HMV radiogram no AM reception

[lesparapluies]

Another frustrating session today! But.. verified that all the wave change switches are working OK. Replaced a few more high resistors, but no difference. Supply voltage is about 25V - slightly higher than the specified 23.5 but not enough to make much difference. The significant drops are west of R32, which is 12V, dropping to 9V west of R18. (so voltages on either side of R18 are 9/12). Voltages on the transistors are (c-b-e): VT3: 9.2/0.8/0.6 VT4: 9.2/3.1/3.0 VT5: 7.36/3.47/3.21 VT6: 10.5/1.72/1.42. So the b-e voltages seem to be within the 0.2V difference normal for germaniums? The 7V on VT5 is cause for concern? All the transistors have been renewed. I have tested a few capacitors but they are all ceramic disc type and all read close to their stated values. Electrolytics have been replaced.If the set is switched off for about 24 hours, it will play AM well for about 20 seconds, after which it quickly fades (I'm going to try to record this as an audio file tomorrow). I was thinking of disconnecting the collector resistors of each of the transistors in turn to try to establish which stage is bringing down the supply, is that wise, or could it cause damage to the previous or following stages? As always, any advice gratefully taken on board.

[lesparapluies]

Sorry I just realised that frismen had asked for voltages across R20 and R25, will check those tomorrow. And looking at my figures, it would seem that the VT5 stage is the culprit?

[frsimen]

There's no need to measure the voltage drop across R20 and R25, you have reported all the important voltages now.

You should check R26 and R28 and the connections between them and R17. I think one of the resistors has gone high in value/open circuit or there is a break somewhere around them. I don't think there is anything wrong with VT5 itself, but these components provide the bias and it is that which is faulty.

Paula

[lesparapluies]

Thanks for this valuable info, will check those next time I get a chance, probably Wednesday. So, if this is the case, would I be right in thinking it's something connected with the AGC circuit, that the faulty component is erroneously reducing the gain and making the station inaudible?

[frsimen]

That’s not quite what is going on. When working properly and in the absence of any signal, VT4 and VT5 are biased for maximum gain by the potential divider formed by R17 and R26+R28. That will make the base voltage of VT5 around 0.1 times the voltage at the left side of R18.

The diode W2 will be very slightly forward biased by the voltage appearing across R28, which will improve its operation as a detector on weak signals. As the strength of the signals increases, W2 will rectify the signal and will develop a positive voltage across R28, which will reduce the bias voltage applied to VT4 and VT5. This reduced bias will reduce the gain of VT4 and VT5.

In your case, the voltage on the base of VT4 and VT5 is much higher than it should be, leading to higher than intended collector current. The gain of VT4 and VT5 might be reduced slightly at this higher collector current, but not by as much as it would were the AGC working properly. The higher than intended collector current of VT4 will stop it working well as a mixer, which is probably the cause of the signals disappearing after a few seconds.

Looking at your measurements again I think the most likely culprit is R26, or the connection from it to R17 or R28.

Paula

[lesparapluies]

Another session tonight; replaced R26, R28 and R17; unfortunately has made no difference (they were all very high, R26 was about 11k5). Readings on VT5 are virtually the same as on the previous posting. One thing I noticed tonight is that placing my meter probe on the emitter of VT4 makes the background noise much louder, not sure whether this is relevant. And it could be just my imagination but I think replacing those 3 resistors has actually improved FM. But same story on AM, strong reception on cold start-up which fades after a few seconds..

[frsimen]

Can you measure the voltage on each side of R26, R27 and R28. It could be that there is a crack in the PCB somewhere along the line, as you should have seen some change in the voltages you measured on VT5.

Another useful measurement is the base voltage on VT5 when AM is working. You will need to have your probes in place before you switch on to catch this.

Paula

[smiler411]

Have you changed C37? It may make bias voltage high if leaking as it will increase emitter voltage of VT4 which will then affect VT5 bias too

[lesparapluies]

Thanks for recent replies. I've been doing some reading up about this type of circuit, especially the AGC line, and I can see exactly why these resistors would be a point of suspicion. When I get some time this weekend I'll look carefully into that area and also examine the PCB for signs of discontinuity. I did notice that the emitter resistor of VT5 was very high (about 730R) but didn't have any spare 560s, so waiting for delivery of them.

[lesparapluies]

Update - I replaced a few more high resistors around VT5 and VT6 last night, no difference. Definitely no breaks on the PCB; everything is connecting which should be. I thought I'd achieved a major breakthrough when I spotted that when I'd replaced C48 at the beginning of this project, I'd put it in the wrong way round (senior moment, forgot it was positive earth) C50 was correctly placed. Took C48 out and replaced with a new one. No difference. The voltage across R26 is 3.1V on the junction with R17 and IFT1, with about 100mV on the junction with the detector diode. Tested C51 out of circuit and appears to be OK. What kind of voltage should I be seeing at the output of W2? I did try to detect voltage on VT5 base at start-up but unfortunately the meter probe slipped off during start-up, will try again tonight. I'm wondering if the wrongly orientated C48 could have compromised another component, maybe VT5, and I should order another? But it seems to function perfectly on FM.....

[frsimen]

From your measurements, there are no breaks in the resistor chain's wiring, but the voltages are wrong. There aren’t too many things that can cause this.

C48 being the wrong way around will not have caused any damage. In the absence of signal, the W2 will not do anything. The 100mV you have measured is caused by W2 being forward biased. That voltage will drop when a strong signal is tuned in, but you won’t see that until you have sorted out the high voltage at the junction of R17 and R26.

Lift one end of R17 and check that it is a 100k resistor. You may have fitted a lower value in error. The radio will still work on FM if you have done this, but on AM, the bias current through W2 will be much higher than it should be and it will not work as intended.

