UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio 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 12th Sep 2019, 4:50 pm   #1
peter_scott
Nonode
 
peter_scott's Avatar
 
Join Date: May 2003
Location: Edinburgh, UK.
Posts: 2,345
Default Yet Another Bush TV22

I was given this set by a friend back in the early 1970s and it worked well for quite a number of years but back in 2014 the line linearity started to go wrong. In fact that wasn't the only thing to fail. It also developed a heater cathode short but AlistairD on the forum very kindly helped out with a nice isolating transformer for that but I still struggled with the linearity, even hand rewinding the linearity transformer and getting the LOPT primaries rewound, all to no avail. At this stage I got fed up and put it aside but reading recently about running current through the overwind to dry it out I thought I'd give that a try.

The set has been in storage in a warm room so I guess I shouldn't have been surprised at the lack of improvement after heating the LOPT for a few days.

When I last ran the set I could get a picture with distorted line scan but on revisiting it I could only get about 2kV EHT that disappeared if I turned the brightness above more than a faint shadow. Discussing it with Jac Janssen he suggested that I try generating the EHT from a voltage doubler off the LOP anode. I tried this and got 3.7kV and a bit more brightness. Jac then suggested that I try removing part of the overwind just in case the shorted turn(s) were near the surface. After a few days soaking in a bath of paraffin I got the pitch off the winding and started carefully unwinding it hoping to find a shorted turn. Unfortunately this was not the case but I continued to use VRat Chris's test of feeding the scope calibration output to the LOP anode end of the overwind whilst viewing the trace on the other end. I did this continuously whilst removing more and more of the winding in the hope that I would see a sudden jump in Q factor if I got to a shorted turn.

No luck! So I now had a LOPT with no overwind and no EHT rectifier heater winding so I connected it back into the set with a silicon voltage doubler and now have a set with a future again.

Lots my nasty temporary fixes (that didn't work) still to sort out and what with a CRT heater isolation transformer to overcome the HKS I guess it might not have quite the video bandwidth that it had when new but we'll see.

Very many thanks to Jac for his suggestions and encouragement.

Peter
Attached Thumbnails
Click image for larger version

Name:	P1040517 (Medium) (2).jpg
Views:	298
Size:	102.4 KB
ID:	190037   Click image for larger version

Name:	IMG_20190912_164110 (Medium).jpg
Views:	319
Size:	83.5 KB
ID:	190039   Click image for larger version

Name:	IMG_20190912_143945 (Medium).jpg
Views:	343
Size:	76.4 KB
ID:	190040  

Last edited by peter_scott; 12th Sep 2019 at 5:18 pm.
peter_scott is offline  
Old 12th Sep 2019, 5:42 pm   #2
peter_scott
Nonode
 
peter_scott's Avatar
 
Join Date: May 2003
Location: Edinburgh, UK.
Posts: 2,345
Default Re: Yet Another Bush TV22

I forgot to mention that with the LOPT overwind removed the EHT from the doubler has risen to 6.5kV.

Peter
peter_scott is offline  
Old 12th Sep 2019, 8:07 pm   #3
FERNSEH
Dekatron
 
FERNSEH's Avatar
 
Join Date: Mar 2005
Location: Gateshead, Tyne and Wear, UK.
Posts: 6,101
Default Re: Yet Another Bush TV22

Hi Peter,
That's an excellent result. With regard to the reduced bandwidth caused by the CRT isolation transformer, why not try inserting a cathode follower between the video amplifier and the cathode of the CRT? A triode connected EF91 or EF80 can be used.

Adjustment of the line drive trimmer might effect a further improvement to the performance of the line output stage.

DFWB.
FERNSEH is offline  
Old 12th Sep 2019, 8:25 pm   #4
peter_scott
Nonode
 
peter_scott's Avatar
 
Join Date: May 2003
Location: Edinburgh, UK.
Posts: 2,345
Default Re: Yet Another Bush TV22

Hi David, Thanks for the bandwidth suggestions. I'm not sure I could find a spare valve in the current chassis but perhaps I could hide a transistor somewhere.

I've got a fair bit of tidying up from my previous messing with the set but I'll post an update if I make much progress on the bandwidth.

