Murphy A484 - residual hum level from speakers

[dazzlevision]

I’ve been restoring this radio - usual Hunts/TCC/Dubilier paper capacitor replacements, etc.

The original mains transformer has in the past been replaced by a Radiospares “Economy” mains transformer. As I had a genuine A484 transformer in my stores, I fitted it. I then noticed a residual hum from the speakers, that seems excessive for an ac mains only radio from a respected manufacturer like Murphy.

Looking at the official Murphy A484 service manual, I see the original transformer had a separate heater winding for the EZ80, whereas they only supplied a replacement transformer with a single heater wiring, which necessitates cutting the link between the EZ80’s cathode pin 3 and heater pin 4. I have fitted the original type with two heater windings.

I’ve also fitted a good 50uF +50uF HT reservoir + smoothing can, so the ac ripple voltage is very low, as you would expect.

I notice that the hum appears as soon as the set is switched on, so suspect it is unwanted magnetic field coupling between the mains and audio output transformers. The audio output transformer isn’t far from the mains transformer and it is inclined at an angle to it and mounted underneath), presumably to minimise coupling (as found in my Bush VHF64).

Do any other A484 owners have any views on this? The A474 shares the same chassis, but in the cheaper moulded cabinet. I also suspect that the original mains transformer proved to be unreliable - any experience of this?

To confirm it is magnetic coupling, I have checked the hum frequency across the speakers before the valves heat up, which is 50Hz (plus a lot of third harmonic distortion), which supports my theory.

[Sideband]

May have always done it. Back in the days when this set was made, a certain amount of hum was expected and accepted as long as it couldn't be heard over the normal volume of the program. I have a couple of sets that have this residual hum at switch-on but in reality, it is so low and becomes un-noticeable at normal volume.

Yes it is induced hum into the output transformer. Not much you can do if the output transformer is chassis mounted but if it's mounted on the speaker, unbolt it and try moving it around if the leads are long enough.

Some manufacturers mounted the output transformer under the chassis which must have helped a lot to reduce the effect.

It could also be mechanical vibration from the transformer itself which is being transmitted through the chassis and into the cabinet (if wood) which acts as a nice baffle making it even louder!

[dazzlevision]

Thanks for your post.

I doubt it is acoustic coupling, as my scope showed the distorted 50Hz waveform at the speaker terminals.

I think it’s louder than it should be for normal listening volume levels. I’ll try adjusting the orientation of the AOPT to see if that makes much of a difference.

[paulsherwin]

Yes, it's surprisingly common. The Pye P43 is notorious for this. Designers in the 1950s seem to have been pretty unconcerned about it, and it's true that listeners then were much more tolerant of background hum then than they are today.

What about a transformer band, I think these are used to reduce the magnetic field.

[DonaldStott]

Originally Posted by merlinmaxwell

What about a transformer band, I think these are used to reduce the magnetic field.

What are these 'transformer bands' and where can I find out more about them?

[ms660]

Originally Posted by DonaldStott

What are these 'transformer bands' and where can I find out more about them?

Search for transformer flux band either on here or on Google.

Lawrence.

[Sideband]

Originally Posted by dazzlevision

I’ll try adjusting the orientation of the AOPT to see if that makes much of a difference.

Ideally the cores of the output and mains transformers should be at right-angles to each other but if you can move the output transformer about, moving it as far away from the mains transformer as possible and rotating it will reduce the hum.

Obviously you have to do this with the set switched on so make sure you insulate the high-voltage terminals!!

[rontech]

Sometimes the transformers are coupled by magnetic transmission through the steel chassis.

I solved this effect in an amp by fitting a sheet of paxolin between the mains transformer and the chassis and using non magnetic ( brass ) fixing nuts and and bolts.

[dazzlevision]

Thanks for all the comments and suggestions offered.

The AOPT is mounted below chassis, fairly close to the mains t/f and its core is inclined at an angle of approximately 45 degrees to it. I removed one of the AOPT's fixing screws and rotated it to the extent of the connecting wiring, which did reduce the hum slightly.

The ends of the two AOPT (self tapping) fixing screws can be seen in the photo - they are along the line of the brown/black twisted pair feeding 6.3V ac to the tuning scale illumination lamp.

I then removed the four fixing screws of the mains t/f and stood it on some insulating material, which had no audible effect.

However, rotating the mains t/f did noticeably reduce the hum level, but mounting it in that position would involve a fair bit or replacement wiring and drilling the chassis. For the moment, I think I'll reassemble the chassis into the cabinet and use the radio for a time, to see whether the hum is objectionable at a normal listening volume and distance.

