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 18th Sep 2016, 1:17 pm   #1
ukcol
Rest in Peace
 
ukcol's Avatar
 
Join Date: May 2003
Location: Harlaxton, Lincolnshire, UK.
Posts: 3,944
Default Circuit idea for discussion.

I pulled a Pye P116U Piper off my "to do" pile earlier this week and got it going. When I had done I decided to change the 220 ohm cathode resistor in the output stage to 470 ohms. This was done to protect the UL41 output valve and its associated components from damage by keeping the cathode current down. The set ran all day at 33 mA of cathode current and the performance in terms of output power before distortion was only slightly down from the unmodified circuit.

Although it was not needed with this particular UL41, I wondered if there was a way of running a UL41 that suffered the leakage problem associated with the deposits on the inside of the glass envelope. So I set about designing a bias stabilization circuit that might address this problem.

I hesitate to use the word “design” as I am not a design engineer but I have had a lot of experience of circuits and have developed a “feel” for them. Had this set had a leaky UL41 I would have built this circuit and played around with it to make it work. As it is I have produced a circuit for the basis of discussion and would particularly like anyone to suggest component values based on calculations.

The circuit is shown below. Dealing with components that are part of the set’s original circuit first:

Rd is the heater dropper, Cg the audio coupling capacitor (“that” capacitor so called), Rg is the grid resistor that originally would have been returned to chassis, Rk the cathode resistor and Ck the cathode bypass capacitor (if fitted). V1 is the UL41 output valve of course.

Circuit description.

A (more or less fixed) negative supply of approximately 100 volts appears at the junction of R2/R4 having been rectified by D1 and filtered by C1,R2 and C2. The AC for rectification is taken for convenience from the top end of the valve heater chain. The current drawn by this negative supply will be very small and so will not upset the heater supply.

Next a positive supply is taken via R5 from the HT line and it is anticipated that the positive voltage at the collector of TR1 will be of the order of 100 volts. This positive voltage will be varied by TR1 and provide the control of the grid bias.

The grid bias will appear at the junction of R3/R4 and will normally be near to chassis potential. C3 provides decoupling for the positive supply. D2 will prevent the voltage at the junction of R3/R4 from going more than about 0.6 volts positive. This would happen while the output valve was warming up and could happen under fault conditions. C5 allows the DC conditions to change whilst keeping that same point at chassis potential with respect to signal.

Z1’s Zener voltage should be chosen so that when added to the 0.6 volts dropped by TR1’s e-b junction we get our “target” V1 cathode voltage. R6 and C4 provide a filter to feed TR1 base and R6 also limits the possible base current.

Rk now, instead of being the auto cathode bias for V1 is now a current sensing resitor for the control circuit. The voltage develop across Rk is compared to the Zener voltage (+ V b-e) and varies the collector current of TR1 and hence the positive voltage at its collector.

If the cathode current of V1 increases TR1’s collector voltage will go less positive and the bias fed to V1 grid will go more negative. On the other hand if the cathode current decreases TR1’s collector current will drop and its collector voltage go more positive causing the grid bias to go less negative.


My suggested components values as basis for discussion and/ or experiment are:-

R1 10k, R2 100k, R3 2M2, R4 2M2, R5 47k and R6 10k.
C1 4u7, C2 4u7, C3 10uF, C4 10uF and C5 1uF (non electrolytic).
D1 1N4007 and D2 1N4148.
TR1 BF459
Z1 11volt 0.5 watt.
Attached Thumbnails
Click image for larger version

Name:	BiasStab2.jpg
Views:	188
Size:	68.2 KB
ID:	130223  
ukcol is offline  
Old 18th Sep 2016, 7:37 pm   #2
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: Circuit idea for discussion.

You still have to have a very high Rg as this is required to keep the previous valves load from being too low so there still remains a risk. Also, the divider taken from HT/-ve supply is a problem as a 10V increase in HT will be a 3V increase in grid voltage. Stability when the transistor is conducting may also be an issue as the feedback path is very complex.

A better option might be to look at a clamp from grid to ground.

Last edited by PJL; 18th Sep 2016 at 7:43 pm.
PJL is offline  
Old 18th Sep 2016, 7:58 pm   #3
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
Default Re: Circuit idea for discussion.

That looks to be horribly-overcomplicating things: if your UL41 is getting leaky/hummy either try the 'flashing' fix or replace it with a UL84.

Must admit, if the UL41 in my Eddystone 840A (which was hummy but yielded to the 'flashing' thing) starts to be problematic again i'm minded to replace it with something like a MJE340 transistor, perhaps with a FET preamp to give the necesary high input-impedance, and a resistor to provide a pseudo-heater for the series string. A number of 1970s TVs did this.
G6Tanuki is offline  
Old 18th Sep 2016, 8:38 pm   #4
ukcol
Rest in Peace
 
ukcol's Avatar
 
Join Date: May 2003
Location: Harlaxton, Lincolnshire, UK.
Posts: 3,944
Default Re: Circuit idea for discussion.

Hi both,

My circuit uses the original grid resistor (Rg) and so I don't anticipate loading the previous stage with a low impedance. I wouldn't (in my ignorance) anticipate a 3 volt increase in HT being a problem as the increase (more positive) on the grid should be fed back and corrected by the error amplifier TR1?

Yes I realize it is a complex solution and there are other ways of addressing the problem. It was more of a thought experiment and possibly an interesting circuit to examine and discuss.
ukcol is offline  
Old 18th Sep 2016, 11:00 pm   #5
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Circuit idea for discussion.

Well, apart from the inelegance of using components based on impure sand, it looks to me as though it would work, and quite well! Cathode current would be approx. Vz1 / Rk.

Transistor circuits always seem to use a lot of components, don't let that put you off!
kalee20 is offline  
Old 19th Sep 2016, 9:02 am   #6
Guest
Guest
 
Posts: n/a
Default Re: Circuit idea for discussion.

Or use a constant current cathode "load", then the cathode will fly up to whatever volts are needed to give the current. OK you loose a bit of swing but not much.
 
Old 19th Sep 2016, 10:19 am   #7
ukcol
Rest in Peace
 
ukcol's Avatar
 
Join Date: May 2003
Location: Harlaxton, Lincolnshire, UK.
Posts: 3,944
Default Re: Circuit idea for discussion.

Quote:
Originally Posted by merlinmaxwell View Post
Or use a constant current cathode "load".....
That's a good idea and has a much lower component count.
ukcol is offline  
Old 19th Sep 2016, 5:57 pm   #8
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Circuit idea for discussion.

Transistor, Zener diode, two resistors, yes much simpler.

But UKCol is trying to overcome the effects of leakage in the UL41, and his circuit can take the lower end of Rg negative to compensate for this, leaving the grid at zero or even a bit negative itself. Whereas a CC in the cathode can't. (Though it may be only a few volts of swing, as merlinmaxwell says).
kalee20 is offline  
Old 19th Sep 2016, 6:49 pm   #9
Guest
Guest
 
Posts: n/a
Default Re: Circuit idea for discussion.

Even an extra 10V of bias would lead to less than a whole decibel of maximum volume reduction, hardly worth noting.
 
Closed Thread

Thread Tools



All times are GMT +1. The time now is 8:13 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.