Project 1969 Philips G22K511 Missing I/F & LOPT, Conversion
At the risk of straying off topic, I was filled with horror when I saw their first SOPS (Self-Oscillating Power Supply) which is I suppose another name for SMPS. One it was explained however, it all became clear. When I started doing private jobs I was filled with even more horror when I came across the IPSALO (Integrated Power Supply And Line Output) in a Salora set (which was actually surprisingly easy to fix.....once I'd worked out what was going on). So really it's just what you get used to.
One of the most incomprehensible chassis of Philips (for me) was the GFL chassis with so many protection circuits that were interdependent, it required the use of a 'service remote control' to help diagnose faults that enabled you to 'switch off' parts of the circuit to isolate the various sections. There was a series switch transistor in the HT supply to the line output stage that used to switch off the line stage in the event of a fault, an auxiliary power supply (to keep the processor and memory going when the rest of the set was shut down.....unless it failed of course and then the service remote was useless.....)! I was glad to see the back of those things.....and I don't mean having my head stuck inside one trying to fathom out what was causing it to shut down.....
I believe the line flywheel discriminator circuit was designed on sound engineering principles. One of the deficiencies of many flywheel discriminator circuits is the lateral shift of the picture when synchronised to the TV station. Adjust the line hold control when the picture is locked and the picture will be seen to move side to side before sync is lost. So my take on the circuit is that the two anti-phase reference pulses from the line output transformer provide a means of keeping the picture centralised within the pull-in range of the line hold control. The positive and negative 320 volt pulses are integrated to form anti-phase ramp waveforms which counter any shift of the relationship of the sync pulse and the reference pulses.
Two years after the G6 single standard set was released into the marketplace in 1969 line discriminator integrated circuits such as the TBA920 became available to perform all the line synchronising and phasing requirements.
Just to say I think the design is very elegant and I do like the fact that correction from the frame is also applied. Unfortunately because the original LOPT is a complex affair to accommodate the flywheel design it’s made the conversion to the PYE transformer a lot harder than it would have been if it had been conventional.
If the line flywheel discriminator is to work properly the positive and negative reference pulses will have to be present. An idea: we know that the windings on the line output transformer are volts-per-turn and will be in the order of 8 to 10V/turn. I found this out while working on the Ekco CT104 line timebase. So why not consider adding a separate centre tapped winding on the line output transformer? The existing pulse windings can be left in place and can take care of the other requirements demanded by the receiver. Wind ten turns on the transformer and measure the peak pulse voltage on the 'scope. It's likely ten turns will produce anything between sixty to one hundred volts.
An example of volts-per-turn. In the Mullard PAL COLOUR TV book page 6.8 details how to make the line output transformer. The EHT overwind consists of 1680 turns, wound at 30 turns per layer. Assuming 6KV is present at the anode of the PL509 19KV divided by 1680 turns gives 11 volts/turn. So it follows the 640V pulse winding will require about sixty turns.
An example of volts-per-turn. In the Mullard PAL COLOUR TV book page 6.8 details how to make the line output transformer. The EHT overwind consists of 1680 turns, wound at 30 turns per layer. Assuming 6KV is present at the anode of the PL509 19KV divided by 1680 turns gives 11 volts/turn. So it follows the 640V pulse winding will require about sixty turns
That's something I had mentioned to Chris but I wasn't sure how many turns would have been required. The PYE transformer he has pressed into service looks like a Weyrad design to me with a single bobbin, The other limb of the core is empty which would be ideal for the pulse windings. The change of the tapping point I suggested for +ve pulse for one of the diodes was looking really promising before the discovery of the o/c ringing coil was found. It's possible he may only need to wind one for the -ve going pulse as we only have -47v to play with from the PYE TX.
Hi Trevor, first of all many thanks for the info. However, I'll pass and leave the rare offer and opportunity of a NOS G6 LOPT, to a fellow collector who has a far more deserving TV than this heap I saved from the bin. Whoever has the LOPT might want to earn themselves a small fortune, as I suspect if that was listed on e-bay it would generate a fair bit of interest.
