The "Muscovite" 6S3P Tube Phonostage - Page 7 - diyAudio
Go Back   Home > Forums > Source & Line > Analogue Source

Analogue Source Turntables, Tonearms, Cartridges, Phono Stages, Tuners, Tape Recorders, etc.

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
Reply
 
Thread Tools Search this Thread
Old 30th July 2012, 12:26 AM   #61
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
In light of further experience I made a minor revision to the value of R6 (see Final PSU) in order to slightly reduce the plate current in the 12AX7 error amplifiers without increasing the loop gain. The value changed from 4.7K to 6.8K and should reduce the plate current to just slightly over 1mA, a slightly better operating point on the plate curves given the relatively low plate voltage across each device. This also reduces screen voltage slightly with a very small resulting reduction in transconductance.

Output noise has dropped very slightly, other changes probably too subtle to pontificate about.

In case it is not obvious the supplies are very much 'in the picture' sonically speaking due to the lack of RC decoupling in this design (a deliberate design choice) but it does mean that significantly altering the performance of the supply and that supply deficits are going to be audible. The PSRR of the cascode input stage is negligible obviously..

Edit: Schematic removed to avoid confusion.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 30th July 2012, 01:04 AM   #62
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
Default Choice of Input Cap for Filament Regulator

Something to be aware of wrt to the input filter cap in the filament supply is that the ripple current rating should be as high as you can reasonably manage, the input cap in mine is running a case temperature of 20 degrees C over ambient - it's rated at 3.8A ripple current, the load current is 2.5A, and I'm thinking this is getting quite close to the limit.

The part I used is here: ECO-S1CA223CA Panasonic Electronic Components | P6547-ND | DigiKey

I've not really found a better choice in a compact capacitor so I will monitor the situation. It may not be such a bad idea to parallel several 10000uF electrolytics instead. I think a pair of these may be a better choice: UPW1C103MHD Nichicon | 493-1807-ND | DigiKey
- less than half the ESR and the same ripple current rating per cap running at effectively half the current of the hard working cap in my unit.

The point is to choose a cap that will not have to be replaced in a decade or more of use.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 30th July 2012, 11:51 PM   #63
diyAudio Member
 
Join Date: Nov 2011
Quote:
Originally Posted by kevinkr View Post
Something to be aware of wrt to the input filter cap in the filament supply is that the ripple current rating should be as high as you can reasonably manage, the input cap in mine is running a case temperature of 20 degrees C over ambient - it's rated at 3.8A ripple current, the load current is 2.5A, and I'm thinking this is getting quite close to the limit.
................
The point is to choose a cap that will not have to be replaced in a decade or more of use.
kevinkr,

Have you considered using OS-CON in that particular location? The data sheet of SA series show ripple current of 9.75A for 1000uF/16V !!!

And, by the way, did you know that 6S3P and 6S4P were designed as pair to work in cascode amplifiers? 6S3P for common cathode, 6S4P for common grid. Pinouts of both tubes are different, but so are the interelectrode capacitances. I have a Russian book from 1964 called "Cascode amplifiers" which describes those tubes. (I speak that language a bit).

Best,
  Reply With Quote
Old 31st July 2012, 12:43 AM   #64
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
Hi Elektroj,
Yes indeed I know about the 6S4P and even thought about using it as the top half in the cascode and choke augmented mu-follower, but went for the 6S3P in all locations as it was simpler and these tubes aren't readily available here in the US.

Oscons are very expensive over here so I did not consider using them, but I will eventually change out the existing cap for a couple of reasonably priced electrolytics that will not have to work as hard. I think the ESR of this cap is a big part of the problem, but truthfully the rise is well within acceptable limits - just that I am a bit uncomfortable with it when it can mostly be eliminated. I think the internal heating will result in shorter life as I do not know how evenly the heating is distributed in the capacitor, but it is operating within ratings..
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 1st August 2012, 01:21 AM   #65
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
Default Line Transients

I made a change to isolate the filament supply from the cathode of the right channel pass tube, believing that this would solve the problem - unfortunately it has not, so I am no closer to understanding the root cause of the problem. Otherwise it is a good idea - it does remove a transient injection path.

Further investigation is required.

