Philips 19TG108U Not a success story.

[FERNSEH]

I acquired this set last week following a visit to forum member Murphy310. I picked up the set from Bob's shop in Sorn on the return trip to Geordieland.

I set about the test procedures on the receiver on the next day. The receiver was connected to a variac to bring up the supply voltage slowly in order to give the HT smooting capacitors just a taste of electricity. Even at 125 volts the valve heaters proved to be OK but there was no HT at the cathodes of the PY82 valve HT rectifiers. It turned out to be that the HT mains dropper sections were OC. For the time being therefore the set will be tested on 200 volts. It took some time to bring up the supply voltage from the variac to 200 volts.

Bad news: it seems that the line output transformer is faulty, the PL81 glows red hot. The 19TG108 employs a version of the well known Philips 17TG100U chassis,the main difference is that the 19" receiver employs printed circuit boards for the timebase modules.
I doubt if there is any chance of the correct Philips line output transformer assembly turning up, so I am considering using a Thorn 850 or 1400 jellypot LOPT. I found a box of jellypots but they were removed from the single standard 1500 receivers.
Many years ago I suggested using a 1400 jellypot as a replacement LOPT in a Philips Stye 70 receiver, the 17TG170U.

The set has been returned to the awaiting repair pile.

DFWB.

[Steve_P]

Pull off the EHT winding and see if it cools down. If it does, use a 1500 Tripler and a capacitor to get the picture size right.

Are you sure you're getting line drive to the PL81?

If it's the LOPT primary, I would tend to go for a 1400 LOPT.

Cheers,

Steve P.

[FERNSEH]

Hi Steve,
I've disconnected the EHT winding. The Philips 19TG108U employs a similar line timebase as the original 17TG100U. The circuit consists of an ECL80 triode section (V611A) and feedback from the line output transformer to maintain oscillation. This means that if the line O P transformer fails there will be no feedback pulse for the ECL80 triode grid.
The ECL80 has two high value resistors (R612 & R623) in the anode circuit, they are OK. It might be a good idea to disconnect the scan coils, they are known to fail.

The line output transformers always gave problems in this range of receivers.


DFWB.

[Heatercathodeshort]

Hello David,
What an odd looking picture of the Jellypots. It looks like you have been on a rabbit shoot when you visited Trevor and that picture looks like your trophies...J.

[FERNSEH]

Hi John,
Don't worry, they aren't bunnies. They are genuine Thorn 625 line only 1500 series jelly pots. Actually, I did not see many rabbits in Ayrshire.

DFWB.

[Tazman1966]

It'll be nice to see if you can graft a jelly pot in.

[Heatercathodeshort]

Originally Posted by FERNSEH

Actually, I did not see many rabbits in Ayrshire.

They keep their heads down up there David. They are very tasty when boiled in a cauldron.. [Rather like chicken, actually very nice.]

The overwind is layer wound I believe and should be easy to rewind. I really should get my wavewinder and underwind machine re-commissioned. I have a couple of transformers that need to be dealt with but it's 37 years since I wound a line output transformer! I will aim for the autumn/winter season as I have no intention in spending our limited summer days and warm evenings in a dusty repair workshop. The Harley Davidson is calling me...That is a really nice receiver David, well worth the effort and quite rare. Regards, John.

[murphyv310]

Hi.
I am not too sure what the 1500 lopts taste like boiled John, there would be jelly everywhere!

David I would give Bob a call as there should be a TG100 lopt in stock if that would work? There are plenty of jellypots too and I am sure some 1400 ones.

[davegsm82]

Please excuse my ignorance, but why are they called 'jellypots'?

Dave.

[oldticktock]

Great question Dave, I've wondered that too, I have a Thorn 950 chassis on the bench at the moment with Jelly pot and it looks nothing like a Jelly pot in the traditional sense. Yet Trader, R&TS and Thorn manuals all refer to them as such.

Hopefully the old hands will reveal all.

