[balou]

Hi,
I want to use a SE5534A (or NE5534A if the SE edition is not available) in a new amp design. The ground channel needs to be unity gain, and without compensation the 5534 series is only stable at gains 3 and higher.
The AD744 also got a compensation pin, and isn't unity gain stable too without compensation. It needs at least 5pF to be stable.

So my question: How can I figure out how much pF I need to make the 5534 stable?


And yet another question: the 5534 and AD744 have different DC offset nulling schematics. Both use pin 1+8 for compensation, but in the schematic from the AD744, the trimpot is connected to pin 4 (V-), and in the 5534 it's connected to pin 7 (V+). The values also differ, the AD744 has 2 MR trimpot and a 1 MR resistor, whereas the NE5534 has a 100kR trimpot and a 22kR resistor.

Are those two offset nulling just two different ways to archieve the same, or are the nulling circuits in those two chips incompatible with eachother?

 

[balou]

In case you wonder about my amp design, here's a preview:

fits in a altoids tin
and no, this is not an inconspicuous bump

 

[majkel]

Looks great! What kind of dscrete buffers are you using? I see some ceramic caps... Do you decouple supply voltages with them to ground or V+ to V-? If there is a virtual ground channel you shouldn't decouple any supply voltage to this but connect them in parallel to the op-amp supply voltage pins.

 

[balou]

it's a monofied sijosae buffer (chosen because of it's small size)

yes, as of now I decouple them to the virtual ground. hm.. you mean to connect them to V-/V+ instead of connecting to ground? this would make sense somehow, to keep gnd free of interferences. the buffer is already only decoupled v- to v+.
oh and I should make full use of the two layers... I'm not going to home etch a board with hundreds of holes anyway...

 

[NelsonVandal]

This is really something! A cool competitor to Mini3. Will this fit in a mint tin with 2 9V's? I was planning on building something like this on a perfboard with 3 AD829's and BD139/140 or MJE243/253 instead of BC327/337, and maybe BC550/560 as input transistors. It seems like there's room for JFET's to isolate the rails, and then it would almost be in the same league as PPA v2 and LISAIII - in a tin! Why not use SMD components (except for opamps) - they're easier to solder and takes less space and could be laid out on both layers.

If you're planning on selling bords you can count me in for a couple.

AD829 is the only opamp I've heard so far without irritating colorations (but I've only used it in ground channel so far).

 

[balou]

General discussion has been moved to this thread:
http://www.head-fi.org/forums/showthread.php?p=3112536
This is also where NelsonVandals questions are answered.

So, anybody got an idea how much compensation is needed for the NE5534 to be unity gain stable?

 

[jcx]

its not hard to find 5534 datasheets which indicate that 22 pF is used for unity gain comp, with a buffer in the loop you may need to increase that

AC compensation circuits, values differ between external comp op amps - no "universal" solution for rolling

there is no "univeral" offset compensation circuit for use between differing part's with os trim pins either, rolling would require jumper/sw selecting different trim arrangements appropriate to each op amp

also the 5534 is an older process op amp with slow pnp - the output really benefits from a ccs pull down

 

[balou]

jcx: thanks - I've read through the datasheet, but the mentioning of 22pF is rather implicitly buried in the technical specs of the TI datasheet, and even more buried in the OnSemi datasheet.

about the trimming - also thanks, I guess if I include a DC comp circuit at all, it will be just for the 5534, I don't think the AD744 or AD829 need much compensation

CCS pulldown: as far as I know, the output on the compensation pin is already in class A. please correct me if this is wrong