UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > Components and Circuits

Notices

Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets.

Closed Thread
 
Thread Tools
Old 19th Nov 2018, 11:10 pm   #1
bikerhifinut
Octode
 
Join Date: Nov 2013
Location: Penrith, Cumbria, UK
Posts: 1,993
Default What if I take my DC feed to an output transformer CT off the reservoir capacitor?

Mulling over the circuit for the rework of the EL34 amp.
Now as things stand, the volts to the output transformer CT are taken after the smoothing choke on a standard CLC filter. All well and good.
But ideally I would like a bit more voltage and straight after the reservoir capacitor off the rectifier would indicate its about right.
As I understand how these things work, a push pull output stage has a kind of natural noise and hum cancellation property, at least of common mode noise, so theoretically taking the power straight off the reservoir cap should be ok.
There are some older designs that show this, either theres no filter or its a simple RC filter to the preceding amplifying stages.
Can anybody see a reason why I can't do this, bearing in mind if I find empirically that theres a bit too much hum I can simply move the wire to the downstream end of the choke.
Might be a silly question but It was something that struck me when reading up on push pull amps.

Andy.
bikerhifinut is offline  
Old 19th Nov 2018, 11:31 pm   #2
Paul JD
Pentode
 
Join Date: Oct 2018
Location: Coventry, West Midlands, UK.
Posts: 220
Default Re: What if I take my DC feed to an output transformer CT off the reservoir capacitor

It's pretty much standard practice on guitar amps for the OT to connect to the reservoir cap so don't see why it should be a problem.
Paul JD is offline  
Old 19th Nov 2018, 11:40 pm   #3
turretslug
Dekatron
 
turretslug's Avatar
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
Default Re: What if I take my DC feed to an output transformer CT off the reservoir capacitor

Tetrodes/pentodes have a high anode impedance and are relatively uninfluenced by a hummy anode supply, at least compared to triodes. The screen grid, however, can make quite an effective modulating electrode, so is best connected post-choke in the typical CLC PSU set-up. This obviously implies that ultra-linear operation is ruled out, or at least "difficult", but I don't suppose many guitar amps used U-L.

Certainly suck-it-and-see- it'll only be shifting a wire or two!
turretslug is offline  
Old 19th Nov 2018, 11:41 pm   #4
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: What if I take my DC feed to an output transformer CT off the reservoir capacitor

In theory if the output valves are closely balanced there should be little hum, even with a significant ripple on the center tap as the plate hum currents phase cancel in each half of the primary. This was one of the many advantages cited for push pull over class A operation. In addition there is no net DC magnetizing flux, so the output transformer cores can be smaller than for class A output transformers.
Argus25 is offline  
Old 19th Nov 2018, 11:44 pm   #5
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: What if I take my DC feed to an output transformer CT off the reservoir capacitor

Cancellation of the hum which arises from HT rail ripple is, as you say, a feature of push-pull amps. But it does depend on the quiescent currents in the two output valves being matched. In the real world there is always some mismatch. The worse this mismatch is, the worse the hum will be.

A more intractable issue is that the HT voltage variation will impose a corresponding variation in the gain of the output stage, and the push-pull configuration will not cancel this. The consequence of this gain variation is that the audio signal will become (inter)modulated by the ripple so, for example, a 1kHz signal will acquire sidebands at +/-100Hz, +/-200Hz etc. The resulting frequencies will not be harmonically related to the fundamental and this will make them sound unusually offensive (it's been known for a long time that intermodulation distortion is much more objectionable than harmonic distortion). Back in the day relatively simple distortion analysers didn't have the selectivity needed to measure these 100Hz sidebands and I suspect that at least some manufacturers believed that the distortion in their amps was actally lower than it really was. I've seen such sidebands present at levels comparable with the harmonic distortion i.e. tenths of a percent.

Cheers,

GJ
__________________
http://www.ampregen.com

Last edited by GrimJosef; 19th Nov 2018 at 11:52 pm.
GrimJosef is online now  
Old 19th Nov 2018, 11:55 pm   #6
bikerhifinut
Octode
 
Join Date: Nov 2013
Location: Penrith, Cumbria, UK
Posts: 1,993
Default Re: What if I take my DC feed to an output transformer CT off the reservoir capacitor

right that does explain why you generally see the supply to CT after a CLC filter.
That decides it, I shall stay conventional.
bikerhifinut is offline  
Old 20th Nov 2018, 1:12 am   #7
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: What if I take my DC feed to an output transformer CT off the reservoir capacitor

Quote:
Originally Posted by GrimJosef View Post
A more intractable issue is that the HT voltage variation will impose a corresponding variation in the gain of the output stage.
True but you would need substantial ripple to amplitude modulate the output from a push pull output stage. So a few volts ripple on a 400V supply would be insignificant in gain alteration of the stage. If you were not happy with the ripple level say if it was over 10 or 20V simply increase the filter capacitor within allowable limits depending on the rectifier (if thermionic) and add a small series current limiting resistor to keep the valve rectifier's peak cathode current down to recommended limits, if silicon rectifiers, don't worry.
Argus25 is offline  
Old 20th Nov 2018, 1:24 am   #8
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: What if I take my DC feed to an output transformer CT off the reservoir capacitor

Quote:
Originally Posted by bikerhifinut View Post
... you generally see the supply to CT after a CLC filter ...
Yes. Among classic hifi amps the major exception that I can think of is the Quad II. They do use the trick though of putting an output transformer winding into the cathode circuit, so you get a version of ultralinear operation while still being able to connect the screen grids, which (as turretslug says) have a big influence on the gain, to the smoothed HT after the choke.

