Winter project.

[Murphyv310]

Hi.
Well I didn't give up and went back to the project. The blocking oscillator transformer was cobbled in and after a few revisions of component values I was able to swing from 4500 to 11,000 Hz. All seemed well enough for a while and I even got the line sync to give a strong lock. That was short lived though as the speed started jumping all over the place. Eventually I found a blocking oscillator transformer for the ringing transformer on a projection TV eht stage. I didn't think it would be suitable but it works fine.
Next it to tidy the stage up and move onto improving the drive to the line scan coils.

[Murphyv310]

Hi.
Today I put the set aside for a while to spend a little time on the PC I want to use for computer generated old TV standards. TBH it was a waste of time, I got no further forward.
I went back to the TV22, after altering some values in the frame stage I've got nearly perfect linearity on 25p standards. I've also fell air to a 7kv DC eht generator that works from 3volts. On the tube for reasonable brightness the eht is 5.5kv.
Focus is excellent at this voltage but regulation is dismal as the unit cannot supply the current required.
I can now though fill the screen on 180 & 240 lines and a still image looks good. The flicker is still present but not hugely irritating, horizontal definition is excellent as it should be using the TV22 front end, the multistandard Aurora and an SCRF as a modulator.
I think I'll not go for all standards on this and limit it to the 25p standards, I'd like to try 120 lines but I'd need advice on how to synchronise a mechanical standard to an electronic TV.

[Murphyv310]

Hi.
Well as usual there is always a problem.
The Philips scan drive transformer has bit the dust. What I have decided to do now is to fit a Thorn Jellypot that I have done in my main TV22. Unfortunately I don't have another one that is intended for 405 lines. Mike (Mikeymushradio) has a 1500 (625 line only) transformer and I will try and see if it will work, it's normally easy enough to "throttle" back the line stage so it wont produce excessive width and EHT by a couple of methods. You can reduce the HT or damp one of the windings.
In the meantime I have replaced the PZ30 with a PY33 and added a B9A base for a PY800 efficiency diode.
Ideally I'd prefer a 950 or 1400 Lopt. If anyone has a spare for sale let me know, i will try the 1500 one firstly.

[Murphyv310]

Hi.
Well as I don't have a suitable LOPT as yet I have been doing some other things on the chassis.
Originally I cobbled in a rather large transformer to supply heaters to the CRT which has a 600ma heater. I have a nice neat little 12v transformer rated at 1.2 amps continuous, I have a "gold" 10 watt resistor of 10 ohms as a dropper, not ideal but neat enough for the job.
I have run up the set today less the line output stage and re-fitted the line blocking transformer I used initially, it is now stable with no drift. I also have been able to work out the values of the charging cap for 180, 240 & 405 lines. I have also found a suitable switch for line system switching.
I'm hoping to finish these mods before working on the line output.
What I did try today was feeding the frame stage from the 555 timer I used for the scope. Disconnecting the frame blocking oscillator transformer and using both triode and pentode of the frame ECL80 gives a good waveform on the scope from the output transformer. This was my intention initially but until I have line scan and EHT its somewhat hard to tell if this would be a success.
I am noting all the changes down on paper and once complete I will post up the full circuit.

[Murphyv310]

Hi.
Well I have to put up my hands and in the spirit of Glasnost have to say there was a man made fault. I decided to ring the Philips scan drive transformer, it rang perfectly. So back it went into the chassis, I still had a very odd scan with just two feint portions of scan on each side of the tube.
After a lot of prodding around all of a sudden we had results. The earth for the 16uf smoother and tagboard, had not a drop of solder, a man made fault from the factory, August 1951.
So we will proceed until I get the jellypot.

[Murphyv310]

   
Hi.
The last few days have been a no no for me. I have done something to my back and I can hardly get about.
Around midday I decided to take two 400mg Ibuprofen which I really shouldn't do as they react with other meds I take. Within a couple of hours I felt a lot better. So I had a couple of hours on and off in the shed. I've been using the Philips projection line scan transformer again, that and the added PY81 and a blocking oscillator circuit from a 1953 Ultra has given me the best results so far.
I did try using a 555 timer as a sawtooth oscillator with some success which no doubt I'll go back to later but the Ultra circuit has by far the best sawtooth with very little overshoot.
I can get full width on 240 & 180 lines. There was some issues with flashover but as can be seen the addition of two pulse caps has tamed the circuit and results are excellent. Take mind the scan transformer is designed for 10,125Hz and its a plus this works and so well on the lower frequency.  
I've quickly sketched out the line stage. The G2 feed to the CRT and the width control is omitted but the Bush width control still works but is rather fierce in its control, from a postage stamp to full width!


The EHT doubler diodes need to be suitably rated as does the their caps, I am now using a PY88 just to balance up the heater chain as the PZ30, PL38, are no longer and subsitiuted with PL36 (30P4 or 19) PY33 & PY88. A cap dropper will be fitted eventually once more mods are done.
Next will be more work on the frame timebase.
Some more pictures on the next report.

[Murphyv310]

Hi.
One thing I tried was using a 555 in the frame stage instead of using the original frame blocking oscillator. I have to say it works very well indeed. In fact it's a straightforward mod. Considerably cheaper than getting the original transformer rewound, not authentic mind you. I was feeding it from a 1881 sync separator chip. Solid lock on both 25 & 50 Hz frame rates.

[Murphyv310]

   
Hi.
The White 1500 Jellypot came today and the red and white one came totally unexpected from Bob.
The Red and white one is the first incarnation of the jelly pot and probably the most useful one for us that want to substitute something better for old transformers.
It did have problems as it could supply 16 to 18Kv without a doubler and it was known to breakdown. In older sets where the EHT is lower these transformers are ultra reliable as they can be "throttled" back either by lower HT or by reducing the the drive, using the VDR in the Thorn circuit is one way and the most ideal or by simply making sure the drive is lower by circuit design.
In the TV22 the line is generated by a multi-vibrator between the the ECL80 Triode and the PL38, you can't do this with the jellypot so either a separate multi-vibrator or blocking oscillator is needed.
In my original TV22 mod fitting the Thorn Jellypot I just reduced the HT to the line stage. 110V works beautifully with excellent width and EHT of 8Kv fed from a single diode connected to the transformer. The set runs down to 240 lines fine.

This time I'll have a stabilised line stage and some degree of reduction of HT. I'll only need 6Kv for the CRT and less power for scanning.
Hope to get results by the weekend.

[Murphyv310]

Hi.
I got very good results yesterday with the Thorn 900 Lopt. At this point I'm running the time base at 90 volts HT, I'm using three 10kv rated silicon diodes in series with the one nearest the Lopt shunted with a 50Pf pulse capacitor. I have this decoupled with a rather too large a value 0.1uf cap rated at 6kv temporarily, this then feeds via a 1.5 meg to the final anode of the tube.
Picture quality is very good line linearity is near perfect bar some left hand foldover. Hopefully over the weekend I'll introduce stabilisation to the circuit and raise the HT, I may need to reduce the line drive somewhat, doing this removes the foldover on the left hand side. I took some quick photos last evening and will post them up later on.

[Murphyv310]

   
   

Hi.
Here we have the LOPT mounted and quickly lashed up.
The picture is 180 lines 25P. I was using my Fuji camera that has a problem with close ups, although macro is selected it still doesn't focus too well.
Still some work to be done on sync and the frame TB.