BRC/Thorn 3500 Chassis
Model: Marconiphone 4711
Original List Price : £000.00 ( to be ascertained)
CRT: A67 120X
The Actual Set
I bought this from the family of the original owners who got it in 1973. It was serviced by the local dealer, their stamp visible on the back of the set. Pretty much untouched since it was retired but sat in the owners lounge all these years, inside its has a thick layer of dust, perfect. The pictures don’t quite relay how big this TV is. It is in fact as big as the 1967 Decca CTV25.
Mazda A67 120X Health Check
As is customary before I do anything else and on which hinges whether the set will be repaired, “is the CRT OK with regards to shorts, filaments and emissions”.
This A67 120X should give a good account of itself with the Red and Blue guns giving a good reading. Green is slightly lower than the other two but may pick up with use.
I’ve had much worse readings on other tubes and those CRT’s have gone on to give excellent results. I always remember the RCA tube in the Baird M702, that had lower readings on all three guns than this tubes green. The Baird went on to give the best picture I’ve seen on a vintage early colour television. With this tube and the excellent 3500 chassis, I expect the results to be at least very good.
That’s it for now on this set until I get the HMV 2703 PSU sorted and off the bench, then this one will get its turn.
powered the PSU up and the rails came up 240, 6.3 & 30V however the 58-65V was missing.
F602 is intact and a check confirms no voltage. R607 is as I’ve encountered on other modules, low at 4.7V likely caused by the Chopper drive transistor VT605 being switched hard-on.
W605 voltage is spot on as the 30V rail is seen on the indicator light so I’m suspecting VT602/VT603 the delay switch and/or mono-stable. I also want to check C609 as I don’t want it taking out the currently working zener.
Time to * spatch-cock……. back soon.
* spatch-cocking the PSU: This is the only way to properly service these modules. It entails opening up the PCB panels and mounting the whole unit in the jig. This way voltage readings, scoping and component replacement is made simple. Picture of this to follow shortly.
Well looks like the gods of the 3K PSU module are with me, I’ve just had my third easy fix. The collector volts of the Chopper transistor were missing from point J. Tracing this back I find a dry joint on the collector of VT604, cleaned and re-soldered I now have all the rails present. This might be two attributable causes to the TV’s retirement, it remains to be seen if there are any other faults lurking on the other panels.
Thunderbirds Are Go!
I will test a few caps that I mentioned and if they check out OK, will leave them and that includes the low voltage Callins. I will also tidy up a couple of components connections. I will replace R620 as a matter of course as that can cause all sorts of funnies.
So far one component replaced, the multi-can
Remove the Line time-base module from the TV and power that up via the jig. Check the rails on that and hopefully have the 8kV pulse on the EHT TX nipple.
C624 the callins cap was leaky so will be replaced, yes an in-circuit test is valid on this cap and confirmed against a stock module.
The line time-base module was removed from the TV, again the Easter dust bunnies have been busy, yes there is a time-base under there!
Next step: Hook up to the PSU module via the jig. And yes I’m going to replace C518(.028uF 800VDC) the Green flyback tuning cap with one of the ones I sourced last year when repairing the Ultra 6714 A .027µF 800VDC Polyprop
There’s a fault on the 4711’s line time-base module.
When the LTB module is hooked up to the now working PSU, R609 (15R) begins dumping the magic smoke and the jig trips.
Remove the LTB module, simulate the line-pulse from -ve W608 and all the rails establish. To absolutely be certain I hooked up a working LTB from stock, connected it to the 5711’s PSU all was well so the LTB module has a problem.
Not come across this scenario before, now to investigate where the fault lies on the 4711’s LTB, something shorted perhaps?
Well its getting late so I think I’m calling time on this fault.
I checked W504, W507, replaced R528 it was o/c, replaced C518 the tuning cap, still no change, it trips. My gut was telling me to check the line output transistor so that was removed and tested, it was OK. However the Mica washer had seen better days. So in with a new one and some fresh compound.
Connected it all back, powered up and yay!!!!! the LTB was running, no tipping and all the rails were present, happy days!
Powered off and set about hooking up the scope to the line driver transistor so as to check the waveform against book. Upon powering up and the flamin thing trips again as before!