It's actually more common to see the OP transformer CT fed direct from the reservoir capacitor in modern hifi amps. The reason is that solid-state rectifiers can now be very hefty and can therefore drive a very large reservoir capacitor. And large capacitors are much cheaper than they once were.

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is online now  
Old 20th Nov 2018, 12:58 pm   #9
G8HQP Dave
Rest in Peace
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
Default Re: What if I take my DC feed to an output transformer CT off the reservoir capacitor

Hum intermodulation from the supply rail is more likely with a triode or UL PP ouput than a pentode PP with clean g2 supply.
G8HQP Dave is offline  
Old 20th Nov 2018, 2:09 pm   #10
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: What if I take my DC feed to an output transformer CT off the reservoir capacitor

Quote:
Originally Posted by Argus25 View Post
True but you would need substantial ripple to amplitude modulate the output from a push pull output stage. So a few volts ripple on a 400V supply would be insignificant in gain alteration of the stage. If you were not happy with the ripple level say if it was over 10 or 20V simply increase the filter capacitor within allowable limits depending on the rectifier (if thermionic) and add a small series current limiting resistor to keep the valve rectifier's peak cathode current down to recommended limits, if silicon rectifiers, don't worry.
Let's check some real numbers. Here's the power supply and output stage of the Radford STA100 amplifier http://www.radfordrevival.co.uk/word...uctions-P3.jpg. A silicon bridge feeds a pair of 200uF capacitors in series i.e. 100uF. Four KT88s (two in each channel) pull a total of 240mA from this i.e. 2.4mC in each 10ms ripple cycle. That generates 7V RMS ripple on the 600V HT rail assuming a 24V pk-pk pure sawtooth waveform (actually it'll be a little less, given the finite impedance of the HT secondary winding, but not very much less).

To check the impact of this 7V RMS ripple I disconnected the negative feedback and ran the amp (nominal capability 100W/ch) at 8W output at 1kHz. The fundamental signal was at 18dBV. The 900Hz and 1100Hz sidebands were measured at -16.1dBV, the 800Hz and 1200Hz ones were at -25.3dBV and the 700Hz and 1300Hz ones were at -37.4dBV. By comparison the harmonic distortion components were at -21.9dBV (2kHz), -23.5dBV (3kHz) and -50.4dBV (4kHz). So the first set of ripple-driven sidebands were substantially larger than the largest harmonic component. Even the second set of sidebands, given that there were two of them, were comparable with the largest harmonic. Bearing in mind that the anharmonic distortion is also more offensive than the harmonic I think it's clear that the ripple-driven features are the herd of elephants in the room.

Assuming the intermodulation was linear I could have reduced the ripple distortion by, say, 10dB by tripling the reservoir capacitors. But 600uF 450V electrolytics would have been quite a big ask back in the 1960s. Instead I fitted a C-L filter (50uF and 2.5H) before the existing 100uF capacitor set. This dropped the 900Hz/1100Hz components by 33dB and the 800Hz/1200Hz ones by 40dB. That really did render them insignificant .

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is online now  
Old 20th Nov 2018, 3:32 pm   #11
G8HQP Dave
Rest in Peace
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
Default Re: What if I take my DC feed to an output transformer CT off the reservoir capacitor

Your IM figures are, within a factor of 2, what a back-of-envelope estimate would suggest. Always good when theory and practice tell the same story!
G8HQP Dave is offline  
Old 20th Nov 2018, 3:39 pm   #12
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: What if I take my DC feed to an output transformer CT off the reservoir capacitor

Thanks, that's good to know. I confess I didn't attempt the theory on this. It's worth pointing out that the distortion levels are not, overall, high. Without feedback each sideband is 34dB below the fundamental. The amp has 26dB feedback so when it's applied each sideband would correspond to 0.1% distortion. The harmonic distortion is a good deal less than this at 8W out but it will increase with output power and so may, at some point, overtake the intermodulation. Then again as the amp moves out of class A it will draw more average current from the power supply and the ripple voltage will increase too.

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is online now  
Old 20th Nov 2018, 5:11 pm   #13
G8HQP Dave
Rest in Peace
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
Default Re: What if I take my DC feed to an output transformer CT off the reservoir capacitor

10V pk AC on 600V DC is roughly 1.7%. This could cause gain of the output to vary by maybe 2.5% (roughly 1.7% from direct transconductance, about half as much again from anode impedance). 2.5% gain variation will cause 1.25% 2nd-order intermods. 1.25% is -38dB. You found -34dB.
G8HQP Dave is offline  
Old 20th Nov 2018, 9:28 pm   #14
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: What if I take my DC feed to an output transformer CT off the reservoir capacitor

Quote:
Originally Posted by G8HQP Dave View Post
10V pk AC on 600V DC is roughly 1.7%. This could cause gain of the output to vary by maybe 2.5% (roughly 1.7% from direct transconductance, about half as much again from anode impedance). 2.5% gain variation will cause 1.25% 2nd-order intermods. 1.25% is -38dB. You found -34dB.
It is about what I thought , a few volts (meaning 3v) ripple is ok, but if its 10 or 20V ripple as I suggested in post #7, it requires a lot of extra capacitance. But if you are stuck with some amp & some power transformer and you want more power output, prior to any specific distortion level, connecting the output transformer center tap prior to the choke is not a bad idea, at the expense of perhaps having to add more filter capacitance depending on the case.

Last edited by Argus25; 20th Nov 2018 at 9:34 pm.
Argus25 is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 8:54 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.