UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > Components and Circuits

Notices

Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets.

Closed Thread
 
Thread Tools
Old 24th Nov 2020, 3:06 pm   #1
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
Default EBC33 strange fault

Running my DAC90 last week, I suddenly became aware of a loud hum.

Quickly sprinting over to switch off (on the basis that the big reservoir / smoothing can might be dying and about to spread itself), hum disappeared immediately on switch off, leaving a second or so of hum-free audio as the sound faded.

The following investigation showed that neither section of the can is at fault. After a few red herrings, the EBC33 is the culprit. Removing the to cap: Hum disappears (in fact the open-circuit grid builds up enough charge to cut off the valve). Shorting the grid to chassis: no hum (except extremely light background).

But, connecting the top cap of the valve itself to chassis by 1MΩ, there's loud hum.

Killing the heater supply by shorting the EBC33 heater to chassis (it's a series chain, so OK - other valves will be slightly over-run, but it's only for a moment...) the hum disappears and clear audio results, at least until the cathode cools down.

Replacing the EBC33 with another: No problem.

So, the hummy EBC33: First, the metallising is intact and continuity exists to Pin 1. Measuring resistance from heater to other pins (and top-cap): O/C with my Fluke 179 (which reads to 50MΩ). However, with 6.3V bench supply connected to heater, resistance heater-grid falls to 1.6MΩ (grid positive) and 18MΩ (heater negative).

'Scoping the EBC33 triode anode when fitted in the DAC90, the hum waveform does look like rectified 50Hz, which is consistent with a one-way leakage, or direct heater emission.

The question is, what could cause this? The grid connection is as far away as it could be from the heater; internally, if any breakdown was going to happen, it would be heater-cathode. And although I didn't notice the problem coming on like someone flicking a switch, it certainly did appear within a few seconds, not over many hours.

Has anyone else experienced this?
kalee20 is offline  
Old 26th Nov 2020, 5:22 pm   #2
vinrads
Dekatron
 
Join Date: Feb 2009
Location: Chesterfield, Derbyshire, UK.
Posts: 3,737
Default Re: EBC33 strange fault

Err no! but thanks for the tip its one to look out for . Mick.
vinrads is offline  
Old 26th Nov 2020, 8:12 pm   #3
Sideband
Dekatron
 
Sideband's Avatar
 
Join Date: Oct 2003
Location: Croydon, Surrey, UK.
Posts: 7,548
Default Re: EBC33 strange fault

Could still be leakage within the valve. The pinch inside holds the electrodes and it holds the grid supports even if the actual connection is taken to the top cap. Leakage between cathode and grid can occur across the surface of the pinch.
__________________
There are lots of brilliant keyboard players and then there is Rick Wakeman.....
Sideband is offline  
Old 26th Nov 2020, 8:21 pm   #4
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,951
Default Re: EBC33 strange fault

It could be that the valve has started to go 'gassy' - can you see if the getter is still nicely silver?

I once had a QQV06/40A go from 'perfectly fine and delivering 40 Watts of RF' to totally-dead-and-getters-gone-white' in just 30 minutes.

When the vacuum's gone, strange things happen.
G6Tanuki is offline  
Old 26th Nov 2020, 9:47 pm   #5
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
Default Re: EBC33 strange fault

Thanks for responses so far, folks!

I'm going to investigate this one a bit more. Unfortunately, the red metallising is in good condition (just when I don't want it to be!) so I can't see the getter.

But, I have some ideas to try, involving measuring the leakage current, seeing if it behaves as a good diode, etc.

