electrolythic caps are you using on the chips?

Sorry! I did not explain correctly.
( I wrote on a rush )


I was trying to say, instead of using only the snubber, I use 100nF in the PSU and 100nF near the chip. With this configuration I get the same result.

Later, I will try a different snubber value on the PSU, 0,5Ohm + 100nF and compare it. 😉

I am also working in a new unregulated PSU.
( 10,000uF + 6,800uF + 4,700uF + 2,200uF + 100nF, for each rail )
I am just waiting for some new caps…





Ps: I used Rubycon 6,800uF caps

 

[carlosfm]

I was trying to say, instead of using only the snubber, I use 100nF in the PSU and 100nF near the chip. With this configuration I get the same result.

No electrolythic on the chip?

 

[XELB]

No electrolythic on the chip?

Ps: I used Rubycon 6,800uF caps

😉

 

[carlosfm]

Too much.
Try 1.000~2,200uf NEAR the chip's pins.
As on my schematic.

 

[XELB]

Carlos, I not using a LM3886 chip for testing!
I am doing these tests with an STK4152II.
Maybe this have something to have with my results.

I am just finishing my LM3886 Bridged PCB, then I can really try as you told.
But for now I am having the same result only with 100nF caps.


BTW, have you tried only with the 100nF in the PSU and 100nF near the chip ?
Have you notice any differences ?

 

[carlosfm]

BTW, have you tried only with the 100nF in the PSU and 100nF near the chip ?
Have you notice any differences ?

Oh no, not again.
Of course I did.
Read this thread, man.

 

[XELB]

Oh no, not again.
Of course I did.
Read this thread, man.

That's amazing because I could not notice any differences!
But this could be due to the STK4152II chip 😉

 

[PKS]

paralleled opa541.

Hello.

I just wondered if anybody could help me:
Im building a mono power amp using two paralleled opa 541.
any recomendations(sorry my poor english! I am a norwegian)
sould i use one as a slave?
Or two paralleled circuits with the same gain?

And... Recomended gain at output and gainstage? ca 8 and 3?
recomended opa in gainstage?
any buffer?

Regards

PK.

Sorry, wrong thread. But im not getting any answers other places.

 

[carlosfm]

That's amazing because I could not notice any differences!
But this could be due to the STK4152II chip 😉

That thing doesn't have the bandwidth and speed to be affected by a normal high capacitance PSU.
It's natural that you may not notice anything when you improve the PSU, but it's also natural that many details in the music escape you unnoticed with that [you name it] hybrid chip.😀

 

[XELB]

That thing doesn't have the bandwidth and speed to be affected by a normal high capacitance PSU.
It's natural that you may not notice anything when you improve the PSU, but it's also natural that many details in the music escape you unnoticed with that [you name it] hybrid chip.😀

Yes, when I compare it with my discrete amp, the STK lacks of detail.

Now I am building a Bridged LM3886 monoblock to compare it with other (modern)STK I bought. ( I will use the same PSU )
Based on this test I will build a power amp and use the spare parts for a mini amp.



By the way, do you Know LeiriTronica ?
http://www.leiritronica.com/website/index.php?id=8

They have a great online catalogue and the prices are good 😉

They sell the LM3886 for 3,12€ + VAT 😎

Just need to register to see the prices.

 

[carlosfm]

By the way, do you Know LeiriTronica ?
http://www.leiritronica.com/website/index.php?id=8

Yes, I've heard about it, a friend from Figueira told me about them.
I never bought anything from them, though.

 

[Padel]

What if....

Hi all,

Sorry if this is the second time that a question like this is looking for an answer, but there are to many posts written about ChipAmps Power supplies and I'm somehow lost!!!

So... we want a low capacitance power supply (1000-1500uF) to enjoy that midrange-treble ''magic'' that made famous all those ChipAmps (LM3875,LM3886,LM4780) but.... we want a high capacitance power supply (~10000uF) so that the supply (voltage) rails does not collapse when the amp has to ''output''
low frequencies, in short, when the output voltage must go up, the supply voltage goes.......down.

The conflict is ofcourse that when you add more capacitance that midrange-treble ''magic'' goes away, and here is where the zobel network comes in to ''fix'' that problem. Please correct me if I'm wrong.

My question is, can I use a high (10000uF) capacitance PS (parallel LM4780) amp for the woofer and low (1500uF) capacitance PS (also parallel LM4780) for the midrange and tweeter section? Do i need a zobel?

All these in a tri-amp speaker system ofcourse.

Padel

-------------------------------------------------------------------------------
Alcohol is the answer but...... I can't remember the question!!!

 

[carlosfm]

Re: What if....

The conflict is ofcourse that when you add more capacitance that midrange-treble ''magic'' goes away, and here is where the zobel network comes in to ''fix'' that problem. Please correct me if I'm wrong.

Yes, that's it.
You need very low impedance up to very high frequencies, something that the big caps can't give you.
Paralleling smaller caps is better but not THE solution.
Bypassing doesn't have the desired effect.
Only the snubber solves the problem.

My question is, can I use a high (10000uF) capacitance PS (parallel LM4780) amp for the woofer and low (1500uF) capacitance PS (also parallel LM4780) for the midrange and tweeter section? Do i need a zobel?

All these in a tri-amp speaker system ofcourse.

Try it.
For the midrange I would still use more capacitance.
Snubberized 4,700uf would do.

Btw nobody has yet tried the latest PSU.
Only NealG has tried the regulated.

 

[Padel]

Hi Carlos, thanks for pointing the right direction. Can you please give me the link where the latest version of the Snuber PSU is?
So in my case (High-cap PS only for the woofer) there's no need for the snubber circuit?

Padel

-------------------------------------------------------------------------------
Alcohol is the answer but...... I can't remember the question!!!

 

[carlosfm]

Hi Carlos, thanks for pointing the right direction. Can you please give me the link where the latest version of the Snuber PSU is?

http://www.diyaudio.com/forums/attachment.php?s=&postid=641774&stamp=1116027333

Btw R3 and R4 can be 1R, that's what I'm using now.

So in my case (High-cap PS only for the woofer) there's no need for the snubber circuit?

No, no, no, you need the snubber.
You will immediately notice tighter bass with the snubber.
The improvement is top to bottom, that's what you gain with a low impedance PSU.😉

 

[methar]

http://www.diyaudio.com/forums/attachment.php?s=&postid=641774&stamp=1116027333

Btw R3 and R4 can be 1R, that's what I'm using now.

How many voltage dropped across this R3 (R4) in case of using 1 Ohm value? As V=IR, so I guess in case I use 25Vac, the output os PS would be slight lower than 35Vdc ( e.g 33-35 depends on the load). Correct?