Marconi 559 hum

[Herald1360]

I wouldn't suggest shorting L19, it won't do anything disastrous, but it may reduce the inductance of L20 which won't help its ripple reducing capability. Just leave the top end of L19 (as viewed in the circuit) open with the speaker connected to the bottom end instead.

[PJL]

I hadn't realised the humbuck coil was still in circuit. Good call!

The transformer secondary would appear as a high impedance without the rectification from the output valve on the primary hence the dependence.

However! it does not explain why there is AC on the HT even when the output valve is taken out.

[PaulR]

I will un - short L19 as suggested.

You are right, PJL, about the AC on the HT line. It seems probable that there is some sort of fault in the mains transformer and I did wonder about it when I first bought the set as it looked as if it might have got hot at some stage. The first thing I did after basic checks was to power it up with no valves to make sure there was heater voltage and HT in the right places. The condition of the case is such that it would not be worth re-winding it or obtaining another so I have decided to leave it and see what happens. It has been running for about 7 hours now and is pleasantly warm as is the old field coil so I am hoping that nothing nasty is happening. The 50Hz hum is reduced now but is still there. It can be felt on the speaker more than heard and I hope that it doesnt become too bad when the speaker is mounted back in the case.

As I said at the beginning the whole set was in poor condition with several old somewhat bodged repairs.

I have a friend who has several boxes of valves and I will try to get round for a rummage for a replacement output valve to try.

Thanks all

Paul

[Billy T]

Ah...

Sounds like you are making progress! With L19 out of circuit there may be some additional filtering required, but there is no longer any great need to 'balance' any filtering effects as you should no longer have any feed-back or feed-forward loop effects. It may be necessary to increase the filtering a little as it was no doubt a finely tuned exercise in hum cancellation originally, and you now have a speaker that almost certinly has a greater LF sensitivity than the old EM jobbie. Most of the old speakers felt stiff as a board compared to a modern soft-suspension model.

I don't think you have a transformer problem either, you would have overheating if any one of the primary, the HT or the rectifier filament windings went to ground, or had shorted turns (you were lucky to get away with shorting L19, you nearly let the smoke out there. It must have been loosely coupled to L20 or it would have protested very quickly. In my experience mains transformers either work, smoke (shorts) or play dead (open windings). Just one shorted turn is often enough to stop the whole circus, or to at least generate enough heat for your nose to be alerted.

Cheers

Billy

[PaulR]

I had disconnected L19 before I shorted it. I left it on for a few more hours yesterday and the mains transformer got hotter to the extent that it was almost too hot to hold if I held the laminations very tightly, but I don't know whether that counts as overheating. There was certainly no smoke.

What sort of extra filtering were you thinking about?

Thanks

Paul

[Herald1360]

Almost too hot to hold tightly is still only around 50-55C, so not excessively hot in transformer terms.

Additional filtering is tricky- especially for the anode supply to the O/P valve, since the 4uF aready there is the max reservoir allowed for the U12. You could try adding a couple of hundred ohms between the rectifier filament and C31, then upping C31 to say 32uF, but you would lose some HT voltage (about 10V or so). C21 can be as large as you like.

[PaulR]

My set has octal valves rather than the ones shown on the Trader sheet. The rectifier in place is a 5Z4G, but I cannot find any reference to its maximum reservoit capacitor.

Paul

[Andrewausfa]

32uF Paul

http://www.r-type.org/pdfs/5z4g.pdf

[PaulR]

thanks - yes I have just seen that. I will try increasing it.

[peteratnith]

Hi Paul, could it be leakage from the 250/0/250 secondary to the heater winding ? Try disconnecting the HT line from the rec. heater winding and feeding it from 2 sil. diodes from the anode connections, with the rec valve out of course. Not for too long as the HT will be a bit high !
We're all dying to see the cure for this!

Peter

[PJL]

I would be inclined to confirm the origin of the hum before changing the electrolytics.

L19 and R25 are definitely out of circuit now and L19 is not shorted?
Don't change C27 as it is designed to resonate with L20 at 100Hz.
You have already raised the value of C26?
Disconnect C25 audio coupling capacitor and see what the hum is like.
Tweak the tone control and see if it changes the hum.
What is the ripple like on C29 when powered-up? Is it predominantly 50Hz?