Chris

[Heatercathodeshort]

The original Thorn 'Jellypot' was a 16kv red unit used in the original 900 chassis. Early versions of the 900 used the previous 850 lopt. This was not a total success and many broke down due to arcing within the transformer. These were later modified for the 950 series using a much lower potential A.C. EHT output. This required an EHT doubler or tripler rectifier tray as apposed to the red half wave unit used in the original models.
The later types were 100% reliable and I have never known one to fail.
The reason they are called 'Jellypots' is due to the insulating material that is used to fill the plastic 'pot' encasing the windings. To put it simply it has the consistancy of 'snot' and sticks to everything! They are without doubt the most reliable flyback transformers every produced and served in the 2000/3000/3500 colour receivers with equal reliability. [I did replace only one in a 3500] Another triumph for Thorn! A very good company. Regards, John.

[murphyv310]

Yes too true. If everyone else had used those line transformers no doubt the TV trade would have died a long time before it did.
Superb tranny's, I have never had one fail ever!!!

[Focus Diode]

Originally Posted by murphyv310

Superb tranny's, I have never had one fail ever!!!

Agree, but they are easy to ruin, as I did to one 1400 LOPT when changing the disc capacitor using too hot an iron.

A lesson learned! I should've cut away the old component then lightly soldered in a new one of course.

Worth bearing this in mind for those not familiar with these LOPTS but one comes along.

A 1400 LOPT would be ideal as these take both 1400 and 1500 triplers as well as being d/s of course. I'm sure I have one which I'll look for tomorrow morning.

Brian

[oldticktock]

Thanks for posting the tip Brian, I was not aware of that pitfall.

Chris

[FERNSEH]

This project is still on going. I'm going to change the line oscillator to the blocking transformer type. Many years ago (1986 to be exact) I converted a Bush TV22 to work with a BRC 1500 jellypot and it was found the only satisfactory answer to the line drive requirement was to change the oscillator to the blocking type. Both the Philips 19TG108 and the TV22 employ the triode section of an ECL80. More than ten years between them.
Today, I found a suitable line blocking oscillator transformer in a 1953 Pilot TV chassis. I'll use it if it is OK.
Now I need to determine what changes are necessary to do the 19" Philips line TB oscillator. It is similar to later versions of the 17TG100U.

DFWB.

[FERNSEH]

Getting some real results now. The blocking oscillator transformer has been wired in. Some adjustment of the value of the resistor which is in series with the line hold control was required to achieve the correct line frequency. It's marked as R613 in the circuit diagram and this is clearly an error as the PL81 screen feed resistor is also marked as R613. A 470Kohm conected in parallel with the 680Kohm resistor brought the line frequency to the correct 10.125Kc/s.
So it follows just like the TV22 I modified to use a jellypot way back in 1986 a independent line oscillator is necessary if you are going to do a line transformer substitution.
Actually a better line oscillator solution would be to rewire the circuit around V611 to become a cross coupled multi-vibrator. Just do away with the line sync pulse clipper which is V611B.

DFWB.

[Heatercathodeshort]

Great fun David! It's amazing what will work in these old 405 line receivers with a little care and fiddling. John.

[FERNSEH]

The blocking oscillator has been taken out and I've rewired the line oscillator circuit to one that resembles the multi-vibrator employed in the Alba T655.
The line drive for the PL81 is perfect for this application, almost a square waveform and 150 volts P - P.
Much, much too much width and EHT, it's 18KV.
So the next task is to decide on how the width and EHT can be controlled. The easy option is series resistors in the anode circuit of the booster diode, something like the arrangement in the early KB VC1 chassis. Other receivers using this simple method of width control were Ekco single and early dual standard models and very early production Pye models 3 and 11.
A stabilised line output stage would be much better of course, something like that found in Ferguson TVs for example. More complication though.

DFWB.

[Focus Diode]

Excellent work indeed. I wonder if the EHT can be tamed if one of the stick rectifiers in the EHT tray is by-passed? I don't know if you've tried Thorn scan coils with the linearity sleeve? I have a spare set if you need them. This of course wouldn't cure the excessive width.

Cheers,
Brian

[bobbyball]

David, didn't you once have a LOPT for the "100" chassis set from me (at Mikey's do)?
Was it dud or used on another set?

Also the tube looks almost too "modern" looking for the set!