Ekco U800 hum problem - UK Vintage Radio Repair and Restoration Discussion Forum

agardiner · 19th Apr 2021, 1:48 pm

Hi,

So I am working on a table model by Ekco, a U800 which is similar to the U834 chassis but with twin speakers.

I have recapped it, both waxies and electrolytics, replaced a faulty UCH81 mixer valve and replaced the metal rectifier with a 1N5408. To compensate for the new diode, I increased its series resistor from 47R to 100R.

When connected to my bench speaker it is working quite well; all bands are working and pulling in stations. However, when connecting it back to its own speakers which are more sensitive, the level of hum is excessive. It only hums once the valves warm up, and is not dependent on volume level, although slightly louder when the volume is turned all the way down.

I measure 3.2V peak to peak ripply on the main smoothing cap, and 0.327V of ripple across C40. This would seem reasonable given a single rectifier diode?

Any tips on what might be the problem please?

Many thanks,
Adrian.

unitelex · 19th Apr 2021, 2:20 pm

Hi Adrian,
Could be noise generated by fast switching of the newly added silicon rectifier.
Suggest trying 630V 1nF - 10nF across the diode. HV ceramic works well in this application.

Best Regards
Chris

agardiner · 19th Apr 2021, 2:42 pm

Thanks Chris, will try.

Would you agree that the level of ripple I can see on the scope would be about normal?

unitelex · 19th Apr 2021, 3:21 pm

Having looked at the U834 circuit, I would check the following:

C43 - chief suspect

Check the S2B switch wafer (contamination carbon) and contacts (clean)

Check currents being fed from the ht to earlier stages, if there is a leaky capacitor in those earlier stages it will cause more ripple generally.

C40 decoupling the anode HT for V3, understand you have changed this but it might benefit from a higher value here, I've often seen improvements with this.

If hum increases at minimum volume it sounds like the hum bucking feedback (R11, R15) is not optimised, maybe it is now not as well matched to the HT ripple due to your diode/cap changes.

C41 can be increased as you don't have a thermionic rectifier current limitation
C42 value can be somewhat critical on some sets with hum increasing if you increase the value of this capacitor due to its interaction with the output transformer/choke and the hum bucking scheme. If you are using higher than the original (60uF) then suggest you try reducing it.

You said you added resistance due to the silicon diode substitution, assume you increased R24. You could try increasing R22 to some extent instead. There might be a balance here.

Best Regards
Chris

unitelex · 19th Apr 2021, 3:22 pm

Where did you measure the 3.2V? C42 or C41?

Nuvistor · 19th Apr 2021, 3:24 pm

Not sure about the ripple voltages.
Could a heater cathode leak in one of the valves be the cause?

agardiner · 19th Apr 2021, 3:24 pm

Quote:

Originally Posted by unitelex

Where did you measure the 3.2V? C42 or C41?

C42. The ripple across C41 was slightly less, but still around 3V.

agardiner · 19th Apr 2021, 3:27 pm

Quote:

Originally Posted by unitelex

Having looked at the U834 circuit, I would check the following:

C43 - chief suspect

Check the S2B switch wafer (contamination carbon) and contacts (clean)

Check currents being fed from the ht to earlier stages, if there is a leaky capacitor in those earlier stages it will cause more ripple generally.

C40 decoupling the anode HT for V3, understand you have changed this but it might benefit from a higher value here, I've often seen improvements with this.

If hum increases at minimum volume it sounds like the hum bucking feedback (R11, R15) is not optimised, maybe it is now not as well matched to the HT ripple due to your diode/cap changes.

C41 can be increased as you don't have a thermionic rectifier current limitation
C42 value can be somewhat critical on some sets with hum increasing if you increase the value of this capacitor due to its interaction with the output transformer/choke and the hum bucking scheme. If you are using higher than the original (60uF) then suggest you try reducing it.

