Lafayette Superheterodyne Radio - 1935

[El Sid]

Just about all of the components are replaced now - lots of cutting, soldering and desoldering! All the RF cans' resistors and fixed caps are changed (forgot to take photos!), and almost all the resistors and caps in the chassis. There are 2 resistors and a cap outstanding - components that I inexplicably didn't order. Often the new resistors just looked wrong because they are so much smaller, but I checked and double-checked the expected current and hence power rating.  Some of the resistors could actually be much smaller, but a modern ½W resistor would just look ridiculous in there! So i even put in 2W resistors just so they looked right.

Not a single one of the old caps was closer than 50% to its correct value, and only one resistor was within 10%.

I know there will be some criticism of the flying leads, but (a) i have replicated what was done before, which had (admittedly shorter) flying leads and (b) was far too lazy to install terminal strips en masse all over the place. I did put some insulation over in the few places where things were very close. Also because of the massive globs of solder in this radio, in many cases I snipped the old lead leaving a bit spare, twisted the new components' leads into spirals, which i then slipped over the old lead and soldered. This saved hours of desoldering and general faffing. Many people don't like this technique, but I wasn't going to spend all that extra time.

So now we have this:


It looks empty compared to what it was before:

[El Sid]

I went component by component to ensure no mistakes (ha!), and at the same time annotated and checked against the schematic. I did find a few differences in component value, and two cases where the circuit is connected slightly differently.

Now all I need to do is figure out the expected current/s in the circuit so I can estimate the size of the resistor to replace the field coil with, and then hook it and the OPT up. Then I can start testing. Power supply first, then power amp, preamp and finally the radio section.

Getting there!

[Family_Dog]

Looking good, El Sid. I also prefer using larger resistors than the minute 1/2w ones available today, the bigger resistors simply "look" better.

No need to worry about neat wiring for radios, RF prefers the shortest paths.

-F_D

[El Sid]

Found a pic of the new components in an IF can:



Compared to what it looked like before:

[marantz123]

Nice work. The radio may need a re-alignment if resistors /capacitors were replaced in the RF/IF sections. If its pulling in stations clearly with good selectivity and sensitivity then its probably aligned well enough for that. The tuning indicator (if string-driven) may be a bit out though but wouldn't be an issue.

[El Sid]

Nice work. The radio may need a re-alignment if resistors /capacitors were replaced in the RF/IF sections. If its pulling in stations clearly with good selectivity and sensitivity then its probably aligned well enough for that. The tuning indicator (if string-driven) may be a bit out though but wouldn't be an issue.

It will definitely need realignment, but I'm not going to go overboard because (a) it's going to be used more as a bluetooth player and display piece than a radio so exact alignment doesn't matter, and (ii) there are no alignment instructions, although it can be figured out.

[Orcish75]

A friend of mine, after seeing the Philips radio restoration I did, asked if I could fix his. So I said I could have a look, but no promises.

Hmm, not sure why you went to so much trouble servicing the radio. You should've just posted it to: Loewe Radio Service, P.O. Box 1320, Cape Town.

[El Sid]

I managed to get some more done on this. I am posting a better copy of the schematic to make things clearer as the discussion gets a bit more technical.

[El Sid]

So after all the recapping and reresistoring, it was finally time to fire up the electronics.

Having already tested the transformer, I put in the 80 rectifier and put it on full limit (40W bulb) on the dim bulb limiter. All went well, and eventually it was unlimited. All was perfect, and I was getting a whopping 450V DC! But that was with no load whatsoever. A glowing valve for the first time:



Next was to test it under load. Firstly, I did some calcs on the estimated size of the resistor needed to take the place of the field coil. It has to be in place because all current returns from ground via that resistor to the choke and hence the centre tap of the transformer. My clue was the specified grid voltage of the Type 47 power valves of -16V, and the voltage divider making the bias voltage (the 2 resistors to the left of the choke - actually 500k and 100k). I landed at 680Ω, 10W. I put 2 x 15k resistors in parallel from the anode pin of each 47 valve base to ground to simulate the 35-40mA each valve would draw. I wired up the output transformer too.

