Douglas Record Player - Information Sought

[af024]

Dear all,

I am attempting to locate some information about a Douglas record player that was kindly offered to me by Barry in this thread:-
https://www.vintage-radio.net/forum/...hlight=douglas
Thank you Barry!

The logo on it appears as that shown on this website:
http://www.douglas-transformers.co.uk/
and the transformers in the amp are also Douglas-badged.

It's a single ECL82 valve amplifier.

Might it have had an Isle of Man connection perhaps?

I can't find anything about the brand, never mind this particular player anywhere.

Fingers crossed that someone here knows something. I'd love to see a circuit.

Regards,


Andy

[af024]

I've sat down to try and sketch the amp circuit out. I attach the result. Sorry for the poor sketch, it was a rushed first rendition.

I must say that I'm rather confused though - just look at the triode anode to pentode coupling - ie a solid link.

Just for fun, here are some voltages:-

Pin Volts
1 gt -1
2 kp 22
3 g1 14
4 h
5 h
6 ap 106
7 g2 108
8 kt 0
9 at 14

So how does this work then chaps?

Also, are the anode volts really low? I'm only seeing about 110vd.c. at the res cap.

Regards,

Andy

[Darren-UK]

They're listed on the Companies House website as being incorporated in 1946 but, so far, that's all I've been able to find.

Any connection with the Isle of Man is extremely unlikely. I've also checked the Isle of Man Companies Register and nothing even remotely like it appears.

[Herald1360]

Hmmm, shades of a Mullard 3-3 here perhaps, though executed with different valves.

No reason not to direct couple At to Gp if the cathode of the pentode is raised up far enough by its Rk for the grid to sit at a high enough voltage for the triode to work. Is that triode anode resistor really 10meg? I would have thought 1meg, maybe.

There seems to be something very odd about the lack of any return path to HT- for the valve cathodes, though.

It's certainly an interesting circuit, there's even an ultralinear tap on the O/P TX primary.

Given the logo situation you describe, I wonder if Douglas IoM is a red herring. Maybe the player was commissioned by Douglas Transformers for some reason (promotional freebie maybe, or just to sell more transformers)?

[AndiiT]

Hi,
It certainly is an intesting circuit!, I re-sketched it using Tiny Cad and was similarly puzzled about the lack of HT -Ve return path for the cathodes too!

Whilst not being too clued up on what the typical component values should be, I am not too sure if the output transfomer tap is "ultra linear" as Herald 1360 mentions or if the tap is just a means of securing the correct screen grid voltage for the pentode section of the output stage.

I am sure that some of our more informed forum members may be able to talk us through the circuit operation soon

Regards

Andrew

[Herald1360]

Originally Posted by Darren-UK

They're listed on the Companies House website

This link seems to have timed out?

You can just start again from here:

http://wck2.companieshouse.gov.uk

it would seem.

[Herald1360]

The O/P transformer tap will feed back a proportion of the anode signal voltage to the screen which has the same effect whether the circuit is push pull or single ended as here. The dc on the screen will just be the same as the anode dc (in fact slightly higher, depending on the winding resistance).

[af024]

Thank you for your replies thus far gents - all very interesting.

Yes it's definitely 10M (brown-black-blue). Amazing though it may seem.

I've taken another close look at the lack of HT- return. My mistake, sorry for sending you all into a spin. There's a hidden sneaky link under the tag board between T1 and T6 (centre bottom on my drawing).

Thank you for having a go at re-drawing Andrew. I'm not sure whether I'm doing something wrong but I can't see said drawing at the moment. Maybe it doesn't allow others to view?

I'm amazed as to how as little as 14V on the anode of the triode section is anywhere near enough to get things going. Mind you, I'm confused about the rest of it too.

Now there's a thing, make a player to sell more transformers. Well you never know!

Regards,

Andy

[Herald1360]

Right. If T1 and T6 are grounded the circuit makes more sense! The 10K and 2K2 at the bottom provide a divider for negative feedback from the output to the cathode of the triode. The 2K2 is the triode cathode resistor, but won't have much biassing effect at the starvation current through the triode.

Going back to voltages, the HT looks very low. I would be suspicious of the rectifier- maybe substitute a 1N4007 plus a 200 ohm resistor and see what happens? Check the dc across the series 47ohm, though, just in case excessive ht current is the culprit possibly via the pentode or perhaps the filter capacitors. Mind you, if the voltage at Kp is across an OK 1K5 resistor, it suggests a low anode current (probably an effect of the low HT, given the direct coupled circuit).

Looking at the ECL82 data it seems like a promising candidate for this sort of circuit- it has a highish negative grid requirement of some -22.5V to start with. If you sit the cathode up by another 20V or so you get a bit of headroom for the triode anode volts. The triode itself is characterised for zero bias with a 10meg grid leak.

[AndiiT]

Originally Posted by af024

Thank you for having a go at re-drawing Andrew. I'm not sure whether I'm doing something wrong but I can't see said drawing at the moment. Maybe it doesn't allow others to view?

Hi,
I haven't posted my re-drawing, I simply did it to get a better look at the circuit! If you wish I can post it but it won't be for a while as I am currently not at the computer that I have saved the drawing on.

