Sony Kp-46wt510 convergence issue, can't shift red up and down

**jmayer** · **10-04-09, 03:24 AM**

My set recently displayed a rather typical convergence issue with RGB seperating, the flash focus tool did nothing to alleviate the problem, and I couldn't move red or blue up/down or left/right in manual convergence. I ordered a kit for my tv from tvrepairkits which I believe is familiar to this forum and known for using quality suppliers.

I've replaced both STK392-560 IC's, using thermal paste, I've re soldered them twice to make sure. Checking to make sure there are no bridged connections.
I've also replaced both pico 5amp fuses on the G board, one was dead, just did them both.
I've removed and tested the 6, 3.15 amp pico fuses under the heatsinks on board D, all test good. I've removed or lifted the 30 resistors under the IC's, all test between 4.7 and 4.9 ohms out of circuit.

Replacing the chips didn't have much of an effect until I replaced the 5amp pico on board D, then Green and Blue came into " alignment" but red was pin-cushioned so it lined up in the center but as you go out to the edges it bends out away from blue/green. I can manual adjust Blue up/down and left/right, Red I can only adjust right/left, up/down has no effect. In manual convergence mode and in the factory menu, the red grid will only respond to horizontal shifts, not vertical, but the numeric values change in the menu.

Just looking for something else to look at, doesn't really seem to be a way to test the voltage on the IC's in with the board installed since they're behind that heatsink, any ideas?

I'm definitely a hobbyist, picked up soldering to repair arcade machines and pinballs, this is my first foray into television repair.

I'm also e-mailing with the support team at tvrepairkits, but the more information the better.

thanks,

jason

**lcaillo** · **10-04-09, 06:17 AM**

Your thread was moved to the Sony sub forum to maintain forum organization and to make information easier for others to find in the future.

If you ran the set with the open fuse in the power supply you very possibly damaged one of the new chips. They don't like having one power supply rail down. You could also have damaged one of the output fuses. These can be very fragile, and should not be removed to test. A fuse should read very low resistance in or out of circuit if it is good. Otherwise it is open and should be replaced. It is also not necessary to remove or lift the resistors. As per advice in the convergence repair thread, they should be tested while the chips are removed and the convergence yoke wires are disconnected. Chances are very good that you damaged a trace or a component in the process of servicing the set.

When I need to test voltages at the chips in these sets, I use a long needle probe and/or use points on other components near the ICs. You can also test the output of the DCU at the connector between the A and D boards.

**jmayer** · **10-04-09, 10:58 AM**

Here's the events of the repair.

I replaced the IC's.
powered the set on, the symptom changed but didn't improve.

I found the dead fuse on the G board, replaced it.
when I replaced the fuse, while soldering a new one on the board I did get a spark
from the tip of the soldering iron to the trace, I'm guessing since it was so soon after powering on that power board had some juice still in it. It wasn't large an there wasn't any smoke, could this have damaged something else? I tested both fuses after this and both tested good, I also replaced the other one just in case.

Powered the set on, green and blue seem to converge, red was pin-cushioned.
I could move blue in all four directions, but red would not adjust up or down.

I tested the 6 pico fuses below the heatsinks in circuit (are these the output fuses your post referred to?).
They all tested good.

I'm using a continuity test in my DMM for these fuses, I noticed in your reply you recommend a resistance test, is it possible continuity would test good, but resistance bad?

I then removed them all from the board to test continuity on. All good so I resoldered them back on testing them again in the board for continuity, all was good.

Powered on the set, no change, red pin-cushioned in same spot, no vertical adjustment.

I then desoldered and resoldered the two IC's without removing the IC's, just to make certain there wasn't any bad joints.

Powered on the set, no change, red pin-cushioned in same spot, no vertical adjustment.

I then downloaded the service manual to see if I could find what specific resistors had any control of the red. It appeared from the schematic, that R8045,51,59,63,66,69,73,78,81,85 were tied to the red. I removed all of these and tested them for resistance, all came back around 4.7, 4.8 ohms. I reinstalled them.

Powered on the set, no change, red pin-cushioned in same spot, no vertical adjustment.

For sanity sake I tested the continuity again of all the 3.15 amp pico fuses, all tested ok. I then lifted the leg of the remaining 20 resistors and tested them for resistance, all tested 4.7-4.8 ohms. Re-soldered them back on. I also desoldered and resoldered all the pins for the IC's yet again, making sure there are no bridged pins again.

Powered on the set, no change, red pin-cushioned in same spot, no vertical adjustments.

This is where I'm at now.

Since the symptoms didn't change from before I started removing and testing the 3.15 fuses and the resistors, I don't think I damaged any components or traces while doing that, just wasted a lot of time.
If I damaged a trace or another component, it would have been before the symptoms stablized, installing the IC's or replacing the G board fuses, but I think I was pretty careful, using a radio shack bulb desoldering iron, I looked each time to make sure nothing was up with the traces.

The only other thing to note, would connecting the RGB cables that run out to the various CRT boards incorrectly cause damage to something? When i first removed the connectors, I didn't do so hot a job of labeling them, so connected the red to the green or something and powered it on, noticing something was way off, I tracked them back to their respective assemblies and labeled them properly.

I'm considering swapping the two IC's to see if the problem shifts to another channel from red, since one handles red and half green, and the other handles blue and the other half green. I'm guessing if there is a problem with the chip, it will transfer when they're swapped? Does this seem a worthwhile step to rule out the chip?

thanks again for responding so fast and taking the time to read this.

cheers,

jason

**lcaillo** · **10-04-09, 11:41 AM**

Most likely you blew a chip by powering it up with only one power supply rail active.

You could also have some bad solder connections to the yokes or otherwise in the circuit. You could also have a bad DCU and you are getting no correction waveform to the IC but this would require a scope to test.

Just change the chips again, or at least the one with the red amp that is not working.

**jmayer** · **10-04-09, 01:47 PM**

Ok, I'll give that a try and post results,
will probably be a week or so before I get another chip and some time.

thanks again,

jason

**jmayer** · **10-06-09, 03:52 AM**

I went ahead and swapped positions of the two IC's before I ordered another chip.
Still getting the same error in the red channel, this leads me to believe it is not a bad chip.
I still can't adjust red up or down.

I also inspected the solder on the D board looking for anything bad, I found a bit of solder which must have dripped and was bridging a transistor on the bottom (i think i did this tonight when swapping the chips), cleaned that up, didn't see any other cracked or cold looking joints.

I get 7 flashes (6 steady and a quick one) on the timer/standby when I power it up after it being off for 10min or more.
(I didn't pay attention to what the flashes were before tonight)

If I turn it off and on again I usually get 4 flashes, rarely will get 6 or 3.

according to service manual.

7 flashes is Q8035,Q8038 shorted
There was an excess bead of solder which had shorted two legs together, I removed the excess solder, but still get these flashes, not sure if it's a memory thing?

6 flashes
+5 line is shorted or overloaded on the A, B Boards.
says symptom for this is no picture, but I get a picture.

4 flashes
+-15 V is not supplied D board
IC8003 is faulty A board
symptoms:
-has entered standby state after horizontal raster
-Vertical deflection pulse is stopped
-Power line is shorted, or power supply is stopped.

3 flashes
IC501 is faulty G board
IC5002 is faulty G board
Symptoms:
-has entered standby mode

I'm approaching the limit of my comfort zone I think, and am considering just having the set serviced professionally since I'm already in over my head in my time vs. repair cost. Just wish I wasn't so stubborn about solving problems.

cheers,

jason

**lcaillo** · **10-06-09, 05:02 AM**

Now that you have done so much work to the board and potentially created extra problem, many techs will simply not want to work on it. Finding problems that someone else may have created in servicing a set can be nearly impossible. It may be cheaper just to buy another D board. The problem is that you don't know if you created a failure in servicing it or if there was something else going on to start with. One could fix or replace the D board only to find that you have a bad convergence generator. If you are over your head in time vs cost, imagine someone else having to check everything you know you did, plus potentially finding something that you don't know that you did. Are you really willing to pay someone for their time to sort it out at this point. I would only take a job like this on an hourly basis, with a hefty up front deposit covering the first couple of hours at least. All of the normal assumptions one makes in troubleshooting a failure have to be reconsidered and may not be good when going behind someone else. For a tech it is a waste of time.

You have a mess. If you want to fix it I suggest you minimize the soldering that you do to the board. Pull both chips, check the resistors and fuses again, and clean up any solder problems. Look for cracks in the board, damaged circuit traces, and connectors that you might have misplaced or forgotten to replace on assembly. The set will run without the convergence ICs if you have the power supplies up, you just won't have any convergence correction. If you can get it to that state again, you know you have a bad IC.

**jmayer** · **10-06-09, 01:26 PM**

Having had to troubleshoot problems behind other people in a software related context, as well as in pinball repairs, I can understand a hesitance to go in behind someone.

"You have a mess. If you want to fix it I suggest you minimize the soldering that you do to the board. Pull both chips, check the resistors and fuses again, and clean up any solder problems. Look for cracks in the board, damaged circuit traces, and connectors that you might have misplaced or forgotten to replace on assembly. The set will run without the convergence ICs if you have the power supplies up, you just won't have any convergence correction. If you can get it to that state again, you know you have a bad IC. "

I've cleaned up the one solder problem I created with the bridged legs, I've inspected the board for damage and broken traces and haven't found anything, all connectors are on, I'm assuming you mean all the molex-type connectors? how do you misplace one? I tested all the fuses again in the board. I'll pull the chips again and test the resistors again, but I'm not sure what you mean about getting it to "that state again", without the IC's installed.

The tv powers up, that's never been a problem. Just the red is out of convergence and can't be moved up or down, it will go left and right. This is the same problem I had after replacing the IC's and the 5amp G fuses. Before replacing them, blue and red wouldn't go up/down or left/right. I'm assuming if I created an issue, it was in that initial step since the symptoms haven't changed since then.

thanks again, I appreciate the time.

jason

**lcaillo** · **10-06-09, 02:35 PM**

You described getting 7 or 6 flashes. This only happens when the set is going into shutdown. Are you talking about the number of flashes before an image comes up or a shutdown code? If you are just counting the number of flashes before getting a picture you are wasting your time. It varies. The LED flashes until the set comes up and the IK pulses from the three cathodes can be equalized. It is not always the same number of times. The flashing LED codes you referenced are diagnostics when the set will not start or goes into protect mode.

Your posts are very confusing. Either you have a shutdown problem /startup problem or the set is running, which is it?

**jmayer** · **10-06-09, 02:48 PM**

Ahh.. gotcha.
Yeah, I was just wasting my time counting the number of flashes before the screen comes up. Like I said, I hadn't ever paid attention to the number of flashes before turning on in the past, always thought it randomly flashed before the tv turned on ever since I first bought it. But when I saw the service manual flash codes chart I thought they were significant before the set comes on.

The set did not power up once after I accidentally dripped a bit solder on that transistor, once I cleaned it, it powers up normally.

The set is running, sorry for the confusion.

**lcaillo** · **10-06-09, 02:54 PM**

So, you swapped the position of the chips and you still have the same symptom on the red vertical? If this is the case, you need to determine if the problem is in the output circuit or the convergence correction from the DCU. You can try moving the position up and down on the red and you should see a change in the d.c. voltage at the red vertical line on the bridge connector between the boards. If you do see a change, try looking at the red v connector to the yoke.

**jmayer** · **10-06-09, 04:34 PM**

Yeah, I swapped the IC positions and still have the same symptom on the red vertical.

I'll check those spots this week and post the results.

thanks again,

jason

Advertisement