You said you added resistance due to the silicon diode substitution, assume you increased R24. You could try increasing R22 to some extent instead. There might be a balance here.

Best Regards
Chris

Many thanks; will look at all of these points.

agardiner · 19th Apr 2021, 3:28 pm

Quote:

Originally Posted by Nuvistor

Not sure about the ripple voltages.
Could a heater cathode leak in one of the valves be the cause?

It had occurred to me, but I don't believe this is the cause; no leaks detected on my valve tester.

unitelex · 19th Apr 2021, 3:32 pm

Check valve bases V3 and V4 (top and underside) for carbonisation and clean any contamination with IPA.

An interesting experiment would be to short V3 anode to ground to see if the hum is coming from V3 or V4

agardiner · 19th Apr 2021, 3:33 pm

Quote:

Originally Posted by unitelex

Check valve bases V3 and V4 (top and underside) for carbonisation and clean any contamination with IPA.

An interesting experiment would be to short V3 anode to ground to see if the hum is coming from V3 or V4

Will do.

ms660 · 19th Apr 2021, 3:46 pm

Quote:

Originally Posted by agardiner

Quote:

Originally Posted by unitelex

Where did you measure the 3.2V? C42 or C41?

C42. The ripple across C41 was slightly less, but still around 3V.

C41 isn't doing its job for whatever reason....53 Ohms (the reactance of C41) and 1k (the resistance of R22) in a potential divider should reduce the ripple voltage drastically, also note that the receiver has a hum cancelling output transformer circuit.

Lawrence.

unitelex · 19th Apr 2021, 3:50 pm

I would also expect the ripple on C41 should be significantly less than ripple on C42
Would check the wiring around T3 R22 C41 and C42

agardiner · 28th Apr 2021, 11:11 am

Solved - I have learnt something.

Thanks again for everyone's suggestions. Working through them, I came to the conclusion that the hum filter circuit must be out of balance. I had the same ripple level of 3.2V across both C42 and C41. The speaker transformer tested fine with no shorts.

The original C42/41 can was 60uF/60uF. I had replaced with a 50/50, although it tested as 56/55. I had already tried making up the values with parallel caps.

However, lowering the value of C42 to 32uF completely cleared up the problem. The ripple across it increased as expected to around 10V, but the ripple on C41 plummeted to only 210mV, proving the circuit was indeed out of balance. A long soak test to ensure C42 was not getting warm, and the set was good to go.

You would not expect that lowering the value of a capacitor can reduce the hum!

ms660 · 28th Apr 2021, 11:18 am

Something not quite adding up there to me.

Lawrence.

19th Apr 2021, 1:48 pm	#1
agardiner Octode Join Date: Sep 2020 Location: Thetford, Norfolk, UK. Posts: 1,741	Ekco U800 hum problem Hi, So I am working on a table model by Ekco, a U800 which is similar to the U834 chassis but with twin speakers. I have recapped it, both waxies and electrolytics, replaced a faulty UCH81 mixer valve and replaced the metal rectifier with a 1N5408. To compensate for the new diode, I increased its series resistor from 47R to 100R. When connected to my bench speaker it is working quite well; all bands are working and pulling in stations. However, when connecting it back to its own speakers which are more sensitive, the level of hum is excessive. It only hums once the valves warm up, and is not dependent on volume level, although slightly louder when the volume is turned all the way down. I measure 3.2V peak to peak ripply on the main smoothing cap, and 0.327V of ripple across C40. This would seem reasonable given a single rectifier diode? Any tips on what might be the problem please? Many thanks, Adrian.

19th Apr 2021, 2:20 pm	#2
unitelex Hexode Join Date: May 2005 Location: Greater Manchester, UK. Posts: 496	Re: Ekco U800 hum problem Hi Adrian, Could be noise generated by fast switching of the newly added silicon rectifier. Suggest trying 630V 1nF - 10nF across the diode. HV ceramic works well in this application. Best Regards Chris

19th Apr 2021, 2:42 pm	#3
agardiner Octode Join Date: Sep 2020 Location: Thetford, Norfolk, UK. Posts: 1,741	Re: Ekco U800 hum problem Thanks Chris, will try. Would you agree that the level of ripple I can see on the scope would be about normal?

19th Apr 2021, 3:21 pm	#4
unitelex Hexode Join Date: May 2005 Location: Greater Manchester, UK. Posts: 496	Re: Ekco U800 hum problem Having looked at the U834 circuit, I would check the following: C43 - chief suspect Check the S2B switch wafer (contamination carbon) and contacts (clean) Check currents being fed from the ht to earlier stages, if there is a leaky capacitor in those earlier stages it will cause more ripple generally. C40 decoupling the anode HT for V3, understand you have changed this but it might benefit from a higher value here, I've often seen improvements with this. If hum increases at minimum volume it sounds like the hum bucking feedback (R11, R15) is not optimised, maybe it is now not as well matched to the HT ripple due to your diode/cap changes. C41 can be increased as you don't have a thermionic rectifier current limitation C42 value can be somewhat critical on some sets with hum increasing if you increase the value of this capacitor due to its interaction with the output transformer/choke and the hum bucking scheme. If you are using higher than the original (60uF) then suggest you try reducing it. You said you added resistance due to the silicon diode substitution, assume you increased R24. You could try increasing R22 to some extent instead. There might be a balance here. Best Regards Chris

19th Apr 2021, 3:22 pm	#5
unitelex Hexode Join Date: May 2005 Location: Greater Manchester, UK. Posts: 496	Re: Ekco U800 hum problem Where did you measure the 3.2V? C42 or C41?

19th Apr 2021, 3:24 pm	#6
Nuvistor Dekatron Join Date: Aug 2013 Location: Wigan, Greater Manchester, UK. Posts: 9,431	Re: Ekco U800 hum problem Not sure about the ripple voltages. Could a heater cathode leak in one of the valves be the cause? __________________ Frank

19th Apr 2021, 3:32 pm	#10
unitelex Hexode Join Date: May 2005 Location: Greater Manchester, UK. Posts: 496	Re: Ekco U800 hum problem Check valve bases V3 and V4 (top and underside) for carbonisation and clean any contamination with IPA. An interesting experiment would be to short V3 anode to ground to see if the hum is coming from V3 or V4

19th Apr 2021, 3:50 pm	#13
unitelex Hexode Join Date: May 2005 Location: Greater Manchester, UK. Posts: 496	Re: Ekco U800 hum problem I would also expect the ripple on C41 should be significantly less than ripple on C42 Would check the wiring around T3 R22 C41 and C42 Last edited by unitelex; 19th Apr 2021 at 3:52 pm. Reason: added suggestion

28th Apr 2021, 11:11 am	#14
agardiner Octode Join Date: Sep 2020 Location: Thetford, Norfolk, UK. Posts: 1,741	Re: Ekco U800 hum problem Solved - I have learnt something. Thanks again for everyone's suggestions. Working through them, I came to the conclusion that the hum filter circuit must be out of balance. I had the same ripple level of 3.2V across both C42 and C41. The speaker transformer tested fine with no shorts. The original C42/41 can was 60uF/60uF. I had replaced with a 50/50, although it tested as 56/55. I had already tried making up the values with parallel caps. However, lowering the value of C42 to 32uF completely cleared up the problem. The ripple across it increased as expected to around 10V, but the ripple on C41 plummeted to only 210mV, proving the circuit was indeed out of balance. A long soak test to ensure C42 was not getting warm, and the set was good to go. You would not expect that lowering the value of a capacitor can reduce the hum!

28th Apr 2021, 11:18 am	#15
ms660 Dekatron Join Date: Apr 2011 Location: Cornwall, UK. Posts: 13,454	Re: Ekco U800 hum problem Something not quite adding up there to me. Lawrence.