Another thing that puzzles me a little about your circuit is the power supply arrangements; as it is shown in your drawing the mains transformer supplies the valve heaters but the HT rail is mains derived using AC/DC (live chassis) techniques, with what appears to be the mains earth as the return path!

Regards
Andrew

[AndiiT]

Hi,
I have just managed to re-draw the circuit again and have posted it on here, I have left the mains transformer and motor out of the circuit and Tiny CAD only has the option of separate valve envelopes but I am sure that anyone looking at the circuit will get the general idea.

Regards
Andrew

[Herald1360]

The power supply is typical of the cost saving tricks used on record players of the time. More typically the transformer would have been the coil on the deck motor with a heater tap, but here it's just a heater transformer. The HT is simply derived from the mains input via a half wave rectifier.There is no isolation between HT- and mains (hopefully neutral, but works just as well if it's the line side).

Maybe just my viewer, but the jpeg of the circuit is almost unreadably low contrast. I had to juggle with it in a graphics package to see the lines.

[Mike Phelan]

I can't read it either!

[af024]

Thank you again for everyone's comments an input. Thank you too for your drawing Andrew, I too am struggling at the moment to make it out so hopefully it's a simple fix to get it to display ok (especially after all your hard work).

Yes the HT is definitely derived straight off the mains and the chassis is connected directly the mains neutral. I am slightly worried about the connection to/from the deck cartridge however since the metal screw holding the cartridge in place is exposed and connected via screened wire to the chassis. There is no isolation at all. I suppose I could connect this to the deck metalwork (floating from chassis) leaving just one side of the cartridge connected directly to the amp chassis? Well either that or add some suuitable caps. Not sure.

I too wondered about the available HT. I tried a 1N4007 in series with an additional 300 ohm and I'm getting approx 275Vd.c. at the reservoir cap. It's a bit on the edge though since the cap is only rated at 275V max. It certainly sounds much much better with the higher HT.

I suspect that the Sentercel is poorly. It doesn;t look like there is excessive current being drawn.

So what do you think is a sensible HT given that the reservoir cap is rated at 275V?

Chris, are you saying that it would be a good idea to consider increasing the value of the 2k2 in the cathode leg of the triode section? At present (admittedly with the low HT) there's only 1V between grid and cathode.

So how does a 10M in the anode of the triode section give anywhere sufficient ummph?

Interesting.

Regards,

Andy

[Herald1360]

If the caps are rated for 275V, I would increase the 300R somewhat. I should think that 250V would be a reasonable figure to aim for.

I wasn't suggesting any value changes for the circuit, it was just a comment that the 2K2 was there for the benefit of the feedback path rather than biassing for the triode.

At normal cathode current the pentode cathode should be at about 75V. Its grid should be about 22V below this about 53V. This is the anode voltage for the triode, which would suggest an anode current of about 20uA. Not unbelievably low, but maybe a 1meg anode load and 200uA would make more sense.

It would be interesting to know the the cathode voltages for the pentode and triode as it is now (I don't think any realistic measurements can be made for the triode anode/ pentode grid because of the high impedance circuitry).

Check first with everything off what the resistance to chassis is from the cathodes of each valve so that the voltages can be translated into meaningful currents.

The screening and screw mount arragement sounds very dodgy. At the very least I'd want to see any accessible metal that needs a signal ground to be stood off from neutral via Y-rated capacitor of less than about 10nF. It would be best to earth such metal properly and rearrange the cartridge signal connections via a couple of such capacitors. The cable screen should go to earth or be disconnected altogether and rely on the cartridge wires being twisted for hum rejection.

[af024]

Hello Chris,

Thank you for your further helpful comments. Are you saying get to 250V ish and do the voltage measurements at that HT?

Regards,

Andy

[Herald1360]

Yes.

At a guess, HT current should be around 35mA, so try increasing the 300R to around 680R. For a quick trial, swap the 680R smoothing resistor with the 300R surge limiter.

[AndiiT]

Hi,
A repost of the circuit, hopefully at higher contrast

Regards
Andrew

[Herald1360]

Yup, it's readable now. Still a bit blurry, though.

I guess the route from drawing package to jpg is a bit uncertain.

[af024]

OK here's an update.

I started off with 1k8 in series with the existing 47R, then a 1N4007. It's a good job I did really since before the valve warms up, the volts at the reservoir cap are approx 300Vd.c.. As the valve warms up it stablisises at around 190Vd.c. I decided to get as close to 250Vd.c. as I could by having the 1k8 in series with a 680R on start up and then shunting the 1k8 out once warmed up. I managed to get to 245Vd.c. at the reservoir cap doing this. Obviously I can have that as a permanent solution. Goodness knows what the characteristic of the old Sentercel was. I guess at start up and cold it offered high resistance which came down as the thing got warm?

Anyhow, with 245Vd.c. at the reservoir cap, I get 230Vd.c supply after the existing 680R resistor. Under these conditions the voltages around the ECL82 are as follows:-

1 0
2 37
3 20
4 h
5 h
6 223
7 226
8 0
9 20

I measured the resistance to chassis from each cathode and the results are below:-

2 1k8
8 2k

So guys, do I need to shunt the reservoir to chassis with a zener perhaps to protect the reservoir cap?

Regards,


Andy