Featured
Latest
BRC/Thorn 3000/3500...
 
Share:
Notifications
Clear all

Forum 141

BRC/Thorn 3000/3500 PSU Module Repairs

162 Posts
10 Users
0 Likes
5,832 Views
Red_to_Black
(@red_to_black)
Posts: 1748
Prominent Member Banned
 

Disregard my last comment as I see C615 is  the mustard and is clearly marked on your board, what about temporarily reducing the value of C614 to see if we can get the voltage output under control with W615 disconnected ? 

 
Posted : 28/09/2016 9:37 pm
Norman-Raeburn
(@norman-raeburn)
Posts: 174
Reputable Member Registered
 

Well guys, What can I say, I admire your determination. I would have launched that panel across the workshop days ago. I look forward to the finished article. Keep up the good work, it's certainly entertaining. All the best, Norman

 
Posted : 28/09/2016 10:08 pm
crustytv
(@crustytv)
Posts: 11869
Vrat Founder Admin
Topic starter
 

Hi Norman,

Well I've let off a few expletives during the course of this repair.embarassed

I'm like a dog with a bone on, I won't and can't let go, I must see it through

Today has been the usual steps backwards followed by forwards. This morning the damn thing was back to the crowbar being fired, this with the feedback amp connected. I then thought OK so lets isolate the feedback again by lifting W615 or R621, that worked yesterday, not this time though!

I decided at this point I would just replace everything in sight on the reverse panel. I used 1% and 5% resistors, the six resistors that remain were removed, tested and were found to be on value, I even replaced the Mullard mustard, though its gone into stock. All the transistors except VT609 and VT601 have been replaced.

psu1.jpg

On the other side I replaced the three main electrolytics, C607. C609 & C19. Before anyone screams the red can I've used for C607 is not good, its NOS, tested spot on value, ESR very good and when rated voltage applied for 20 mins showed no leakage and no heating. For testing its perfectly fine.

psu3.jpg

The original C609 which tested good a few days ago, tested open today, likely died with all the powering up and running.

c609.jpg

This evening I have the feedback amp back in circuit, the 58-65V rain is sitting at 71.2V and the only way to achieve this is with R629 ( max clockwise) & R631 (max anti) a few turns back from there on either, fires the crowbar.

psu2.jpg

psu4.jpg

What's really frustrating is I can find a shed load of info about and advice about a low 58-65V rail but nothing regarding a high one.

This particular article on the main site is excellent reading about the function of the PSU. Many tips on servicing and fault conditions not a lot about the situation I face.

CrustyTV Television Shop: Take a virtual tour
Crusty's TV/VCR Collection: View my collection
Crustys Youtube Channel: My stuff
Crusty's 70s Lounge: Take a peek

 
Posted : 29/09/2016 10:03 pm
Red_to_Black
(@red_to_black)
Posts: 1748
Prominent Member Banned
 

Have we got something strange like leakage of the PCB itself, or maybe a duff chopper driver transformer ( I know, I know they are bullet proof ) ?

There is sod all left to check/change, and my suggestion of temporarily subbing C614 for say a 1500pF ?

 
Posted : 29/09/2016 10:28 pm
crustytv
(@crustytv)
Posts: 11869
Vrat Founder Admin
Topic starter
 

Hi R2B,

No harm in checking T602 you're right though, probably very unlikely to be faulty. As for PCB leakage, I would be surprised if its that I've repaired far worse cooked ones ( HMV 2703), the only real damage to this module is lifted tracks on VT605 requiring jumper wires.

The cap idea is as you say is an option worth exploring but not sure how it leads us to the cause, probably just me being thick again. Its late I will pick this up again in the morning.

CrustyTV Television Shop: Take a virtual tour
Crusty's TV/VCR Collection: View my collection
Crustys Youtube Channel: My stuff
Crusty's 70s Lounge: Take a peek

 
Posted : 29/09/2016 11:21 pm
crustytv
(@crustytv)
Posts: 11869
Vrat Founder Admin
Topic starter
 

Red_to_Black said
my suggestion of temporarily subbing C614 for say a 1500pF ?  

As you suggested I replaced C614(2000pF) with a 1500pF. I reset the factory EHT preset (R631) to mid point and the engineer SET EHT (R629) to mid point.

The result? The 58-65V rail comes up squarely on at 60V 

psu5.jpg

Adjusting R629 & R631 I can now swing the rail above and below 60V as you should be able to.

 

The voltages on the transistors are still pretty much what they were before except, the feedback amp has improved. b=33.3V, e=32.6V, c=1.7V. Not too bothered by the base and emitter as they reflect the slightly higher 30V rail, the collector though at only 1.7V is wrong that would normally be 10V.

