Suspect Mains Transformer in 1930s Radio P.S.U. - Page 2 - UK Vintage Radio Repair and Restoration Discussion Forum

'LIVEWIRE?' · 19th Apr 2018, 8:25 am

Briefly replying to posts #15-20 I was fairly certain that the imbalance between the two halves of the primary winding ware normal for reasons stated. Jerry, I'm 100% certain there are no mains filter caps. I've double checked, and the Power supply is exactly as I've drawn it. Mike, I can check the valve heater windings. I have several low-voltage D.C. supplies, and can no doubt find another transformer with a low-voltage A.C. secondary. I hadn't thought of disconnecting the heater center tap from chassis, so, thanks for that tip. At the moment I don't have a known working valve radio to try via my lamp limiter, so that, for the time being, isn't an option. Finally, I've just looked at my consumer unit again, and the item which trips is definetly a Wylex R.C.D., which trips @ 30mA. There is a yellow test button on the R.C.D., and I've never known it to 'nuisance-trip!'

Station X · 19th Apr 2018, 8:40 am

Have you tried short-circuiting the "output" of the test lamp? This would mimic the worst possible fault on a radio without an earth connection. ie a short-circuit mains transformer primary winding.

'LIVEWIRE?' · 19th Apr 2018, 10:33 am

No, Graham, I hadn't thought of that. By so doing, this would, of course put the 100W Bulb straight across the mains, so it would draw about 400mA @ 240v.

Boater Sam · 19th Apr 2018, 10:36 am

Stick an earth on the chassis and disconnect in turn the line and then the neutral. If the rcd trips, you then know which is causing the problem.

With only 2 wire mains lead and no earth, tripping the rcd is not logical. Investigate your lamp limiter, has it got a return to earth instead of the neutral?

Station X · 19th Apr 2018, 10:46 am

This fault seems to defy the laws of physics, so either something we've been told is incorrect, or some fact has been omitted.

One obvious question is does the RCD trip with the set connected to the lamp limiter, but with the set not turned on?

HamishBoxer · 19th Apr 2018, 10:53 am

Good point Graham, an internal short on the Volume On/Off pot switch?

Station X · 19th Apr 2018, 11:05 am

That shouldn't trip an RCD though, but I'm clutching at straws. I thought it might be a useful diagnostic test.

HamishBoxer · 19th Apr 2018, 12:05 pm

The problem is to a degree,trying to make an educated guess were if it was on the bench in front of you no doubt it would be quickly diagnosed by some of us.

I put some of us to allow for those who are not too technical or spent there life doing repairs.

Hence,that is quite right that they join this forum for advice.

'LIVEWIRE?' · 19th Apr 2018, 1:14 pm

I've re-checked the wiring to my lamp limiter, and all is in order, so will get back to the radio soon.

julie_m · 19th Apr 2018, 1:58 pm

Given the way an RCD works, only one of two things can trip it: Something flowing out of the live and not returning via the neutral (perhaps taking a shortcut to Earth through someone's body), or something going into the neutral that did not come from the live (e.g., from another circuit not fed via the same RCD, or a charged capacitor dumped across N-E). An appliance connected only to the live and neutral pins of a socket should not be capable of tripping an RCD.

A short-circuit from neutral to earth could also trip an RCD, as it will shunt away some of the current that would otherwise return via the neutral through the RCD.

If you have a Megger, try measuring across L-E and N-E in turn. Whatever is amiss almost certainly involves the Earth connection at the output of the lamp limiter.

Refugee · 19th Apr 2018, 3:56 pm

I would start with the insulation tester at the plug pins on the limiter.
For all we know there could be something like a dead slug in the limiter.

crackle · 19th Apr 2018, 5:28 pm

If there was a leakage from primary winding to chassis, and the transformer secondary heater winding was also connected to chassis via the centre tap, could that cause an imbalance on one leg (the one nearest the chassis leakage) of the mains supply.

Mike

HamishBoxer · 19th Apr 2018, 5:35 pm

Has the primary (mains side of transformer) been checked to the chassis with an Ohm meter for leakage?

Sorry if I have missed it if so.

julie_m · 19th Apr 2018, 5:37 pm

But then, where's the leakage current going? An RCD trips when there is a difference between the currents in the live and neutral. There must be an alternative current path; either from the set to Earth, or (less likely) from somewhere not supplied through the same RCD to neutral.

TrevorG3VLF · 19th Apr 2018, 5:54 pm

As mentioned previously, a twin lead will not give an unbalance and will not trip an RCD. The only possibility is that the set is wired line to earth. Look in the plug.

Silicon · 19th Apr 2018, 7:05 pm

Some RCDs were sensitive to an imbalance of the positive and negative pulses in the AC waveform.

