Featured
Latest
Philips 21TG100U Li...
 
Share:
Notifications
Clear all

Forum 141

B&W TV Philips 21TG100U Line timbase fault.

82 Posts
10 Users
84 Likes
5,585 Views
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
Topic starter
 

In fact the line and frame sync fault was and still is to a certain extent caused by the line flyback pulse finding it's way into the signal circuits. Strangely, swopping the two EF85 valves in the IF amplifiers has almost cured the fault but more work is required to completely clean up the vision signal path. The attachment shows that the employment of the 210/300 series line output transformer is a viable proposition. 

Philips 21TG100U Pic 1

To reduce the EHT and width a 330 ohm resistor is connected in series with the anode of the PY81 boost diode.  In order to resemble the line drive waveform of the 210 series the mark space ratio of the drive to the PL81 grid will be changed to something like 40:60, the latter figure being the conduction time of the line output valve.  The cathode current of the PL81 is 200mA during the latter part of the scanning stroke, the mean value is 150mA over the 98.8microsecond period is still rather high for that valve.

Till Eulenspiegel.

 
Posted : 09/04/2021 3:43 pm
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
Topic starter
 

The Philips 19TG108U is a development of the 17TG100U. The example I have here needs a replacement line output transformer.  Might use it for spare parts.

Philips 19TG108U

 Its got a brilliant AW47-90 CRT which would be better off fitted in the Ferguson 705T Senator.

Till Eulenspiegel.

 
Posted : 11/04/2021 3:02 pm
Lloyd, Lloyd and Lloyd reacted
Katie Bush
(@katie-bush)
Posts: 4884
Famed Member Registered
 
Posted by: @till

Might use it for spare parts.

Ooh, that's naughty! I say give a chance and see if you can make it live again? Then if all else fails, rob the poor thing's grave.

 
Posted : 11/04/2021 4:44 pm
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
Topic starter
 

Hi Marion,  this is the set I tried a jellypot line transformer in, it wasn't a successful project. Maybe the colour TV line output transformer could be tried out again in this set, but with a beefier line output valve, a PL36 or PL504.  How about a solid state line output stage? Remember the strange Thorn 1500 with the transistor frame timebase?  Or am I dreaming that one.

This TV and radio restoration game isn't getting any easier Seems gone are days when one pick up a set, blow away a light layer of dust, change a few capacitors and voila! a working set.

Meanwhile, back to the 21TG100U. The very high amplitude line drive waveform at the grid of the line output valve. The anode of the oscillator valve receives it's supply from the boost HT rail.

Philips 21TG100U wfm PL81 G1

 Till Eulenspiegel.

 
Posted : 11/04/2021 5:39 pm
crustytv
(@crustytv)
Posts: 11869
Vrat Founder Admin
 
Posted by: @till

Remember the strange Thorn 1500 with the transistor frame timebase?  Or am I dreaming that one.

Not a dream 👍 

https://www.radios-tv.co.uk/community/black-white-tvs/1500-chassis-solid-state-field-and-modified-cct/#post-89382

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 : 11/04/2021 5:56 pm
Katie Bush
(@katie-bush)
Posts: 4884
Famed Member Registered
 
Posted by: @till

This TV and radio restoration game isn't getting any easier Seems gone are days when one pick up a set, blow away a light layer of dust, change a few capacitors and voila! a working set.

I know that feeling! I guess as time goes by, the best ones have been picked over and taken already, and of those left, they deteriorate further. Eventually, there won't be many left in restorable condition. Then, add to that we are all getting older, and we're losing our numbers as time progresses. I made a bit of a point of saving as many sets as I could find, before tackling any restoration - I just hope I haven't left it too late to get stuck in.

Also add in the factor of things that we (I at least) still think of as modern, which are now appearing in places like eBay. I'm talking 1980s Sony, Panasonic, and similar era SS sets, the sets that don't quite have the appeal of the older marques. To be honest, I prefer to see valves rather than transistor/IC circuits.

 
Posted : 11/04/2021 6:15 pm
Lloyd, Lloyd and Lloyd reacted
Lloyd
(@lloyd)
Posts: 1897
Prominent Member Registered
 

It’s these damn Philips sets that’s the problem! Far too clever for their own good! That 385U has been leading me up the garden path with a sync fault, and no matter what I do I can’t seem to pinpoint it.

great sets when they work! I also have a Philips 1708U, which works well, apart from an occasional frame linearity glitch, another annoying intermittent fault!

 Regards 

Lloyd 

 
Posted : 12/04/2021 12:16 pm
PYE625
(@pye625)
Posts: 5121
Famed Member Registered
 

Like the little girl in a certain story or nursery rhyme....

"When she was good she was very very good... But when she was bad, she was HORRID ! "

 

To understand the black art of electronics is to understand witchcraft. Andrew.

 
Posted : 12/04/2021 4:44 pm
Lloyd, Lloyd and Lloyd reacted
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
Topic starter
 

Following my policy of glasnost (openness) I must report this line output transformer substitution job isn't going very well at all.

