Roberts RM50 - modulation hum on VHF?

[dazzlevision]

I get a hum on weaker VHF signals when tuning in a station. BBC Radio 2, 3 & 4 VHF signals from Oxford and Mendip are pretty weak where I live in North Wiltshire (although on the top of a hill).

For an internal VHF aerial, the RM50 (not a large mains only radio) uses a single length of insulated wire threaded around the back cover and terminated in a Belling-Lee TV aerial plug - not even a telescopic aerial!

I have tried wiring an RF bypass capacitor across the mains transformer's primary winding and moving the mains cable wiring (as far as lead lengths allow), to no effect.

Before I look for a fault, I'm wondering if this is normal for weak VHF signals in this model, which uses Mullard LPxxxx modules for the RF and IF stages?

[G6Tanuki]

I've got a RM40, which shares many features with your radio.

Mine hummed. There was a lot of ripple across the big cylindrical blue electrolytic, and it was half-wave ripple!

One of the rectifier diodes was open-circuit. I replaced the rectifier and the capacitor [which seemed to have lost quite a bit of capacitancr, possibly because of having to handle higher peak-currents due to the half-wave rectification] and humminess was cured.

I use mine connected to a 5-element Yagi in the attic, and with a bunch of coaxial -stub filters to notch out the nuisance BBC National stations below 95MHz which otherwise can cause crossmodulation.

[dazzlevision]

Originally Posted by G6Tanuki

I've got a RM40, which shares many features with your radio.

Mine hummed. There was a lot of ripple across the big cylindrical blue electrolytic, and it was half-wave ripple!

One of the rectifier diodes was open-circuit. I replaced the rectifier and the capacitor [which seemed to have lost quite a bit of capacitancr, possibly because of having to handle higher peak-currents due to the half-wave rectification] and humminess was cured.

There is virtually no audible 50 or 100Hz hum from the speaker with volume at minimum, but I will 'scope the two dc rails.

[dazzlevision]

Well, I've 'scoped the two dc supply rails and they are fine.

I've also bridged the LT rails, VHF module varicap tuning voltage feed and the AFC decoupler electrolytics, but no change.

Fitting a 1nF ceramic across the above electrolytics also made no difference, as did fitting it across the BY164 full wave bridge rectifier and 1N4148 half wave rectifier diodes.

The scope shows the buzz is 100Hz and rather rich in harmonics. Earthing the chassis when the scope probe croc clip is connected does substantially reduce the hum level.

I'll try an RF filter in the mains lead next.

[dazzlevision]

Before trying an RF filter in the mains feed, I finally remembered that I had a spare RM40 chassis in my loft. Circuitry wise, the RM40 is very similar to the RM50, so I thought I’d get the RM40 chassis powered up to compare modulation hum levels between the two.

The RM40 and RM50 use a BY164 bridge rectifier to supply the higher current consuming audio stages. The RM40’s BY164 also directly feeds a series regulator transistor, for RF and IF stages.

The RM50 has two parallel connected 6.5V/0.3A scale illumination lamps which require a separate 6Vac winding on the mains transformer. As a result, the 6V winding also feeds a low current half wave rectifier and reservoir capacitor mounted on the mains transformer, which is then connected in series with the main BY164 derived power supply, to feed the series regulator for the RF and IF stages.

The RM40 and RM50 PCBs have virtually identical copper tracks, apart from a connection between the two dc rails, which places them in series in the RM50.

Looking at the official Roberts RM40 service sheet, it shows four 2.2nF ceramic capacitors connected across each rectifier in the BY164. However, the actual PCB doesn’t have these fitted, even though the PCB is punched for them. Instead, a 100nF polyester type is connected across the ac feed to the BY164, as fitted on the RM50 PCB. This might have an adverse effect on RF suppression in the power supply stage.

In any event, I altered one RM40 PCB track and the wiring to the mains transformer, to be the same as the RM50 and connected it to the RM50 mains transformer.

The result was a very significant reduction to the tuning hum level on the modified RM40 chassis.

I’m still inclined to think the effect is modulation hum, so my next move will be to fit the four ceramic disc capacitors (as shown on the RM40 circuit), to see if that reduces hum levels on the RM50 chassis.

[paulsherwin]

Don't like that 1N4148. Try replacing it with a proper fast rectifier like a UF400x - pennies on eBay. The bridge would also be quieter with UFs too.

[Boulevardier]

Is the hum there on AM bands, or just VHF?


Mike

[dazzlevision]

Originally Posted by Boulevardier

Is the hum there on AM bands, or just VHF?
Mike

Just VHF.

For good measure, I have already opened up the LP1185 VHF IF/detector module and replaced two electrolytics in the FM detector stage.

[dazzlevision]

Originally Posted by paulsherwin

Don't like that 1N4148. Try replacing it with a proper fast rectifier like a UF400x - pennies on eBay. The bridge would also be quieter with UFs too.

Yes, I could do that, but I have fitted the four 2.2nF disc ceramics around the BY164 diodes and it has cured the problem in the RM40 chassis.

I have plenty of these, removed from more modern discrete bridge rectifier circuits in CTVs, to fit in the RM50 chassis, which I expect will cure that as well.

[dazzlevision]

I'm pleased to report that, after removing C7 (100nF 250V Mullard metalled Polyester "tropical fish" type)and fitting the four Murata blue dipped 2.2nF disc ceramics - as originally used in the RM40, the "modulation hum" effect on VHF has been eliminated.

Roberts used one of the holes originally destined for a 2.2nF ceramic as a termination for the black wire from the mains transformer (-ve end of C5, 470uF) to the BY164's +ve PCB track. I removed this wire in order to fit the 2.2nF capacitor.

Nearby, on the same track, the red lead from the positive terminal of C6, the 4700uF reservoir electrolytic is terminated. I removed this wire and fitted a (Philips type) PCB solder tag in its place and then soldered the red and black wires to it, giving a neat and secure termination.

[paulsherwin]

Good result.