Decca AMG111 hum

[PaulR]

I am having great trouble in sorting out a hum problem with this very simple record player amp.

It was advertised as not working and the first thing I noticed was that t he fuse in the Pressac mains connector was blown. I had one of these in the 1960s and one of the smoothing caps failed then so I replaced the can in this one as it was original. I also had to replace the valve socket as I could not get a reliable connection no matter what I tried. As I had to do this I also replaced the remaining capacitors and any out of spec resistors. I have put a new Ronette copy cartridge in it wired for mono.

It sounds good but with what I think is an unacceptable amount of mains hum. The HT is well smoothed with 3 x 33uf electrolytics and I added another to the reservoir cap to see whether that made any difference. The resulting HT is very smooth and I cannot see any ripple on the 'scope - but it still hums independently of the volume control.

The deck had that odd feeling of unearthed metalwork so replaced the two wire Pressac with a 3 core IEC socket.

The voltages are a bit low presumably due to increased internal resistance in the metal rectifier and I did try replacing it with a diode and 200 ohm resistor but that brought the anodes up rather high so I reverted to the original rectifier.

I have tried a replacement ECL82 from my box of old valves but that didn't help.

This thing is so simple I feel there must be something staring me in the face but I just cannot see what it is. Can anyone suggest anything I can try?

Thanks

[cathoderay57]

Hi Paul. When you tried using the diode instead of the metal rectifier, did it still hum?
Assuming it did, I recommend that you now measure the voltages compared to the service data i.e. anode pentode=200v; g2 pentode 175v; k pentode 14v. Anode triode 90v; k triode 1.25v. As a first step it's worth seeing these results to find out if the dc conditions are correct. After that try connecting grid 1 pentode to chassis i.e. short out R7 and see if the hum disappears. Cheers, Jerry

[crackle]

You May have introduced an earth loop.
I think the recommended action is to install the earth to a fixed point, then wire from that to all other metal parts, and the amp ground. Ensure there is not a double earth connection to any signal path.
For instance if you have earthed the record deck then ensure that the pickup wiring is only earthed at the amplifier end.

Mike

[PaulR]

Thank you for the replies.

Yes it did hum when I used the diode. One of the reasons for removing it again was to see whether the hum reduced. I will take the voltage readings and report back.

Part of the reason for earthing it was to try to reduce the hum. It did help with a "buzzy" component of hum, but not with the smoother hum if that makes sense. I connected it to one of the two component earthing points on the chassis and not directly to the deck which is earthed through the screen of the pickup wiring. The deck can be completely removed as the mains and signal leads have plugs at the amp end and it makes no difference whether it is connected or not.

Originally I connected the earth of the new caps to one of the two component earthing points and I changed it to the bolt that fastens the new tag strip to be nearer the original earth of the can but this made no difference.

I will report back with the voltages tomorrow.

Thanks

[cathoderay57]

Ok. The point of earthing the pentode grid is to see if the hum is still present I.e. might be a problem with the pentode stage. If the hum disappears it is originating in the triode stage. The other thing worth doing is disconnecting the negative feedback link from the speaker winding of the output transformer and see if that makes any difference. The other possibilities are that one or both of the other electrolytic capacitors has lost capacity, or that there is heater/cathode leakage in the ecl82 although you say you tried another valve with the same result so maybe that's unlikely unless both valves have had a hard life. Cheers, Jerry

[PaulR]

I checked the voltages with the MR in place and they were all very low so I have reinstated the diode and resistor which has brought them back to nearer where they should be:-

Triode anode - 101v
Triode cathode - 1.34v
Pentode anode - 216v
Pentode cathode - 13.74v
Pentode screen - 184v

Shorting the pentode control grid removed the hum entirely.

Disconnecting the feedback loop has no effect

[Nuvistor]

One small possibility is a leak between pins 3 and 4 of the ECL82 valve holder, I.e g1 and heater.
If the valve base is ceramic or hard plastic probably not but if paxolin a definite possibility.