The Chopper drive is still a problem, base is a little high at 1.6V where it should be 1V, emitter is correct, the collector though is only 2.8V as opposed to where it should be at around 10-11V. Subsequently due to the chopper drive the junction of R607/C608 is still pulled down to 6.3V instead of 12V and is getting very hot. So although the 1500pF has got the 58-65V rail to where it should be, its not a solution and the monostable is still not working. I really don't know what its telling me or what on earth is going on in this PSU anymore..

CrustyTV Television Shop: Take a virtual tour
Crusty's TV/VCR Collection: View my collection
Crustys Youtube Channel: My stuff
Crusty's 70s Lounge: Take a peek

 
Posted : 30/09/2016 10:58 am
Red_to_Black
(@red_to_black)
Posts: 1748
Prominent Member Banned
 

The main idea was to see if the monostable on time could be shortened and the 60V rail brought down to safer levels and also to see if the feedback amp and  loop was capable of actually doing its thing, which it is, now with the psu sort of semi-working as it should you can now investigate in the areas that are not doing what they are supposed to be.

This was always meant to be temporary to enable the supplies to come up in a safer less temperamental way giving you a better opportunity and a starting point to fault find, now we have reduced the mark/space ratio albeit artificially I would have expected the 12V to be more or less correct, I will have a think about this a bit later, also if/when JC sees this he may have some ideas too.

This is actually telling us quite a lot.

 

Edit: with Smps you have a reference Voltage or another way of saying a starting point, now a sample of the output voltage is compared to the reference, if the output sample is below then the PSU ups the output until they match, likewise if it is above the output will lower until again they match, this is the basis of regulation.

 Now what we needed to do was see if the problem is with our reference, or with the feedback, and like all loops we end up going round in circles, the basic idea is to break the loop in a safe (for the psu) controlled manner, and try to make sense of what is or is not happening.

 With this particular psu the feedback had simply bottomed out and ran out of road so to speak and unable to control the output any lower than it was even though it was still too high, this was also inconvenient because it kept crowbarring, also with the feedback amp bottomed out we couldn't tell if it was capable of working, the voltage readings here did not really help either because it could have been something in that part of the circuit that was faulty.

Now we can at least take some dynamic readings with the output in the right ball park and see what is or is not working.

 
Posted : 30/09/2016 2:01 pm
crustytv
(@crustytv)
Posts: 11869
Vrat Founder Admin
Topic starter
 

Excellent Baz thumb_gifI get the gist of what we're up to now, soz for being a bit slow on the uptake. Being at this so long now and what with all the reading until my eyes an brain pop, made me a bit fuddled on clarity.

I'm too intrigued what Mr Thorn (JC) will make of all this too.

CrustyTV Television Shop: Take a virtual tour
Crusty's TV/VCR Collection: View my collection
Crustys Youtube Channel: My stuff
Crusty's 70s Lounge: Take a peek

 
Posted : 30/09/2016 2:44 pm
Red_to_Black
(@red_to_black)
Posts: 1748
Prominent Member Banned
 

I have had a little think (dangerous I know grin_gif), with the psu giving near the correct output I would have expected VT08 Collector to be giving near the book voltage readings, I know our 30V supply is a little bit higher but the voltage difference between emitter and base is almost book values, and about as near as can be expected especially as that part is adjustable.

Now it should follow that as the output is near the book value then the M/S ratio should also be near the book too, and if that is the case then it should also follow that the 12V line via R607 should be also somewhere near the book value too, although I am not expecting that to be exactly 12V.

Hmm hmm_gif need to have a another think.

 
Posted : 30/09/2016 6:37 pm
Jayceebee
(@jayceebee)
Posts: 2014
Prominent Member Registered
 

Hi all, sorry to have been AWOL but had a service call from hell for the last three days. Very early starts and very late nights/early morning finishes, my brain is also mashed and may not be much help at the moment. Anyway back to the important work.

Good thinking with C614 but the drive pulses must be needle thin now. I too would have expected the voltages on VT608 would be somewhere near the manual so that is a surprise. I would like to see what the collector of VT605 looks like and also the DC on the 58-65v rail to see if there is excess ripple, recommend you AC couple for the latter?

John.

John.

 
Posted : 30/09/2016 9:47 pm
crustytv
(@crustytv)
Posts: 11869
Vrat Founder Admin
Topic starter
 

Jayceebee said
I would like to see what the collector of VT605 looks like. 

vt605col.jpg

Jayceebee said
also the DC on the 58-65v rail to see if there is excess ripple, recommend you AC couple for the latter?

John.  

 60VrailDC.jpg

CrustyTV Television Shop: Take a virtual tour
Crusty's TV/VCR Collection: View my collection
Crustys Youtube Channel: My stuff
Crusty's 70s Lounge: Take a peek

 
Posted : 30/09/2016 10:06 pm
Red_to_Black
(@red_to_black)
Posts: 1748
Prominent Member Banned
 

Now that is a bit odd, the M/S ratio does not look really changed much at all, the on time still looks the same on the C of VT605, just the peak looks to be a bit lower, it still looks as though the monostable itself is not working correctly, which seems to be confirmed by the low 12V line via  R607.

