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 > Other Discussions > Homebrew Equipment

Notices

Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members.

Closed Thread
 
Thread Tools
Old 20th Jan 2018, 11:57 pm   #21
Philips210
Nonode
 
Join Date: May 2007
Location: Redruth, Cornwall, UK.
Posts: 2,562
Default Re: Wanted: Circuit for HV PSU

Hi Al.

That's an impressive looking and very useful power supply and thanks for posting the details, I will have a read through the circuit description later. When I can find the time, I'll probably have a go at building your circuit.

Regards
Symon.
Philips210 is offline  
Old 21st Jan 2018, 12:53 am   #22
Terry_VK5TM
Nonode
 
Terry_VK5TM's Avatar
 
Join Date: Oct 2010
Location: Tintinara, South Australia, Australia
Posts: 2,324
Default Re: Wanted: Circuit for HV PSU

Can I mention one common mistake in Al's design (and perpetuated in many other designs).

The volt meter should go after the mA meter to read the true output voltage.

In a HV psu I doubt it would make a great deal of difference, but in low voltage designs it can lead to reading differences of up to a couple of volts on either side of the Amp meter.

A couple of volts may not sound like much, but if you're running a 5v circuit and only getting 3v after the amp meter, it can lead to a lot of head scratching.
__________________
Terry VK5TM
https://www.vk5tm.com/
Terry_VK5TM is offline  
Old 21st Jan 2018, 1:18 am   #23
julie_m
Dekatron
 
Join Date: May 2008
Location: Derby, UK.
Posts: 7,735
Default Re: Wanted: Circuit for HV PSU

But if you put the ammeter upstream of the voltmeter, then the ammeter is going to be measuring the current drawn by the voltmeter on top of the current flowing through the load .....
__________________
If I have seen further than others, it is because I was standing on a pile of failed experiments.
julie_m is offline  
Old 21st Jan 2018, 10:56 am   #24
mole42uk
Nonode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Resolfen, Wales; and Bristol, England
Posts: 2,588
Default Re: Wanted: Circuit for HV PSU

Isn’t that why the Farnell power supplies have a dual-scaled meter with a selector switch to read either current or voltage?
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 21st Jan 2018, 11:41 am   #25
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Wanted: Circuit for HV PSU

I don't think a switched meter helps the fundamental problem does it ? If you have the voltmeter switched in and set the voltage to 5V, then you switch the voltmeter out and the current meter in, the actual voltage on your load will no longer be 5V because now you have the current meter in, which you didn't when you set the voltage. This is one of the real beauties of modern digital voltmeters - they typically draw hardly any current at all, so if you put them after a current meter they won't change its reading. One solution using old technology was to set up a potentiometer circuit. When a voltage potentiometer is balanced it draws zero current https://en.wikipedia.org/wiki/Potent...ing_instrument.

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is offline  
Old 21st Jan 2018, 12:58 pm   #26
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Default Re: Wanted: Circuit for HV PSU

Thanks for all your responses.
V before I or I before V? (Nothing to do with spelling!) I suppose it's just another example of trying to have one's cake and eating it: where do all those currents go?

Al.
Skywave is offline  
Old 21st Jan 2018, 1:03 pm   #27
eddie_ce
Hexode
 
eddie_ce's Avatar
 
Join Date: Aug 2010
Location: nr. Hannover, Germany
Posts: 372
Default Re: Wanted: Circuit for HV PSU

Guys,
many thanks for all your constructive comments.

I have put together the attached circuit and I would be extremely grateful for your comments and suggestions. I have merged one or two ideas from others into the circuit so can claim no originality for it.

I am not sure how much voltage to expect at TP1 which makes further calculations a bit dodgy. I would like to achieve an output range of about 150 to 265V and imagine I will have to experiment with Rtop and Rbottom.

Components have been chosen because I have them in my cupboard.

Does this circuit seem realistic and viable?

Many thanks, as always
Attached Files
File Type: pdf PSU1.sch (2).pdf (42.3 KB, 129 views)
__________________
Eddie
BVWS Member.
Friend of the BVWTM

Last edited by eddie_ce; 21st Jan 2018 at 1:09 pm. Reason: spelling
eddie_ce is offline  
Old 21st Jan 2018, 1:45 pm   #28
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Arrow Re: Wanted: Circuit for HV PSU

Quote:
Originally Posted by Terry_VK5TM View Post
Can I mention one common mistake in Al's design (and perpetuated in many other designs).

The voltmeter should go after the mA meter to read the true output voltage.

In a HV psu I doubt it would make a great deal of difference, but in low voltage designs it can lead to reading differences of up to a couple of volts on either side of the ammeter.
Terry: valid point; thank you.

Inspired by your post, I did a few sums for the two cases of a H.V. and a L.V. PSU. In both cases, I assumed an ammeter with a native FSD of 1 mA and an internal resistance of 1000 Ω

Case 1: L.V.
Assume load current= 1-amp. For FSD of 1-amp., the mA meter needs a shunt of 1 Ω. So volt-drop across mA meter = 1v. Hence, for a voltmeter reading of 5v., 4v. across the load.

Case 2: H.V.
Assume load current= 100 mA. For FSD of 100 mA, the mA meter needs a shunt of 10 Ω. So volt-drop across mA meter = 1v. Hence, for a voltmeter reading of 100v., 99v. across the load.

Case 1 and Case 2.
With the voltmeter across the load instead and assuming it also has a native FSD current of 1 mA and an internal resistance of 1000 Ω, a resistor is obviously required to be in series with that meter to produce the required FSD. However, the max. current drawn by that meter for both cases will still be 1 mA: negligible compared to 1-amp. and 100 mA drawn by the load.

When I produced my design, did I take Terry's considerations into account? Can't recall now - all too long ago. But Terry's point is well-worth bearing in mind for future reference.

Al.
Skywave is offline  
Old 21st Jan 2018, 1:56 pm   #29
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Arrow Re: Wanted: Circuit for HV PSU

Eddie: the EL86 has a max. anode voltage of 200-v. Your design will substantially exceed that value. I would also be inclined to choose a choke input filter, anyway. And a potential divider for the screen-grid voltage of the EF80 might be a better choice, too.

Al.
Skywave is offline  
Old 21st Jan 2018, 3:29 pm   #30
eddie_ce
Hexode
 
eddie_ce's Avatar
 
Join Date: Aug 2010
Location: nr. Hannover, Germany
Posts: 372
Default Re: Wanted: Circuit for HV PSU

AL,
I was under the impression, that having a floating heater supply and an assumed max anode voltage of 350V and a minimum cathode voltage of 150V I would just be in the safe region of 200V.

So is the 200V rating an absolute value irrespective of the potential of the valve's other electrodes?

If so, it looks like EL34s.
__________________
Eddie
BVWS Member.
Friend of the BVWTM
eddie_ce is offline  
Old 21st Jan 2018, 4:07 pm   #31
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Arrow Re: Wanted: Circuit for HV PSU

Eddie - my remark was based on data from the National Valve Museum:

http://www.r-type.org/exhib/abv0002.htm

plus the Mullard datasheet. (As below as an attachment).

If you can obtain them, my favourite for series-pass regulator valve duty has always been the 12E1, but they tend to be a bit pricey. OTOH, the venerable 807 is a bit lower on Va, Ia and Pa max., but usually substantially cheaper. On account of their popularity in the valve Hi-Fi world, EL34 valves are becoming seriously expensive.

Al.
Attached Files
File Type: pdf EL86-Mullard.pdf (81.0 KB, 60 views)

Last edited by Skywave; 21st Jan 2018 at 4:19 pm.
Skywave is offline  
Old 21st Jan 2018, 4:36 pm   #32
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Wanted: Circuit for HV PSU

Quote:
Originally Posted by eddie_ce View Post
AL,
I was under the impression, that having a floating heater supply and an assumed max anode voltage of 350V and a minimum cathode voltage of 150V I would just be in the safe region of 200V.

So is the 200V rating an absolute value irrespective of the potential of the valve's other electrodes?

If so, it looks like EL34s.
I followed the link to Al's datasheet for the EL86 and on the back page it says that the limiting value for Va is 250V, not 200V. This is, by convention, measured relative to the valve's cathode so you're right, with 350V on the anode and 150V on the cathode you'll be fine. You might worry a bit about what happens at switch-on which will depend, among other things, on whether you use a solid-state HT rectifier and on whether you're feeding a large, so slow-to-charge, capacitor on the output terminal. But I'd be surprised if this produced any real trouble. Cold, the valve is rated for 550V Va !

Cheers,

GJ
__________________
http://www.ampregen.com

Last edited by GrimJosef; 21st Jan 2018 at 4:42 pm.
GrimJosef is offline  
Old 21st Jan 2018, 5:07 pm   #33
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Arrow Re: Wanted: Circuit for HV PSU

Quote:
Originally Posted by GrimJosef View Post
I followed the link to Al's datasheet for the EL86 and on the back page it says that the limiting value for Va is 250V, not 200V.
Yes indeed: the limiting value. But no respectable design will use that as a 'design centre' parameter. Besides, the said datasheet states that the 200v. is for the valve operating under Class-A conditions, which I believe is applicable in Eddie's design.

As for those voltages, yes, it can safely be assumed that they are relative to the cathode - although I didn't see that explicitly stated.

Al.
Skywave is offline  
Old 21st Jan 2018, 5:11 pm   #34
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Wanted: Circuit for HV PSU

Quote:
Originally Posted by eddie_ce View Post
... Does this circuit seem realistic and viable?
Mostly it looks good. But there is one serious flaw. You cannot put a 10uF capacitor across the 85A2 gas-discharge valve. Valves of this type have a 'negative resistance' region in their characteristic curves and too much parallel capacitance turns them into relaxation oscillators. It is often claimed that they will destroy themselves like this, even explosively !

This is perhaps the main drawback with gas-discharge voltage references. They are inherently noisy and it seems like a good idea to smooth this noise away. But you can't do it with just a capacitor (if you use a solid-state Zener diode you can, but I have seen too many blown Zener diodes to trust them in valve circuits). Steve Bench, in my earlier reference, suggests (his page 3) that you use no more than 10nF across the gas-discharge tube. Other people are braver and will risk up to 100nF. If you want to quieten the reference further then one way to do it is to use a triode-pentode instead of the pentode error amplifier. the pentode section still acts as the amplifier, but the triode is used as a cathode follower with the gas-discharge tube setting its grid voltage and its cathode wired to the pentode's cathode. Now you can put a large capacitor across the common cathode resistor to eliminate the reference noise without blowing the reference valve up.

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is offline  
Old 22nd Jan 2018, 1:08 am   #35
Terry_VK5TM
Nonode
 
Terry_VK5TM's Avatar
 
Join Date: Oct 2010
Location: Tintinara, South Australia, Australia
Posts: 2,324
Default Re: Wanted: Circuit for HV PSU

Cheers Al

Same thing happens using DMM's/multimeters etc when measuring current.

Caught me out once
__________________
Terry VK5TM
https://www.vk5tm.com/
Terry_VK5TM is offline  
Old 22nd Jan 2018, 8:44 am   #36
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,636
Default Re: Wanted: Circuit for HV PSU

There's this circuit that has been posted before - http://www.parcgwyn.talktalk.net/valvetester.htm
uses an EL81, but any circuit that uses line OP valves has got to be better than using EL84's. For not so simple variable valve PSU's see here - http://diyaudioprojects.com/mirror/m...ench/reg1.html

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is offline  
Old 22nd Jan 2018, 10:08 am   #37
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Wanted: Circuit for HV PSU

Quote:
Originally Posted by Diabolical Artificer View Post
... but any circuit that uses line OP valves has got to be better than using EL84's ...
It depends a bit on what we mean by 'better' and how close to the limit we are trying to get. If we need to supply high currents (hundreds of mA) then line output valves can be quite cheap options for this. But if our current requirements are more modest then they may not be so good. I recently built a supply where I was aiming to get the best possible regulation against variations in current demand. This meant the supply had to have the lowest possible output impedance. Roughly speaking the output impedance will be 1/gm, where gm is the mutual conductance of the pass valve, divided by the closed loop gain of the error amplifier. Using a pentode as the error amp a gain of 100 isn't too hard to achieve. Then it becomes a matter of finding a pass valve with a nice high gm. At high currents the gm of line OP valves can be decently high. But if we only run them at a low current, because that's all the load needs, then the gm will fall away and the supply's output impedance will rise. In the end it becomes a matter of running the pass valve at a point where it's working reasonably hard. If the current demand is low then a smaller pass valve may well be better than a larger one.

I should say that with an EL86 and an EF86 delivering 250V/50mA I was able to achieve an output impedance of well below an ohm. At this point the resistance of the umbilical cable to the device I was powering started to matter !

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is offline  
Old 22nd Jan 2018, 5:20 pm   #38
short wave
Pentode
 
Join Date: Mar 2013
Location: Leeds, West Yorkshire, UK.
Posts: 246
Default Re: Wanted: Circuit for HV PSU

How about using "part" of a Farnell circuit for a "E350" also uses EL34s (as requested). Here is the manual..
regards S-W
short wave is offline  
Old 22nd Jan 2018, 6:54 pm   #39
eddie_ce
Hexode
 
eddie_ce's Avatar
 
Join Date: Aug 2010
Location: nr. Hannover, Germany
Posts: 372
Default Re: Wanted: Circuit for HV PSU

Short Wave
Building an E350 is still on the to do list, I have most of the parts but lack time. So I want to knock up a relatively simple supply over the next couple of weeks.

GrimJosef
Many thanks for your helpful comments, I did not know about the perils of putting a capacitor across a stabiliser. I have now redesigned and am using an ECF80, the triode as a cathode follower as you suggested.

I'm away for a few days, but when back would like to re-post the modified circuit for members' scrutiny.

Is there a formula to calculate the expected off-load voltage at TP1 on my circuit? This would make further calculations a lot easier.

I really appreciate the time you all take to reply to this thread.
__________________
Eddie
BVWS Member.
Friend of the BVWTM
eddie_ce is offline  
Old 23rd Jan 2018, 12:41 am   #40
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Arrow Re: Wanted: Circuit for HV PSU

Quote:
Originally Posted by eddie_ce View Post
Is there a formula to calculate the expected off-load voltage at TP1 on my circuit? This would make further calculations a lot easier.
Well my understanding goes like this. With no-load at TP1, the reservoir capacitor will charge up to the peak value of half of the full secondary winding voltage (mains xfmr.). Hence, 300 x 1.4 = 420 v.d.c.

At low load currents, the voltage to the anode of the pass valve will approach that figure: so there could be problem then (Va max. rating). That's one of the reasons why I usually advocate using a choke input filter: the off-load / low-load voltage (at your TP1) is lower; the inherent regulation of the cct. is better. And at low O/P voltages & with substantial load currents, the pass valve has an easier life.

Al.
Skywave is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 2:48 pm.


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.