Civic/Melodyn R4900

[dsergeant]

This 1960s eastern European set which has been discussed on here in the past had a duff mains switch. With that replaced the radio actually works very well with loads of stations on MW and FM with just the built in aerials. But there is an issue with the EM87 magic eye.

I have a rather poor circuit which I think I downloaded from here.

Initially at switch on the magic eye lights up green but with no movement of the green segments which remain joined. After a minute or so the green light goes out leaving just the heaters glowing. Measuring the voltages there is around 190V going in on pin 6 and around 140V on the deflect on pin 7 but when the green light goes off that rises up to the same 190V as on pin 6. The two resistors, 100k and 2.2M, read correct.

Does this look like a duff EM87 or should I be looking for something else? Not sure really how these things work.

Dave

[paulsherwin]

It sounds as if the EM87 has an internal leak or short.

[Chris55000]

. . . There's an internal drawing of how these EM87 indicator valves work in the August/September 1963 issue of "Practical Wireless" (download it from "World Radio History"), but as regards your fault – it is simply a defective indicator valve – the contact spring from the target anode electrode assembly parts company with the transparent coating of stannic oxide that the green phosphor is sprayed onto – the "target" electrode in these valves is both a rectangular mask plate and the phosphor itself!

. . .A replacement EM87 valve should sort your problem out!

Chris Williams

[dsergeant]

Thanks Chris, EM87 on order.

[dsergeant]

An update.
As suggested a new (old stock...) EM87 sorted out that side, and the magic eye worked fine. After an initial test all seemed well so I put the covers back on......

However now I have noticed a couple of things:
1. The FM is not very sensitive, needs checking out.
2. More serious, after a few minutes from switch on the sensitivity on MW/LW drops quite a bit, which is also confirmed by the movement on the magic eye reducing. I suspect a bias voltage is drifting, possibly duff resistor, but have not investigated yet.

Any suggestions for where to look first?

Dave

[G6Tanuki]

I would start off by puting a high-impedance [DVM or VTVM, not an old analog AVO] voltmeter on the AGC line and see what the voltage is before and after the problem occurs.

After hat, meter the cathode voltages of the frequency-changer and IF amp to see if they shift significantly - that would help tie down whether the problem is in the IF or FC stage.

Could be leakage, too, in the MW/LW/FM selection switches; some of the paxolin used in Eastern European production was not too good, and if you have HT and low-voltage coexisting on adjacent contacts in the switch, leakage is always a possibility.

[dsergeant]

Been doing some checks today. It seems the HT is slowly dropping after switch on, starting at 250V after the bridge rectifier which drops after five minutes to 230V. This in turn results in the HT to the mixer and if valves dropping to 210V and the screen voltages of the two valves also dropping. The ECH81 screen goes down to 45V.

The screen resistor, 47k, of the ECH81 measured correct but was burnt, I replaced it but that changed nothing.

I have not changed the smoothing cap but I was wondering if this drop is a feature of the selenium rectifier. The rectifier doesn't get hot but does seem to be dropping a lot of volts. Maybe I should replace that with silicon diodes, then would have too much HT!

It still works fine on R4 LW at these lower volts but the magic eye is certainly showing the drop in gain.

Dave

[dsergeant]

I have now fitted a silicon bridge to replace the selenium one, HT is now stable, slightly higher but that is expected. But I am still finding the screen volts of the ECH81 dropping just the same during the first few minutes, from around 100V to 50V, the 47k screen resistor runs hot. Grid of the heptode is fed from the AGC line and is around -1V, rising a bit to around -.5V, not sure if that is correct. Whatever, the screen current is obviously increasing. HT of the heptode is correct, around 200V, so presumably anode current is about right and stable.

Where should I look next? Faulty ECH81?

Dave

[G6Tanuki]

Screen decoupling capacitor going leaky??