Philips Iberíca (Copresa) BE-592, Bad OP TX

[LincolnG]

Got one of these with an intermittent fault on the Output TX (EL41 PP). The primaries are intermittently open circuit. (3,5k/3,5k I presume)
I can get a new output transformer from Ali, but that wouldn’t come with the feedback windings (DC ohms 0.9, 0.6, 0.3, 0.3, [2.3 total]. Could this be taken off a secondary tap with some resistors.
Or I have to find someone in Spain to rewind the primaries of the original. Or a replacement
Any suggestions?

[Ed_Dinning]

That looks like one of Philips special really complicated designs with several feedback loops.
The transformer will need to be very carefully reverse engineered for a rewind unless you are very lucky and the primaries are the outer windings.

You could try a secondary tap kludge but I suspect it will mess up stability and tonal qualities.

That must be about the most complex feedback system I've ever seen

Ed

[LincolnG]

I have this Philips BE 592-A with a bad audio output transformer. This has an EL41 Push-pull output.

It is intended to replace the transformer with a 8k 6V6 (push pull) transformer.

The problem is that the output transformer has an additional winding, which I'm not totally sure of its entire purpose.

The new transformer will be 4k 0 4k and 0 4r 8r

Starting with the EL41 cathodes which go through the winding to chassis (orange), can this be replaced with a single resistor and capacitor, or do I need to split the cathodes with each having a capacitor and resistor and what values.

Or would I be better to somehow preserve this winding

[Silicon]

The third winding is almost certainly used to generate three feedback signals.
It has five terminals.

Two signals have the same phase but the third is in the opposite phase.
They are connected to the tone control potentiometers and to the grid of one of the output valves.

I can't see an easy way to generate the feedback signals with resistors and capacitors.

[LincolnG]

My first concern is the cathode circuit (orange) without it, nothing will work at all. Without the others, I think it should do something, if only on the external input.

[Wellington]

My feeling is that it would be a shame to modify/remove the feedback arrangements to accommodate a replacement output TX. Philips sets often sound lovely – and that may apply especially to this one, since it has push-pull.

Have you attempted any further diagnosis and/or surgery on that defective output transformer? Sometimes (sorry if you already know this) you can discover breaks in the lead-out wires that run to the terminals, or even a bad joint where they are soldered to the terminals. Further surgery, if you're lucky (and willing to attempt it) may reveal an easily-accessible break in the primary winding – though being a Philips set, I suppose the output TX is covered in that yucky black pitch.

The intermittent nature of the fault is unusual. Does it only happen when the transformer is warm? It could indicate something like a easily-fixed poor connection or something more terminal (shorted turns within).

Originally Posted by LincolnG

Or would I be better to somehow preserve this winding

I think there may be a way to do that. Doing a quick forum search reveals a post by Leon Crampin (post #3) back in 2008, where he mentions his "backwards transformer" modification which allowed the feedback winding on the existing transformer to remain actively in circuit. Basically (I think), you would disconnect the primary of your defective, original TX and connect in your new TX. You would then connect the secondary winding of your new transformer across the secondary of the old transformer – this would feed the speaker as usual, but also use the secondary windings on your old TX to do the feedback. (This would mean finding space for a second transformer in your radio).

It's also described here, along with some useful stuff about feedback arrangements in Philips sets!

(Disclaimer: I write this as an amateur, merely suggesting what might work from reading around. The sources I've found don't discuss push-pull circuits, and I don't know if this difference would have any implications for your set.)

An alternative approach, if necessary, would be to get your defective transformer rewound. Ed Dinning, who replied upthread, does rewinds.

[LincolnG]

One leg of the secondary reads 500k ohms, the other open circuit. Putting on the HT, the volts arrive but not in force. With the EL41s in circuit, there is very low volume sound in between wild crackling. Indecently the original speaker was bad with a rubbing voice coil, shorting to chassis.
I did some research about getting it rewound and I found a guy in Barcelona that does exactly that. However, he will only rewind the whole thing a new to be able to guarantee the integrity of it as a unit and talking about 150€ to do so.
I got an 8k El84 transformer from Ali for 15€.
As suggested, I’ll take off the transformer and if the secondaries are on the outside I’ll look to see if there is anything obvious, if there isn’t, remove them.
The cathode circuit also goes through this “feedback” winding and I intend to replace the cathode winding with a resistor and decoupling capacitor. Looking at schematics and data sheets the value of the resistor between 82-150 ohm (Rk) and a bypass capacitor between 25-50uf (35v). The value of Rk could be split between two resistors to accommodate the path (yellow) to the volume pot.
The AGC / Bias (blue) via R19 & C36 (pre) will continue to work normally.
Via R10 (pink) and 47K (green) around the centre tap (salmon) seems to serve the EM4, perhaps some sort of automatic tone control and the bias of the detector. At this moment I am not considering the EM4 since they are expensive and purely cosmetic, if later I do – it should work well enough getting its signal from R21 (AGC) as in other sets.

[LincolnG]

From the original transformer I tool it apart and removed the primary windings.

Then I spent time rebuilding the power-supply and changing all the old wax/paper capacitors. The AZ41 rectifier, gets replaced with a pair of diodes under the valve base, so it’s easily reversible.

The cabinet was in a shockingly bad state, somebody had decided to cake it in loads of thick varnish. I sanded that all back, re-stained and lacquered it. The speaker cloth was replaced, the speaker bag patched up. The speaker had to be replaced, the voice coil was scraping on the magnet and it sounded horrible.

All the valves replaced with N.O.S. since the last owner decided to remove them all and the knobs too. I couldn’t find 4 matching Bakelite knobs anywhere, so I got 4 oak door knobs, stained and polished them after drilling holes to fix them on the shafts.

The radio had a few faults that I had to work through, but now it’s working perfectly, although on an external amplifier/speaker. It works very well on my new roof antenna.

https://drive.google.com/file/d/1o0ETlT5CUFvJEoJ7LqgAqgcLklsVAfCx/view?usp=drive_link

Then I bought a new EL34 /84 push-pull transformer from Ali. The idea is to parallel the secondary of the old transformer with the new one, to drive the feedback windings, some of the resistors values might have to be reduced, we’ll see what happens when it arrives. Once it's all working I can always investigate removing the feedback windings with some other arrangements.

Next job is to test all the high value carbon resistors, to see which will have to be replaced.

[AlanC]

Strange and interesting circuit! It looks like the grid of the lower EL41 is taken straight to HT- via a resistor - no audio is coupled here - and their cathodes go to chassis via a tap on that feedback secondary. How I read that (please feel free to tell me I'm wrong!) is that this transformer winding is key to the push-pull process - it's a self phase-splitting push pull output stage!

Alan

[LincolnG]

Originally Posted by AlanC

It looks like the grid of the lower EL41 is taken straight to HT- via a resistor - no audio is coupled here -

Alan

To the Negative Bias via R29. Any audio here is coupled by C40 or C26, although C26 is a very small value, only 390 pF. The inverted signal for the lower EL41 seems to be provided from the cathode?

What part the transformer's second secondary might be playing here; apart from decoupling the cathodes I have no idea!

Nobody else has been able to shed any light on it either

Hopefully when I get the new o/p transformer, things might become clearer