At this point I’m tired, cannot think and can hear the ghostly chuckle of LLJ telling “honeybunch” to come look at this poor sod playing with a 3.5K for fun!
Back at this tomorrow, I can see my night time dreams will be filled with LTB modules going round and round and round my head.
Just to be certain of the events last night, this morning I powered up the PSU with the LTB connected, yes they both power up. Power off, power back up, it trips.
Just to be sure no one thinks this is a jig problem, if I connected another working LTB to the PSU it stays up and can be power cycled as many times as you like. Therefore there is an intermittent fault condition on the 4711 LTB.
So as Andrew and I had both started to think caps, I looked at C525, C526, C505 and C511. The first two were completely o/c the latter two were double in value.
Replaced them, no change.
The hunt continues…………….
The two on the left ( C525, C526 o/c) the two on the right (C506. C511 gone high)
The LTB is working we have 8kV pulse and scope shows the oscillator running.
Just to recap, when the 4711 Line time-base module is powered via the 4711 PSU on the Jig, intermittently (more times than not) it will trigger the crowbar causing the unit to operate the cut-out. If I disconnect the 4711 Line time-base from the PSU, the 4711 PSU just runs and runs without fault, the crowbar does not fire.
As you recall from yesterday I tried another spare line time-base module running from the 4711 PSU. This other line time-base runs perfectly fine. This lead me to believe there was an intermittent fault, perhaps a cap breaking down on the 4711 line time-base. However after checking likely suspects I could see I was getting nowhere fast.
Had a break from it early this afternoon and pondered, I thought what I have not tried is another spare PSU on the jig powering the 4711 line time-base. If there is truly a fault on the LTB then it should fire the crowbar, well no matter how many times I power cycled it didn’t, it stayed up, rock solid.
The rabbit hole has been fully navigated.
It would appear when the 4711 PSU is partnered with the 4711 line time-base and powered via the jig, the 4711 PSU intermittently objects.
The 4711 PSU and line time-base modules have now both been independently tested and work. They are both ready for inclusion back in the TV.
5 components thus far; PSU multi-can LTB: C525,C526 (o/c and replaced), C506,C511 both high and replaced.
A critical check on the I.F. panel, C132. If this goes s/c then it takes out the 30V supply on the PSU. I will also remove the video panel to see if there’s any damage from the leaked multi-can which sat right above.
The boards on this set are barely visible.
Removed the video to inspect for electrolyte damage from the can above, thankfully there’s none.
Video Board Before
Video Board After
Next, check that cap I mentioned above. I might also remove the chroma board and Sound/Frame board to get the worst of this crud off before powering up.
C.B.D = Chroma Before Dust
C.A.D. = Chroma After Dust
The I.F. panel being vertically mounted was less dust ridden, looks like C132 has already been changed from a callins in the dim distant past.
No leakage at rated or +30%
Mother Hubbards Chassis Cupboard looking bare
Next stop, all boards back in and then power the set up on full mains. Its a non event at the present. All the panels refitted, all connections made and checked, fitted in the in-line meter to monitor EHT, hooked up the Test Card generator and powered on……… A big fat nothing!!!
its the cut-out that’s o/c. I will fit a new one but that’s a job for tomorrow as I will need to relocate a solder station to the display room. I’m tired, its getting late and I’ve been at this set on and off for most of the day. Hopefully tomorrow there will be light on the tube and we can see what state the rest of the set is in. Lucky time is not money anymore, I would be out of business before I even started.
Replaced the o/c cut-out now we have power into the set.
Another non event. Crt heater glows, 58-65V rail sits at around 59.7V, no sound, no raster.
EHT meter confirms no EHT.
Disconnected tripler and I could draw a spark from the EHT TX nipple so the 8kV pulse is still present. Wondering if the tripler is toast.
Also noticed a whiff of the magic smoke escaping from the LTB, though unable to see from where at present.
Just did another check and now the EHT TX is not producing the pulse. Back to the workshop PCB bench. This is going to be one of those sets, I can tell.
Time to get methodical about this but at the same time speed things up a little. Suspicion and nagging thoughts of the past few days. I started to suspect all was still not well with the 4711 PSU and LTB, despite indications to the contrary. As you will recall when they were partnered on the jig they were not happy to co-exist.
