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 > Specific Vintage Equipment > Vintage Television and Video

Notices

Vintage Television and Video Vintage television and video equipment, programmes, VCRs etc.

Closed Thread
 
Thread Tools
Old 17th Jun 2018, 4:16 pm   #1
1100 man
Octode
 
1100 man's Avatar
 
Join Date: Apr 2016
Location: Ventnor, Isle of Wight, & Great Dunmow, Essex, UK.
Posts: 1,377
Default Hot LOPT's!

This is a problem I've now had on several 405 & D/S sets. The set works fine with a good picture but the LOPT gets much hotter than I would like. I do wonder how hot some of them got originally, especially being jammed up against a very hot LOP valve and boost diode and then being covered by a tin screening cover. Maybe this was one reason for premature failure on certain models?
Anyway, I've never done much investigation before: my solution in the past was to mount a small cooling fan fed from a capacitor dropper. This works fine and keeps the transformer nice and cool.

The Pilot TV that I've gradually been rebuilding has a similar problem and I've spent many hours trying to get to the bottom of it.
After half an hour, it's uncomfortably hot. After an hour its way too hot to touch. Not good.
The EHT, at 12kv is a bit low (specified at 16Kv).This drops to 10KV as the transformer gets hotter.

I thought it might be due to leakage from the wax coating on the overwind so this was dissolved off and re-coated with transformer varnish. Same with the primary. New LOP valve, boost diode, & EHT rectifier have been tried. All passive components are new anyway. I've experimented with the value of the tuning capacitor across the transformer. Nothing has made the slightest difference! I even removed the thing from the chassis to see if there was leakage to the core:- there wasn't!

The set has turned into a test bed for experiments with black level clamping & gated AGC systems and so has now been on for well over 100 hours I would guess so the transformer has had plenty of time to dry out. During this time It's been fan cooled and has worked fine like that.

I conclude, therefore it must be the transformer itself. I can understand the primary getting hot as it has to pass the HT current of the line stage, some 150Ma, through much of it's winding. But the overwind also gets just as hot so maybe it's due to internal leakage within the winding between layers that is the problem?

I doubt if I'll get to the bottom of it, so it will just have to be fan cooled like all the other ones!
Does anyone else have similar overheating problems?

All the best
Nick
1100 man is offline  
Old 17th Jun 2018, 4:31 pm   #2
Ed_Dinning
Dekatron
 
Join Date: Jan 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 8,171
Default Re: Hot LOPT's!

Hi nick, I wonder if the core was not up to it on 625 and had high losses. Which standard is it worst on?

Looks like you eliminated dielectric losses when you re-potted it. It could be a case of accountants finger, in that the windings are simply too small a diameter, copper content reduced to save a few pennies. This may stack up with the EHT reduction as copper has a large tempco.

Ed
Ed_Dinning is offline  
Old 17th Jun 2018, 4:31 pm   #3
kan_turk
Hexode
 
kan_turk's Avatar
 
Join Date: Mar 2014
Location: Dublin, Ireland
Posts: 396
Default Re: Hot LOPT's!

Yes, yes and yes

I used to think I was doing something wrong when it happened with experimental setups but I have experienced it with bog standard sets known to be working correctly - I never found the answer and have, reluctantly, adopted the same strategy as yourself ie employ a cooling fan - can't say I'm happy about it - I just don't like components running hot but maybe the truth is that this was normal back in the day - just nobody bothered - anyway its slightly comforting to know that someone else is in the same boat

Rgds
John
kan_turk is offline  
Old 17th Jun 2018, 5:30 pm   #4
simpsons
Octode
 
Join Date: Feb 2007
Location: Harrow, London, UK.
Posts: 1,483
Default Re: Hot LOPT's!

Could it be that RBM in their A640 chassis and other manufacturers placed the LOPT right at the bottom of the set just above air vents. Convection currents, TV sets placed on a carpet apart, would flow past the tx helping it keep cool.

Screening cans are a necessity though, I wouldn't fancy 3.5kv pulse from the line output top cap instead of a pacemaker, that is as nature gave me!!

Last edited by simpsons; 17th Jun 2018 at 5:32 pm. Reason: Didn't use windows spell check
simpsons is offline  
Old 17th Jun 2018, 7:45 pm   #5
1100 man
Octode
 
1100 man's Avatar
 
Join Date: Apr 2016
Location: Ventnor, Isle of Wight, & Great Dunmow, Essex, UK.
Posts: 1,377
Default Re: Hot LOPT's!