Again firing it up gradually, things looked quite good - on the anodes of the valves I was getting about 270-280V, and about -10V on the grids. Given the 250V max plate voltage specification for the 47s and the required grid voltage of -16V, i realised I had to drop more voltage across the "field coil" resistor. I experiment and settled on a 390Ω, giving a total of 1070Ω. Looks good:

Anode/plate voltage: (note the 2x15k load resistors below my hand simulating the valve current draw)


Grid voltage:


Noting of course that I will need to tweak these values once the radio has its full complement of valves in - I'm not sure what the actual total current draw will be.

More to follow after dinner....

[El Sid]

Next up now that the power supply was working was the power amp stage.

I took out the dummy loads, moving a pair of 15k resistors to the supply going to the rest of the circuit to simulate the 40mA or so of estimated draw from the rest of the valves. I put 10Ω resistors in series on the plate winding connections on the output transformer so I could measure the anode current of the valves. Then, with some nerves, in went the 47s! Again powered up with the gradual reduction of limits, carefully shecking grid and anode voltages, until at no limit, things were pretty perfect. The plates at around 240V, grids at -16V. Checking the grid currents: (÷ voltage by 10 to get current)



So we have 29mA, which is pretty close to the target 31mA, and very well balanced for an 86-year-old radio!

Now that the DC situation was working, time to put in a signal. Holding breath........

Booyah! Yellow trace in, blue trace on OPT speaker output:


I was so very pleased with myself.

I thought I'd check out the inputs to both valves to make sure the phase inverter was working. Now the phase inverter is quite unusual - it's a paraphase of sorts, but implemented about the power valves themselves - there is a voltage divider on the output of the first valve (resistors circled in green) that is fed back to the grid of the second valve.


Looking at the balance: (yellow=first valve, blue=second)


Clearly a bit of adjustment required. So I reduced the one resistor by putting another across it until things looked right:


All looking good at this stage. Next step then was to put in the Type 56 preamp/detector valve, and see if that works. Damn - first roadblock. The valve is dead. Thanks to @kubusi (and with an offer from @charles too), a replacement is on its way. Next installment soon I hope...

[El Sid]

The 55 double diode triode valve duly arrived from @kubusi, and in it went. Voltages and currents checked out OK, so then I tested it as an audio preamp (i.e. the triode section) by putting a signal on its grid, and it worked as expected, if a bit noisy. Next up was to inject an AM signal at the 520kHz intermediate frequency on the grid of the final IF stage - the leftmost valve (the 58 IF/RF pentode) in the pic below. This will establish if the valve is working, if the final IF transformer is working, and if the detectors on the 55 are working.



And the result - working! AM signal on grid is the yellow trace, the output on the speaker terminals in blue. I adjusted the variable caps in the IF can to maximise the signal. The one didn't make too much difference, and the other reached its limit while the out put was increasing. So I think I may need to increase its capacitance by putting another cap across it. I then checked the next stage by similarly injecting the signal onto the 58 grid. It still worked, but not particularly well, and I ran out of time to continue. When I get back to it I'll start by swapping the valves - hopefully that's the issue. But good progress anyway.

[Family_Dog]

Great stuff, El Sid! Glad you were able to find the 55 valve, I also looked for one for you in my collection but no joy.

Looking forward to when it is up and running, hope you have an AM transmitter close by.


-F_D

[El Sid]

I put in another new (old) 58 for the second IF stage, and voila! Working well, getting a really strong signal on the output.



Using my AVO to align the IF transformers - works so well! Seems appropriate...

[Family_Dog]

I would have been so jealous of that AVO if not for the fact that I already have two of them, a Model 8 and a Model 7, both in perfect condition as yours is above!!


-F_D

[El Sid]

First IF stage also working now, so all that remains is the oscillator. The 56 valve unhelpfully has "DUD" scratched onto the base, which doesn't bode well.