So this is what happened next:-
- TV was run with the 4711 PSU but with the 4711 LTB swapped out. Then Installed a known working spare LTB, still nothing.
- Next I tried a known working PSU and reinstalled the 4711 LTB, still nothing.
- Finally I removed both the 4711 PSU and the 4711 LTB, then installed the two known working spare PSU & LTB, life at last.
As you can see below, with the two spare modules installed I was rewarded with EHT( running at 24kV) but I have a frame collapse and no sound. Hard to photo and the line is not as bright as the photo. I’m not surprised by the frame collapse, half expected as much especially as it appears to be the original board untouched ( transit screws still present) and it has a few of those large Callins.
But WTF! with the 4711 PSU and LTB.
The 4711 PSU checked out OK on the jig with all rails present. The LTB checked out OK when powered by another PSU on the jig, all waveforms correct and 8kV pulse present. But clearly both are incapable of running from the tests above.
Back to the bench with these two and some………
TV running with replacement PSU and LTB modules.
But I’ve done nothing!
Some steps forward, the set is now working with the 4711 PSU quite why its decided to start playing ball, your guess is as good as mine.
The LTB is still a donor module.
The frame collapse has also cleared though I still don’t have sound.
I cannot tune in to the test card generator, looking at the noise, its out of focus.
Next I will try the 4711 LTB, I’m fully expecting that to not work.
All very odd.
Still running with the 4711 PSU and donor LTB. The 4711 PSU seems to be behaving itself.
I now have the line freq correct and tuned in the test card. The raster is decidedly purple.
Back to the 4711 line time-base module.
I’m starting to wonder if the EHT transformer has expired or has a wiring problem. Running it on the jig with a working PSU, I have the 30V rail, 60V rail and the oscillator is running but there’s no 8kV pulse on the EHT transformer nipple. I’ll do a thorough check of EHT T/X connections.
With regards to the 4711 LTB under test, I think I can discount the EHT transformer from suspicion.
All the wiring was checked to and from the transformer, it was fine. At the weekend I had removed VT504 to test and it was OK, yesterday I decided to connect the CPV unit and AVO to monitor VT504 collector volts. Whoa!! way down at around 50V, that may explain why the 8kV is missing.
Now this is where it gets a bit vague for me, I’m not sure what voltage I should have when the modules are running via the jig and not under load. What I do know is that to correctly set up collector peak volts on a live running TV set, with the CPV unit and the AVO, you should be aiming for 490V.
At this point I decided to remove the working line time-base module that I had temporarily installed in the Marconi and check what its CPV reads. As this LTB is working and producing the required EHT, It should then provide me with a clear indicator of what VT504 CPV should be producing on the jig. No surprise it read higher at around 200V.
That was yesterday, armed with the previous days findings this morning I put the 4711 back on the jig with the aim of trying to locate why the collector volts were low. To my surprise the 4711 VT504 CPV was now at 200V not the 50V I had yesterday! Further checking at the EHT TX nipple and I can now draw the 8kV spark, so its working.
I’m leaving it a while, disconnecting from the PSU and jig then will check again later today. Something would appear to be intermittently causing this fault and its going to be a pig to locate.
4711 line Time-base, CPV showing at 50V, 30V rail, 58-65V rail present, Osc running
Today CPV at 200V
Update: Progress & Set-Back
As mentioned above, yesterday the 4711 line time-base module had been tested and run for a few hours on the jig, VT504 CPV was stable at the 200V unloaded state. Time to move this into the Marconi.
Today I managed to have the Marconi running with its own PSU and own line time-base modules, hooray! EHT was respectable at about 22kV (I have the EHT modestly set for the time being) and the sickly lack of green test card was presented on screen, all good thus far. Whilst considering my next move which was to tackle the likely candidates for the no green fault (C224, C803, VT211/2 or W206), I turned my back to the screen, upon returning only to see the three gun dots fading. A glance up at the EHT meter confirming EHT dropping away along with my prior optimism of making progress.
This latest treat was traced to F603 having blown, stuck another one in and that blew straight away. OK so now we have another fault. With the H.T fuse popping the likely cause is the line time-base,”no s***t Sherlock” I hear you all cry. Reading service data suggests the likely candidates for this scenario being the line drive transistor VT503 or the CRT first anode reservoir cap C523. If its the latter then in all likelihood W505 will require replacing too.