Edit.
Is it a wired chassis or PCB, if a PCB it could be a leak on that.

[PaulR]

Thanks Frank. The valve holder is new. It is hard plastic rather than paxolin. It is a wired chassis.

I am wondering about the valve now. The replacement I tried was an old one pulled from a P+P amp

[PaulR]

I think it must be the valve. I tried disconnecting the heater whilst it was switched on and the hum immediately went. That is a pity as a new valve will cost more than the record player did!

[vidjoman]

You said the valve holder is new. Is there a mistake in the wiring made when it was changed.

[PaulR]

I don't think there was a mistake. I took "before" pictures and it looks ok.

[Nuvistor]

Confirm the heater wires and G1 wires are dressed as before.

[Trevor]

Are the heater windings earthed either via a centre tap resistor combination or on one side of the heater this can hum if the heater is left floating
Trev

[cathoderay57]

The hum is coming from the Triode section then. Before you give up on the valve it's worth checking/bridging with new/replacing the other 2 x 33uF HT caps (you said you'd done the reservoir already). Also try shorting the slider contact of the vol control to chassis just in case R1 has gone o/c or has a dry joint. Occasionally the carbon track inside the pot develops a dry joint with the earthy end contact and this test will eliminate that as poss cause. I have a few ECL82 so will test you a decent one and send PM with test results and price. Cheers, Jerry.

[PaulR]

Thanks for all the replies.

I found another ECL82 and the hum is still there so I think it must be something with the wiring as I doubt that three valves would have the same fault. I cannot really understand why the hum went as soon as I disconnected the heater lead from the transformer though.


I did replace all three of the 32uf smoothing caps. I just bridged one of the new ones to increase the capacitance to 64uf. This seemed to reduce the hum marginally so I left it in.

I have tried earthing the slider of the volume control which did not help. One side of the heater winding is earthed so there is a single heater wire with the other side of the heater and indicator bulb earthed.

I think this is one of those times where I need to leave it for a few days and come back to it with fresh eyes.

Many thanks

[cathoderay57]

OK, well done for persevering. When you get back to it try disconnecting the coax lead from Pin1 (Grid Triode) and earthing Pin 1. If the hum stops then you have pick-up from the tone control circuitry or a bad screen on the coax. The only other thing I can think of at this point is bad earthing on the metal casings of the volume control and tone control. Try extra earth wires there. Cheers, Jerry

[Ambientnoise]

I presume it is 50Hz hum and not 100Hz buzz which dominates ? You said one side of the heater supply is earthed so it could be that the heater current is producing a voltage drop in the wiring which is effectively added onto the input. You could try disconnecting the heater from earth, taking the two heater wires twisted together directly to the transformer, then adding a link from one side of the transformer heater windings AT the transformer to earth. Instead, You could also try putting say two 1k resistors across the heater and earthing the centre tap.

[nicam49]

Hi, would it be feasible to run the heater from a 5volt dc wall wart so no AC present to introduce hum?

[cathoderay57]

Another thought would be to fit a screening can around the ECL82. Try some cooking foil earthed to the chassis first to see if there's any hum improvement and if it fixes the problem replace the socket with one with a skirt and can.

[PaulR]

Thanks for all the extra thoughts.

I have tried earthing the triode grid without actually disconnecting the screened lead but it made no difference. I have checked the screening for continuity to earth and that is fine. I will try disconnecting it completely as suggested.

It does seem to be 50Hz. It only has a half wave rectifier. I wonder whether the heater supply is the cause somewhere but I cannot see what it is at the moment. I feel that it should work as designed but I could try the twisted pair idea to see whether it helps. Similarly with the idea of a separate DC supply for the heaters, it should work as designed.

I have already tried a screening can on the valve and unfortunately it made no difference.

I think that the cause must be something to do with the heater supply somehow because the hum stopped when it was disconnected. It is basically one wire so I don't know what is the problem, but it may become glaringly obvious when I look at it again. I have already checked for stray bits of wire or solder connecting it to a grid but I cannot see anything.