Have you guys seen this...yet?

[Peter Daniel]

So, why are lots of people using 1000uF or even less?

Beacause "it enables to trace avalanches of fff." I'm not sure what that means, but definitely the freshness of sound is there

http://www.sakurasystems.com/products/47amp.html

 

[carlosfm]

So, why are lots of people using 1000uF or even less?

Forget it.
I use regulated PSU with big caps before the regs.

Unfortunately these chips, with unreg. PSU, only sound good with 1000~1500uf per rail/chip.
Believe me, increase the capacitance and it affects midband and treble.
But with this low capacitance bass always suffers, except with very sensitive and easy to drive speakers.

 

[Peter Daniel]

I use regulated PSU with big caps before the regs.
B]

I tried it too, but the freshness of sound was gone. No more regulated supplies for me. I might as well forget I ever used them.

 

[hifi]

a PSRR of >100db @ 100Hz also help it get away with those smal caps

/Micke

 

[jan.didden]

Beacause "it enables to trace avalanches of fff." I'm not sure what that means, but definitely the freshness of sound is there

http://www.sakurasystems.com/products/47amp.html

That's advertisement, that doesn't count.

Jan Didden

 

[jan.didden]

Forget it.
I use regulated PSU with big caps before the regs.

Unfortunately these chips, with unreg. PSU, only sound good with 1000~1500uf per rail/chip.
Believe me, increase the capacitance and it affects midband and treble.
But with this low capacitance bass always suffers, except with very sensitive and easy to drive speakers.

Yes, bass will suffer, less with sensitive speakers. I can also understand that midband and treble will be affected. The PSRR is less at higher frequencies, and the increased supply ripple contains increased high-frequency harmonics to the line frequency. And when you get to the output level limit, say with strong bass signal with superimposed mid/high you get 100Hz modulation of the output (because the supply, which determines the max output level, is modulated).
How can it sound better with more junk on the output signal?

Jan Didden

 

[jan.didden]

a PSRR of >100db @ 100Hz also help it get away with those smal caps

/Micke

Indeed, but it does nothing for high-level output modulation and the PSRR is less for the higher frequency ripple harmonics. How can that improve the sound?

Jan Didden

 

[Upupa Epops]

Jan, guys haven't quite clear in this. Try it little bit explain , but maybe it will be little hard, inertia wiews are sometimes great .

 

[carlosfm]

How can it sound better with more junk on the output signal?

Indeed, but it does nothing for high-level output modulation and the PSRR is less for the higher frequency ripple harmonics. How can that improve the sound?

Indeed it's technically very odd, I know.
I was very sceptic too, before making my first GC.
I made tests with capacitance.
You would have to try it and listen, as these chips are not "politically correct".

With regulated PSU, I use small cabs on the chip (33~100uf).
On the output of the LM338 regs there's only a 47nf film cap.
Before the regs it really doesn't make much difference the capacitance you have.
I use 10000uf.
The amp drives speakers that the unregulated low-capacitance one could only dream of.
And with a smaller trafo.

My speakers are not easy to drive, even by good commercial amps.
A friend of mine that sells hi-fi (and high-end) told me once that he didn't sell many Epos 11 (or 12) speakers because, although excellent speakers, it was difficult to pick an amp that goes with them.

 

[jan.didden]

Indeed it's technically very odd, I know.
[snip]With regulated PSU, I use small cabs on the chip (33~100uf).
On the output of the LM338 regs there's only a 47nf film cap.
Before the regs it really doesn't make much difference the capacitance you have.
I use 10000uf.
The amp drives speakers that the unregulated low-capacitance one could only dream of.
And with a smaller trafo.[snip]

Yes that makes sense, what can these regs give, 5 amps or so? At the input of the reg you would need just enough capacitance to prevent the reg to drop out at heavy passages, 10kuF sounds OK.

Jan Didden

 

[carlosfm]

Yes that makes sense, what can these regs give, 5 amps or so?

Well... sort of.
From the datasheet:

"The LM138 series of adjustable 3-terminal positive voltage regulators is capable of supplying in excess of 5A over a 1.2V to 32V output range.
The current limit circuitry allows peak currents of up to 12A to be drawn from the regulator for short periods of time. This allows the LM138 to be used with heavy transient loads and speeds start-up under full-load conditions."

Seams quite good for a class AB amp that plays music.

 

[soundNERD]

hmm, seems pretty good, parellel them and you could drive a pretty large amp with them.

Is there a negative version?

 

[jan.didden]

Well... sort of.
From the datasheet:

"The LM138 series of adjustable 3-terminal positive voltage regulators is capable of supplying in excess of 5A over a 1.2V to 32V output range.
The current limit circuitry allows peak currents of up to 12A to be drawn from the regulator for short periods of time. This allows the LM138 to be used with heavy transient loads and speeds start-up under full-load conditions."

Seams quite good for a class AB amp that plays music.

Indeed. 12 amps peak in 4 ohms with a supply of say 50V dc + and -, that's a healthy 200W++. Not bad for an integrated regulator.
Has anybody tried these regulators with a separate external, high speed etc error opamp?

Jan Didden

 

[carlosfm]

Has anybody tried these regulators with a separate external, high speed etc error opamp?

No, but tell us more about that.

 

[jamesjung21]

capable of supplying in excess of 5A over a 1.2V to 32V output range.

I thought 5A is possible only up to 25V....

And then the max current rating drops from 26V...

 

[carlosfm]

I thought 5A is possible only up to 25V....
And then the max current rating drops from 26V...

Where did you see that?
It may have escaped me...
What is important is that VIN-VOUT is less than 10V.
Like on this graph, on page 3 of the datasheet:

 

[dave1968]

Isn't this the same amp at the bottom of this page?: http://www.hilberink.nl/amps/lm3875schema.htm

Now that is a funny read.

That is funny. Hey Karma. They said your amp looks like spaghetti
MMmmmmmmmmmmm, spaghetti

well sure i was in a hurry new job and all. heh im a few hundred
away from the first piece of a mark levinson system i have had my eye on.

ware is the marble saw you ordered?