Packard Bell AGC issue

[miniman82]

Need to ask about this one, because I've never had an issue with AGC circuits before. This chassis seems to be pretty close to a CTC-9, the tube layout and most of the schematic match up aside from the AGC portion. It's still a 6AW8A triode section doing AGC amp duty, but the voltages are really screwy. According to the CTC-9 schematic I have here there should be +.8v on the plate, +96 on the grid and +138 on the cathode. Instead I have -104/+137/+147, with a portion of the negative plate voltage going through a resistor to form the RF AGC line. So the negative voltage is causing the tuner to be completely cut off, proven by connecting an external power supply to the AGC line: when I do that I get a picture, without it I get a blank raster. So obviously something is wrong with the AGC circuit, and I'm leaning towards to the abnormally high grid voltage. Can't nail it down though, because on the schematic I have the grid goes directly to the AGC pot which is grounded on one end, but on the actual chassis it goes to some other control's wiper.

Anyone have some random PB schematics laying around I could compare this to? Only number I see on the chassis says "149190", but that sounds more like a serial number than a model number to me. Any help is appreciated!

[DaveWM]

did you scope the plate for the horz pulse?

[miniman82]

I suppose if that was missing it would account for the really low plate voltage. I only looked at the DC points on it, but the meter was trying to do some funky stuff while reading the plate. I'll scope it when I get home, if it's absent I suppose the first culprit is the coupling cap coming from the FBT?

[DaveWM]

yea that is where I would start, but there may be some other stuff in there as well. Should see a 600v pp at the plate. See if there is a 10meg resistor from a B+ to the RF agc

[miniman82]

Schematic calls for a +350 AGC pulse, I'll get back to you in a few hours.

[miniman82]

There is a 600v pulse at horizontal rate on the plate and it stays there rock solid, what should I check next? I was assuming the high grid voltage was making it so the plate voltage never got close enough to 0 to pull in a picture, but I'm not really sure why that is without a decent schematic to follow. If it were a true CTC-9, I'd blame a bad AGC pot right away since that is what sets the grid voltage. On this chassis I'm not sure since I don't know where else the AGC circuit goes.

[tom.j.fla]

If your P.B. is from 1957/58 is likely a 98C1 chassis Sams 386-3. Pix of front and chassis would help. If a little later it may be 98C3/4 from 1960 or 61. All the best,Tom

[miniman82]

This one is likely later, it has a giant wafer switch on the tuner to run the panel full of individual channel indicator lamps if that helps narrow it down. I would guess 60 or 61.

[tom.j.fla]

If said P.B. is early to mid 60's, then look at Sams 546-1 and 614-2. I don't remember ever seeing one with individual channel lights. If VHF only then look at 546-1 first. All the best, Tom

[DaveWM]

look around for any AGC filter caps .1 ish and check for any high ohm resistors going back to the B+ from around the plate of the AGC. the grid bias looks like -10 which may be about right. If your reading 600v PP then the output should be positive not negative.

[madRad45]

Looks like 614-2 might be the correct Sams, which is for Chassis 98C6. I see the "Computer Lamp Switch" for the individual channel lamps on the schematic - which I just scanned. PM me the best way to get it to you.

[miniman82]

Cool, I'll send you my email.


Just had a real duh moment in the garage: in case the schematic don't match, follow the wires! So I followed the grid of the AGC tube to the very next pot, which happened to be on the back of the chassis instead of the tube side like my book says. That ended up being the AGC pot and now it works, so I guess the other would have to be the noise inverter? No idea, but something in vertical isn't right now so I have to tackle that next. Thanks for the push in the right direction, sometimes I guess I need to talk things out to get right. lol

[DaveWM]

IIRC there is a RC network in there that you want to check for drifted resistors, its in the feedback loop, look for where the sync pulse is fed into the vert multivibrator for a drifted resistor. You say vert issue, lack of height? sync? lin? Of course you want to check the can cap on the cathode 1st as its a likely suspect if you are short on scan.

oh the resistor I am remembering that gave me so much trouble was a 39k from the cathode (pin 3)of the 6GF7 to output from the vert integrator network (just before the sync pulse was coupled to the plate of the vert osc thru a cap). that one was pretty critical.

[miniman82]

It's got a really bad vertical fold, but I've already replaced all the electrolytics. The vertical tube was really weak, so it got replaced with an NOS one same thing with the audio out tube. Not that I care about audio at this point, but it is one of those chassis that forms a voltage divider through the 6AQ5 audio tube, so I figured why not take care of it. I don't know if the linearity pot has just gone bad or I'm not adjusting the right one, but no matter what I do right now I can't get it to look right. I'm tempted to blame the output transformer, but I don't have another I can sub in at the moment. There's also a ,001/2kv cap still on the PCB that's original, I suppose that could be bad. It looks like it feeds vertical pulses from the plate of the output back to the multivibrator, so it's probably semi critical. I'll have a look at the resistors and swap out that cap.

[DaveWM]

if it has those blue plastic caps they are prob ok, all mine were. look for that 39k resistor and any others that are 5% marked. IIRC it was pretty resistor drift that caused the issue. really doubt a vert out xformer, check the vert osc plate supply from boost, look for drifted high 2.2 meg and check pot for height.

You do have the conv board plugged in right?