RESOLVED - WG6100 - deflection board issue due to heat?

Sprout

Well-known member
Joined
Jun 15, 2010
Messages
1,398
Reaction score
118
RESOLVED - WG6100 - deflection board issue due to heat?

Working out the last kinks in resurrecting a Tempest. Just got a fully refurbished HV cage and neck board in the mail from Andrewb (thanks much!), and dropped in a deflection board that I'd rebuilt from a carcass.

HERE is what I'm seeing after it's warmed up for a few minutes.

All diodes and transistors check out on the deflection board (P314, no input protection PCB). New filter caps. All connectors re-flowed. Have checked resistors around the X/Y transistors - nothing open, nothing that looks burned up. LV2000 installed.

New chassis transistors for X/Y -existing power transistors. All check out fine, consistently.

When I have a fan blowing on the deflection board, it's rock-solid, and looks excellent. No jiggling in the bottom/right corner, no general wackiness on the right side of the screen. Can play multiple games without issue - looks great.

So - should I let it heat up, break out the compressed air/freeze spray, and start hunting? Looking for clues as to what may cause this particular behavior to give me a place to start... Does this indeed seem like a deflection board issue, or do I have a flaky chassis transistor, or something else?

Appreciate any tips/insight!
 
Last edited:
Sanity check - Have you confirmed it isn't your game board? (Do you have a scope or other monitor you can use to confirm its not the game board?)

It does look like a deflection board issue, but the fact that it's intermittent is a little curious, as I've never seen one intermittent quite like that.

One thing I have seen that can cause that distortion is a bad R712 (3.3 ohm, 1/2 watt). It sounds strange, but I've seen it more than once, where they go open without burning up. You might try replacing it, just for kicks.

Aside from that, it's hard to tell if it's power supply related or not. It doesn't look to clearly be affecting only a single axis, which suggests it's PS-related, but sometimes a bad transistor on one axis can screw up the PS as well, which gives the two-axis type of distortion you're showing in the vid.

The fact that its intermittent also suggests a bad connection somewhere. One culprit on these is the solder joints on the connector header pins. But not just reflowing them (which is definitely an issue if they are original), but you can get cases where the traces crack right around the reflowed solder joints, right where the joint meets the trace. So the breaks can be hard to see, as it's the trace that actually cracks, not the solder.

So, you might want to just tone out every connection with your DMM, from the top of the pin on the parts side, to some other spot on the trace, on the solder side, away from the pin, so you know you're testing the whole path. This also applies to all of the joints around the LV2000.
 
Hmm, I think it's NOT heat related as it stops instantly. Heat remains for a while

I agree with Andrew's header/tracks cracks suspicion

A GND pin most likely because as Andrew said it's not X OR Y related seemingly but both / All
 
To address questions:

Output looks fine on my scope. Only occurs with the 6100 connected.

All headers on deflection board reflowed, and individual wire connections checked/ohmed out before the board was ever installed, but can certainly eyeball them again.

I've lightly sanded all header pins as well, although they didn't look damaged or pitted - just to ensure a good connection.

I should have mentioned additional work - all electrolytics replaced on the game boards, ARII rebuilt/recapped, big blue pulled/replaced.

Fan is blowing ONLY on the deflection board area. If I turn on the fan, power up the game, it's solid. If I turn off the fan, it starts to degrade in about 30 seconds, and gets to the point shown in the video within a few minutes. If I turn the fan back on, it slowly gets better, to where it doesn't completely lose the image - just has the jiggles...

Heat causes expansion, so a cracked trace could still be a culprit, but was leaning more towards something that was failing under load/stress/heat given the dramatic difference the fan makes between jiggling/loss of image, vs. solid vectors/no jiggle.

I'll definitely check that resistor and the headers/wires again - I believe R612, R613, R712, and R713 were all 3.3's, and all checked out previously.

I'll update the thread as I work through this otherwise!
 
Stick a piece of cardboard down the middle of the deflection board, and only have the fan blow on half of it, to at least try to isolate which side is appearing to be heat-sensitive.

You can also try touching each transistor on the board (or looking at them through a thermal camera, if you had one), to see if one is getting noticeably hot. (Aside from the 50V AC at the input fuses, there is only ~30V DC on the deflection board, which won't hurt you.)

I still lean slightly more toward the cracked joint or resistor theory over the bad transistor, but it could be either. (Also, replace R712 anyway, even if it measures ok when cold.)
 
I'm thinking game board. Reseat roms and vector ram? I haven't seen that in a monitor before. I'm not an electronics wiz but I have seen a fair share of 6100 monitors. I can certainly be wrong.

While the monitor is running, try GENTLY wiggling the top of the black heat sinks on the deflection board to check for a cold solder joint. You'll either need to pull the monitor or have a mirror or an assistant.

I'd like to see what the end result is. Be sure to post back your findings. Some really good advice up above from others as well. I understand that it only occurs without the fan but I think that is a red herring.
 
Is the LV2000 regulating. :)

ding ding ding ding... we have a winner...

As I was letting it heat up to do some more testing, I noticed the -26 LED go dim on the LV2000...

Put my meter on it, and instead of being around -26, it was pretty consistently around -21.5 or so. The +26 was rock solid by comparison. While in this state, I was getting "normal" video, albeit with the wiggles/jiggles shown in the video.

As I waited for the LED to dim, I kept my leads on the -26, and once it dimmed (which coincided with the completely scrambled video), saw the voltage drop - to about -3.6v. Yeah, that might be a problem...

Grabbed the freeze spray/compressed air, and had my wife watch the screen - any time I popped that area of the LV2000, it immediately came back up, jiggles minimized, and we were back to relative normalcy, until it heated back up. Did this several times, and it consistently brought it back up.

Then tested the -33 input - both the -33 and +33 never dropped below about 31.5 respectively. This confirms the issues appear isolated to the LV2000.

So - looks like I got a bad part, and need to get it swapped out!

Thanks for the help and suggestions; I'll follow-up once I get the replacement installed.
 
Ok, well, I *thought* we had a winner...

Swapped in the replacement LV2k - same issues once it heated up.

In checking the -26v output, it was still eventually getting pulled down to -21v, and then having the same issues as the other one. Checking Q103 (2N3792, which was new) always showed it as good, but since that is the ONLY thing hanging off the output, I swapped it out anyway. That seemed to do the trick - monitor came up, and stayed up. Put multiple games on it - even after everything had gotten quite hot around back, vectors were good, zero issues.

So - happy about that. However, now my gameboard decided to go down - new topic to address that posted here...
 
I have heard of this before, on a friends machine, where one light on the LV2000 goes out along with part of the screen

I made a test rig for him to narrow it down, as he was interstate. IE 2 x Transistors on a stand alone heatsink (IE: 2N3792 & 2N3716)

Plugged it in to the deflection board and only one position righted the problem

Ended up being a hairline crack in the transistor holder. He replaced the holder which fixed it
 
Last edited:
Back
Top Bottom