Also what should be your 10V feedback line across R637 seems to be being possibly loaded down by the monostable stage. something seems awry around that part, now I know all the parts have been replaced here, but something else must be amiss.

RE. the voltage across R607, with the output artificially lowered. the voltages around the transistor VT608 E and B are now not far from the manual values, all else being equal the voltage around the 10k should also be similar to the manual, unless it is being affected by the monostable stage.

 
Posted : 30/09/2016 11:10 pm
crustytv
(@crustytv)
Posts: 11869
Vrat Founder Admin
Topic starter
 

The Troublesome Final PSU, the end of this saga & what a surprise!

After a restless night with the 3K power supply circuit rattling around my head, I awoke late for me. A swift cuppa and then back into the workshop for another day's battle with the PSU from hell. Today was a full of doubt day, one where I was convinced it must be either an own goal or I've replicated a man made error that was already present.

Painstakingly I went over every change I had made, every component I had introduced. Were the diodes orientated correctly, were the transistors installed the correct types and were the pin-outs mapped accordingly, nothing was wrong. I then compared the circuits to the physical taking note of any variances, again nothing that would cause the fault condition this PSU was exhibiting.

Around three O'clock with virtually nothing left in the bag of ideas I was scoping the anode of W615, it looked so wrong and I was starting to embark on yet another rabbit hole when the door bell rang. Who could that be disturbing my train of thought? Off I scuttled to the door and to my surprise it was Dr. Thorn, none other than our very own John (Jayceebee). I've never been more happy to see anyone and my pained expression must have said it all. I think days of this PSU had aged me as I proclaimed, "John great to see you, come on in" dragging him off to the workshop. I explained I had spent the morning and early afternoon beating a rather nice dent in the bench with my head, John pulled up a chair "Right lets be having a look at this".

Over the next couple of hours he went over the entire board lifting components, taking meter readings and every now and then going, "hmmmm, odd, very odd". Out came the scope, waveforms checked followed by further head scratching. Now at this point I was a little relieved as I fully expected John to walk in sit down and within a short period of time either find the fault or point out where I had introduced one or two of my own. As this had not happened my misery loved to find some company but what was this damned elusive fault?

As a last ditch desperate measure T602, the chopper drive transformer came under the glare of our suspicion, I reminded John that R2B had mentioned this part the other night but I had stated, "They never go bad". It was removed but when it was ohm'd out against a stock item, it matched. We replaced it anyway just to be sure, it didn't change a thing.

At this point John looked at me and said, "you're not going to like this but can we please try this module in a set". The disbelief on my face must have said it all, John asked what troubled me about that. I explained two things, one that the PSU has to have 1500pF for C614 to stop the crowbar from firing, that the collector volts on the feedback were so far off book. Two without C614 changed the SET EHT presets have to be wound to their maximum opposites and even then it does not always stop the crowbar from firing.

John explained that he believed I had fixed the original fault days ago and there was in fact nothing wrong with the PSU, it just didn't like the jig! I asked how could that be I'd fixed seven dead thorn PSU's in the jig, five of them this year, they all ran happily, why would this one decide it was not going to work in the jig.

John said he was convinced it was the case, I however remained unconvinced. The spatch-cocked PSU was duly put back together and C614 was reverted to its original value of 2000pF, off we went to the TV room where all the TV sets reside. I decided to use the HMV 2703, its PSU module was removed and this troublesome PSU installed in its place. What happened next?.............

Nothing untoward! No tripping, the set came up the rail was checked at 58V, it was adjusted to 60V, all was good........fright

The PSU was removed put back on the jig and it tripped, the crowbar firing..... So folks for some unknown reason one PSU out of eight I've repaired on this jig was having none of it. Why this is the case I have no understanding of, neither does John. The only difference being John had the foresight to not believe what he was seeing, I however did not hence the days of torture.

Earlier in the afternoon in relation the T602 and long before this had been ascertained, I mentioned a phrase of Sherlock Holmes, "Once you eliminate the impossible, whatever remains, no matter how improbable, must be the truth." Something I will remember and when put through a trial such as this PSU has given me, one lesson I'm likely never to forget.

Ivor (itmog57) pointed me an LLJ article today where Les talked about doing battle with a Thorn 3K PSU in a customers home. How it defied logic, sapped his usual diagnostic powers and how he rushed off home to tell "Honey Bunch" how awful it had been to him. Only now do I understand!

CrustyTV Television Shop: Take a virtual tour
Crusty's TV/VCR Collection: View my collection
Crustys Youtube Channel: My stuff
Crusty's 70s Lounge: Take a peek

 
Posted : 01/10/2016 9:43 pm
Cathovisor
(@cathovisor)
Posts: 6390
Famed Member Registered
 

Excellent 🙂

I refer the honourable gentleman to post #103.

 
Posted : 01/10/2016 10:05 pm
Red_to_Black
(@red_to_black)
Posts: 1748
Prominent Member Banned
 

Result !cheers

JC did mention to try it in a set a few days ago.

Is anything physically different about that PSU ? like components omitted/added or changed compared to the other versions, Thorn could be sods for making slight hidden changes.

I must say I was beginning to wonder if a fault had developed with your jig in between you repairing the last one and this one, speaking of which have you tried the jig since on a previously known working board that worked on the jig ?

I bet you are glad that one is finished!

Well done to both of you cheersbirthday

 

Oh and thanks for a most interesting thread. electro_gif

 
Posted : 01/10/2016 10:08 pm
crustytv
(@crustytv)
Posts: 11869
Vrat Founder Admin
Topic starter
 

Red_to_Black said

I must say I was beginning to wonder if a fault had developed with your jig in between you repairing the last one and this one, speaking of which have you tried the jig since on a previously known working board that worked on the jig ?

OH yes I've tried four other PSU modules in the jig this evening, they all work OK, the jig is working fine, nothing wrong. It just refuses to allow this one last PSU to work. I too am intrigued as to why.

I will continue to work with the jig as it affords me the luxury of working on a PSU and LTB at the bench. Anyone who knows these modules, knows you cannot work on them in the manner you need to in the set. Back in the day field engineers would just replace a few easy to get at components, if that did not fix the fault they would panel swap leaving the bench tech to Jig repair the faulty module.

I've actually had a plan but just never got around to it of to building a full Thorn 3k test rig. I have a spare 3500 chassis and CRT, I've been toying with the idea of removing the harness and building a bench mounted test rig of my own, a sort of fully spatch-cocked 3K affording easy access to all panels and with the CRT facing forwards.

Until then the mini jig will do, if I get spurious results that cannot be explained, I will ensure the PSU is adequately safe and test in a set. 

CrustyTV Television Shop: Take a virtual tour
Crusty's TV/VCR Collection: View my collection
Crustys Youtube Channel: My stuff
Crusty's 70s Lounge: Take a peek

 
Posted : 01/10/2016 10:31 pm
Red_to_Black
(@red_to_black)
Posts: 1748
Prominent Member Banned
 

Sounds like a plan, one of the shops where I worked had a G11 jig similar to what you propose, it was made up of a less than perfect tube and a channel selector where only a few buttons worked, ie. a bitsa made up of all the grotty bits but it served its purpose well.

having said that the G11 PSU could be ran on the bench with just a temporary mains lead soldered to it and a 60W lamp soldered on the output, you just had to careful where you put your fingers.electro_gif

 
Posted : 01/10/2016 10:41 pm
sideband
(@sideband)
Posts: 4215
Famed Member Moderator
 

I never worked on Thorn sets to any degree...certainly never any of the colour sets as I was at Philips by then. It's a pity I didn't say something a week or so ago because I had a nagging doubt about this PSU not working in the test jig....!

I've come across this sort of thing a few times in the past where, under normal circumstances you can load up a particular power supply with a dummy load and everything works fine. Then there is one that refuses to work under any circumstances but you check it in the set and all is well.......steamy_gif

 

Well done for sorting it out....a more restful night tonight!

 
Posted : 01/10/2016 11:22 pm
Jayceebee
(@jayceebee)
Posts: 2014
Prominent Member Registered
 

Well, we got there in the end. Quite why it doesn't like the jig I don't know but it does have it's place, as Chris rightly says servicing a 3000 psu in a set is fiddly and time consuming. I've always been wary in putting my ultimate trust in so called servicing "Aids" even simple diagnostic flashing LEDs or complex stuff such as Philips "Compair".

I'm pleased I managed to convince you to try it in a set, I could certainly detect your unease but these psu's never failed catastrophically as did some parallel chopper circuits designs used by some manufacturers. Failure of the small value resistor current sensing resistor in the emitter in parallel types often caused total carnage.

John.

John.

 
Posted : 01/10/2016 11:25 pm
Brian Cuff
(@briancuff)
Posts: 2063
Member Rest in Peace
 

Well Chris, top marks for tenacity! It's a pity that it turned out that you had unknowingly repaired the PSU earlier in the saga but the thread really did make for interesting reading. I have had little to do with solid state stuff at my end of the hobby but that's what's great about it - Radio/TV covers a huge range of technologies from spark to too much compression so there are bits of that enormous spectrum which will appeal to different individuals.

Well done. thumb_gifwelld_gif

Forum Memorial

 
Posted : 02/10/2016 8:02 am
Page 8 / 9
Share: