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 > Specific Vintage Equipment > Vintage Test Gear and Workshop Equipment

Notices

Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders.

Closed Thread
 
Thread Tools
Old 1st May 2021, 4:28 pm   #1
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Pye 8440 valve-stabilised bench power supply

So, having picked up this Pye 8440 power supply https://www.vintage-radio.net/forum/...d.php?t=178658 from Woodchips (thanks again !) I’ve now set about reviving it. The first thing was to trace out the circuit (see attached pics). I've split it into four, partly in the hope you'll still be able to read it at 800 x 600 and partly because these sections are relatively distinct from one another.

I’ve been as careful as I can, and without examining every last detail it looks OK to me, but I’d be very grateful to anyone who could go through it and check for ‘that can’t possibly be right !’ errors. The only components for which I don’t have values are the HT+ smoothing choke and the mains transformers’ secondaries (in most cases neither voltage nor current) but if you see any other missing labels, again please let me know.

The basic architecture is straightforward. Mains power is supplied directly to the primary of T1, via a wander-plug voltage selector in the neutral line. This winding also acts as an autotransformer for the primaries of the other two mains transformers, T2 and T3, which don’t have their own primary voltage selection. T1 simply powers the valve heaters. But there are 15 of these, with a total current of 11.6A at 6.3V and 4A at 5V as well as provision of two isolated 5A 3.15-0-3.15V supplies to terminals on the front panel. So it needs to be large, and it is. The mains live supply to T2 and T3 is delayed using a DLS16 switch (V14), thereby giving time for all the valve heaters to warm up before HT is applied.

T2 has just one secondary, providing the AC for the 0-500V 250mA HT+ supply. The output of this supply is delivered in four switched ranges, 0-125V, 125-250V, 250-375V and 375-500V. To limit the dissipation in the regulation circuitry the primary of T2 is also switched so that the secondary generates a suitable output.

T3 has three secondaries each of which feeds a metal rectifier bridge. The first provides power for the fixed -250V HT- supply. The second and third generate independent HT supplies used only inside the unit. One of these powers the HT+ supply’s error amplifier. The other powers the screen and control grids of the HT+ supply’s pass valves.

The HT- supply derives its reference voltage from an 85A2 (V15) and compares this with a fraction of the -250V output in a 12AX7 (V8) long-tailed pair error amplifier. A pot on the rear panel of the unit, VR1, determines the fraction and can therefore set the output to -250V exactly. The error amp feeds another 12AX7 (V9) which, in turn, drives the EL81 (V7) pass valve. There is a 100k pot across the -250V output providing a variable negative supply rated at a nominal 1mA. The main (fixed) -250V output is rated at 30mA. The whole HT- supply is protected by a 100mA fuse immediately after the metal rectifier. When I opened the holder for this there was a 250mA fuse in it. That might explain why the filter choke, which comes after the fuse and which was rated at 75mA, had failed open-circuit.

The 0V rails in the HT+ and HT- supplies are commoned together. They’re connected to the metal chassis by a 0.1uF 1000V capacitor. The chassis is safety-earthed.

The HT+ supply uses two parallel-strapped 5R4 rectifiers (V12, V13) feeding a choke-input filter. This contains a relatively low value paper-in-oil reservoir capacitor (in fact all of the HT reservoir and smoothing capacitors in the 8440 are paper-in-oils). The HT+ pass valves are 5 EL34s (V1-5) wired in parallel. Following them is the output meter which can be switched between one voltage (0-500V) and two current (0-50mA and 0-250mA) ranges.

The HT+ pass valves’ control grids are driven from the HT+ error amplifier and driver. The latter is another EL81 (V6). The error amp is a long-tailed pair made up of two EF91s (V10, V11). V10 uses a 470k+22k potential divider across the HT- supply as its reference. V11 is fed by a fraction of the difference between the HT+ and HT- outputs, the fraction being determined by a range-switched resistor ladder and also by the HT+ voltage control pot. The final valve is an ECC81 (V16) with its two triodes paralleled and wired between the HT+ error amp’s HT rail (anodes, via 1k loads) and 0V (cathodes). The control grids are set at a voltage somewhere between the HT rail and HT- by an unlabelled pot inside the supply.

