SS phase splitters?

[M Gregg]

Does anyone have,

Any circuits for SS phase splitters to drive tube sections? (I'm thinking discreet FET based)i.e. Van Scoyoc cross-coupled splitter.
And what if any problems were encountered?


Regards
M Gregg

 

[ingenieus]

If you have a tube circuit, it can probably be adapted. Here is an article on how triode tubes can be replaced directly by high voltage MOSFETs.

MOSFET Follies - letting the Solid State Demon into your tube amp

 

[Eli Duttman]

It's no secret that I like MOSFET voltage followers. As a "concertina", AKA split load, phase splitter is a tweaked voltage follower, a MOSFET doing that job rates to be quite satisfactory. Look here, where Jeff Yourison is working (with a little help from yours truly) on a PP 6Y6 amp that employs a MOSFET "concertina" phase splitter.

While FET current amplifiers are highly satisfactory, I'm very dubious about FETs as voltage amplifiers. The cross coupled phase splitter is a somewhat complex bit of circuitry. Voltage amplification is occurring, along with the generation of the 180^o out of phase signals. I can see depletion mode (for self biasing ease) MOSFETs source followers at the I/P of the transformerless version, but linear triodes being employed in the voltage gain positions. Perhaps the DN2540N3-G would be suitable in the voltage follower "slots".

 

[ingenieus]

Here is am improved version of the Van Scoyoc cross-coupled splitter published in 1957:

http://www.next-tube.com/articles/cci/cci_EN.pdf

I ran it through LTspice, and it works perfectly. The only thing is, I don't get the bit about "cathode voltage of power tubes must be increased by 20 volts." My limited knowledge of tubes does not help, especially since the simulation works without this. Why is that necessary? To improve the voltage swing range? In practice, would this be implemented with voltage divider resistors off B+?

 

[Eli Duttman]

The only thing is, I don't get the bit about "cathode voltage of power tubes must be increased by 20 volts."

The phase splitter's O/P cathode followers and the O/P tubes' control grids that are DC coupled to them "sit" at +20 VDC. Correct operation of a tube, with few exceptions, requires the control grid be negative with respect to the cathode. Implicit in the remark is a cathode (self) biased set up. Since the control grids of the O/P tube pair are at +20 V., instead of 0 V., the cathode has to be raised above ground 20 extra volts to achieve the desired operating point.

If the cap. coupled O/P cathode followers are replaced by DC coupled ZVN0545A source followers, cap. coupling to the O/P tubes' grids can be employed and any valid biasing method will work. 🙂

 

[Wavebourn]

Here you go, just a concept:

 

[Cookj]

Novotone - Projects 35 and 37

Published in Electroniqe Pratique

Cheers

John

 

[tinitus]

I think this one is from a Japanese site

 

[MiiB]

'The last circuit is a LTP, to me one of the best ways to make a phase splitter. I would make it with two lsk170's or something like that, then cascode it with transistors that can withstand the high(ish) voltage.

Or maybe looking to Calvin's fine buffer-thread make it with a CFP pair as input + a shielding cascode. Then you can run the output tubes without any coupling caps.

 

[ingenieus]

Novotone - Projects 35 and 37

Published in Electroniqe Pratique

Cheers

John

I can read Dutch and Flemish, but my French is non-existent. It's a shame because it seems like the site contains a lot of useful information.

I eventually managed to figure out that one needs to click on "Schéma de l’amplificateur" to see the schematic of the amplifier. Here are the actual links. 😉

http://www.novotone.be/_site/projets/Projet35/Fig02.pdf

http://www.novotone.be/_site/projets/Projet37/Fig02.pdf

 

[ingenieus]

Correct operation of a tube, with few exceptions, requires the control grid be negative with respect to the cathode.

Thanks for that. The light suddenly came on and it nearly blinded me. I should read on tubes some more. 🙂

 

[ingenieus]

I have run a simulation of the circuit with MOSFETs in place of the tubes. It works, but the output is only 5Vpp, so some more work is needed. Since the OP has reacted yet, I will leave it on the back burner.

 

[kenpeter]

My 2¢, Pointlessly equal impedance hybrid concertina with Mu/2 gains...

 

[rsumperl]

Good morning,
I'd like to try the above circuit. Any guesses on how much swing? I'm building a sweep tube amp that will probably have 40-45v of bias.

Thanks,
Ray

 

[M Gregg]

Thanks for the input so far,

Great ideas and circuits.🙂

I was wondering how easy/hard it would be to implement, (reduce space) with some of the more complex splitters. Nice to see other examples.
Theoretically I guess you could build a whole tube circuit with FETS I wonder if anyone ever has and what it sounded like (Off topic) FET with OP Tx.
Back to phase splitters the small 6V6 is interesting.

Regards
M. Gregg

 

[mos57]

Those days i have presented a new type of phase splitter. There are also version specified for tube circuits.

Have a look at this thread:

http://www.diyaudio.com/forums/solid-state/238405-fully-differential-phase-inverter.html 😎

Francesco

 

[tinitus]

I can read Dutch and Flemish, but my French is non-existent.

try with google translator 😉

Here are the actual links.

wow, a wimpy BS107

btw, would a simple SS 'phase splitter' handle multiple parallel output tubes better than if a tube 😕

 

[M Gregg]

wow, a wimpy BS107

btw, would a simple SS 'phase splitter' handle multiple parallel output tubes better than if a tube 😕

The Velleman K4000 comes to mind with a 12AX7 in concertina driving 4 EL34 per channel. You may have a point!

Regards
M. Gregg

 

[tinitus]

not so sure
driving capacitance also seems problematic with solid state designs
maybe its all the same, a Murphy's Law

anyway, what is the reasoning for using solid state devices ?

gosh ! I only know the difference of tubes and SS by the looks

 

[kenpeter]

P