phase splitter concerns ?

[coresta]

Hi Tubholics , why is a tubed phase splitter a lesser solution vs an Xformer (an UTC for example) one ?

[AndreasS]

Dear Pierre,

the tubed phase splitter has lesser costs...

Regards Andreas

[SY]

Because it transfers the responsibility of competent design from the hobbyist to a professional.

A competently designed phase splitter will handily outperform an interstage many times its size, weight, and cost. But a competent phase splitter is not trivial to design. If one wishes to use a transformer for phase splitting, a balanced input transformer will also do a much better job, and bring a few extra advantages to the table (like galvanic isolation from the rest of the system).

[Salas]

Originally posted by coresta
Hi Tubholics , why is a tubed phase splitter a lesser solution vs an Xformer (an UTC for example) one ?

Originally posted by SY
Because it transfers the responsibility of competent design from the hobbyist to a professional.

A competently designed phase splitter will handily outperform an interstage many times its size, weight, and cost. But a competent phase splitter is not trivial to design. If one wishes to use a transformer for phase splitting, a balanced input transformer will also do a much better job, and bring a few extra advantages to the table (like galvanic isolation from the rest of the system).

When a very low output impedance is needed so to drive a DHT PP in large swings and in grid current region with a good slew rate and bandwidth, isn't a mosfet buffer, followed by an xformer, followed by an RC connected phase splitter the descending order of best to so-so solutions?

[Bandersnatch]

hey-Hey!!!,
Your assumption that a TX phase splitter is better is in error IMO. Now if you *WANT* the colourations a TX core involved, designing a tube circuit to do it will probably be a frustrating exercise. There are even different colours for the differently configured cores and materials too.
cheers,
Douglas

[SY]

Depends on the swing. If it's more than 200V p-p, then maybe, maybe not, but I'd question the design choice of that output tube.

edit: Anent Douglas's comment, I assume you're interested in accuracy of output tracking the input, not signal processing. If the latter is the case, forget everything I said.

[Andrewbee]

What about the Sycltron (sp)? I remeber it being a cross-coupled design with transistors in the cathodes and an input transformer.

[SY]

I've never used it for really big swings- in my screen-drive amp, it's used as an input stage, with an input transformer, then drives a conventional 6SN7 diff amp (quite similar to the second stage of the Crystal Palace). That stage can swing over 200V differential output at 0.15% THD, while not suffering the reactance and LF issues of an interstage (which renders feedback difficult).

In any event, the key is the input transformer, which only has to swing a volt or two and can easily be kept outside the feedback loop.

[Miles Prower]

Originally posted by coresta
Hi Tubholics , why is a tubed phase splitter a lesser solution vs an Xformer (an UTC for example) one ?

It's not. A couple of really good electronic splitters are the differential with active tail loading, and the cathodyne. These are definitely better, and unlike an IST, can be used inside a gNFB loop. You will find these in use for precision applications like o'scope deflection amps and instrumentation amps. ISTs are never used there.

Other splitter topologies like the paraphase in all its various iterations aren't so good since these have different levels of harmonic distortion between phases, and are useful for quick 'n' dirty solutions where sonic quality isn't so important.

The only other place an IST is preferable is RF circuits where the IST is usually a double tuned RF xfmr.

[richwalters]

With Morgan Jones comments on amplifier stage bandwidths, phasesplitter transformers don't stand much of a fidelity chance when put against concertinas and other p/s configurations. In my recently completed leviathan amp they wouldn't touch the performance I am after. The concertina is the one I've used for repeatability.

rich