Thanks,

Peter
peter_scott is offline  
Old 12th Sep 2019, 8:46 pm   #5
FERNSEH
Dekatron
 
FERNSEH's Avatar
 
Join Date: Mar 2005
Location: Gateshead, Tyne and Wear, UK.
Posts: 6,101
Default Re: Yet Another Bush TV22

Hi Peter,
If your RF deck is equipped with EF80 valves then the video amplifier valve can be replaced with a PCF80 triode pentode. With EF91 RF decks a BF337 transistor could be used instead. Easy to hide the emitter follower transistor among the under chassis wiring.

DFWB.
FERNSEH is offline  
Old 12th Sep 2019, 9:01 pm   #6
beery
Heptode
 
beery's Avatar
 
Join Date: Jun 2006
Location: Ware, Hertfordshire, UK.
Posts: 734
Default Re: Yet Another Bush TV22

Hi Peter,
Oh dear,
how can you put a trippler in such a rare set etc. etc. I mean they hardly ever come up for sale

Before you removed the overwind, did you rule out EHT leakage from the heater winding of the EY51 (or a gassey EY51) by using a stick diode? The heater winding has been known to leak through the paxolin to the transformer's core. Later TV24 sets (where the EHT is arranged to be slightly higher) had a ceramic ring at the top of the LOPTy to insulate the heater winding, so it must have been a problem back in the day.

I guess that if it was simply the overwind at fault, then maybe it could have absorbed so much moisture that the change in 'Q' as it was unwound was gradual, unlike with a shorted turn.
The fact that the line linearity now looks ok is a sure sign that something was wrong with the overwind. I say put the line output cover back on and no-one will know

Please see the attached horror...

Cheers
Andy
Attached Thumbnails
Click image for larger version

Name:	tv22jel1.jpg
Views:	273
Size:	55.3 KB
ID:	190049   Click image for larger version

Name:	tv22jel2.jpg
Views:	275
Size:	64.6 KB
ID:	190050  
beery is offline  
Old 12th Sep 2019, 10:04 pm   #7
FERNSEH
Dekatron
 
FERNSEH's Avatar
 
Join Date: Mar 2005
Location: Gateshead, Tyne and Wear, UK.
Posts: 6,101
Default Re: Yet Another Bush TV22

Hi Andy,
It's the first Jellypot transformer conversion. who would ever do such a thing to a TV22?

DFWB.
FERNSEH is offline  
Old 12th Sep 2019, 10:06 pm   #8
Lloyd 1985
Octode
 
Lloyd 1985's Avatar
 
Join Date: Jun 2008
Location: Coningsby, Lincolnshire, UK.
Posts: 1,775
Default Re: Yet Another Bush TV22

It’s not a horror if it works!

I’ve got a TV22 here with the primary shorted, and I’ve also been gifted a TV22A Lopty, I wondered how difficult it’d be to graft that in? I know my overwind is ok, as it was producing a cracking picture right up to when the primary went I could just bite the bullet and attempt to make a new primary!

Regards
Lloyd
Lloyd 1985 is offline  
Old 12th Sep 2019, 10:28 pm   #9
peter_scott
Nonode
 
peter_scott's Avatar
 
Join Date: May 2003
Location: Edinburgh, UK.
Posts: 2,345
Default Re: Yet Another Bush TV22

Hi David, It is the EF91 deck so I think it'll be the hidden transistor.

Hi Andy, I'll just have to creep away and hide in a corner. I spent a stupid amount of time trying to fix the linearity but the set probably got fed up with messing in all the wrong places of its anatomy so it then said to itself "I'll stop this idiot with an HKS." It didn't recon on Alistair coming up with the isolating transformer so as a last ditch it decided to kill the EHT.

When I still had a picture the linearity at its worst was as below.

Peter
Attached Thumbnails
Click image for larger version

Name:	5368-tv4.jpg
Views:	154
Size:	48.1 KB
ID:	190058  
peter_scott is offline  
Old 12th Sep 2019, 11:08 pm   #10
FERNSEH
Dekatron
 
FERNSEH's Avatar
 
Join Date: Mar 2005
Location: Gateshead, Tyne and Wear, UK.
Posts: 6,101
Default Re: Yet Another Bush TV22

The TV22 was one of the first TVs to employ the line output valve as a switch rather than as a linear amplifier. The line output valve conducts and the sawtooth waveform is generated in the line output transformer. The first 40% of the scan comes from reclaimed energy so it would seem from your picture the reclaim diode or associated circuits weren't functioning.

