Help with Pultec/2520 frequency response?

Hey folks,

Attached is a schematic for a solid state Pultec I built. I'm not sure, but I've heard this was a drawing reverse engineered from an original solid state api 2520 amped Pultec. Don't quote me on it though.

Anyway, very hard to find info on this, so . . . .

My question is about a low frequency bump I'm getting. This is with it in bypass or engaged but with all filters rolled off. It's the same. It's also the same regardless of whether I use an output transformer or not.

I've tried changing the output capacitor and this moves this bump down below 20 HZ, but I'm wondering if it's still a problem.  I don't want to use up headroom with whatever is following the Pultec. Not sure where  the bump is coming from. I'm feeding the pultec from a 600 ohm output, so that's not the issue.

Here's frequency response with a 100uF output cap (didn't have a 150uF cap on hand)

And here's a 470uF cap.

The bump is now down below 20Hz, but it's still there.

resonance in the output circuit.
why you see a 47 ohm resistor in parallel with output capacitor in an API 325 card (later ones).
I believe Abbey has discussed this.

Is this with a Triad HS-56 output?

Thanks folks,

Here's the link I think gridcurrent is referring to: http://groupdiy.com/index.php?topic=40582.0

I'll give that 47 ohm resistor a try.

Rad. Okay, that evened out the resonance. I used 49 ohms because that's what I had.  Now the resonance is only around .5 dB bump and it's below 20Hz with a 470uF cap.

By the way, I'm using a JLM 111DC transformer on the output. http://www.jlmaudio.com/shop/jlm111dc-output-transformer.html But wait, that's a gapped transformer batman.  What happens if I omit the output cap since I have a gapped trasformer that can handle DC? (I guess I'd still need the resistor so I don't get that resonance?)