The imbalance effectively creates a net DC current through the RCD.

It might be possible to create a DC current through the RCD if the full wave rectifier is open circuit (or short circuit) between one of the anodes and the cathode.

'LIVEWIRE?' · 19th Apr 2018, 7:59 pm

I started this thread by stating that I had checked from both sides of the primary to chassis with my TMK Leakage tester, the readings obtained being >200Megohms, as they were between Primary & secondary. The TMK tester applies 500v to the item under test. I don't have a conventional 'megger', but find the TMK meter perfectly good enough for these tests. To answer your point, Silicon, the 1821 valve rectifier was plugged in whilst powering the radio, as was the 6v pilot bulb. No other valves were fitted. Trevor, the set is definitely not wired line to earth. I 'm using the original 2-pin mains connector, which only connects to the Tx primary, and to which I fitted a length of new twim mains lead and 15A Plug (with a 2Amp Fuse!). Refugee, I only recently built the limiter, using a new switched 13A socket, separate 13A swirch wired to short out the series filament lamp, which is mounted in a new batten lamp-holder. The mains lead I fitted is admittedly a length of 3-core already fitted with a 13A plug, which I saved from some scrapped item. It should be O.K., but I'll check that again ASAP.

julie_m · 19th Apr 2018, 8:15 pm

If the RCD is responding to an asymmetric load due to half-wave rectification, then you should be able to trip it with a 1N4007 diode in series with a 100W tungsten filament lamp.

But then, a transformerless set would also trip such an RCD; as might some modern hairdriers, and other short-term-use appliances which use simple half-wave rectification to get a half-heat setting.

MotorBikeLes · 19th Apr 2018, 9:34 pm

Trevor #35, what you say is NOT correct. If the radio was purely resistive, you would be correct, but where you have AC , inductors and or capacitors, ANYTHING can happen. A big capacitive delay in 0ne part of the circuit can cause an RCD to trip. As others have said, the RCD does NOT sense leakage to earth, but it DOES sense current difference between the two conductors, which is USUALLY due to a leakage to earth from either L or N.
Les.

G8UWM-MildMartin · 19th Apr 2018, 9:37 pm

It sounds to me as if the RCD also feeds other equipment with a significant amount of suppression capacitance both L-E and N-E and the "spike" when the transformer is switched is being divided asymmetrically, causing a transient residual current. In other words, the transformer is OK and the tripping will stop if some other appliances are unplugged (not just switched off with a single pole switch), though I'm surprised it hasn't happened before.

19th Apr 2018, 8:25 am	#21
'LIVEWIRE?' Rest in Peace Join Date: Mar 2006 Location: N.W. Oxfordshire(Chipping Norton) Posts: 7,306	Re: Suspect Mains Transformer in 1930s Radio P.S.U. Briefly replying to posts #15-20 I was fairly certain that the imbalance between the two halves of the primary winding ware normal for reasons stated. Jerry, I'm 100% certain there are no mains filter caps. I've double checked, and the Power supply is exactly as I've drawn it. Mike, I can check the valve heater windings. I have several low-voltage D.C. supplies, and can no doubt find another transformer with a low-voltage A.C. secondary. I hadn't thought of disconnecting the heater center tap from chassis, so, thanks for that tip. At the moment I don't have a known working valve radio to try via my lamp limiter, so that, for the time being, isn't an option. Finally, I've just looked at my consumer unit again, and the item which trips is definetly a Wylex R.C.D., which trips @ 30mA. There is a yellow test button on the R.C.D., and I've never known it to 'nuisance-trip!'

19th Apr 2018, 8:40 am	#22
Station X Moderator Join Date: Jan 2003 Location: Ipswich, Suffolk, IP4, UK. Posts: 21,289	Re: Suspect Mains Transformer in 1930s Radio P.S.U. Have you tried short-circuiting the "output" of the test lamp? This would mimic the worst possible fault on a radio without an earth connection. ie a short-circuit mains transformer primary winding. __________________ Graham. Forum Moderator Reach for your meter before you reach for your soldering iron.

19th Apr 2018, 10:33 am	#23
'LIVEWIRE?' Rest in Peace Join Date: Mar 2006 Location: N.W. Oxfordshire(Chipping Norton) Posts: 7,306	Re: Suspect Mains Transformer in 1930s Radio P.S.U. No, Graham, I hadn't thought of that. By so doing, this would, of course put the 100W Bulb straight across the mains, so it would draw about 400mA @ 240v.