DFWB.
FERNSEH is offline  
Old 13th Sep 2019, 8:55 am   #11
peter_scott
Nonode
 
peter_scott's Avatar
 
Join Date: May 2003
Location: Edinburgh, UK.
Posts: 2,345
Default Re: Yet Another Bush TV22

I can't now remember all the bits I messed with. When I come to tidy it up something may click.

Peter
peter_scott is offline  
Old 13th Sep 2019, 10:08 am   #12
beery
Heptode
 
beery's Avatar
 
Join Date: Jun 2006
Location: Ware, Hertfordshire, UK.
Posts: 734
Default Re: Yet Another Bush TV22

Hi Lloyd,
my first TV22 was like that. I remember Gerry Wells told me it was only good for display, however since it also had a big chunk out of the front of the case it was no good for that either.
So, the only course of action was to re-wind the inner windings. The set always performed well after that, but I eventually swapped the chassis with anther set so that I could do a nicer restoration by re-stuffing waxies.

Hi David,
maybe I should power up your old frankenstein chassis sometime. It has lost its RF unit, the focus unit, scan coils, CRT clamp, CRT and cabinet - so maybe a nice job for Christmas then.

Yes the circuit is ground breaking, but when it goes wrong the failure mode results in a runaway situation.
The grid if the PL38 is biased by using grid current to act as a diode clamp to charge the grid coupling capacitor during the positive part of the drive waveform, clamping it to the cathode voltage. When the waveform drops down it swings the grid negative. This of course was to become normal in line output stages that followed from the TV22, but the PL38 was not really designed to work this way.
The failure mode with the primary has to do with something first affecting the level of the drive pulses, it could be leaky caps, a damp lopty, rusty laminations, a low emission ECL80 or any number of things. The result of a low drive waveform is the PL38 spending too much time in the linear area causing the tranfsformer to saturate and get hot whilst producing even less feedback to the ECL80. In this situation the transformer gets shorted turns quite quickly due to overheating and once that happens the PL38 gets killed.
Another weak point is the heater cathode insulation of the PZ30 which was not really intended to double as an effciency diode.

When I restore a TV22 I always fit a NOS ECL80 was the weedy cathode area of the triode section does not last all that long. You can always use the ECL80 you have replaced in a less destructive circuit...

Of course damp lopty sindrome is more common in bakelite sets where the plastic case can hide the fact that a set might have been stored in very damp conditions.

Cheers
Andy
beery is offline  
Old 13th Sep 2019, 10:20 am   #13
FERNSEH
Dekatron
 
FERNSEH's Avatar
 
Join Date: Mar 2005
Location: Gateshead, Tyne and Wear, UK.
Posts: 6,101
Default Re: Yet Another Bush TV22

Hi Andy,
Later production TV22s used the PL81 as the line output valve.
I've tried the PL36 with no success whatsoever, one would have thought the newer valve would be an ideal replacement for the PL38.

I'm sure we can find all the bits needed to recommission the Frankenstein TV22. I can have the CRT securing band made for you.

DFWB.
FERNSEH is offline  
Old 13th Sep 2019, 12:35 pm   #14
peter_scott
Nonode
 
peter_scott's Avatar
 
Join Date: May 2003
Location: Edinburgh, UK.
Posts: 2,345
Default Re: Yet Another Bush TV22

I had a look at the vision response this morning and the isolating transformer is not making any obvious reduction. I can get the 2.5MHz bars ok but yet to see the 3MHz. The alignment isn't very good and the video amp appears to saturate very easily such that turning up the gain chops off the sync pulses.

Peter
peter_scott is offline  
Old 13th Sep 2019, 2:57 pm   #15
peter_scott
Nonode
 
peter_scott's Avatar
 
Join Date: May 2003
Location: Edinburgh, UK.
Posts: 2,345
Default Re: Yet Another Bush TV22

Got the alignment a bit better but the LO should be at 63MHz but in practice I'm struggling to tune the sound IFs there so had to go a little lower.

The vision response is fine even with the HKS transformer.

