Project [Sticky] 1969 Philips G22K511 Missing I/F & LOPT, Conversion

That is rather odd then.... unless going open under load.

Would do no harm to temporarily short it out maybe?

If the HT were being pulled down, wouldn't R2115 protest ?

It may have done as it was open circuit earlier.... funny that.

This is weird.

Just checked the DS/IF and L2681 on that is 5.7R

Unless the PFL is being called upon to be amplifing a crazy HF signal that is pulling down that much HT and cooking the 10k in the process....

The pad of L2681 although looking innocent was in fact open. Stitched the post to the cct and now we're back in business. Why that pad decided to go open at this particular point, just further confiirms how much this TV hates me!

The print on that IF PCB reminds me of the Sony ICF-2001D that I repaired after battery leakage, the acid had gotten under the green solder resist and travelled along, eating through tracks and weakening others. Although your PCB hasn’t been subjected to leaky AA cells, it does look like it’s been a little damp, and has begun corroding under the solder resist, probably not enough to really worry about, as the tracks are fairly thick, but may be why that pad gave up.

Regards,

 Lloyd 

Also looks like some of the valve base joints could be on the dry side.

A side effect of all the battling with this G6 kind of makes me want someting a bit more challenging than a comfortable and cosy RBM A640. I have my eye an a HMV badged Thorn 3500... but so do several others I expect, so say no more. 😉 

Posted by: @pye625

this G6 kind of makes me want someting a bit more challenging than a comfortable and cosy RBM A640. I have my eye an a HMV badged Thorn 3500...

I doubt that'll offer much of a challenge, it's already working from one of the pictures, and the biggest challenge on 3K/3.5K sets is the PSU. After that, they're rock solid.

I could be wrong but even though it's unlocked the video appears much cleaner to me than it was with the DS IF. 

You're not wrong on two counts, John 👍 

1) That the video quality is way better than what we were getting out of the DS/IF configured in SS mode.

2) That despite having a good board and good pulse coil, we would not achieve lock, correct.

So as John pointed out in a pervious post, I need to find out why I have such high voltage (42V) on the cathode of the PCF802 when it should be only 5.8V.

So you're now back at the point you were at a couple of weeks ago but now with hopefully a correctly shaped pulse WF21 from the EF80 on the IF board, cleaner video and correctly operating brightness and contrast control. The high triode cathode on the PCF802 reactance stage is I suspect due an imbalance of WF22/23 from the LOPT. Try running with the reference pulses disconnected from pins 20/21 of the LOPT and see if it's anywhere near the quoted 5.8v, it will not be exact but with the small amount of direct sync provided by R4077/C4011 you may just about achieve a weak lock by adjusting L4501/2.

If you do get the cathode volts a lot lower then it is probably because the the flywheel feedback pulses are widely imbalanced and you will need to keep experimenting with R4070/2. You will probably struggle to get WF23 to it's correct value but whatever you can get it to try and make WF23 the same amplitude by changing the value of R4070 and adjusting the flywheel balance pot. Good luck.

Hi John, can you clarify the following:-

Are you asking me to remove the PYE manufactured LOPT winding outputs (as shown in white) from the 20/21 posts? Or are you asking for me to remove the wires that I have added from the G6 cct to posts 20/21, as shown in yellow, or do you mean All of them.

As you can see from the photo, post 20 has the PYE manufactured winding output connected, and a wire I added linking post 14 to 20. Post 21 has the PYE manufactured winding output connected, and a green wire I added, that's all there is on 20 & 21.

Hi Chris, sorry for the late reply, hello from Edinburgh and my circuit and notes are a couple of hundred miles away. You need to disconnect the lead that feeds the -ve 47v pulse from the LOPT and I’ve just remembered the one from your newly added tag strip as we found a better tap off point for the +ve going pulse. Leave in place the link from 14 to 21. Hope this helps

Ok that now makes sense, when you said this morning to remove the pulses from 20/21, I thought what....there's nothing on '20' to remove other than the link from '14'.

So I'll remove the -47V pulse from 21 (green) and the orange from pin 1 that is providing the +ve pulse to the new tag board. 👍 

 I’d forgot we had changed the taping point for the +ve pulse so sorry for the confusion. 

Nah, it's my fault, I should have put two and two together and realised what you meant. I've just become so paranoid about this TV now, I've become overly concerned I don't create 'Man made' errors on top of the problems we're already facing. It's certainly a confidence sapper this TV.

Posted by: @jayceebee

Try running with the reference pulses disconnected from the LOPT and see if it's anywhere near the quoted 5.8v, it will not be exact but with the small amount of direct sync provided by R4077/C4011 you may just about achieve a weak lock by adjusting L4501/2.

OK, pulses removed and the cathode of the PCF802 instead of being at 42V is now 13.3V.

No lock possible, in fact it's a much worse mess on the screen now than what I had before.

So I guess its back on with the pulses and experiment with R4070/2 values. Currently R4070 is a 22k and R4072 is a 220K, this is going to be a pain trying to findi the balance values.

Leaving R4072 as a 22K, but reducing R4072 from 220K to 180K, this is where we're currently at. What follows now are the waveforms for 22 & 23, plus how they should appear from the manual, followed by the resultant effect on screen.

Hmm, I knew we would struggle with WF22 and it doesn't look good but back on Page 41 of the thread WF23 looked really promising when you moved the take off point, almost as the manual but looks to be inverted to me now. Not sure what's going on.

Well, as I've said before, this TV is possessed with a spirit of not wanting to play ball.

The waveform you mention that looked good back on page 41 was with the with R4072 as a 220K, this time I'd replaced it with a 180K. So I've put the 220K back in, and the waveform is better, but still looks inverted. Also, as I mentioned to you before, if I leave the alignment tool in the core and depending on its angle, it can really help the lock. Below is with R4070 22K, R4072 back to 220K and not forgetting the 8.2K on a tag strip directly off pin 7. The tool wedged in the core. The results speak for themselves, the waveform is still not great, but a lot better.

Below is the waveform you referred to on page 41, and yes, it was the right amplitude. This was with the exact same components on the timebase as we have now but remember that waveform was the with the DS/IF PCB & pulse coil, now we've the new SS IF and its pulse coil, are these having an impact?

To be honest, I've no flaming idea how to sort this, it's a total bag of ......