According to the 'developments in TV receivers' section in one of the red R&TV servicing books (they really are invaluable!) this set uses a 'de-saturated' LOPT.
The winding is split into two parts separated by the boost cap. The magnetic field generated by the DC current of the LOP valve cancels out as it passes through both parts of the winding thus reducing core saturation. This is supposed to give greater efficiency to drive the then new 21" 110 degree tube. An added benefit is that the line whistle is much more subdued than in a conventional transformer.

This set is 405 only and the transformer seems well made and easily removed. Just two screws, four push on tags and the two top cap connections. You could change one in the field in under 5 minutes! Maybe they expected unreliability!

Left to it's own devices, in it's present state it would self destruct if left on for much more than an hour or so (without fan cooling) so it can't have run that hot when new.

I've always considered the shield was there to reduce RF radiation from the line stage rather than as a protection against shock. Many shields leave the top caps exposed anyway!

Looking at the extract from the circuit below. HT current flows through the width coil and then through winding b-a on the LOPT. Then, conventionally, through the efficiency diode and then through winding d-e to the LOP valve anode.
Cheers
Nick
Attached Thumbnails
Click image for larger version

Name:	DSCN0678.jpg
Views:	143
Size:	83.3 KB
ID:	164784  

Last edited by 1100 man; 17th Jun 2018 at 7:52 pm. Reason: more text added
1100 man is offline  
Old 17th Jun 2018, 9:31 pm   #6
Techman
Dekatron
 
Techman's Avatar
 
Join Date: Nov 2009
Location: Lincolnshire, UK.
Posts: 4,985
Default Re: Hot LOPT's!

My own personal opinion - all the signs point to their still being dampness within the overwind. It takes years for it to infiltrate the windings with it having the pitch round it from manufacture, and it takes months for it to be driven out. Personally, I wouldn't use any form of permanent varnish sealant on the windings until I was absolutely sure that all traces of absorbed moisture had been driven out. I would then think about applying the new sealant perhaps during a particularly dry week sometime in the middle of summer, and after the set had been running for the previous few hours, after making sure the EHT was still up to spec.

The above is just my personal opinion, remembering the line tranny failures that I sometimes had in sets stored in the back of my old workshop which had no overnight heating back in the late 70s. An added fan is not a bad idea in some respects, but these sets when they were used as everyday sets were sometimes left running all day and half the night with only the normal rate of tranny failure, so shouldn't need any sort of fan.

The screening can is just that, RF screening, certainly not for any sort of shock protection - you only have to look at the old BRC 1400 and where the line output valve was when the chassis was hinged open!
Techman is offline  
Old 18th Jun 2018, 5:24 am   #7
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Hot LOPT's!

Going back to the basics of why a transformer dissipates heat; there are the three losses, I^2R losses (copper losses) and core losses of eddy currents and hysteresis.

Potentially there are dielectric losses too and these would be expected to be very low and primarily influence events during flyback time when the rate of change of voltage is highest, not during active scan time.

So we could deduce from this where the losses are that are heating up your transformer likely are:

Firstly, the bulk of the losses are very unlikely in the overwind - why... because the thermal coupling of the overwind as a physical structure to the transformer body and winding/insulation it sits on, is poor.

Therefore if the overwind was the primary heat source it would be smoking to significantly hotter than the transformer body (except perhaps for a lopty immersed in oil or a thermally conductive potting agent where most of it assumes a uniform temperature). So Loptys with free air overwinds, that are heating up, it is probably not the overwind.

The primary winding I^2R losses are almost certainly responsible.

For loptys with ferrite cores the eddy current and hysteresis loss will likely be very low. laminated iron core ones low losses too, but there will be some heating.

The lopty is a high Q tuned circuit (during flyback only) but in this time the winding currents (so copper loss) are low. And it is a combined test of the losses from a combination of I^2R losses (low in flyback) core losses, dielectric losses, any losses in moisture in the overwind. So to get even half to 2/3 of normal EHT, those losses would have to be relatively low, or the flyback resonance peak would be damped right out.

So in my view, the inescapable conclusion is that the gauge of the wire (on the loptys you are referring to) on the primary is too thin and its I^2R losses that occur during active scan time are those that are generating the bulk of the heat. So I agree With Ed, about the diameter of the copper wire, its simply too small on the primary and the DC resistance too high there.
Argus25 is offline  
Old 18th Jun 2018, 9:50 am   #8
1100 man
Octode
 
1100 man's Avatar
 
Join Date: Apr 2016
Location: Ventnor, Isle of Wight, & Great Dunmow, Essex, UK.
Posts: 1,377
Default Re: Hot LOPT's!