I’ve got three issues with the circuitry:

1. I don’t know what V16 is for. It appears to act as a variable current drain on the HT+ error amp’s HT supply. It will also be affected by variations in HT-. Given that a pot’s been provided to set the current flowing through it, it would be handy to know where to set it. I’d be grateful for any help with that.

2. As already noted, the HT- supply input choke has failed and when this happened the -250V will have disappeared. If the 100mA fuse had blown the same thing would have happened. Given that the HT+ supply uses the HT- one as its reference, I’m not sure that the HT+ circuitry will have failed safely when the HT- disappeared. Would the EL34s or the rectifiers have been at risk ? Should I be checking for damage to any other components ?

3. The data sheet for the DLS16 http://www.r-type.org/pdfs/dls16.pdf contains an ‘Important Note’ which says “To prevent any potential difference existing between the moving contact and heater, the two should be tied externally”. In the 8440 the moving contact is connected to mains live. The heater is connected to 0V, which is common to both the HT+ and HT- supplies and is not connected to safety ground !! In the event of a short between the moving contact and the heater, mains live could appear unexpectedly on the supply’s outputs. I’m shocked (no pun intended) about this. How hard would it have been to add an extra isolated winding to T1 for the DLS16 heater ? I’m actually going to fit a separate small 6VAC mains transformer inside the chassis to do this job. Can anyone see a problem with that ?

Apart from the above I’m optimistic that I can get this working again. A significant problem is lifting it onto the workbench though. With its slide-on safety cage the whole thing weighs 43kg !

Cheers,

GJ
Attached Thumbnails
Click image for larger version

Name:	Pye 8440 mains heaters earth.jpg
Views:	131
Size:	40.9 KB
ID:	233018   Click image for larger version

Name:	Pye 8440 HT- supply.jpg
Views:	116
Size:	39.1 KB
ID:	233026   Click image for larger version

Name:	Pye 8440 HT+ supply pass & meter.jpg
Views:	112
Size:	42.5 KB
ID:	233027   Click image for larger version

Name:	Pye 8440 HT+ error amp.jpg
Views:	116
Size:	50.8 KB
ID:	233028  
__________________
http://www.ampregen.com

Last edited by GrimJosef; 1st May 2021 at 4:53 pm.
GrimJosef is online now  
Old 1st May 2021, 7:41 pm   #2
turretslug
Dekatron
 
turretslug's Avatar
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
Default Re: Pye 8440 valve-stabilised bench power supply

I wonder if the "important note" re. the DLS16 was introduced after the event, so to speak- it might be that the device designers felt that the need to avoid potential difference between heater and switch elements was obvious, but it ended up being implemented in circuits that assumed it could be used like a high-isolation relay with unfortunate outcomes. Another use of a device that makes my toes curl slightly is in those voltage stabilisers like the VR150 that included an interlock link between base pins that sometimes got used as a primary-side link. Maybe I'm just too deeply immersed in modern expectations of safety isolation! I'd certainly agree that the separate heater transformer scheme sounds wise.

I assume that V16 provides something like open-loop compensation by its shunt action for supply variation impinging on the less-than-ideal characteristics of the comparator circuitry and would be adjusted for minimum output voltage change as the supply voltage was varied over a specified range? I've noticed that, as valve-based HT supplies (in particular, but also more generally) seek to go from "goodish" to "very tight" standards of regulation, the design needs to compensate for the fact that active elements such as comparator and reference can no longer be regarded as ideal, unchanging and consistent between samples, and open-loop and feedforward techniques that need individual adjustment make an appearance.

Good luck with it, I find apparatus like this decidedly intimidating from both electrical and physical stances,

Colin
turretslug is offline  
Old 2nd May 2021, 1:02 am   #3
trobbins
Heptode
 
trobbins's Avatar
 
Join Date: Oct 2009
Location: Melbourne Australia
Posts: 898
Default Re: Pye 8440 valve-stabilised bench power supply

I'd certainly be making a safety and performance mods list - I see it as an expected restoration action, given that the aim would be to make the supply operational and to put it to use once again.

One common safety issue has been looming mains AC and secondary side wiring together, which I am not a fan of. So rewiring the neutrals to be 'unswitched' would be a start. I concur about the delay relay heater being isolated by an interposing isolated transformer.

Given the output stage grids are referenced back to the -250V rail to MR2 rail difference, then V16 makes sense. Perhaps some trial and error calibration loops will make its importance more obvious.

There would seem to be a few 'protection' related mods to try and mitigate future faults, such as more assessment of fuse location and value, and ss diode protection of the 5R4's, and some form of interlocking of the EL34 screen/bias supply if T2 goes off-line, and even some form of over-current fold-back or limit or inhibit.
trobbins is offline  
Old 2nd May 2021, 10:39 am   #4
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Pye 8440 valve-stabilised bench power supply

Thanks for the comments.

trobbins might well be right about the DLS16 warning being a later addition. The date codes on the 8440's capacitors are from the first half of 1958, so I guess it will have been produced a little later than that. The data sheet with the warning is from 1961. But the DLS16 had been on the market since 1949, so perhaps designers should have been familiar with it ? The mechanism is quite simple. The attached pic shows the moving contact, which is in the shape of a P-clip, with the heater wire running vertically up through the centre of it. As it heats up the P-clip opens, connecting the vertical pin on the right behind it to the one in front of it. Obviously if the heater should snap there's a real risk of it touching the moving contact.

The wiring loom looks a bit higgledy-piggledy. Components aren't positioned as close to one another as they might be, so connecting wires can be long, and they also seem unusually thin (presumably with correspondingly thin insulation). Of course what looks higgledy-piggledy to me might be some enormously subtle layout designed to minimise noise and drift . There is some degree of matching wiring colour to function, but it's not quite universal. Most, but not all, of the 0V wires are black. Most, but not all, of the -250V wires are blue. Fortunately the loom is 'laced' with wire buckles, which can be undone, so there is the option of separating wires to trace them and also of changing them if desired. For aurthenticity's sake I'm minded to leave as much as possible in place, and fortunately most of the mains wiring is run in sheathed (in fact metal-mesh screened !) twin-core. But, again not all, and the screens tend to be connected to 0V, not the safety grounded chassis. So there is another risk there if the insulation should fail.

As for V16, I suspected it could be used to minimise some undesirable aspect of the performance. Its physical position, and that of its set-point pot, look a bit as though they've been shoe-hormed into spaces in a pre-existing layout, and that might be supported by the fact that V16 is the highest of the valve numbers, each of which is hand-stamped into the chassis. Then again, V15 is the 85A2, which must have been there right from the start.

Cheers,

GJ
Attached Thumbnails
Click image for larger version

Name:	DLS16 mechanism.jpg
Views:	87
Size:	28.6 KB
ID:	233057  
__________________
http://www.ampregen.com
GrimJosef is online now  
Old 3rd May 2021, 7:46 pm   #5
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Pye 8440 valve-stabilised bench power supply

I've just discovered a couple of errors in the third of the circuit diagrams in post #1 - the one that shows the HT+ supply, pass valves & meter.

The 15k resistor which acts as the internal load for this supply was missed off the first version of the diagram (without it, and in the absence of an external load, there's no DC path from the pass valves' cathodes back to 0V).

I was aware that this 15k resistor existed and when drawing the circuit out I indvertently attached its value to the load resistor for the raw supply, before the pass valves (this load is needed to keep the choke-input filter voltage under control in the absence of any other load). That resistor should be 33k, not 15k.

I've attached an amended version of the circuit. Please disregard the version in post #1.

Cheers,

GJ
Attached Thumbnails
Click image for larger version

Name:	Pye 8440 HT+ supply pass & meter.jpg
Views:	61
Size:	43.0 KB
ID:	233181  
__________________
http://www.ampregen.com
GrimJosef is online now  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 3:34 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.