self split and feedback

[Andy Le Blanc]

A bit more bench time today with a 50w self split push pull amp.
I made an assumption and treated the thing like a simple SE amp.
I sourced some feedback from the OPT secondary and applied it to the
un-bypassed cathode of the 6j5 that drives the PP circuit through a 250k pot.
reducing the series resistance and increasing the feedback, being positive
feedback, eventually created instability and oscillation.

This phase inversion like any other long-tail inverter is based on a un-bypassed
cathode resistor shared by both sides of the inverter, in this case the power tube cathodes
of the PP circuit, there's a lot of inherent negative feed back, that isn't necessarily desirable.
Most cathode coupled inverters cancel even order "musical" harmonics.

So I used the positive feedback to increase the subjective gain of the amp
to match the gain the amp had when the cathode resistor of the 6j5 was fully bypassed
This was still well below the point of instability.

the amp sounds more open, the harmonic color is better.
Although the amp is not clean I don't think its a problem, I'm continually surprised
at the harmonic content
I noticed the low end was not as tight as I like, a bit more to go with it.

[chopstuck]

Andy, I bought four old Davis 106 amps on ebay (sounded like a good idea at the time). I gutted one an built a "Dirty 18 " in it but should really try the Paul Ruby zener mod. It sounds edgy-non musical.

I tried just cleaning another one up and trying the old circuit. So I replaced the E-caps and mounted a 1/4" phone jack in the Mic spot. Needed a 33k plate resistor on the input to tame it a bit. I don't know about using the second triode where it is. I disconnected the phono input and just have the pot there.
The self-split sounds OK at lower volumes but this amps sputters as you turn it up. I was wondering if I can get a good tone out of this configuration. I have enough gain. I was going to use the other half of the 6EU7 triode for a cathodyne but if I can make it sound good with the self split I can use it for something else. I measured 0.5vdc on the driven grid after the blocking cap.
This seems odd. Is it ?

[roberto]

the amp sounds more open, the harmonic color is better.
Although the amp is not clean I don't think its a problem, I'm continually surprised at the harmonic content
I noticed the low end was not as tight as I like, a bit more to go with it.

You could try to remove bass content from PFB, this should improve the situation.

[tictac]

Make the feedback adjustable by replacing the feedback resistor with a 25k pot with a 2k resistor in series; you'll be able to adjust the low end response and should be able to tighten up the bass a bit. I did this with a pair of Dynaco amps and it really works well...

TT

[chopstuck]

Tictac, are you suggesting to increase the nfb by decreasing the resistor value - 27k max vs 35k?

Would decreasing the 0.1 "black cat" decoupling cap to .05 or .022 have a advantage by decreasing bass content to the driven grid ?

I would like to keep it self split - just clean up the sputtering OD tone.

[chopstuck]

Well I just cleaned up and replaced all the e-lytics, all the ceramics and used some tried and true preamp resistor values and the amp shines.
Couldn't tell it's self split until you "cross the line" to the unbalanced PI thing.
Which also can sound really nice.

I just have volume, tone and an adjustable NFB on the front panel.
I'm labeling it "HAIR".


I don't think the Davis design group had this in mind when they built it.

You should see people's faces when this thing warms up.

It's all a lot of fun.

[chopstuck]

Andy, I love my little Davis and was thinking of a 6L6 or EL34 based self split version of a tweed TV pro ( 5C5 I think) off the top of my head...

Have you made any headway with yours ?

I would really like to see your schematic.

I guess by the time the high powered amps came out self split was obsolete.
Seems everything that old is new again.

I have a theory about why this particular model amp sounds good.
I have yet to test it.

[Andy Le Blanc]

that davis is just about what I did, few differences, I used a vol/tone like early fender deluxe, and I tried the much simpler single shared cathode resistor and ground the control grid of the inverting side, like a moonlite?...
and I did it with kt 88... gonna have to get back to it soon, fun amp for what it does, like zen amp, as simple as possible