I'll be back...
kalee20 is offline  
Old 27th Nov 2020, 8:31 am   #6
Robert Gribnau
Heptode
 
Robert Gribnau's Avatar
 
Join Date: May 2017
Location: Konongo, Ghana
Posts: 510
Default Re: EBC33 strange fault

Quote:
Originally Posted by Sideband View Post
Could still be leakage within the valve. The pinch inside holds the electrodes and it holds the grid supports even if the actual connection is taken to the top cap. Leakage between cathode and grid can occur across the surface of the pinch.
As far as I know, the two diodes are situated in the lower half of the valve, while the triode is situated in the upper half of the valve. The reason for having the control grid lead coming out at the top of the valve is to minimize the capacitance between the control grid and the other electrodes. It therefore seems unlikely to me that the grid supports run all the way down and into the pinch. This ofcourse doesn't rule out cathode to grid leakage, but I don't think it can happen in/on the pinch.
__________________
Robert

Last edited by Robert Gribnau; 27th Nov 2020 at 8:43 am. Reason: typo
Robert Gribnau is offline  
Old 2nd Dec 2020, 8:37 pm   #7
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
Default Re: EBC33 strange fault

Quote:
Originally Posted by G6Tanuki View Post
It could be that the valve has started to go 'gassy' - can you see if the getter is still nicely silver?

When the vacuum's gone, strange things happen.
Can't see the getter unfortunately, the pretty red Mullard metallising is intact!

However, I am confident that gassiness is not the issue.

With heater powered at 6.3V DC, 5 minute warm-up allowed.

Contact potential / space charge tests (measured with 10MΩ DVM:

Vk-g: -1.7V
Vk-d1: -1.9V
Vk-d2: -2.3V

Which seems a bit high... Measure short-circuit currents by strapping an ammeter between cathode and relevant electrode:

Ik-g: 45μA
Ik-d1: 50μA
Ik-d2: 62μA

Taking grid or either diode anode positive with respect to cathode, significant current flows (milliamps, as expected). Take these electrodes 100V negative with respect to cathode, measure the reverse current:

Ir g: < 0.1μA
Ir d1: < 0.1μA
Ir d2: < 0.1μA

If the valve had been gassy, I would have expected significant reverse current flow. There isn't. So I think vacuum is good.

Measure leakage heater-cathode at 100V DC test voltage:

Valve cold:

Ih-k (cathode positive): < 0.1μA
Ih-k (cathode negative): 0.1 - 0.2μA

Valve hot:

Ih-k (cathode positive): 50-490μA
Ih-k (cathode negative): 50-470μA

In each case, the current starts off at the smaller figure and steadily grows, levelling off at the larger figure.

Removing heater power: in each case, current falls to <10μA in 5 seconds. This is consistent with a leakage resistance (impurity in the heater insulation slurry mix? As resistance falls with rising temperature, for most insulators). As the leakage current is nearly the same in both polarities, this tends to discount any thermionic effects.

I have done more tests, but writing up is tomorrow's instalment!
kalee20 is offline  
Old 3rd Dec 2020, 12:20 am   #8
Refugee
Dekatron
 
Refugee's Avatar
 
Join Date: Apr 2012
Location: Worksop, Nottinghamshire, UK.
Posts: 5,549
Default Re: EBC33 strange fault

I am wondering if there is a little bit of the heater sticking out of the top of the cathode tube that has lost some of its insulation.
You might be able to see it by looking between the screening and the top cap.
Refugee is offline  
Old 3rd Dec 2020, 1:59 pm   #9
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
Default Re: EBC33 strange fault

Quote:
Originally Posted by Refugee View Post
I am wondering if there is a little bit of the heater sticking out of the top of the cathode tube that has lost some of its insulation.
You might be able to see it by looking between the screening and the top cap.
I think Refugee’s on the right track. Unfortunately, I can't quite see enough, even with the dim glow from the heater providing internal illumination.

Sod's law dictates that just when you actually want a bit of flaking-off metallisation, it's adhered well and nearly pristine...

As yesterday's tests found significant leakage between heater and cathode, measuring current between heater and grid is not so simple – cathode to grid current is obviously dead easy from a nicely-emitting cathode, so there is definitely a path heater to grid via the cathode.

Leakage between heater and cathode is NOT the original cause of the hum in the DAC90, because the cathode is AC grounded (just to be sure, I had connected 4700μF between cathode and chassis and it made not a scrap of difference).

So, I came up with the circuit below to measure current from heater to grid NOT via the cathode. Applying a few volts reverse bias between cathode grid, guarantees no current here, as from yesterday’s measurements. Any grid current, therefore, is directly from the heater.