Paula

[smiler411]

If the resistor is correct is it possible you have a short circuit somewhere? The voltages on the transistor and the fact that it works on FM suggest the transistors themselves are all fine so wouldn't change them again.

May need to disconnect parts of the bias circuit to try and determine where the high voltage is coming from by isolating different sections

[lesparapluies]

Well.. another fruitless session tonight. R17 is definitely a 100K, it's one of the ones I changed anyway. Went over the entire PCB with a strong light and a magnifying glass, plus a continuity meter, and no problems there. On switch-on the base of VT5 is about 1.5V and the signal disappears after it reaches about 2.5V. It's only possible to get the 10 seconds or so of normal function if the set has been switched off for several hours. The collector voltage on VT5 is barely above 6V, less than half of what it should be (on both FM and AM) so it's a wonder it's functioning at all - and yet, as always, FM reception is near perfect. Something which may be significant is that the voltages on VT6 fluctuate quite a lot, and at times the figures on base and emitter don't meet the 0.2V difference which you would expect. I'm learning a lot about the theory behind this type of design but one thing I can't grasp is that, in the first IF stage, VT4, the two transformers (FM/AM)are switched by S2, but in the next 2 stages they both seem to be in circuit at the same time, effectively connected in series, is that normal? I'm wondering if there is a faulty ceramic cap somewhere. I'm trying not to think it may be a faulty cap inside one of the cans. I may just get this set working while there are still a few stations on AM.....

[lesparapluies]

Originally Posted by smiler411

If the resistor is correct is it possible you have a short circuit somewhere? The voltages on the transistor and the fact that it works on FM suggest the transistors themselves are all fine so wouldn't change them again.

May need to disconnect parts of the bias circuit to try and determine where the high voltage is coming from by isolating different sections

I'm not sure the voltages on the transistor are correct - 6-7V on the collector of VT5. Could I disconnect the AGC line altogether and try it or would that mess up everything around VT5? I'm willing to give anything a try, I feel the answer is staring me in the face but I'm not seeing it!

[frsimen]

The voltage on the collector of VT5 are too low because it is drawing too much current. the same is true of VT4. VT6 will also have slightly lower voltages than listed, but they won't be so far out.

The switching of the IF transformers on the first stage is quite common and won't be causing this problem.

It looks like either VT5 or VT4 has collector to base leakage which gets worse after it warms up. You can easily prove if the cause is VT5 by lifting one end of R20. When you switch the radio on, monitor the base voltage of VT5 and if it stays low, the culprit is VT5. There won't be any signal, but you will have identified the trouble.

If that doesn't sort the problem, leave R20 disconnected and lift one end of R13. Try measuring the voltage on the base of VT5 again. If it stays low this time, the culprit is VT4.

Reconnect R20 and confirm that the problem stays away, just in case both VT4 and VT5 are faulty.

You may have been very unlucky and fitted a bad transistor earlier in the repair.

Looking at the circuit, none of the ceramic capacitors can cause these odd voltage measurements, nor would a faulty capacitor inside the IF cans.

If you have one, you could try a PNP silicon transistor as a temporary measure to replace VT4 or VT5 if they prove to be faulty. Something like a BC558 would do for a quick check.

Paula

[lesparapluies]

As always, very helpful answer, thank you, will get on to that at the weekend

[Pfraser]

Would some tests with freezer spray be any help?

Suggestion for a low-cost test: put a screwdriver or similar tool in the freezer. Retrieve and wrap in thin plastic. Apply to parts.

[lesparapluies]

Originally Posted by frsimen

The voltage on the collector of VT5 are too low because it is drawing too much current. the same is true of VT4. VT6 will also have slightly lower voltages than listed, but they won't be so far out.

The switching of the IF transformers on the first stage is quite common and won't be causing this problem.

It looks like either VT5 or VT4 has collector to base leakage which gets worse after it warms up. You can easily prove if the cause is VT5 by lifting one end of R20. When you switch the radio on, monitor the base voltage of VT5 and if it stays low, the culprit is VT5. There won't be any signal, but you will have identified the trouble.

If that doesn't sort the problem, leave R20 disconnected and lift one end of R13. Try measuring the voltage on the base of VT5 again. If it stays low this time, the culprit is VT4.

Reconnect R20 and confirm that the problem stays away, just in case both VT4 and VT5 are faulty.

You may have been very unlucky and fitted a bad transistor earlier in the repair.

Looking at the circuit, none of the ceramic capacitors can cause these odd voltage measurements, nor would a faulty capacitor inside the IF cans.

If you have one, you could try a PNP silicon transistor as a temporary measure to replace VT4 or VT5 if they prove to be faulty. Something like a BC558 would do for a quick check.

Paula

Did some checks tonight and got the following results: With R20 disconnected, voltage on the base of VT5 = 292 mV. With R20 and R13 disconnected (I'm assuming you didn't mean R15, the collector of VT4?) voltage on base VT5 = 295mV. With R20 back in but R13 still unconnected, base VT5 = 3V, with R13 and R20 both back in place, base VT5 = 3V

[frsimen]

Good news, you've found the fault!

VT5 is faulty, or you might have wired the replacement incorrectly. Check that first, but if it looks to be fitted correctly you need a new transistor for VT5. A PNP silicon transistor may work if you don't have an AF1xx to hand. Be careful not to overheat the transistor when you solder in the replacement. Germanium transistors are less robust than silicon when it comes to heat.

Paula

[lesparapluies]

Thanks, will get on to that this week. It's probably getting on for 50 years since I soldered a germanium transistor and yes I had forgotten that they are more delicate than their silicon relatives, so will bear all that in mind