I am still not convinced that a few volts ripple should have so much impact.

[Billy T]

Does it still hum with the rectifier out? If it does, that is really the first priority, because based on your previous comments, it is not going to go away any time soon.

Anything else is probably a waste of time and effort because the foundations have to be right to start with. Hum with no HT is unlikely to go away just because you work on the live side of the set, though you may stumble over the answer in the process. Fortunately no time woking on radios is ever really wasted, it is too much fun.

Cheers

Billy

[PaulR]

PJL - L19 is completely disconnected. I added .47 to C26 and replaced C27 with a 0.047 when I re-capped the set.

There is 12v AC 50Hz on C29. I am attaching some photos of the oscilloscope trace taken from the speaker connection with and without C25 connected. The hum is more or less pure sine wave with it connected and the tone control doesnt make much difference. Without it the ripple is a lot less and the tone control does make a difference. The oscilloscope was set to 10ms /cm and 0.1v/cm. The references to min and max cap relate to the capacitance of the tone control.

Billy - It only hums with no rectifier when the output valve has warmed up.

Paul

[smiler411]

I would have thought that as the hum is only present with C25 connected it is indicating the fault is not with the output stage. Is it possible V4 has a heater to cathode leak, its cathode is not connected to ground so hum pickup is possible here

[electrogram]

could it not be that the mains transformer is inducing hum into the output transformer ? is the output transformer mounted on the chassis ? if so you try removing it and see if it reduces the hum, if there are a lot of short connections to the optx rest it on a piece of sponge on the chassis which will mechanically isolate it to some degree, this will save a lot of unsoldering/soldering

[ms660]

The problem with this set has got me confused...I would make sure everything that's fitted is fitted correctly as it was intended to be and then start trouble shooting with everything up and running, all valves in and nothing unplugged/disconected etc, and see if you can pin down the source of the hum..first check all voltages references to see if they are within spec. then use the 'scope working from the speaker back toward the front end of the audio/det, any hum/ripple measurements done with the 'scope should be done with everything functioning on the set ie: all valves (good) in and working.
You can sometimes go round in circles unplugging valves when the power supply is of the type that is employed in that set. (CT for bias etc)
There is the possibility as already mentioned by a previous member, that the hum has become more noticable with the more modern speaker installed, it's hard to say without being there.

Good luck.

Lawrence.

[PJL]

What happens to the AC voltage on C29 without C25 connected?

Billy, there is most likely excessive leakage in the mains transformer but the hum is only present without HT because the output valve is acting as a rectifier. A normally biased beam tetrode would be quite tolerant of HT ripple as it is almost constant current.

The tone control is a problem here because it couples the AC voltage (hum from the HT) on the anode to the grid and this is why the hum is effected by this.

However, most of the hum is coming from earlier stages which is either:
1. Pick-up, possibly from any modification of the wiring
2. More likely, inadequate decoupling of the HT for the triode

I would try increasing the value of C21.

However, I do think you need to track down the leakage current. The most likely is one end of the HT winding to the screen. The way to check this is to simply measure it with an AC ammeter from the HT CT to chassis (might be a good idea to protect the meter with a 1K series resistor) but you will need to disconnect all secondary connections first.

[PaulR]

When I did the work on the set I put a .01 cap between V5 anode and the variable tone cap to protect it from the HT because the fixed vanes are very warped and appeared to rub on the moving ones. Having said that there wasn't any leakage and the tone control actually works. I don't think that makes any difference but I thought I ought to mention it.

I used 10uf to replace all the smoothing caps in the can. Should I try increasing C21 further?

When you say to disconnect all the secondary connections is that everything from the HT winding and CT or just from the CT itself? We keep coming back to the mains transformer!

[PJL]

See if doubling C21 helps.

The mains transformer does seem the only likely suspect. How long have you had the set and has it been damp?

[PaulR]

Ok I will try that. I only bought it a few months ago, but it does seem to have been damp. There was surface rust on the chassis and the veneer is lifting.