That’s where I’m at, more to follow……….
Time to get at the components on the LTB, this involved a little disassembly of the module.
It was a 50/50 chance as to who the culprit might be for F603 blowing, I removed the line driver transistor, assuming it would be the weakest link, no it was OK.
OK then it must be C523 the .022uF 1.5kV cap then. Indeed it was s/c
Did it take out W505? No its OK and a rather robust fella compared to what was fitted as original. Must have been changed in the past.
So in with a new cap, refit the transistor and diode then see if I can get a little further on this time, maybe tackle the missing green. Or will something else go pop?……………… More later.
Back in business, running on the original 4711 PSU and LTB. I’ll leave it running a little while to see if any gremlins rear their heads. If it seems to be behaving itself I will then look into the missing green. Oh and not forgetting there’s still no sound output but that’s been the least of my worries with this set.
Looking at the time I think this repair will continue tomorrow
Finally running on its own PSU & LTB
No green is not back its intermittent. Not sure what the cause is yet, It might be the components I listed earlier but I have a suspicion about the convergence panel. I might be mistaken but I’m sure the green went when I hinged the panel up,I cant recreate it so I might be wrong. Any way good news is the set is still up and running but I’m not counting my chickens just yet, I think this set still wants a fight.
That is it for tonight, more to follow………….
Good opportunity to clean of the crud and clean up the edge contacts.
Two of the A1 switches have been removed and bridged, one rather poorly. These will have new switches installed. I note the uprated mods to R751 and R752 have been carried out at some point in the past so no need to address that. I will replace C765, C766 & C767 and R790. The latter should be a 1.2M but has been replaced with a couple in series.
Finally not forgetting the Green A1 pot which is under suspicion for the intermittent no Green fault.
More to follow……………..
Replaced the components mentioned earlier and before checking the A1 pot I put the panel back in the set. With the new A1 switches I turned off Red and Blue. Green is there but very low (photo below makes it look brighter than it is), the A1 pot has no effect, whereas when the red and blue are selected their A1 pots have full low to high swing.
At this very point the tuner channel decided to pop out of its mechanical lock and refused along with any of the others to lock. Any of you familiar with repair old thorn sets will be familiar with the dreaded Thorn tuner rot!!. Upon removing the tuner all the little plastic bits came tumbling out.
So as if I didn’t have enough on my plate with this set before I can go any further with the green fault, I’ve now got to repair the tuner so as to get a signal
Tuner dismantled ready for repair.
Repair completed, cleaned and a little clock oil added to the threads. M6 cup washers installed to replaced rotted Thorn plastic ones. Now to reassemble and get back to the colour problem.
Ever felt like you’re going around in circles……Test Card circles?
Tuner is now working once again and I’m back to where I was a few hours ago.
OK on with the colour problem.
Well although the Green A1 pot was erratic changing it made no difference. I measured the A1’s at the CRT base Red=119V, Green=115V and Blue=122V.
Now I have fitted the new A1 switches I can see what’s going on.
Switching off Blue and Green A1’s at the convergence, leaving just the red gun
Switching off Red and Green A1’s at the convergence, leaving just the blue gun
Switching off Red and Blue A1’s at the convergence, leaving just green.
Yikes, that’s the best I can photo it, very dim.
Remember also that when I dabbed Reds output onto the Green video board channel red was bright.
Green is down but but to be honest even so I expected better than I’m getting even at that level.
I’ll give it a clean and balance and if that does not resolve it a bop.
Clean and balance did very little to improve so I did a rejuvenate. This has now restored green to a decent emission. I’ll monitor for 15-20mins if its sits happy where it is now, I will try the set again. If it drops off, i’ll give it another try.
Well its seems stable…………….
Green is back and we have grey-scale. However this set really hates me! Whilst in the process of sorting this CRT problem the colour has now completely gone.
So……… next step? Disable the colour killer and see what’s decided to put its legs in the air this time. I can see this is going to be a looooooooooooooooooong repair.
Disabling the colour killer ( clip in an 82KΩ at junction of C323/24 to ground) reveals the reference oscillator free running.
I’m done tussling with this set and calling it quits for tonight. Its been a long day the TV has thrown a fair few tantrums at being woken up. Tomorrow I will start work on the decoder.
