Paillard 768

[Richardgr]

Following a change in circumstances with my work, I have found myself stuck in Switzerland for 6 months, working remotely for a company that needs me in the country boundaries for data privacy reasons. I decided to drive, and packed a rudimentary tinkering station, and the mandatory old Tandberg recorder, packed with oodles of cracked Rifa caps.

I could only stretch the Tandberg out so far, so I have decided to try my hand at getting a basket case radio running again. On an auction site I came across a very sorry looking Paillard from around 1936 with the speaker hanging out the back and obvious damage to the case. Close inspection of the speakers made me think it was looking worse than it really was.

I checked the Radiomuseum page ...
Paillard 7 68
... and I saw that it is quite an interesting schematic. The output power comes from a 2A3, with an 80 rectifier, and 2A6, 2A7 and 2 58's making up the remainder. Since the 'buy it now' price was cheap just for a 2nd hand 2A3, and I had 4 days of the Easter break looming in front of me, I decided to take the plunge.

It is quite elegant, but it is also very large considering the size of the chassis (35 wide by 25 deep by 23 high). The cabinet is 40 wide by 30 deep by 53 high.

It is not obvious how the cabinet was damaged. It is not due to water. I'm hoping it has not been dropped, for the sake of the tubes. The base of the cabinet is completely detached, and 2 knobs are missing. Veneer is not so bad, but one deep scrape. Thanks to this site, I have seen worse presented here and I think that once it is glued again, it might look like a different proposition.

The speaker was placed on the tubes, but is in excellent condition.

Some rust on the chassis, but nothing major. Dust, something's nest, but reasonable.

Now I need to plan the work to see if I can get it running again. The risk it was dropped is quite high, so I suppose that means focus on the chassis and get confidence in the major components.

I don't have a dim bulb tester with me, but I do have a variac and a good supply of fuses. Interestingly it has a 20mm glass fuse holder on the back that looks original.

The schema is a bit unusual since there are chokes on the ground - presumably for the directly heated tube. Also there are 5 controls on the front, 2 are potentiomers with switches, but I don't see them all on the schematic. I really want to make sure I have the schematic exactly right before I do any surgery, since I underestimated the complexity of a previous project, and was never able to recreate it again.

The schema has 'zf 120khz' written on it, which I guess is significant. Is that the intermediate frequency?

I post a few pics of the chassis in the next post.

[Richardgr]

A few more pics.

It's definately been tinkered with a few times. One time was a repair, late 60's from the pink EroFoil caps. Then at some time it looks like it was just used as a speaker.

The chassis has a metal plate for the underside, which is a nice touch. The wave change switch moves the dial up and down, and the tuning capacitor is driven by a sort of clutch working in a groove on the tuning dial wheel.

There is also some sort of signal strength indcator, in that small light over the dial.

The wax caps will be replaced, and one of the can caps has a small cork jammed in a hole on the top, but I will try to get it going slowly on the variac and leave them for later.

The biggest hassle, by the looks of it, is there is a lot of perished rubber wire running in metal braid that has been soldered to the chassis. Someone has gone to a lot of trouble to shield the wiring and tubes.

I will not be needing the 'pick up' input, or the external loudspeaker output, so I was thinking I could remove a bit of the wiring to simplify it a bit. Plus the big 12uF can under the chassis can go, and make it easier to fault find the power supply.

Does anyone know if it is OK to post a schematic from The Radiomuseum? Otherwise I will draw it again.

[Wellington]

"Swiss Made"

What a stylish set! I'm glad you rescued it, and hope you can get it going.

Originally Posted by Richardgr

Does anyone know if it is OK to post a schematic from The Radiomuseum?

I'm not quite sure what you mean - are you asking if it's OK to post a copy of the schematic on this forum? See Rule B8 here. As it's a Swiss set - I suspect it would be OK from the perspective of this forum. But possibly not from the perspective of RadioMuseum. But I'm not sure, there are others on this forum who are more familiar with that site.

[Herald1360]

Radiomuseum definitely objects to the posting of schematics downloaded from their site on public forums. Your own reverse engineered version should be OK as long as the set is not included in the service data available top right in this forum.

[Ed_Dinning]

Hi Richard, those leaf switches are notorious for poor contacts. The can be cleaned with very fine sandpaper between the contacts, which also need retensioning

Ed

[G6Tanuki]

The signal-strength indicator is probably a little neon-bulb - typically connected to either the screen-grid or anode circuit of an AGC-controlled valve - and whose brightness/pattern varies as the AGC action reduces current through the valve with increasing signal-strength.

Some were just a simple neon; others had the electrodes rod-shaped a bit like a "Jacob's Ladder" so the glow lengthened/shortened, yet others had circular electrodes so you got a glowing orange spot that got bigger/smaller as signal-strength varied.