Edit: I have removed these older schematics to avoid confusion.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 1st August 2012, 05:01 AM   #66
diyAudio Member
 
Join Date: Nov 2011
kevinkr, take a look at this book, there might be some hints re. series tubes:
Handbook Preferred Circuits Navy Aeronautical Electronic Equipment : National Bureau of Standards : Free Download & Streaming : Internet Archive
  Reply With Quote
Old 1st August 2012, 04:04 PM   #67
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
Hi Elektroj,
Thanks for the link, great stuff, some very interesting designs.. I've downloaded it and am reading it, at first blush it does not appear that I have done something egregiously bad, but apparently I have.. LOL

Have a few additional ideas to try tonight, revolving around somewhat input voltage dependent suspected parasitic oscillations in the pass tubes. The problem is not always there, seems to be mostly at times where high line conditions are common. Installed lossier rectifier tube and the problem just went away at least temporarily.

Further investigation required.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 2nd August 2012, 01:08 AM   #68
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
Default High Voltage PSU Updates 6BD5GT/6V6 (Improved Schematic)

Last night I replaced the 5AR4 with a 6087 which dropped the plate voltage by roughly 20V, the bad behavior stopped immediately and did not recur for the balance of the evening. Since the 20V drop in HV (more than 10% reduction in headroom across the pass element) also slightly reduced the transconductance I surmised that the combination of my circuit design and high line was a recipe for instability.

I have added a 1kV ceramic cap from the screen grid to chassis ground and a ferrite bead within 1cm of the plate lug.

The ferrite bead can be any with 100 ohms or more loss at 100MHz. These just slip over the plate lead and must be close to the socket.

Whether or not this actually works as a complete fix remains to be seen, so far it seems to, but many more hours of use under differing line voltage and ambient temperature will be required to be sure.

I reinstalled the 5AR4 and so far for about an hour it has behaved. I'll report back as to whether the problem is fixed or not later.

Edit: 2013-05-12: Most current version of the HV PSU
Attached Images
File Type: png 6V6_PSU.png (74.6 KB, 439 views)
File Type: png 6BD5GT_PSU.png (70.8 KB, 375 views)
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 4th August 2012, 05:22 PM   #69
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
Talking PSU Sorted

Happy to report that the screen cap and plate ferrite have resolved the power supply issues reported in the last few posts.

The dual supply despite the added complexity does seem to bring something worthwhile to the equation, the much reduced inter-channel cross-talk has resulted in significant gains in both image width and depth - the most in fact that I have ever heard in my system. Noise levels are extremely low, and as noted this pre-amp is the least microphonic I have ever designed, clearly the 6S3P-EV is well suited for this application.

Those wanting to get really extreme could build it fully dual mono with separate rectifiers and filtering - I am not certain of the point of diminishing returns, but I've probably surpassed it even with this design.

The other potential improvement would be to heat the 12AX7A filament with regulated DC which might very slightly improve the performance by eliminating the error amplifier tube filament thermal response (DC output wander) to varying line voltage. The observed wander is only a few mV typically with a good 12AX7A so this is definitely at the point of diminishing returns. I did not use the DC filament supply to the pre-amp to power these tubes because the required DC offsets (bias) are quite different - the 6S3P is not supposed to operate with the filaments positive relative to the cathode although I do this currently and have not observed any ill effects.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 8th August 2012, 12:05 AM   #70
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
Talking Self-Inflicted Misery

Umm, someone didn't do a very good job of soldering the junction between the two 220uF/400V series connected caps that comprise C3, that someone of course would be me..

Tonight I basically discovered that I had completely missed the joint and I'm not sure exactly how that happened, and it was not in a spot where the lack of solder was apparent until I really looked. Oops..

This interesting omission on my part resulted in intermittent popping and the occasional full disconnect which is what apparently triggered the stability issue I have been writing about.

So probably about 50% of the odd behavior I have been chasing was self inflicted in the most direct sense, and the rest may not have been entirely design related.

The interesting thing at the end of this is that this pre-amp despite the ridiculously high gain is also ridiculously quiet. It now appears that the cartridge and transformer along with a little external 60Hz pick up are the dominant noise sources in the phono path. You pretty much have to max out system gain to hear what noise there is.. I'm relatively happy about this - that was one of the design goals.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote

Reply


Hide 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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Phonostage with tube moving coil legarem Tubes / Valves 4 13th October 2011 04:42 PM
freaky new phonostage Vinylsavor Tubes / Valves 35 30th November 2010 07:30 AM
Russian 6S3P=WE417? huskydawg9 Tubelab 8 24th March 2010 01:32 PM
rjm 2 tube phonostage hacknet Analogue Source 6 10th April 2006 05:27 AM
Pearl phonostage Mark777 Pass Labs 0 6th July 2003 11:18 AM


New To Site? Need Help?

All times are GMT. The time now is 11:34 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2