I did contact Mike Barker, and he offered to rewind my primary, but in the end I even decided against that too. This TV is just an exercise in experimentation, what is possible, with what I have to hand and has been modified accordingly. I've no intention of investing my time in reverting to spec, certainly not the valve derived EHT regulation side of things. So removing the 25kV overwind would be a crime against a NOS G6 LOPT and I wouldn't do that.
On a related note and possible good news. Another pulse coil looks to have been sourced, however it is broken. It may or may not be a simple fix, we shall have to see, so the project may not yet dead in the water. This TV might still live with its PYE LOPT, Solid State EHT & Focus, Mike Phelan Beam limiter and DS in SS mode IF, watch this space.
No problem at all, I suppose your G6 is like my Frankenstein TV 22 that I was slated for over the years. It will run on 180 to 625 lines, has flywheel sync, Video amp with cathode follower, BRC Jellypot LOPT, fully stabilised EHT, line flyback suppression, heater capacitor dropper, and many other small tweeks.
Ive not followed your thread that close Chris. I've done many LOPT substitutions even in the trade. The GEC LOPT is a good all rounder too. Incidentally I don't like feeding a tripler directly from the PL anode, if you are getting 25KV from there there is a tuning issue and you'll land up with issues on the primary. You need a small secondary winding to give about 1kv, approx 1000 turns in phase and series to where the original over wind was connected to, this will allow you to tune the LOPT and reduce the voltage on the PL anode and in the rest of the circuit. Over 8Kv on the PL anode is too high.
Radiomuseum Member. Collection HERE
Ive not followed your thread that close Chris. I've done many LOPT substitutions even in the trade. The GEC LOPT is a good all rounder too. Incidentally I don't like feeding a tripler directly from the PL anode, if you are getting 25KV from there there is a tuning issue and you'll land up with issues on the primary.
Wise man for not following the thread, at 42 pages it would probably kill any enthusiasm for TVs stone-cold dead.
The project started in Feb of this year, and has gone through many changes, mainly due to lots of missing parts incredibly turning up! However, its suffered a fair few set backs and failures too. Out of the three pulled LOPTs I found 2 had o/c primaries, the third was OK but only lasted a couple of months, even my deep pool of determination is almost depleted.
Once the only good primary failed I'd had enough of G6 LOPTs and replaced it with a PYE 697 is in its place, so the line stage is pretty much a PYE 697/G6 hybrid at this stage and rock solid, and works remarkably well with very little changes required. However, the project ground to a halt last week due to the G6 pulse shaper coil on the I/F going o/c, this resulted in me unable to get a line lock. I didn't fancy modifying the flywheel discriminator, so I'm hoping the donor coil that John has sourced on my behalf will allow me back into the game or the flywheel will need changing to the 697's implementation.
G6 Update; The TV that refuses to lay down and die!
Through a kind donation from John (many thanks John) who approached Glyn on UKVRRR, to see if he had a spare Pulse Coil for a G6 IF. Glyn did, and John delivered it along with the single standard board it came from this evening. Yes, unfortunately the pulse coil was broken off from its base. This seems to be a common theme with the G6 IF, that's four boards I've seen now with this exact problem. Amazingly, a friend who often sends me boards contacted me yesterday and said he had a batch he will be sending me shortly, you could have picked my jaw off the floor when I spotted a SS G6 IF among them, sadly the pulse coil is nowhere to be seen. Still, a source of spares is always welcome. This pulse coil certainly seems to be the Achilles heel of the G6.
Anyway, the board was put through the PCB cleaning process, and it came up much better than I thought. There are three diodes, two transistors and one cap missing. I may well install this board at a later date, however at this point I don't want to introduce any further problems to sort. We know the current DS IF configured as SS is sort of working. I propose to continue with that and once we've ironed out all the other faults then consider installing this SS IF.
Now that leads on to the all important and two critical questions, 1) is the pulse coil usable in other words no open sections and 2) can it be repaired. Well I can answer 1, yes the sections have continuity, I've ohm'd them out and have identified the coils. L2712 is the 14R section, the other two are L2710, L2711, both these are 33R. I've not labelled these as they can be either.
Now the tricky bit, soldering hair fine wires to the base posts and then securing the centre core to the base, micro-surgery here we come.
Once again, many thanks to the kind efforts of John and Glyn, hopefully I'll be back in the game with this recalcitrant G6. 👍
So there's a real possibility this set could be restored to "as-built" condition in the future? That's amazing.