[ms660]

Originally Posted by Richardgr

The schema is a bit unusual since there are chokes on the ground - presumably for the directly heated tube. Also there are 5 controls on the front, 2 are potentiomers with switches, but I don't see them all on the schematic. I really want to make sure I have the schematic exactly right before I do any surgery, since I underestimated the complexity of a previous project, and was never able to recreate it again.

The schema has 'zf 120khz' written on it, which I guess is significant. Is that the intermediate frequency?

The voltage developed across loudspeaker field coil in the HT-ve rail is used to bias the output valve via the potential divider that's connected across it.

So far as I can make out the extra switch connects the PU input to the input of the 1st AF amplifier, it's possible that the switch is ganged with the RF/IF gain control so that the switch is operated (connecting the PU to the amplifier) when the RF/IF gain control is set to minimum gain (maximum resistance)

The RF/IF gain control varies the cathode bias on the RF amplifier and IF amplifier valves, both cathodes being common to that control.

The tuning indicator is in series with the HT feed to the anodes of the RF and IF amplifier valves.

120kHz on the schematic is the IF for sure.

If that's any help.

Lawrence.

[Richardgr]

Thank you all, lots of useful pointers. I have built a Tubelab TSE kit for directly heated tubes, but I ran out of time to get that running before I left home, so I am lacking experience with a DHT.

My first base is to be able to power up wth no tubes to be able to verify some voltages and see if the transformer is functional, and there are no safety issues with the wiring.

The power supply is isolated, so I don't see any reason why I do not start out with an earthed power lead. There are 2 caps in series with primary, one across the mains switch (0.01uF) and one across the fuse (0.1uF), so I have a couple of X2 caps I can use for them. The caps go to the heater centre tap, which I think is significant for how the heater supply floats.

[ms660]

Originally Posted by Richardgr

The power supply is isolated, so I don't see any reason why I do not start out with an earthed power lead. There are 2 caps in series with primary, one across the mains switch (0.01uF) and one across the fuse (0.1uF), so I have a couple of X2 caps I can use for them. The caps go to the heater centre tap, which I think is significant for how the heater supply floats.

The two capacitors shown in the schematic (one from the fuse and one from the power switch) should be replaced with Y Class types, not X Class types.

The center of the artificial center tap across the valve heater winding connects to chassis but the connection dot at that point on the schematic is missing, another connection dot that's missing is where the HT side of the primary of the output transformer crosses the HT+ve rail.

Lawrence.

[Richardgr]

Thanks again! How does it work with the secondary center tap? Is there meant to be a ground at the end of the 0V rail, by the first coil that has the aerial connection?

[ms660]

Originally Posted by Richardgr

Thanks again! How does it work with the secondary center tap? Is there meant to be a ground at the end of the 0V rail, by the first coil that has the aerial connection?

All the current flowing through the amplifying valves flows through the loudspeaker field coil and the choke and back to the the center tap of the HT winding, that means that due to the voltage drop across those two coils the chassis will be +ve with respect to the center tap of the HT winding, in other words the center tap will be -ve with respect to chassis. The output valve is a directly heated type so the filament is the cathode, both sides of the filament are connected to chassis via the two low value resistors that are connected across the the heater winding on the transformer, in other words the center tap of the HT winding is -ve with respect to the filament (cathode) therefore that -ve voltage can be used to bias the output valve.

There's normally an Earth connection to chassis in receivers of this type, usually via a socket or binding post on the rear chassis apron, or less commonly via a fly lead that's connected to chassis.

Looking at the schematic I would reduce the value of the 0.1uf that's connected between the fuse and chassis to 0.01uf to match the one that's connected between the power switch and chassis, both should be Y Class types as mentioned earlier.

Lawrence.

[Richardgr]

Started to poke around a bit to get a better idea how the circuit is implemented. One thing that makes it difficut to follow is a few of the capacitors are wrapped in metal foil and grounded seperately. Also a lot of the wiring has metal braiding, which is soldered to the chassis in multiple locations. That braided wiring is rubber cored, which is quite brittle where it is exposed, but I am bargaining on it being low voltage and do not intend to disturb it, so hope that I can look at the minimum to get some signs of life, to start with.

I went carousing the second hand shops yesterday, and found a ready supply of filament lightbulbs in a 'Brockihaus'. They even had bayonet bulbs, which surprised me. There was also an E27 ceiling fixture, so Bob was my uncle, for the princely sum of one quid!

