diyAudio (
-   Tubes / Valves (
-   -   phase splitter with Tr ? (

sam8888ysl 11th October 2008 04:04 AM

phase splitter with Tr ?
2 Attachment(s)
I wanna to build a new PP AMP (6528PP )
I had built some PP AMP (5687 PP 7236PP 12B4PP 6AQ5 PP......) with tube phase splitter
and works well (not excellent)

and there are some Tr 600:600 without CT
I hand some idea but i do not know them work or not
so if any command
thanks very much


Colt45 11th October 2008 06:02 AM

two of em will work, I don't think that resistor abortion will though :xeye:

teemuk 11th October 2008 10:39 AM

The one with single transformer and a resistor-derived center tap (top right) looks like the one I would use if I would build something like this. Using two transformers instead of one seems a bit redundant. It's a quite lossy circuit, though: First you have the losses of cathode follower (gain is always less than unity), then you loose half of the amplitude again due to transformer coupling arrangement. e.g. If input at primary side is 2Vpp then outputs at "phase-splitted" secondary are 1Vpp signals that have opposite phases.

Shoog 12th October 2008 08:09 AM

My personal experience is that interstage phase splitting works very well with a few reservations.
The easiest and best way to implement it is to splash the cash and get yourself some Lundahl LL1660. This will represent an investment for life. This is the most difficult job for any transformer in a valve amp to perform, and this is why there are no cheap options available.
If you try to press gang any other type of transformer into the role then you will need to go parafeed and push some serious current through it to avoid interwinding capacitance induced high frequency roll off. Step down is definately a good idea if you go this path, and its well worth generating twice the gain just to chuck away in the transformer. In this case a cathode follower is probably a good idea, but of course you run the double penalty of a cathode follower and the need for an extra gainstage and hence associated cap coupling.

I have done this with mains toroidals. The result sounds fantastic, but lacks a certain amount of sparkle due to a bit of high frequency roll off. My advise is buy the Lundahl.
Another option is to shift the phase splitting to the input, do a bit of step down, and drive a LTP into the finals. I have done this twice with mains toroidals and microphone transformer, and it works well. The big advantage with this approach is that the input see's no DC and so needs no gap. A 2:1+1 step down should be adequate.


sam8888ysl 15th October 2008 06:19 PM

2 Attachment(s)
thanks a lot

well as bellow
its my next power amp ( just in mind)

Shoog 15th October 2008 08:15 PM

That looks like a good and imaginative design.
Two points though, are you certain that the drop over the driver choke's will settle down to 40V. You might need a bit of a trimmer pot just to fine tune it.
You are not getting the advantage of a LTP driver arrangement, This may well not be a huge issue as the input transformer should give accurate phase splitting - but you may find the drivers losing some of that balance. A good LTP will maintain that balance through the driver stage.

You should be able to derive the negative rail by building a bridge rectifier. I did this on my last project and generated the neg rail with silicone diodes, choke input should keep the voltage down. The positive rail was vacume You should be OK dropping the voltage resistively if the front end is drawing minimal current.

Good luck.


sam8888ysl 16th October 2008 03:40 PM

I will add a resistor above the choke 80H maybe2-8K depends.
if no good i'll try LTP latter

Shoog 17th October 2008 08:02 AM

If the choke on the -160V supply side is good then you should preserve most of the LTP functionality. If I were doing it I would plump for a CCS in its place, though I can see you want to stay away from silicone as much as possible. If you did try a CCS it needs to be rated at about -300V.


sam8888ysl 22nd October 2008 08:44 AM

2 Attachment(s)
i tried these Tr
with 6J6 to drive 7236
as below
it works pretty good i like them better than tube phase splintter i had made before
but to match paired is very important
i had buy 24 to match paired
its quite cheap about 2 EUR each

Shoog 22nd October 2008 11:28 AM

If you can generate a little bit of negative rail (from the heaters possibly) and then insert a CCS into the tail then balance should be less of an issue. Alternatively if you add a choke into the tail as you originally suggested then you should get almost as good a result without the neg rail.

The problem with using balanced valves is that they will not stay balanced so you either have to replace them at short intervals or live with the inevitability of degradation of sound quality over time. It is always my intention to make my amps robust and adjustment free. If that means a slight sacrifice of absolute sound quality this is a compromise I am quite willing to make.

In your original idea for using a choke as the tail, if you have voltage to burn then placing a resistor between it and your negative rail will stiffen your LTP arrangement.


All times are GMT. The time now is 12:16 AM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Copyright 1999-2016 diyAudio

Content Relevant URLs by vBSEO 3.3.2