Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Help diagnosing phase-splitter
Help diagnosing phase-splitter
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Thread Tools Search this Thread
Old 4th July 2008, 11:50 AM   #1
pilli is offline pilli  France
diyAudio Member
Join Date: Aug 2006
Default Help diagnosing phase-splitter

I have built, for my Stax headphones, the amplifier described at http://www.bonavolta.ch/hobby/en/audio/hdgsta.htm
(diagram attached).

The thing makes music, and rather well, but I am not completely sure that everything is right:

For one thing, the oscilloscope shows higher amplitude on one branch, and the trimmer doesn't adjust beyond a certain level that I don't find "enough" (say 5% amplitude difference, otherwise gross distortion).

And the sound has something strangely disturbing that I cannot define well, but that has me develop a headache after an hour of listening.
Oscillation? Couldn't detect any on the 'scope or with the "AM radio" method, but who knows...

And then, maybe it can sound even better.

So here's the questions.
(Just primary pattern-matching.
Thanks for your patience, I am not very experienced with tubes (barely more with transistors...), but I'll learn if tought well )

The circuit looks like a phase splitter with each "phase" driving an output stage for one of the stators (of course Staxes work in balanced mode, and, fwi understand, the few schemas I've seen around all apply this same basic architecture).

"My" splitter looks like a "cross-coupled" structure, except that the output from the first triode on the Stax diagram is from the anode, instead of being a cathode-follower (like described in tubecad or bonavolta or other rare drawings of this Van Scoyoc thing.)

Is it supposed to work anyway (and the (dynamic?) inbalance I observe is just supposed to be there)?

If not, and if I decide to move the "cross" to the cathodes, should the feedback be applied somewhere else?
Should anything else change?

Thanks in advance,

Last piece of info: my output tubes are 6S4 because I had started one of Gilmore's designs, then found this one easier to start with.
Also, my coupling caps are 1,5uF, because "those were lying around when I reached for 400V caps".
Don't think this has an influence on the input stage.
Attached Images
File Type: gif hdgsta1a.gif (15.6 KB, 309 views)
  Reply With Quote
Old 4th July 2008, 01:18 PM   #2
SY is offline SY  United States
On Hiatus
SY's Avatar
Join Date: Oct 2002
Location: Chicagoland
Help diagnosing phase-splitter
Wow, there's some very odd things about this circuit. The second stage tubes have only 50V across them. Stage current is VERY low. This is not a low distortion amplifier.

Your observation is correct, they have the first two stages connected as a cross coupled cascode instead of the "normal" biological amp.

Can you provide a DC voltage map for your build? Are the voltages symmetrical?
"You tell me whar a man gits his corn pone, en I'll tell you what his 'pinions is."
  Reply With Quote
Old 4th July 2008, 01:27 PM   #3
ray_moth is offline ray_moth  Indonesia
diyAudio Moderator Emeritus
ray_moth's Avatar
Join Date: Jan 2004
Location: Jakarta
It is a very unusual design and is unlike any other that I've ever seen. I agree, the way it's drawn looks at first glance like a Van Scoyoc splitter but studying it more makes me think it's a cross-coupled SRPP. I don't think it would help to turn it into a true Van Scoyoc because then you would lose gain. In any case, a Van Scoyoc is basically unbalanced and wastes tubes. That's why it's rare.

I'm unfamiliar with Stax headphones but, from the description in the article, I guess they're electrostatic?

I do agree with the author that some form of grounding would seem better at the junction of those 100uF cathode bias bypass capacitors. I also feel that the 300k plate loads for the second 12AT7 are very high and could be a cause of serious noise.
  Reply With Quote
Old 4th July 2008, 01:59 PM   #4
pilli is offline pilli  France
diyAudio Member
Join Date: Aug 2006
Thanks for these quick replies!

Yes, those are electrostatic "headspeakers" (ideal application for tubes, I guess.)
I have used them for a while with simple step-up transformers and I did not get that headache (I mean, I don't think it just "headphone tiredness") that makes me suspect my build.

The voltages I get are "consistent" with the diagram

I attached a picture with some voltages I remember (writing from the office , I'll measure more this evening)


- The DC voltages I get are symmetrical, or can be made quite symmetrical with the trimpot; but then the sinusoids are much less symmetrical (I can post some images later...) in amplitude.
I went for a compromise, also based on the source site:
"...the adjustments can be done statically or dynamically.
(...) in an unbalanced design, it is better to adjust the trimmer so that the "lower" branch has a slightly higher voltage. In this way, the amp is adjusted for DC but also for AC.
With a sinusoidal generator and oscilloscope, it is pretty easy to adjust the trimmer so that each branch of the final stage input receives a signal in phase opposition but identical amplitude."

- Output voltages react well to changes to B+, so I get them to 0V

- I was surprised how silent it is, given my building-skills are far from impeccable. Maybe noise balances out.

- I did put the resistor between the back-to-back caps.

- I have slightly lower supplies (components I had available)

In any case, would you say that a simple mistake in drawing the diagram is ruled out?

I mean, it's different enough from a "normal" cross-coupled phase splitter to say that it is not supposed to be wired differently?