Thanks for your informative reply, Hugo. In this transformer, the 'overwind' is on a separate limb so has very limited thermal coupling with the primary. I suppose the ferrite core will conduct heat to a limited extent.
The overwind (left hand bobbin in the picture), does seem to be generating as much heat as the primary which is what led me to conclude that it could be inter- layer leakage causing the problem?
There is plenty of energy within the transformer as the heater of the U26 EHT rectifier is glowing very brightly.
I had wondered if incorrect line drive could be causing the TX to generate excess heat? However, the width & linearity are good and there are no striations so I decided that probably wasn't the cause either!
The slightly low EHT could also point to excess losses within the overwind I would think. Hmm..

All the best
Nick
Attached Thumbnails
Click image for larger version

Name:	DSCN0580.jpg
Views:	225
Size:	95.0 KB
ID:	164822  
1100 man is offline  
Old 18th Jun 2018, 11:37 pm   #9
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Hot LOPT's!

Nick,

Overwinds tend to be wave wound too, I doubt the overwind is the primary heat source, it would really be smoking if that were the case.

Also, in flyback mode, the current peaks in the overwind, have a very brief duty cycle (maybe less than 10% of scan time) when the EHT diode conducts on the peaks of the high voltage flyback waveform.

If there was excessive resistive or dielectric losses in the overwind, that wouldn't likely happen, the peak voltage would not be there.

If there were energy losses during the flyback voltage rise (rather than just on peaks), in the overwind, due to say moisture I would expect ionization or arcing within it and obvious failure. But also the duty cycle of flyback vs scan time is low, so it would unlikely be the cause of the energy loss (heating).

What I'm trying to suggest is this:

If the overwind can produce a short duration peak of any reasonable amount of EHT, because the duty cycle of that peak is so short, then it is unlikely going to cause any significant amount of energy loss during the bulk of the time, which is active scan time.

During active scan time, the polarity of the voltage across the overwind is reversed and voltage across the overwind much lower than during the flyback peak, perhaps only 1/10 to 1/20 of it, which is where the overwind voltage spends at least 80 to 90% of its duty cycle. Also during scan time, the rate of change of the induced voltage is low compared to flyback time. So, simply, if the overwind supports any reasonable EHT, you can ignore losses in it for the whole of active scan time. Which is the second reason, apart from the thermal coupling issue, why your overwind is not likely the problem. I think it has to be the primary (+ any core losses), because it dissipates energy during active scan time, which is most of the time.

During the scanning cycle the primary current has to magnetically energise the transformer core for the right hand side of the scan and return energy from the core to the power supply for the left hand side of the scan (damped current), so the energy exchanges there are the more lossy ones. The flyback winding just extracts a small amount of energy for a brief period of the cycle.

An infra red photo would be interesting. Ferrite cores do have losses and some can get quite hot under the right circumstances.

Hugo.

Last edited by Argus25; 19th Jun 2018 at 12:04 am.
Argus25 is offline  
Old 19th Jun 2018, 5:51 am   #10
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Hot LOPT's!

A sentence in the above post was worded poorly, end of second to last paragraph, would better read: " The overwind just extracts a small amount of energy for a brief period of the scanning cycle during flyback"

Also you might find out out where most of the heat is coming from by attaching a small bead temp probe to the core and or the primary. You could measure the temp of different parts of it with a point and aim laser guided temp probe, many are quite cheap these days. Obviously don't attach anything to the overwind unless it was turned off.
Argus25 is offline  
Old 20th Jun 2018, 12:45 am   #11
1100 man
Octode
 
1100 man's Avatar
 
Join Date: Apr 2016
Location: Ventnor, Isle of Wight, & Great Dunmow, Essex, UK.
Posts: 1,377
Default Re: Hot LOPT's!

Hugo,
I have a 'point and shoot' temperature measuring device so I left the set on for 50 mins and monitored the temperature of various parts of the LOPT during that time.
The overwind seemed to be rising at exactly the same rate as the primary which seemed strange until I measured the top section of the core. This was continually the hottest part of the transformer. The lower section of core was significantly colder.
The picture below shows the temperatures of different areas after 50 mins. Wax was starting to bubble out of the core of the overwind by this time!
So it would seem that the top core is the bit getting hot and heating up both the primary & overwind.