All leads were kept airborne to minimise leakage:

100V applied, heater negative to cathode: Ig = 1μA
100V applied, heater positive to cathode: Ig = 0.1μA

So there is some forward current flow, albeit small.

Cutting heater power: current decays to 0.1μA in <5 seconds (heater loses temperature much faster than the cathode tube).

Whacking up heater voltage to 10V (mentally apologising to all valves, everywhere), forward current to grid increases to 8μA.

All this is quite consistent with direct emission from heater, getting picked up by the grid. Hotter heater has greater emission hence greater current. And, even just 1μA in 1MΩ grid leak would give 1V of hum on the grid in DAC90. The numbers add up...
Attached Thumbnails
Click image for larger version

Name:	EBC33 circuit.jpg
Views:	102
Size:	49.6 KB
ID:	221743  
kalee20 is offline  
Old 9th Dec 2020, 7:41 pm   #10
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
Default Re: EBC33 strange fault

Last test: Circuit set up as shown. Steady-state voltages as shown... All reasonable.

But, voltage on anode has huge, asymmetric hum, 4.5V p-p. This reduces as the balance potentiometer is twiddled towards the centre, nearly disappearing at near central position (resistances actually measured 249Ω + 255Ω). Either side of this and hum rises. Waveform looks the same (looks like a half-wave rectified crummy sinewave). However, what's interesting is that using the second scope channel, it is actually the other half-cycle that gets rectified.

Shorting the 1MΩ kills the hum, which definitely indicates that it is a small grid current causing the problem. Postulated that this occurs when the heater goes sufficiently negative of the grid, thus encouraging heater-grid emission.

So, last test: Disconnect the potentiometer wiper from 0V, connect it to a DC voltage source, so that the heater is biased +ve with respect to 0V. With balance potentiometer at one end, increasing the voltage from zero reduced hum till it is just a tip of a sinewave which is evident. I was expecting the hum to disappear at about 9V (peak of 6.3V) because at this point, the heater would at no point of the AC cycle be going negative of 0V, but it actually went with just 3.5V of bias. Last photo shows the hum just about to vanish.

I'm about as confident as I can be, that there's direct heater emission which is getting picked up by the grid. (It probably gets picked up by other electrodes too, but less harm is done). In a circuit which biased the heater line sufficiently positive, it would not give any trouble. (And likewise, if powered by DC).

Why it seemed to come on relatively quickly, I can only guess that some insulating oxide fell off the heater where it poked out of the end of the cathode tube.

Quote:
Originally Posted by Refugee View Post
I am wondering if there is a little bit of the heater sticking out of the top of the cathode tube that has lost some of its insulation.
You might be able to see it by looking between the screening and the top cap.
Which is a good point! I had a peer in, difficult to see, so winding up the heater to 12V for a moment, the cathode was glowing bright orange, enough to see that the heater is completely enclosed by the cathode tube (at least at the top end). Which makes it rather odd - but I've exhausted this issue I reckon!
Attached Thumbnails
Click image for larger version

Name:	Circuit.jpg
Views:	104
Size:	45.2 KB
ID:	222208   Click image for larger version

Name:	Hum 1.jpg
Views:	48
Size:	67.0 KB
ID:	222209   Click image for larger version

Name:	Hum 2.jpg
Views:	49
Size:	44.6 KB
ID:	222210  
kalee20 is offline  
Old 9th Dec 2020, 7:55 pm   #11
paulsherwin
Moderator
 
paulsherwin's Avatar
 
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,784
Default Re: EBC33 strange fault

For some reason, these double diode triodes (not only the EBC33) seem to develop odd faults. Maybe it's something to do with the construction.
paulsherwin is online now  
Old 9th Dec 2020, 11:43 pm   #12
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: EBC33 strange fault

I would agree with Sideband, a common failure mode for valves including the UL41 and you might be able to measure the leakage resistance when the valve is still warm. You may also be able to 'fix' the fault using a gas hob spark generator by inducing a few sparks between the grid and all the other electrodes tied together.
PJL is offline  
Old 10th Dec 2020, 11:05 am   #13
dominicbeesley
Octode
 
dominicbeesley's Avatar
 
Join Date: Nov 2004
Location: Hebden Bridge, West Yorkshire, UK.
Posts: 1,885
Default Re: EBC33 strange fault

It seems odd that it has happened of a sudden unless something has physically changed. Could it be a flake of cathode material falling on an exposed part of the heater? Or could the cathode tube have slipped?