OK, the EHT has been tamed and the picture width just needs a slight increase to give the required over-scan for those early 110 degree CRTs. The aspect ratio is 5:4 and the instruction books inform the user to adjust the width for over-scan. But that fold-over at the left side continues to cause concern. The picture can be corrected by  adjustment of the line hold control but long term stability can't be assured. 

To prove once and for all that the fold-over is not a result of mis-timing of the synchronising, a separate line oscillator module has been constructed. This simple device will determine if it is a sync timing fault or the fault lies in the output stage proper.  That being incorrect matching of the existing scanning coils to the 210 line output transformer.

Unlike the Thorn designs from 1959 on in which the circuits of the line output stages hardly changed at all, remember early 900 series models employed the same line output transformer that was used in 700 625 convertible models and 800. 850 dual standard models. the clever jellypot replaces the conventional transformer.     Throughout the sixties, model for model, Philips 110 degree CRT models employed different types of line output transformer. We had the 17TG100U, 19TG112 405 only models followed by the twin-line convertibles.  Next came the dual-standard 19TG155 series replaced in 1965 by the Style 70 models. Then came the 210 dual-standard and 625 only 300 series. 

The line oscillator module was connected between sync separator and grid of the line output valve.  Now there no doubt about it, the left side picture fold-over is a result of incorrect scan coil matching.  Now there's another component we should have saved from those TVs we smashed up all those years ago, scanning coils!

Philips 21TG100U Line osc

 

Till Eulenspiegel.   

 
Posted : 27/04/2021 5:32 pm
MurphyV310
(@murphyv310)
Posts: 453
Honorable Member Registered
 

 

IMG 20210405 141406 resized 20210405 033751697
IMG 20210402 160504 resized 20210402 050941707

Hi.

Interesting thread David, I understand your pain with this. I have done a good few LOPT Transplants over the years with good success. We realise that the LOPT may or may not be a good match for the scan coils and the the oscillator "Could" pose some issues.

One of my early swaps was on a ropey TV22 that was destined for the bin. It had a duff LOPT, a down to air CRT and other maladies. I had a Thorn 900 jelly-pot, one of the very early ones with the higher EHT. Like you David I grafted this into the set, I removed the PZ30 & PL38 and substituted one half of the PZ30 with a PY33, a PY81 for the recovery diode and a PL36 for line output, a blocking oscillator was added using a 30FL1 the pentode section as sync separator. The issue I had was foldover on the left and the HT required was only 95v for 9kv EHT & Full width. Later I added 9, 1kv avalanche diodes in series as an EHT stabiliser, I then increased the HT to 115v to the line stage, This gives a stable 9kv at both zero beam current and full brightness and contrast. To cure the foldover I reduced the tuning of the LOPT by lowering the tuning cap value. This is done as it shortens the flyback time which is why the EHT rises as the value drops, having stabilised EHT this mod was successful.

I fitted a 1400 LOPT to a Murphy V659 and used the Thorn stabilisation circuit with 100% success, no foldover or other problems. I've also substituted an unknown LOPT into a Pye V4 with no issues all that needed to be added was a few turns of insulated wire round for the AGC and flywheel reference pulses.

The latest mod was a Pye VT4 with yet another duff LOPT, this time the transformer was a 1500 625 line Jelly-pot. I made a home made doubler and fitted the transformer on a plate in the well of the original TX. I had issues with the refernce pulse but I managed to fix it. There was Foldover on the left and the right was cut off, as the flyback time was far too long, so a drop in in tuning cap value to 68pf cured that issue but EHT was 20KV so I rewired the scan coils to parallel configuration and fed the line stage via a 500 ohm 10 watt resistor for perfect results.

 

In your case David you need to add in stabilisation as the 210 TX is designed for stabilisation, or reduce the HT to the line stage, the tuning cap needs to be reduced which will cure the Left hand side Foldover, the EHT will of course increase but that can be tamed by stabilisation or lower HT, You can try the avalanche diodes trick too. The 210 has both 405 & 625 tappings for the scan coils so try it on the 625 line tapping this may give a better match to the coils, it's really time consuming to all this but usually will work out in the end.

Good Luck.

Two pictures of the VT4 Mod above

Cheers,
Trevor.
MM0KJJ. RSGB, GQRP, WACRAL, K&LARC. Member

 
Posted : 27/04/2021 7:47 pm
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
Topic starter
 

Hi Trevor, the first set I modified to accept a jellypot transformer was a Bush TV22, that was in the summer of 1985. All the necessary parts came from an unfortunate BRC 1500 that happened to be on the scrap pile.  The 1500 line blocking transformer was used because the original feedback circuit from the output transformer didn't work properly. The ECL80 line oscillator valve was retained. A PY81 boost diode was added to the circuit because it was considered the peak voltages might be too high for the PZ30 damper diode. EHT came from a doubler circuit which uses two BY182 diodes.  The set is still in existence and resurfaced in the vintage-radio forum as the "Franken Bush".

Another set to receive the jellypot treatment was a Decca DR123 23" console. That set can also be found in the vintage-radio forum.