Looking at the construction of the Tx, there is a metal mounting plate at the top & bottom. It has the usual threaded brass rods to clamp the top & bottom together. So in effect there is a shorted turn around the entire core. Could this be causing a problem?
What other reasons would cause only the top half of the core to generate excess heat?
Many thanks
Nick
Attached Thumbnails
Click image for larger version

Name:	DSCN0679.jpg
Views:	164
Size:	45.4 KB
ID:	164906  
1100 man is offline  
Old 20th Jun 2018, 9:18 am   #12
Hybrid tellies
Nonode
 
Hybrid tellies's Avatar
 
Join Date: Jan 2004
Location: 1966-1976 Coverack in Cornwall and Helston Cornwall. 1976-present Bristol/Bath area.
Posts: 2,965
Default Re: Hot LOPT's!

On my Decca Bradford 10 the LOPT is mounted under the PL509 and PY500A valve platform and just above the tripler but even with this arrangement the LOPT's on this chassis still ran warm and did suffer failures mostly due to shorted turns on the main winding.
This is the naughty bit on my set ever since I have owned it from 1977 the LOPT outer screen shield is missing and I never got around to replacing it. The LOPT is still the original and seems to run cooler than it does in most Bradford chassis with that outer screening shield fitted There does not seem to be any other adverse effects with this screening shield missing but I suspect the radiation on the AM Long and Medium wave bands is higher than it should be and thankfully I don't yet have a heart pacemaker fitted
__________________
Simon
BVWS member
Hybrid tellies is offline  
Old 20th Jun 2018, 9:30 pm   #13
Maarten
Dekatron
 
Join Date: Dec 2007
Location: Haarlem, Netherlands
Posts: 4,184
Default Re: Hot LOPT's!

If the wax was bubbling out, the overwind may have been hotter than your measurements indicate.
Maarten is offline  
Old 20th Jun 2018, 9:58 pm   #14
1100 man
Octode
 
1100 man's Avatar
 
Join Date: Apr 2016
Location: Ventnor, Isle of Wight, & Great Dunmow, Essex, UK.
Posts: 1,377
Default Re: Hot LOPT's!

Hi Maarten,
I measured a surface temperature of 45C but I expect the temperature deep within the winding was much higher than that causing wax to be forced to the surface. Certainly too hot!!

Hybrid tellies,
If it makes you feel any better, I've got a box of screening metalwork from all my TV's!! I've never re-fitted them to any of my sets partly because the transformers get more air, partly because some of them have cooling fans in the way and partly because none of my projects have ever been fully finished... Even ones started 20 years ago In fact most of them get used 'al fresco' rather than being put back in their cabinets! They are all a 'work in progress'!

Cheers
Nick
1100 man is offline  
Old 20th Jun 2018, 10:00 pm   #15
Philips210
Nonode
 
Join Date: May 2007
Location: Redruth, Cornwall, UK.
Posts: 2,562
Default Re: Hot LOPT's!

Hi Nick.

I was wondering if the two cores are the same and the LOPT can be easily dismantled, then you could try swapping the two cores to see if the problem shifts to the other core. Look out for any core spacers though. It's more likely though there's a winding fault.
I also thought about removing the EHT overwind and run the set without it to see if there's still a temperature rise in the LOPT.

Regards
Symon.
Philips210 is offline  
Old 21st Jun 2018, 12:14 am   #16
1100 man
Octode
 
1100 man's Avatar
 
Join Date: Apr 2016
Location: Ventnor, Isle of Wight, & Great Dunmow, Essex, UK.
Posts: 1,377
Default Re: Hot LOPT's!

Hi Symon,
When I took the LOPT apart last year to remove the wax coating and re-varnish it, I couldn't get the core apart. I didn't want to use too much force. So I don't know what's going on where the core ends go into the windings.

I've just partly dismantled it again to the point where the bare transformer can be run outside the set on longer wires. So the tagboard and mounting brackets have been removed and wires attached direct to the wires coming out of the windings. All very fragile!!
At least now I can easily see the top core and measure it's temperature.

It's now been on for 40 mins. Definitely the hottest part of the transformer is the core just where it goes into the overwind, 50C. The other end of the top core is at 35C. The overwind is slightly cooler than the core at 45C. So it still seems like it's the core that is getting too hot.
Further investigation required, I think!! But not tonight.

Cheers
Nick
1100 man is offline  
Old 21st Jun 2018, 1:05 am   #17
Refugee
Dekatron
 
Refugee's Avatar
 
Join Date: Apr 2012
Location: Worksop, Nottinghamshire, UK.
Posts: 5,549
Default Re: Hot LOPT's!

