|
A lesson learned... the hard way
Yesterday, I changed a lot of resistors on the color and deflection boards of this set. When taking ohm readings at the tube sockets, the path taken is from the tube socket to ground or to B+. Easy way to find resistors off spec. However, there are many in there that won't be in the path between the points mentioned above. After checking several, I found quite a few that were more than 20% off from the marked value. I found several that had doubled or trippled in value. Many of them looked toasted, and i'm sure many of those probably drift in value as the set warms up. Altogether, I changed out about 40 resistors. I felt pretty good thinking that things would work well once finished.
Not so.
I powered the set up and waited to see how much smoke would pour out. No smoke... sound came in and then screen lit up. Looking good so far. Naturally, after changing that many resistors, I figured I'd have to make adjustments. The picture was terribly out of sync. I couldn't get a lock on the vert or horz. Tried putting in another sync tube... still no good.
I powered it down and did a quick check of resistance reading at the sync socket. The readings were WAY OFF. One of the pins should read 4 meg to going to ground... I got a reading of 30K. The other pin readings weren't much better. Well crap... I instantly figured I put resistors in the wrong places.
Back to the bench.
First, I looked at each resistor and compared to Sams. Then I went back and checked again. Then I checked under the PCB to make sure I didn't short any of the printed lines together with solder. Everything looked fine. I then traced each printed circuit from the tube socket to see if I could find a place in which I wired something wrong. Everything still looked good. Then I unsoldered resistors and caps surrounding the sync tube socket and checked them individually. They all checked fine. Hmmm... I'm running out of things to check.
Lastly, I decided to check the tube socket itself. Come to find out, something was wrong here. I was getting resistance readings in places I should get no reading. I took a piece of wire and worked it under the socket... thinking I'd find something wedged between the socket and PCB... perhaps shorting socket pins togther. Nothing turned up.
Finally, I unsoldered the socket the removed it completely. It looked fine. Took readings across the pins... nothing. Okay, so this means the socket is fine.
Hell... there's nothing left to check! Or is there?
I'm sitting there staring at the PCB. Keep in mind that the caps and resistors surrounding the area where the socket goes have been removed. For the hell of it, I placed my probes between a couple of the connections... one of which only has a few solder connection points... a couple of resistors and a pin to connect a wire. I looked at the meter and found a reading of 200K ohms. Okay... that's not possible... this meter must be malfunctioning. Got another meter. Same reading. What the hell??? How is the PCB conducting through the board? I checked from that same point on the board to chassis ground... got 500K! This would explain my sync circuit not working right... but why?
I got some alcohol and a q-tip to make sure there wasn't any dirt... even though it looked clean. Hell, it should be clean... I sprayed that socket a couple of days ago. Wait a minute... I sprayed that socket! There was oil in that spray! So, I cleaned the board with the alcohol and noticed my resistance readings changed. Now they're reading about 1 meg. I can't get it any better than that as of yet.
Sooo... I'm guessing that the oil in the spray has soaked into the PCB where the socket goes... probably thru fine hairline cracks. That's the only thing I can think of that would allow conductance thru the board.
And all this time I was expecting to find I wired a resistor in the wrong spot!
So, when you guys spray tube sockets, make sure you're using an electrical spray that DOES NOT have oil in it. It could really screw up your PCB.
__________________
Charlie Trahan
He who dies with the most toys still dies.
|