The 21TG100U. Having proved that it is not the line oscillator or the sync causing the foldover condition the the oscillator module can be removed and the circuit rewired to the original manufacturers specification.

Till Eulenspiegel.

 
Posted : 27/04/2021 8:19 pm
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
Topic starter
 

Tried out today a set of scanning coils made by Plessey. The results were even worse, reduced width and even more picture foldover on the left side.  Next, fit a set of the earlier Philips 110* scanning coils, the brown potted type. Results the same as the AT1030 coils that the set came equipped with. It's likely these coils were replacement for the older potted type. The only solution to the transformer substitution is find a set of scan coils from a Philips 210 or 300. Even the scan coils from a Pye 368 should be OK because that chassis employed the 210 line output transformer.

Philips 21TG100U Scan Coils 110deg Plessey
Philips 21TG100U Scan Coils 110deg 1
Philips 21TG100U Scan Coils 110deg AT1060

Till Eulenspiegel.

 
Posted : 28/04/2021 4:37 pm
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
Topic starter
 

At last I'm getting somewhere with this set. I'm now using the French Philips scanning coils along with reconnecting the coils to the 625 tapping on the line transformer secondary. Now the left side foldover has gone. Also, the boost HT voltage has come down to a more realistic 600V instead of the overly high 890volts. In the original 17TG100U circuit the line linearity coil was in series with scanning coils. The linearity coil is now in the same circuit position as the 210 chassis, that is in series with the line transformer secondary windings.  The attachment shows just enough picture width but remember the HT supply to the boost diode is reduced by the 340 ohm resistor. That resistor can be replaced with a chain of series resistors to permit the adjustment of picture width. 

The line oscillator module can be removed and the original oscillator circuit reinstated.  The EHT voltage must be checked to confirm it is not too high, >18KV. 

What is clear about these line transformer substitutions is that  the matching of the scanning coils is absolutely critical is some circuits, it certainly was in this Philips set.

Philips 21TG100U Scan Coils 110deg FD09200

Till Eulenspiegel.

 
Posted : 29/04/2021 10:09 am
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
Topic starter
 

Managed to increase the picture width by replacing the existing 100pF flyback tuning capacitor with one of 180pF. The value of the "S" correction capacitor will need changing to suit the inductance of the scanning coils and line output transformer.  The PL81 seems to be holding up despite the high cathode current.

Philips 21TG100U Scan Coils 110deg FD09200 2

Till Eulenspiegel.

 
Posted : 29/04/2021 5:18 pm
PYE625, Nuvistor, ntscuser and 6 people reacted
Nuvistor
(@nuvistor)
Posts: 4594
Famed Member Registered
 

Looking good. 👍 

Frank

 
Posted : 29/04/2021 8:47 pm
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
Topic starter
 

Hi Frank, this has been quite a job but we believe the set was worth saving, the Mullard AW53-88 CRT is brilliant.  A no sound fault occurred recently which was caused by the 2.7megohm anode load resistor of the triode section of the PCL83 output valve having gone open circuit. A well known fault in these sets.

Just checked the EHT voltage - it's 17KV that's perfect.

Till Eulenspiegel.

 
Posted : 29/04/2021 9:18 pm
MurphyV310
(@murphyv310)
Posts: 453
Honorable Member Registered
 

Hi.

That's better David.

The Murphy V659 always suffered with LHS Fold-over with two different LOPTs. When the overwind failed I got sick of these oil filled transformers. After fitting the 1400 Jelly-pot the Fold-over was cured so even on the original set up there was a mismatch and this resulted in a nasty LHS Fold-over.

Cheers,
Trevor.
MM0KJJ. RSGB, GQRP, WACRAL, K&LARC. Member

 
Posted : 29/04/2021 9:33 pm
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
Topic starter
 

To finish the job properly I've ordered from RS Components a quantity of 75ohm 5watt Vishay resistors to make up the width control resistor chain which will be in series with the HT feed to the boost diode. The resistors will housed in the line output transformer box and connected to the existing width control tags. Also needing attention is the value of the "S" correction capacitor, at present it is 1 microfarad. It's likely a smaller value is required. If that is the case rather than do any alterations to the main chassis an additional capacitor can be fitted in series with one the line scan coil leads. 

Till Eulenspiegel.

 
Posted : 30/04/2021 4:45 pm
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
Topic starter
 

The 75ohm resistors arrived earlier this week and four will be used in the width control circuit.  RS part number 846-9898.   Might still consider using the original AT1030 scanning coils, but only if the correct width and EHT can be achieved. 

Philips 21TG100U width control resistors

 Remove the temporary width control resistors.

Philips 21TG100U width control 1

Till Eulenspiegel. 

 
Posted : 08/05/2021 2:18 pm
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
Topic starter
 

The width control resistors are now wired into the circuit.  To supply HT to the anode of the boost diode will the former connection for the line blanking pulse can be used.   

Philips 21TG100U width control resistors 2

Till Eulenspiegel.

 
Posted : 12/05/2021 11:03 am
Page 4 / 5
Share: