Cool simple "clean" CFA

[thimios]

OK, here comes the layout...

8.5 x 3 inch, double-sided. Not really simple any more
However - still pretty cool

Placing the order tomorrow... takes about a week to produce.
And then we'll see...

P.S. Big ones are actually under the board - 3D renderer just can't show them the right way Anyway, gives an impression.

Give us the pdf for home etching.

 

[vzaichenko]

Ah, Mr. Precise Etching, sorry I didn't do it right away - I will drop PDFs here later today

 

[vzaichenko]

PDFs

Hi Thimios,

Here go the PDFs in the following order:

1) Top traces;
2) Bottom traces;
3) Top silk.

I did not mirror anything, so let me know is you need more "views"

Cheers,
Valery

 

[thimios]

Hi Thimios,

Here go the PDFs in the following order:

1) Top traces;
2) Bottom traces;
3) Top silk.

I did not mirror anything, so let me know is you need more "views"

Cheers,
Valery

Ok i believe that is all i need,just post the schematic according to this.
PS is this according to post#21?
Thanks again.
Thimios.

 

[jwilhelm]

Is this one safe to run at a little higher voltage rails or is 52 around it's limit? Looks like a fun project!

 

[vzaichenko]

Ok i believe that is all i need,just post the schematic according to this.
PS is this according to post#21?
Thanks again.
Thimios.

No, this one is exactly per post #37

Cheers,
Valery

 

[vzaichenko]

Is this one safe to run at a little higher voltage rails or is 52 around it's limit? Looks like a fun project!

No, it's not the limit - I will actually run it at around 52-56 (got a couple of 1KW transformers ), but it can be higher.
Let me know, how much you need - I will check if some adjustments are needed.

 

[jwilhelm]

I have a 63 volt supply to test with. I might make up a lower voltage one but I prefer higher voltage amps.
Just a suggestion here but you've got a wonderful protection system designed. Ever thought of adding a couple pads to connect your overcurrent sensing?

 

[vzaichenko]

OK, just checked the parameters, for 63V rails:
R5, R8 = 63K
R38, R39 = 8.2K

All the rest is perfectly fine - no adjustments required.

Overcurrrent sensing - you're right, however, the PCB is rather packed. In a few other designs I just soldered a 2-pin plug right on top of the big resistors
Although, I will think where to place it in the refined version of the PCB...

 

[jwilhelm]

OK, just checked the parameters, for 63V rails:
R5, R8 = 63K
R38, R39 = 8.2K

All the rest is perfectly fine - no adjustments required.

Overcurrrent sensing - you're right, however, the PCB is rather packed. In a few other designs I just soldered a 2-pin plug right on top of the big resistors
Although, I will think where to place it in the refined version of the PCB...

Yes that board does look kind of full. Thanks.

 

[vzaichenko]

Thimios, wait! Jeff's question has inspired one good idea I wanted to avoid CCSs for the input pair, but using them will make the amp more tolerant to the rails variation / fluctuation. I have found an elegant way of implementing them.

Schematic and PCB will be slightly updated!

 

[jwilhelm]

Sounds like this cool, simple clean amplifier is becoming even less simple.

 

[vzaichenko]

Just a very little bit less (6 parts added). But now it's virtually rail voltage independent
At least in the range +/-45V to +/-70V no adjustments required.
It's also more tolerant to rail "sagging" caused by high power consumption or mains fluctuations

 

[vzaichenko]

Version 02

Here goes v02.

I have also added the "BIAS" plug for overcurrent sensor - in the middle of PCB. Size is the same, just "cleaned up" some space

Have fun!

 

[jwilhelm]

Has anyone tested this amp?

 

[vzaichenko]

Has anyone tested this amp?

Hi Jeff,

I have just received the PCBs - 3 minutes ago I believe I will be the first to try

Cheers,
Valery

 

[jwilhelm]

Okay. I'll wait to order boards until you test then.

 

[vzaichenko]

Yep

 

[kimschips]

Hi all, anything news? I am exited to hear from Valery post 56

 

[vzaichenko]

Slowly but surely

Working hard here
Very close to testing. Arranging appropriate local heatsinks, changing the PSU on my test bench, so... most likely - tomorrow