• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Has anyone used an opamp to split phase for a tube amp?

I just threw together one and it looks great on the scope. Almost no phase shift and pretty linear, provided you keep it at 2:1 gain. Seems like you could put a tube driver stage after it and drive the outputs. Then use global feedback. I know it would be a hybrid, but would that be a bad thing.

It has far less phase shift than a 12AX7 stage.

Has anyone tried this?
 
Here's an example. I built one amplifier channel on a breadboard four years ago and haven't done any more, but it performs surprisingly well. Output power is about 30W with a regulated PSU.


That's really interesting, solid state amp with tube outputs. How does it sound on speakers?

Mine would be simpler than that. Just a tube stage between the opamp and outputs.

You should see the atrocious performance of a tube paraphase splitter on my scope. Anything past 5Khz and it just falls apart. At 20Khz the phase shift is so bad it looks like some kind of joke circuit. It's not the loading of the scope leads, it's actually that bad.
 
I know it would be a hybrid, but would that be a bad thing.

IMO, absolutely not! Performance is the key, not any specific technology. IME, tubes do several things better than SS. However, carefully crafted combinations of "hollow state" and SS frequently do a better job than either technology does alone.

FWIW, I believe the 12AX7 triode should be restricted to voltage amplification duties, as its low gm and high RP make it a very poor driver of downstream circuitry.
 
That's really interesting, solid state amp with tube outputs. How does it sound on speakers?

Mine would be simpler than that. Just a tube stage between the opamp and outputs.

That design didn't have a notable sonic footprint that I can recall, despite running in class AB2 and using a $12 speaker line matching transformer as OPT. The apparent complexity stems from the fact that voltage gain in 6080 tubes is only about 1.0 in this application. All of the transistors in the middle of the schematic are just implementing a pair of high-voltage op-amps in inverting configuration. I'll be interested to follow your amplifier development story.
 
I built it today on breadboard and it really sounds good. Sounds way cleaner than the original paraphase type splitter. The new circuit sounds very clean in the highs. I've yet to figure out a way to get the global feedback into the ic's input, but I think this is a good start. I used a stereo magnavox EL84 amp as the test bed. I'll get a schematic up soon. Each scope channel is connected to the EL84 grids with global feedback disconnected.



Using a TL082 and 12AX7. Right at output tube grids. 20Khz.
IMG_20160404_203407_zpsdbsasgpv.jpg



Using it's original paraphase at output tube grids. 20Khz
IMG_20160404_220856_zpsgoubyi8l.jpg


The circuit on breadboard
IMG_20160404_203351_zpsz1lmkuns.jpg
 
repair guy,

You have shown why tubed paraphase splitters are out of favor. The current consensus is that split load, AKA "cathodyne", AKA "concertina", and differential (LTP) splitters are best.

I definitely want to see your schematic. FWIW, I'm thinking that it may be possible to drive the non-inverting opamp I/P with the signal from the 'X7 and apply the NFB to the chip's inverting I/P or vice versa.
 
Here's the schematic so far. I didn't draw the opamp power supply, it's regulated +15, -15. I know it needs some optimizing, but it works really good. I might change out the 12AX7 for a 12AU7 for less gain. If anyone has any better design for the resistors, and how to add the global feedback I'd appreciate it.

IMG_20160405_001339_zpsvdeii1eq.jpg