(In that case it is worth trying to understand if and how it works...)

(I'll be back in a few hours with real data from the scope.)

Attached Images
File Type: gif hdgsta1a v.gif (15.2 KB, 257 views)
  Reply With Quote
Old 4th July 2008, 03:18 PM   #5
SY is offline SY  United States
On Hiatus
SY's Avatar
Join Date: Oct 2002
Location: Chicagoland
Help diagnosing phase-splitter
Understand that the waveforms might well be unsymmetrical- the 2nd harmonic distortion of each side is likely to be quite high, the idea being that it cancels in the differential output. That will be difficult to verify without a differential input to your scope, but we might be able to eyeball some stuff from scope photos and get an idea.

Further in the "what were they thinking?" department, one notes that the feedback is nearly ineffective, maybe 4 or 5 dB at the most, maybe even less. Why bother?
"You tell me whar a man gits his corn pone, en I'll tell you what his 'pinions is."
  Reply With Quote
Old 4th July 2008, 09:10 PM   #6
pilli is offline pilli  France
diyAudio Member
Join Date: Aug 2006
...phew! this took some time!
(VERY careful poking probes into my p-to-p mess!!)

So here is what I can measure, if I can take some more time from you guys:

This image is at the two coupling capacitors "input side" (that is "before" the cap)
(Refer to the diagram of my first post.)

Input is a sine wave at 1KHz, 2Vpp.

The upper trace is the upper branch, out of pin 1 of the second stage. The lower trace is out of pin 6.

As you can see, there's no way the amplitude of the lower branch can increase.
In the "best compromise" the upper trace is about 38Vpp, and the lower less than 24Vpp. And both visibly distorted.

If I turn the trimpot, it distorts, but the two amplitudes never match.

In the "best compromise", the DC in the second-stage plates are:
"upper" = 90-95V
"lower" = 95-105V

The first stage plates (I wasn't sure of the earlier measurements) are worse than what I thought: upper "input" plate is at 68V, lower at 71V, which means, as SY pointed out, that the second stage only has a tiny A-K voltage across it...
Attached Images
File Type: jpg input.jpg (85.7 KB, 222 views)
  Reply With Quote
Old 4th July 2008, 09:17 PM   #7
pilli is offline pilli  France
diyAudio Member
Join Date: Aug 2006
...and this is what comes at the outputs.

Some difference in amplitude, some distortion too, I would say.
(edit: maybe it is clipping with 2Vpp input...? I'll try with a lower input.)

I only have one probe with 10x so I had to take them one at the time...

The left image is the "lower" branch, looking like 320 Vpp.
The right image is the "upper" branch, at about 400 Vpp.

...hey, are they in phase?
Should they be?
Attached Images
File Type: jpg output.jpg (48.5 KB, 221 views)
  Reply With Quote
Old 4th July 2008, 09:24 PM   #8
EC8010 is offline EC8010  United Kingdom
diyAudio Moderator Emeritus
EC8010's Avatar
Join Date: Jan 2003
Location: Near London. UK
They only appear to be in phase because of the way you've triggered your oscilloscope. If you displayed Ch1 and Ch2 simultaneously you'd see one was of inverted polarity.
The loudspeaker: The only commercial Hi-Fi item where a disproportionate part of the budget isn't spent on the box. And the one where it would make a difference...
  Reply With Quote
Old 4th July 2008, 09:45 PM   #9
pilli is offline pilli  France
diyAudio Member
Join Date: Aug 2006
Originally posted by pilli
(maybe it is clipping with 2Vpp input...? I'll try with a lower input.)
...ok, I repeated with 1Vpp and there is MUCH less distortion.

But still the same difference in amplitude between the two "phases".

thanks for the clarification. I could have thought of that, the scope doesn't know any "absolute beginning" of periodic signals better than "synchronize on a trigger" (or something like that, anyhow, I got it...)

  Reply With Quote
Old 4th July 2008, 10:57 PM   #10
wakibaki is offline wakibaki  United Kingdom
Join Date: Jan 2008
Despite what SY says about 2nd harmonic (and his suggestions are worth taking seriously), my first intuition on looking at this circuit is that it's symmetrical, and I would expect to see symmetrical DC and AC voltages around the circuit.

This would be my primary focus; where is the assymmetry being caused? By comparing across the limbs of the circuit you should be able to see where things are out of kilter, although the cascade makes this a bit more difficult... the side with the greater gain is probably nearer correct than the low side, gain being easier to lose than ...gain. (sic)

  Reply With Quote


Help diagnosing phase-splitterHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Is this a phase splitter? tom-vdl Solid State 9 15th October 2009 08:21 PM
phase splitter with Tr ? sam8888ysl Tubes / Valves 9 22nd October 2008 11:28 AM
Phase Splitter Name Gold_xyz Tubes / Valves 17 21st February 2008 09:48 AM
HELP; DIY PP amp - it's phase splitter, or buy Chineese. WE91 Tubes / Valves 19 30th October 2007 04:01 PM
jensen-11P-1 as a phase splitter Stefano Pass Labs 13 10th March 2004 09:28 PM

New To Site? Need Help?

All times are GMT. The time now is 10:58 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.79%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio