Phase splitter choices on input of PP amp

[andyjevans]

I'd like to know what are the choices for a phase splitter on the input of a PP amp so it could be driven by a preamp with gain using RCA inputs. Amp in question is an EL84 in PPP triode, so 8 output tubes in all. Choices I can think of include....

• input transformer like 1:1+1 or 1:2+2
• solid state splitter
• valve concertina without the gain stage in front
• ??? something else

I'd like something simple and transparent that could be driven with a 26 DHT preamp in this case, but this is a universal application so feel free to talk around choices as much as you would like.

 

[vinylkid58]

??? something else

Long-tail pair. https://www.valvewizard.co.uk/acltp.html

DC coupled: https://www.valvewizard.co.uk/dcltp.html

You can current sink the tail as well.

jeff

 

[andyjevans]

Thanks for mentioning the long tailed pair. It's one option. But I'd like an option which doesn't provide gain - just splits phase. Or if an input transformer it would be unity gain or x2 for instance. If I drive the output stage with a DHT 26 stage I won't need additional gain.

 

[20to20]

Cathodyne.. Dynaco style. No gain.

 

[wg_ski]

No gain, but also limited output swing since the outputs are developed simultaneously. Takes more voltage to drive a triode than a pentode or even UL. Might want something that can run on a higher rail than a 12ax7, and elevate the heater if necessary. Once you get above 130 VDC on a cathode it’s pretty well mandatory.

 

[Fuling]

I've always had good results from LTPs using high-ish mu triodes such as ECC81, 6SL7, EC86 and the triode sections from ECL84.
Sometimes I've used Hammond 156C chokes in the tails instead of solid state constant current sources but thats just a gimmick.

 

[jcalvarez]

Andy, what is the gain and max output swing of the 27 preamp? Sorry I'm not familiar with it.

 

[andyjevans]

No gain, but also limited output swing since the outputs are developed simultaneously. Takes more voltage to drive a triode than a pentode or even UL. Might want something that can run on a higher rail than a 12ax7, and elevate the heater if necessary. Once you get above 130 VDC on a cathode it’s pretty well mandatory.

The EL84 in triode should be easy to drive - around -11V bias. That's a good point about H-K voltage. A mosfet or input transformer would avoid this.

My 26 DHT stage has a gain of around 8 and output impedance of around 7.5K.

 

[euro21]

With this output impedance the #26 unable properly drive the (generic) SE-PP transformer.
IMHO must to use low gain tube, which have enough large input impedance.

Why not use parafeed CT choke driving?

 

[andyjevans]

Hello Bela - as you say the 26 has a higher than desired output impedance. A parafeed CT choke looks good, thanks for that.

Where would I find one off the shelf? Or could it be possible to use one side of an interstage, like for example a 124B

https://www.hammfg.com/files/parts/pdf/124B.pdf?t=1697661929

 

[andyjevans]

A parafeed CT choke doesn't look a common part. I found this one. Still researching.......

https://www.sowter.co.uk/acatalog/E_A_Sowter_Ltd_PHASE_SPLITER_CHOKES_22.html

 

[euro21]

Maybe LL1667/1668. I have LL1668, but never tried this mode.
Peerless 17173, Magnequest EXO-173 are especially CT plate choke.

 

[Fuling]

I use Edcor XSM 15k:15k as parafeed phase splitters in a project. At first I used them as 1:1+1 interstage transformers, then one half of the primary winding on one of them went open circuit so I rewired them to use only the secondaries.
Works fairly well but again, it's a bit of a gimmick. A pair of triodes as an LTP would do the job much better.

 

[bondini]

Perhaps a 1:1+1 input transformer, secondaries feeding a pair of MOSFET source followers, then on to the EL84s?

 

[Merlinb]

A whole webpage on unity gain phase splitters:
https://www.tubecad.com/2021/04/blog0533.htm

 

[waltube]

I use a LPT with 6H30 at 400Vdc and 10 mA each section
Good swing good ThD
Not simple to fix the right point
Also a good Zout

Walter

 

[FlaCharlie]

Thanks for mentioning the long tailed pair. It's one option. But I'd like an option which doesn't provide gain - just splits phase. Or if an input transformer it would be unity gain or x2 for instance. If I drive the output stage with a DHT 26 stage I won't need additional gain.

Showing my ignorance here . . . could the Hammond 1140-LN-C, which we have both used "in reverse" as a 1:4 SUT be used to invert phase and, to keep it unity gain, just add a voltage divider on the outputs?

 

[andyjevans]

https://www.hammfg.com/files/parts/pdf/1140-LN-C.pdf?t=1697661928

I have 2 pairs of 1140-LN-C so interested. It works in step-down but that's not what I'm looking for - unity or step up. Maybe another in the range would work?

The 124B on the other hand might work as a centre tap splitter as mentioned before.

 

[FlaCharlie]

Yeah, looking again I see that using SUT mode with that model wouldn't work because the single primary becomes the secondary. I guess you just need one with dual secondaries which can then supply one in-phase and one "out of phase" / opposite phase to the output tubes.

Or would it be possible to create an artificial center tap on the secondary??

 

[Windcrest77]

Just spit-balling here... The advantage of a transformer is that it is passive. So you could have a switch that will allow the amp to accept single ended or an XLR balanced input. That XLR is already coming in with anti phase ready to go. All better DAC's these days have a balanced out already phase split, it seems to me PP amps these days don't even need a phase splitting stage if you have a good DAC with XLR outs.