19th Apr 2018, 10:36 am	#24
Boater Sam Banned Join Date: Dec 2013 Location: Middlewich, Cheshire, UK. & Winter in the Philippines. Posts: 3,897	Re: Suspect Mains Transformer in 1930s Radio P.S.U. Stick an earth on the chassis and disconnect in turn the line and then the neutral. If the rcd trips, you then know which is causing the problem. With only 2 wire mains lead and no earth, tripping the rcd is not logical. Investigate your lamp limiter, has it got a return to earth instead of the neutral?

19th Apr 2018, 10:46 am	#25
Station X Moderator Join Date: Jan 2003 Location: Ipswich, Suffolk, IP4, UK. Posts: 21,289	Re: Suspect Mains Transformer in 1930s Radio P.S.U. This fault seems to defy the laws of physics, so either something we've been told is incorrect, or some fact has been omitted. One obvious question is does the RCD trip with the set connected to the lamp limiter, but with the set not turned on? __________________ Graham. Forum Moderator Reach for your meter before you reach for your soldering iron.

19th Apr 2018, 10:53 am	#26
HamishBoxer Dekatron Join Date: Aug 2007 Location: W.Butterwick, near Doncaster UK. Posts: 8,935	Re: Suspect Mains Transformer in 1930s Radio P.S.U. Good point Graham, an internal short on the Volume On/Off pot switch? __________________ G8JET BVWS Archivist and Member V.M.A.R.S

19th Apr 2018, 11:05 am	#27
Station X Moderator Join Date: Jan 2003 Location: Ipswich, Suffolk, IP4, UK. Posts: 21,289	Re: Suspect Mains Transformer in 1930s Radio P.S.U. That shouldn't trip an RCD though, but I'm clutching at straws. I thought it might be a useful diagnostic test. __________________ Graham. Forum Moderator Reach for your meter before you reach for your soldering iron.

19th Apr 2018, 12:05 pm	#28
HamishBoxer Dekatron Join Date: Aug 2007 Location: W.Butterwick, near Doncaster UK. Posts: 8,935	Re: Suspect Mains Transformer in 1930s Radio P.S.U. The problem is to a degree,trying to make an educated guess were if it was on the bench in front of you no doubt it would be quickly diagnosed by some of us. I put some of us to allow for those who are not too technical or spent there life doing repairs. Hence,that is quite right that they join this forum for advice. __________________ G8JET BVWS Archivist and Member V.M.A.R.S

19th Apr 2018, 1:14 pm	#29
'LIVEWIRE?' Rest in Peace Join Date: Mar 2006 Location: N.W. Oxfordshire(Chipping Norton) Posts: 7,306	Re: Suspect Mains Transformer in 1930s Radio P.S.U. I've re-checked the wiring to my lamp limiter, and all is in order, so will get back to the radio soon.

19th Apr 2018, 1:58 pm	#30
julie_m Dekatron Join Date: May 2008 Location: Derby, UK. Posts: 7,735	Re: Suspect Mains Transformer in 1930s Radio P.S.U. Given the way an RCD works, only one of two things can trip it: Something flowing out of the live and not returning via the neutral (perhaps taking a shortcut to Earth through someone's body), or something going into the neutral that did not come from the live (e.g., from another circuit not fed via the same RCD, or a charged capacitor dumped across N-E). An appliance connected only to the live and neutral pins of a socket should not be capable of tripping an RCD. A short-circuit from neutral to earth could also trip an RCD, as it will shunt away some of the current that would otherwise return via the neutral through the RCD. If you have a Megger, try measuring across L-E and N-E in turn. Whatever is amiss almost certainly involves the Earth connection at the output of the lamp limiter. __________________ If I have seen further than others, it is because I was standing on a pile of failed experiments.

19th Apr 2018, 3:56 pm	#31
Refugee Dekatron Join Date: Apr 2012 Location: Worksop, Nottinghamshire, UK. Posts: 5,554	Re: Suspect Mains Transformer in 1930s Radio P.S.U. I would start with the insulation tester at the plug pins on the limiter. For all we know there could be something like a dead slug in the limiter.

19th Apr 2018, 5:28 pm	#32
crackle Rest in Peace Join Date: Mar 2011 Location: Basildon, Essex, UK. Posts: 4,100	Re: Suspect Mains Transformer in 1930s Radio P.S.U. If there was a leakage from primary winding to chassis, and the transformer secondary heater winding was also connected to chassis via the centre tap, could that cause an imbalance on one leg (the one nearest the chassis leakage) of the mains supply. Mike

19th Apr 2018, 5:35 pm	#33
HamishBoxer Dekatron Join Date: Aug 2007 Location: W.Butterwick, near Doncaster UK. Posts: 8,935	Re: Suspect Mains Transformer in 1930s Radio P.S.U. Has the primary (mains side of transformer) been checked to the chassis with an Ohm meter for leakage? Sorry if I have missed it if so. __________________ G8JET BVWS Archivist and Member V.M.A.R.S

19th Apr 2018, 5:37 pm	#34
julie_m Dekatron Join Date: May 2008 Location: Derby, UK. Posts: 7,735	Re: Suspect Mains Transformer in 1930s Radio P.S.U. But then, where's the leakage current going? An RCD trips when there is a difference between the currents in the live and neutral. There must be an alternative current path; either from the set to Earth, or (less likely) from somewhere not supplied through the same RCD to neutral. __________________ If I have seen further than others, it is because I was standing on a pile of failed experiments.