Picture centring suffers from vignetting if I try to push the picture over to the right and moving the ion trap magnet doesn't help much.

Peter
Attached Thumbnails
Click image for larger version

Name:	P1040522 (Medium).jpg
Views:	151
Size:	63.5 KB
ID:	190095   Click image for larger version

Name:	P1040521 (Medium).jpg
Views:	152
Size:	60.2 KB
ID:	190096  

Last edited by peter_scott; 13th Sep 2019 at 3:09 pm.
peter_scott is offline  
Old 13th Sep 2019, 4:41 pm   #16
beery
Heptode
 
beery's Avatar
 
Join Date: Jun 2006
Location: Ware, Hertfordshire, UK.
Posts: 734
Default Re: Yet Another Bush TV22

Quote:
Originally Posted by FERNSEH View Post
I'm sure we can find all the bits needed to recommission the Frankenstein TV22. I can have the CRT securing band made for you.
DFWB.
Hi David,
I think the front CRT securing band and the back panel are the only bits I won't have. The rest is all somewhere. For the RF unit, I can update the top deck to work with a later TV22 EF80 RF unit that I have spare.
I even have the cabinet with a chunk missing. Maybe I should fix it with fibre glass and get the cabinet sprayed a nice Bush Ivory colour

Cheers
Andy
beery is offline  
Old 14th Sep 2019, 8:39 am   #17
peter_scott
Nonode
 
peter_scott's Avatar
 
Join Date: May 2003
Location: Edinburgh, UK.
Posts: 2,345
Default Re: Yet Another Bush TV22

As of last night I had two faults that I was trying to fix.

1. Inability to centre the picture left right.
2. A slow rolling distortion of the frame scan.

Number 1. still eludes me but having replaced most of the decouplers in the set I was trying to think where the mains could be corrupting the scan and wondered if the heater voltage in the frame oscillator and output could be somehow coupling. So this morning I swapped the two ECL80s over and to my delight the fault disappeared.

It's a bit strange though because the cathodes in the frame timebase are both at ground potential so you might expect any leakage from the heater to be invisible.

Peter
peter_scott is offline  
Old 14th Sep 2019, 10:28 am   #18
peter_scott
Nonode
 
peter_scott's Avatar
 
Join Date: May 2003
Location: Edinburgh, UK.
Posts: 2,345
Default Re: Yet Another Bush TV22

Sorry folks, I was deceiving myself in thinking that swapping the ECL80s had cured the fault. I was just unlucky that the mains frequency had corresponded with my frame rate at the time.

I still have the slow frame distortion. I can see it on the frame output anode and I have about 2 volts of mains ripple on the HT but if I add more decoupling I can reduce that to about 1.2 volts but it doesn't seem to alter the extent of the distortion.

What HT ripple should I expect in this set?

Peter

Last edited by peter_scott; 14th Sep 2019 at 10:46 am.
peter_scott is offline  
Old 14th Sep 2019, 10:43 am   #19
wd40addict
Heptode
 
wd40addict's Avatar
 
Join Date: May 2009
Location: Stevenage, Herts. UK.
Posts: 790
Default Re: Yet Another Bush TV22

Do you have the 0.75uf hum reduction capacitor still in circuit?

My findings were it reduces the ripple on the HT line from about 0.7v rms to about 0.15 rms. The set will work fine without it, but you can see the picture expand and contract slightly as the Aurora generated 50Hz beats gently with the mains 50Hz. With it in circuit this effect becomes virtually invisible.
wd40addict is offline  
Old 14th Sep 2019, 10:54 am   #20
FERNSEH
Dekatron
 
FERNSEH's Avatar
 
Join Date: Mar 2005
Location: Gateshead, Tyne and Wear, UK.
Posts: 6,101
Default Re: Yet Another Bush TV22

Hi Peter,
Is the 0.75mfd ripple canceller capacitor still present? It's the large TCC capacitor mounted on the smoothing choke.
The off-set picture can be caused by a DC component in the scan coils, is the 2mfd on the scanning coils tag board OK? If the scan coils are wound on an iron core then it's possible the core could be magnetised. This condition has been experienced in the Pye LV20.

DFWB.
FERNSEH is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 1:44 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 - 2020, vBulletin Solutions, Inc.
Copyright ©2002 - 2020, Paul Stenning.