So that coil has three windings as shown on the SS diagram. Also an interesting point in your earlier post there is what appears to be a thicker tinned remnant of a wire on the base pin 4 but nowhere to be seen on the old coil winding itself but is present on the replacement. I now wonder if this snapped completely off the old coil, looking closely at pin 3 am I seeing possibly two wires to this connection?
Not sure if I understand John, from my point of view the Single standard pulse has three coils (L2710,11,12) with 4 lead-outs, the Dual standard only having two coils (L2711,12), would have only 3 lead-outs, as I had. See below.
It's true that the circuit only shows two windings but the board layout for your DS IF does in actual fact show three windings which only adds to my confusion. Plus, your original coil only shows one wire as a twisted pair.
Now I may be very wrong here but are you sure the 14 ohm winding is L2712, are the winding resistances in the manual? I have a sneaking suspicion it could possibly be the unused L2710. A close up of the coil base from the SS IF might help me decide.
Plus, your original coil only shows one wire as a twisted pair.
But surely that is correct, a dual standard pulse coil will only have one twisted pair? Two coils joined, equals one twisted pair. The Single standard Pulse has three coils, so has two twisted pairs? I must be thick, as I'm just not following you.
Now I may be very wrong here but are you sure the 14 ohm winding is L2712, are the winding resistances in the manual?
Yes and yes
Ok, the manual says it all. Just me overthinking things and not being able to express my own thoughts correctly.
OK moving on, the coil has now been checked, orientated and repaired. Phew! That stretches the limits of my steady hand coordination under a scope.
However, John informed me yesterday even with the new coil he very much doubts, and would be surprised if I get lock up. This being due to the spurious 45V we had on the PCF802 cathode. More work to be done in this area due to the PYE LOPT conversion. At least we can continue now knowing the Pulse coil is eliminated from suspicion.
Whilst I had the Single Standard board on the bench to repair the Pulse coil, I thought I might as well install the missing components.
These were as follows :-
- 1st and 2nd Vision IF's; TR2142, TR2143 - 2 x BF196
- X2151, X2153; 2 x OA91 AGC amp cct,
- X2152; 1 x BA154 Luminance output cct
- C2047; 1 x 10uF (should be 12.5uF but 10uF will suffice) 450V again LUM out cct
Not sure why but something also made me look inside CAN 'E'. Glad I did as the little ferrite core in L2748, in the chrominance detector had popped out, luckily retained within the can. I secured it with a blob of blutack.
The IF board is looking rather swish now, from when it first arrived. The only couple of repairs needed were for missing pads on R2115 and R2126. Whether there are any underlying faults with this board remains to be seen at a later date.
It would be rather nice to see how that IF performs but for anyone that hasn’t been following closely it’s not just a case of plugging in the SS board. Several changes had to be made to the circuit and wiring, not least the rather crude and hairy brightness control that the DS has, this required a rather beefy pot as it one end to HT and feeds the screen grid of the PFL200.
Note that the repaired ringing coil will need it’s connections swapping also as the coils are wired differently between the two chassis.
I've been pondering this all day, and I think it's probably the better way to go. I've decided to bite the bullet, remove the DS/IF and install the single standard IF instead. Revert the cct changes we made to accommodate the DS/IF, such as the brightness pot etc. I'll still need to break into the lum output and insert the Beam limiter cct, but that can sit on a little PCB attached to the rear of the SS/IF much like my little interface board.
By doing this, I think we eliminate some mod problems in regard to Contrast, and I eliminate that horrid high voltage on the brightness by going back to the SS implementation, which was much calmer. Also, I really don't feel inclined to remove the Pulse coil from the SS/IF, then faff about orientating it and extending legs, just so I can shoehorn it onto the DS/IF.
Ok, I may end up having to fault find unknown lurking IF problems on this new board, but I think ultimately that will be preferable, who knows maybe I'll be lucky and it will be fine. I notice nearly all the cores on this SS IF are still waxed, untouched, unlike the DS/IF which looks to have been twiddled within an inch of its life along with all the one-ended presets.
The SS board is installed and hooked up, all I need to do now is revert the cct for the brightness.