And yes I agree, there’s nothing wrong with red and blue that warrants a clean and balance so there wasn’t any danger of me even thinking about following that advice.
Before doing anything I powered the set on with the colour killer disabled and was greeted with colour lock. Took a photo to document and in the time it took to take that photo the oscillator had drifted and I had baber-pole. I also noted that the 7.8kHZ ident is most definitely off as evidenced by the extreme left side of the test card.
Not a lot wrong with this now. It could be as simple as the crystal but I’ve yet to find a case for me where that is true though there’s always a first time so I will stick a new one in just to save a lot of effort. If I still cannot adjust to lock then its likely those creaking old components on the decoder, then Its scope time as it could be a fault in the burst gate amplifier circuit VT301 and VT302 or in the phase discriminator W302/W303 or the polarity splitter VT304, which supplies the gating pulse to switch on the burst gate amplifiers.
Colour lock before drift off
More to follow……………..
Before diving in and wasting my time swapping the crystal when my gut is screaming “its not the crystal !!! ” I decide to first check with the scope, the burst gate (collector of VT302). Lucky I did as it reveals although the burst signal is present, its a weak one at about 2.5V.
At that level I guess there’s no surprise the reference oscillator cannot lock to the transmitted burst. Whats nagging me though is it does for about 40 seconds at initial power on, although subsequent power cycles always produces no lock.
I did try to adjust R354 to widen the gating pulse but its at its max already, turning the other way just reduces the burst even further. Time to get the meter out and take a voltage map around VT301, VT302 also while I’m at it VT304, W302/3 hopefully I’ll then have a clearer picture of what’s going on in there.
Decoder voltage map
() = expected
|VT301||Burst Gate & Amp||.968||5.07 (7.8)||27.5 (29)|
|VT302||Burst Gate & Amp||27.5 (29)||27.6 (29)||.005|
|VT303||D.C. Amp||2.4||1.8 (1.1)||1.9 (4)|
|VT304||Reference Oscillator||12.2||11.7 (10.5)||18.7 (20)|
|VT305||Emitter Follower||11.2||10.6 (11)||20.2 (18)|
|VT306||7.8kHz Amp||2.9||2.3 (2.5)||20.2 (19)|
|VT307||Switched Amp||25.6||26.3 (27)||21 (17)|
|VT308||Polarity Splitter||.971||1 (1.35)||26.9 (27.5)|
|VT309||Chrominance Amp||2.7||2 (min 1/max 3.7)||24.5 (min 24/max 28)|
|VT310||Delay Line Driver||12.5||11.9 (12.5)||23.2 (24.5)|
Well the only things to me that look wrong are VT301/2 (though nothing too drastic) and the DC amp but mainly the DC amp.
The amplified burst is compared with the reference oscillator signal in phase discriminator L301-W302-W303 and the appropriate correction is applied to oscillator via the DC Amplifier VT303. If VT303 is not working properly then correction will not be occurring, so is this the cause of the inability to lock?
I will check those diodes too and look at the gating pulses
This set is starting to annoy me now, the EHT just disappeared. A check on PSU and the 58-65V rail has gone. Removed the module, checked it on the jig and its confirmed, the PSU has now developed a further fault. Back to square one!
I think I can see why this set was laid up.
The PSU fault is again intermittent, just went to the bench to start diagnosing it, powered up on the jig, it works, all rails present. To confirm I wasn’t going bonkers I put it back in the set, it runs happily for a while then goes off. Is it the LTB PSU love hate affair returned?
Tried the set with a spare PSU nothing! Tried it with spare LTB, it works.The unlocked colour still persists with these new modules just in case anyone felt the original LTB might be related.
Finally just for the hell of it, in with a spare decoder to prove the point and we have locked colour. Of course the resulting picture is not perfect as the set has not had any set ups done. One thing I do notice and whether it can be addressed when the Marconi’s own panels are sorted, there would appear to be green bleed. Quite noticeable on bubbles green shirt, considering the problems I had with the tube, I’m not at all surprised.
I’m going to take a break from this colour fault and go back to the PSU and LTB. This time they are going to get a full service rather than a do just what’s necessary.
As Arnie says “I’ll be back”