My next (well my first) experimental step would be to give it a good shake, holding upside down then run it in the test jig keeping it upside down?

I love these conundrum threads!
dominicbeesley is offline  
Old 10th Dec 2020, 2:19 pm   #14
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
Default Re: EBC33 strange fault

Quote:
Originally Posted by paulsherwin View Post
For some reason, these double diode triodes (not only the EBC33) seem to develop odd faults. Maybe it's something to do with the construction.
Yes Paul - I had a hummy UABC80 some years ago (yes I know this is a triple-diode triode!) Hum on FM, but disappeared on tuning to a signal. I never investigated it though.

Quote:
Originally Posted by PJL View Post
I would agree with Sideband, a common failure mode for valves including the UL41 and you might be able to measure the leakage resistance when the valve is still warm. You may also be able to 'fix' the fault using a gas hob spark generator by inducing a few sparks between the grid and all the other electrodes tied together.
Well, I've come across UL41's with internal leakage - but tests in subsequent posts to Sideband's, indicate that leakage is not the issue here. It's one-way conduction!

Quote:
Originally Posted by dominicbeesley View Post
It seems odd that it has happened of a sudden unless something has physically changed. Could it be a flake of cathode material falling on an exposed part of the heater? Or could the cathode tube have slipped?

My next (well my first) experimental step would be to give it a good shake, holding upside down then run it in the test jig keeping it upside down?

I love these conundrum threads!
It's well thought of. I can't see any of these: the cathode tube wouldn't slip down without buckling the cathode lead. And I can't see a flake of cathode material breaking off and spontaneously heading upwards towards the top of the cathode tube (valve was in normal orientation at the time), and entering, in the manner of a basketball ball. But, I've got nothing better to offer!

I shall try giving judicious shakes and thumps in test condition - the set-up still exists. I'll report back!
kalee20 is offline  
Old 10th Dec 2020, 4:04 pm   #15
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: EBC33 strange fault

Love this thread , unravelling a seemingly intractable mystery. It takes a certain relentless curiosity and imagination , along with patience and clarity of insight , to get this far!
__________________
Al
Al (astral highway) is offline  
Old 10th Dec 2020, 4:10 pm   #16
Guest
Guest
 
Posts: n/a
Default Re: EBC33 strange fault

What are the measurements on the good valve?
 
Old 10th Dec 2020, 4:11 pm   #17
Al (astral highway)
Dekatron
 
Al (astral highway)'s Avatar
 
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
Default Re: EBC33 strange fault

Could you pass a one-shot pulse of current for <5 mS between G and heater (Heater pins shorter together to make one electrode) and if that doesn’t work then between G and everything else? It might clear something that shouldn’t be there?
__________________
Al
Al (astral highway) is offline  
Old 14th Dec 2020, 9:36 pm   #18
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
Default Re: EBC33 strange fault

Been on the job tonight!

Dominic - setting it up, running it, re-creating the grid-current fault with rectified hum, and turning the valve upside down and flicking it, tapping it with a pen, and bouncing it gent on the bench, makes not a scrap of difference!

Astral - trouble is, even with 100V, there's only 1μA flowing. That's with grid positive. And with grid negative, there's even less - it is as it should be.

So, trying to get a pulse of current to flow and zap off anything that shouldn't be there, is almost impossible. I could raise the voltage, but I fear that nothing would happen till I raised it to something which would cause a flashover in even a good valve. (And I'm not inexperienced in the high-voltage zap technique, having used it successfully on UL41's).

Merlinmaxwell - good valve measures O/C both ways, grid-heater.
kalee20 is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 1:27 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.