Have you tried to remove the core while it is hot?
If wax is holding it you should be able to get the core out.
Those point and shoot thermometers vary according to the texture of what you are pointing at.
I have also got one. They measure the average temperature of an area that gets larger as you move back from the item you are testing. This is a likely conflict as there is EHT hissing at you while you are measuring the over-wind.
Refugee is online now  
Old 21st Jun 2018, 4:45 am   #18
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Hot LOPT's!

Quote:
Originally Posted by 1100 man View Post
It's now been on for 40 mins. Definitely the hottest part of the transformer is the core just where it goes into the overwind, 50C. The other end of the top core is at 35C. The overwind is slightly cooler than the core at 45C.
The heat always moves from the hotter to the cooler part. So its not looking like the overwind is the main heat source, or the primary for that matter. There may be some heat added from those structures, but it everything is clearly being heated by losses in the core, as is likely some of the primary heat too.

The odd thing is the asymmetry that one half of the core is hotter than the other half and as noted there could be a difference in the ferrite halves. If the gap was greater in the core half where the primary is, the flux leakage could explain some of the measurements, but not the top to bottom asymmetry

Ideally the core would be running cold/cool. The times I have seen ferrites heat like this occur when the flux density repetitively peaks very high for part of the magnetic cycle.

If you could get the core halves apart it might be possible to increase the air gap spacer and not run into any excessive linearity problems. But you don't want to break the halves either. The probability of getting it apart intact will be better if its hot. It is all very interesting. Do you have any other ferrite cores to substitute/try ?

I guess it is possible, that there are some shorted turn in the base of the overwind close to the former it sits on and nearer to one core half than the other, if that were true it would be surprising that its generating a reasonable EHT, but I guess its not impossible.That could possibly explain the asymmetry. Can you point the temp probe at the top and bottom overwind surfaces near the former ? The more I think about this, it would explain the temperature distribution you have measured and the bubbling wax in the overwind too.

Last edited by Argus25; 21st Jun 2018 at 4:54 am.
Argus25 is offline  
Old 21st Jun 2018, 10:31 pm   #19
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Hot LOPT's!

Nick,

If overwind turns were shorted out, the current and energy loss in them, which would largely be generated during scan time, must be below the fusing temperature of the thin wire. Since the turns and generated emf are proportional, then any number of them shorted out would produce a similar current, but more power dissipation for more turns. There could be half a shorted layer near the bobbin.

One thing you could try, find some wire of a similar diameter to the overwind's wire, wrap a loop of it around the former to get the same diameter, solder the ends together, and see if the EHT drops a few kv (or it kills it) and see how hot it gets. If the short (if this is the cause) in is the layers near the bobbin, it may be possible to unwind that bottom layer from the side and release the overwind from its bobbin and repair it.
Argus25 is offline  
Old 22nd Jun 2018, 12:29 am   #20
1100 man
Octode
 
1100 man's Avatar
 
Join Date: Apr 2016
Location: Ventnor, Isle of Wight, & Great Dunmow, Essex, UK.
Posts: 1,377
Default Re: Hot LOPT's!

I've been doing some further investigation this evening. The first conclusion is that my very cheap point & shoot thermometer is not useful for spot temperature measurements. I was suspicious of the results so I have gone back to my trusty 50 year old digital temperature measuring device, namely my finger!!

Probing various parts of the transformer after it had been on for 30 mins with said finger revealed that in fact the hottest part was the overwind. The two places where the wax had erupted can be seen in the picture. These were even hotter than the rest of the overwind. The core was noticeably cooler and the primary was hot but not as hot as the overwind.

When I was doing 'ringing' experiments last year with LOPT's, I put a shorted turn of wire round the winding which killed the ring completely. I had never tried this with a working transformer powered up however.
I put a turn of 1.5sq mm pvc insulated wire round the overwind and soldered the ends together and switched on. The transformer struggled manfully into life but there was barely a glow from the U26 heater and a large, dark picture. After 30 seconds, smoke started to come from the insulation on the shorted turn: I switched it off!!

So I would doubt that there are any shorted turns in the overwind as I'm sure it would melt pretty quick if there were. I still wonder if there might be many small leakages between layers though?
I don't doubt a rewind would cure the problem but a cooling fan is much cheaper!!
This LOPT has spent very many years in very wet conditions- the rest of the set was totally rusty and full of mud!! So it can probably be forgiven for being slightly unwell!

All the best
Nick
Attached Thumbnails
Click image for larger version

Name:	DSCN0683.jpg
Views:	170
Size:	94.1 KB
ID:	165015  
1100 man is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 7:53 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.