[sparkymike]

Would the LT brown and black wire for the panel light have any effect, passing close to that valve?
Mike.

[jonnybear]

Just a thought as you have refitted the original transformer have you tried reversing the mains input, and putting the link back on the EZ80, which may have some effect on AC ripple.
John

[dazzlevision]

Originally Posted by sparkymike

Would the LT brown and black wire for the panel light have any effect, passing close to that valve?
Mike.

Unlikely, as the hum is present as soon as the radio is switched on (from cold) and the twisted pair has a very low radiated magnetic field.

[dazzlevision]

Originally Posted by jonnybear

Just a thought as you have refitted the original transformer have you tried reversing the mains input, and putting the link back on the EZ80, which may have some effect on AC ripple.
John

I have left the wire link between pins 3 & 4 on the EZ80 valveholder, as the genuine Murphy A484 transformer I fitted is the original type, with a separate winding for the EZ80's heater (presumably to keep the heater to cathode potential low).

I will try revising the polarity of the mains supply.

[thermionic]

.......Just a mad idea....what about deliberately injecting a 50Hz signal into the grid of the output valve to cancel out the induced hum? Would this work?

Cheers. SimonT.

[Sideband]

Originally Posted by dazzlevision

Thanks for all the comments and suggestions offered.

The AOPT is mounted below chassis, fairly close to the mains t/f and its core is inclined at an angle of approximately 45 degrees to it. I removed one of the AOPT's fixing screws and rotated it to the extent of the connecting wiring, which did reduce the hum slightly.

The ends of the two AOPT (self tapping) fixing screws can be seen in the photo - they are along the line of the brown/black twisted pair feeding 6.3V ac to the tuning scale illumination lamp.

I then removed the four fixing screws of the mains t/f and stood it on some insulating material, which had no audible effect.

However, rotating the mains t/f did noticeably reduce the hum level, but mounting it in that position would involve a fair bit or replacement wiring and drilling the chassis. For the moment, I think I'll reassemble the chassis into the cabinet and use the radio for a time, to see whether the hum is objectionable at a normal listening volume and distance.

It's all a question of compromise. Murphy would have been well aware of the induced hum and probably arrived at the final position commensurate with wiring layout, chassis size and acceptable hum level.

[dazzlevision]

Originally Posted by thermionic

.......Just a mad idea....what about deliberately injecting a 50Hz signal into the grid of the output valve to cancel out the induced hum? Would this work?

Cheers. SimonT.

As the magnetically induced voltage appearing across the speakers is a 50Hz fundamental with substantial third harmonic content, injecting a 50Hz sine wave at the g1 of the EL84 (with a suitable phase relationship) would be difficult to achieve, I would think.

From memory, Bush coupled a point in the series heater chain to the HT line in their Bush TV66 TV set, through a big 0.5uF wax dipped paper dielectric capacitor and a series resistor, in order to achieve some HT ripple cancellation. A similar technique was used in the Bush VHF64's initial production, but was dropped in favour of a "choke" type audio output transformer tapped primary design approach (the AOPT was also inclined at a suitable angle to the mains transformer).

[dazzlevision]

From Sideband: "It's all a question of compromise. Murphy would have been well aware of the induced hum and probably arrived at the final position commensurate with wiring layout, chassis size and acceptable hum level."

I'm sure you are correct, with the additional constraint of managing costs.

[Chris55000]

Hi!

As the magnetically induced voltage appearing across the speakers is a 50Hz fundamental with substantial third harmonic content, injecting a 50Hz sine wave at the g1 of the EL84 (with a suitable phase relationship) would be difficult to achieve, I would think.

It's not as difficult as it sounds – search for my post entitled "Hum Cancellation using Wheatstone Bridge Techniques?", I posted a circuit from W.W., October 1955 describing the technique employed!

Chris Williams

[dazzlevision]

Originally Posted by Chris55000

Hi!

It's not as difficult as it sounds – search for my post entitled "Hum Cancellation using Wheatstone Bridge Techniques?", I posted a circuit from W.W., October 1955 describing the technique employed!

Chris Williams

Interesting articles.

I note that the Murphy A484 uses negative feedback in the EL84 power output stage, by feeding back the signal across the AOPT's secondary winding across a 100 Ohm resistor inserted into the earthy end of the volume control. This is one of the hum reduction techniques referred to in the Wireless World July 1961 article “Negative Feedback and Hum” (pages 382 – 385).