19th Apr 2018, 5:54 pm	#35
TrevorG3VLF Rest in Peace Join Date: Apr 2011 Location: Matlock, Derbyshire, UK. Posts: 1,378	Re: Suspect Mains Transformer in 1930s Radio P.S.U. As mentioned previously, a twin lead will not give an unbalance and will not trip an RCD. The only possibility is that the set is wired line to earth. Look in the plug.

19th Apr 2018, 7:05 pm	#36
Silicon Nonode Join Date: Jun 2007 Location: Coulsdon, London, UK. Posts: 2,167	Re: Suspect Mains Transformer in 1930s Radio P.S.U. Some RCDs were sensitive to an imbalance of the positive and negative pulses in the AC waveform. The imbalance effectively creates a net DC current through the RCD. It might be possible to create a DC current through the RCD if the full wave rectifier is open circuit (or short circuit) between one of the anodes and the cathode.

19th Apr 2018, 7:59 pm	#37
'LIVEWIRE?' Rest in Peace Join Date: Mar 2006 Location: N.W. Oxfordshire(Chipping Norton) Posts: 7,306	Re: Suspect Mains Transformer in 1930s Radio P.S.U. I started this thread by stating that I had checked from both sides of the primary to chassis with my TMK Leakage tester, the readings obtained being >200Megohms, as they were between Primary & secondary. The TMK tester applies 500v to the item under test. I don't have a conventional 'megger', but find the TMK meter perfectly good enough for these tests. To answer your point, Silicon, the 1821 valve rectifier was plugged in whilst powering the radio, as was the 6v pilot bulb. No other valves were fitted. Trevor, the set is definitely not wired line to earth. I 'm using the original 2-pin mains connector, which only connects to the Tx primary, and to which I fitted a length of new twim mains lead and 15A Plug (with a 2Amp Fuse!). Refugee, I only recently built the limiter, using a new switched 13A socket, separate 13A swirch wired to short out the series filament lamp, which is mounted in a new batten lamp-holder. The mains lead I fitted is admittedly a length of 3-core already fitted with a 13A plug, which I saved from some scrapped item. It should be O.K., but I'll check that again ASAP.

19th Apr 2018, 8:15 pm	#38
julie_m Dekatron Join Date: May 2008 Location: Derby, UK. Posts: 7,735	Re: Suspect Mains Transformer in 1930s Radio P.S.U. If the RCD is responding to an asymmetric load due to half-wave rectification, then you should be able to trip it with a 1N4007 diode in series with a 100W tungsten filament lamp. But then, a transformerless set would also trip such an RCD; as might some modern hairdriers, and other short-term-use appliances which use simple half-wave rectification to get a half-heat setting. __________________ If I have seen further than others, it is because I was standing on a pile of failed experiments.

19th Apr 2018, 9:34 pm	#39
MotorBikeLes Nonode Join Date: Dec 2012 Location: Kirk Michael, Isle of Man Posts: 2,350	Re: Suspect Mains Transformer in 1930s Radio P.S.U. Trevor #35, what you say is NOT correct. If the radio was purely resistive, you would be correct, but where you have AC , inductors and or capacitors, ANYTHING can happen. A big capacitive delay in 0ne part of the circuit can cause an RCD to trip. As others have said, the RCD does NOT sense leakage to earth, but it DOES sense current difference between the two conductors, which is USUALLY due to a leakage to earth from either L or N. Les.

19th Apr 2018, 9:37 pm	#40
G8UWM-MildMartin Heptode Join Date: Jan 2004 Location: Stockport, Greater Manchester, UK. Posts: 827	Re: Suspect Mains Transformer in 1930s Radio P.S.U. It sounds to me as if the RCD also feeds other equipment with a significant amount of suppression capacitance both L-E and N-E and the "spike" when the transformer is switched is being divided asymmetrically, causing a transient residual current. In other words, the transformer is OK and the tripping will stop if some other appliances are unplugged (not just switched off with a single pole switch), though I'm surprised it hasn't happened before.