I powered up the radio with no tubes and the 60W bulb in series to see how the transformer was. Not much point continuing if that is not functional. That checked out OK, but I had some very strange behaviour with the dim bulb. It came on quite bright at 150v on the variac, then flickered dull and bright a lot. Subsequent starts seemed to reduce the problem, and eventually I could power up to 235 (my line voltage) with the radio set to '250V', and the dim bulb was OK. This power supply is a bit unusual, and I think that there was an electrolytic in series with the heater to the rectifier and the centre tap from the secondary, even with no rectifier. I intend to just use new electrolytics under the chassis anyway, but the one I could test measured OK for capacitiance. (One does look suspect though). Unloaded, the voltages looked OK, with the transformer being 700V CT.

The circuit does not the match the schema from Radiomuseum in some details. There is an extra winding on the transformer for the two 58's and 2A7. There is only one choke.

I am now reverse engineering the circuit as much as I can, and have got stuck on the identification of the old components. I have worked out some of the naming convention - the 1 meg resistors have a brown body (1), black tip (0) and green spot (+ 5 '0's). However there must be capacitors on that tag board, and I need help with those. I have lifted the board so I can see how it is wired underneath, so should have a schema soon.

There are 11 components left to right. Listing the tags, I have:
1) ? 4K resistor?
2) 25uF electrolytic. What vintage?
3) 500K (was able to measure this - OK)
4) 100K (measured OK)
5) ?. Measures as 10uF ?
6) 100K (measured OK).
7) 0.1uF 1000VDC EroFoil. There is a second 0.1uf linking to tag 9.
8) 15K tucked underneath. Looks like 3 or 5W
9) Two 51K in parallel for 25.5K, 3 or 5W. + 0.1uF to tag 7.

[Richardgr]

Is there a good place to read about how the loudspeaker works, and how to test it? I'm guessing that 4 pin plug is a common way for wiring to the speaker assembly back than. The transformer is the OPT, and there has been some tampering at some stage, I think so that the radio could be used jaust as an external speaker.

[ms660]

It works like a conventional loudspeaker, but in this type the magnetic field for the poles is provide by an energized coil (the field coil) instead of a permanent magnet.

The resistance between the connections to the speech coil when disconnected from the transformer should be low, typically around 3 Ohms or so, the resistance between the connections to the field coil will be much much higher, I would expect somewhere between 1.2k and 2.5k.

Lawrence.

[Cobaltblue]

Some manufacturers used plugs to aid servicing.

What you have there is an Energised speaker the two wires disappearing in the back are the electromagnet to create the magnetic field required.

The other two are the primary of the output transformer.

Both should read in the hundreds of ohms the field coil may be higher a couple of K some of them.

The flying lead looks to be on the secondary perhaps for an extension speaker?

Cheers

Mike T

[PJL]

I think your estimates of carbon resistor wattage are very optimistic, these old resistors are mostly 1/2W.

[Richardgr]

Hi Lawrence, the penny has dropped regarding your comments about the field coil loudspeaker, Now I realise the seconf choke on the schematic is the field coil transformer.

[ms660]

Originally Posted by Richardgr

Hi Lawrence, the penny has dropped regarding your comments about the field coil loudspeaker, Now I realise the seconf choke on the schematic is the field coil transformer.

No problem but remember a field coil isn't a transformer in the sense of an output transformer but it will have a transformer action so far as reducing the effects of the field coil ripple current (hum) in the speech coil, either by a hum cancelling winding or a shading ring, from what I can see from the photo's your loudspeaker might have a shading ring fitted, a shading ring is in effect a shorted turn that's placed between the field coil and the speech coil, various references about that on the web.

Lawrence.

[ms660]

Originally Posted by ms660

No problem but remember a field coil isn't a transformer in the sense of an output transformer but it will have a transformer action so far as reducing the effects of the field coil ripple current (hum) in the speech coil, either by a hum cancelling winding or a shading ring, from what I can see from the photo's your loudspeaker might have a shading ring fitted, a shading ring is in effect a shorted turn that's placed between the field coil and the speech coil, various references about that on the web.

Sorry....I missed the edit time window....should be ....."from what I can see from the photo's your loudspeaker might have a shading ring fitted or it could be relying on the separate choke or both".....

Lawrence.

[Richardgr]

Some progress.
I have bypassed the eletrolytics for now, using serial 20uF 400V because I did not have 12uF in the high voltage. The new caps are mounted off the board so I can reuse them more easily if and when I fit the right size caps.
There were a few wax capacitors, which were swapped out.
The metal film Erofoil caps were quite bulky, and I fancied using them as coupling caps in a single ended build, so I 'harvested' those, replacing with Panasonic metal films.

There was a blackened cap that was identified as 1uF on the schematic. I was wondering if anyone could identify it, and let me know if it is an electrolytc, and if so, which end is +.

I am getting closer to starting up with a variac with the tubes installed, but need to build something to support it so I can test voltages and protect the tubes.

I am fortunate in some ways this is not an original set in great condition. It had been restored somewhat in the late 60's, so I can rest easy at night after substituting some modern parts here and there.