19V200 HV problems

davemcl

Member
Joined
Mar 29, 2010
Messages
128
Reaction score
0
Location
New Jersey
19V2000 HV problems

I have a V2000 that came out of an omega race running in an asteroids game. Game board known to be good. I'm not a newbie to the hobby, but a newbie to working on the vector monitors.
I've read the FAQ & Guide forwards & backwards a couple times, searched the monitor forums & don't seem to find this problem anywhere. The def. board & HV board have both been recapped, transistors & diodes seem to check out OK & voltage readings on the deflection board at P500 all checked out OK with the exception of pin 2 which read a little high, I think around 33.6V instead of 30V. All voltages were tested at P900 of the HV board & checked out,, pin 8 read about 39.7V instead of 38V listed in the manual. Voltages were tested with all connectors plugged in at the back of the connectors with a digital multimeter. I've turned down the brightness & contrast on the deflection board to the minimum to get a decent picture & turned the focus adjustment at R909 counterclocwise to the point that the picture is still in focus. I don't presently have a working high voltage probe, so I haven't messed with the votage adjustment at R905 on the HV board. The HV board has a new HV diode in the oniginal holders, dielectric grease on both ends. I can play 4-5 games & then the 1.5A fuse at F102 blows. I did have 1 time last week when the fuse at F600 blew, but after replacing, it hasn't blown again. This time, the fuse blew at F102 & I saw some smoke come from the HV unit. It was mounted in the metal cage mounted on the monitor frame, perforated cover left off so I could see what was going on.
I noticed the resistor at R900 was hot enough to fry an egg on & that some of the dielectric grease had cooked out of the HV diode end by the holder on the red anode wire side. I also notice that a small peice of the rubber or plastic cover on the HV trasformer (flyback) chipped off. I'm 99% sure it wasn't that way before I reinstalled the monitor this time. I have an extra HV board, but before I take the flyback off there & install it or just pull the HV diode & put it on that board, I'd like to know what's causing this. I'm thinking maybe the CRT is getting too high a voltage (higher than 14,800V) the manual says 14,500V give or take 300V, but since no probe, I can't test it. Would the excessive voltage cause R900 to heat up & maybe cause F102 to blow as well? I'm stumped, I'm hoping one of the experts can shed some light or make some suggestion where to look or what to check for faults.
I should mention that I've tried the monitor on a working astroids deluxe game that never gives me any problems with the G05-802 installed in it & tried it with another working asteroids board in a different machine. It does the same thing regardless, so I think it is isolated to being an issue with the monitor. The big caps on the defletion board have not been replaced, but appear to be alright.

Thanks
Dave
 
Last edited:
Hope no one minds if I bump this back up to the top. At least it makes me feel less dumb that no one has responded. I'll let this question stand for a while until I take the shotgun approach & replace all the transistors & test all the diodes on both boards one more time. I am working on borroring a HV probe as well. When I can test the anode voltage, I'll know a lot more. Either way, I'll continue to document my findings, so the info. will be here if someone else has a similar promblem with a vector monitor in the future.
Thanks for reading....
 
also....if a moderator could edit the thread title to read "19V2000" instead of 19v200, it would be helpful for someone doing a search in the future. Sorry, the keys on these 'puters are a bit too small for me & I sometimes mistype.
 
If r900 is the largest resistor at the top of the high voltage unit,they do tend to run hot.Have you re-flowed the solder joints on the connector yet? Also I think it is pin 5,but you can set your 90 volts there,without needing a high voltage probe.
 
Thanks for the response vctrcollectr. I reflowed the connectors when I recapped the HV board & deflection board, so they should be good. I think that you hit on something with the 90V adjustment. I think it might have been a little higher than 90V at pin 5, I'll have to put it back together & recheck. Are you saying that the pot at R905 is an adjustment for the 90V? The way I was reading the FAQ & Guide, I interpreted it to mean that the pot at R905 was an adjustment for the 14,500 V going to the anode cup on the CRT & that that should be within 300V of that number for proper operation. I probably misread it, I'll give that part a re-read. I got a hold of a working HV probe to check the anode voltage from a friend, so I hope to give this another look. I'll report back my results.
 
Last edited:
The pot adjusts the 14,500 volts and the 90 volt line at the same time.From what I understand,if you adjust it to 90v,then it should be close to the 14,500 that you need.
 
On p 41 of the FAQ & guide it states:
On the Wells-Gardner 19V2000 monitor, the anode voltage is supposed to read 14.5KV High
Voltage. That's 14,500 volts with the beam current at zero. R905 adjusts this output voltage. To
read this voltage, your meter needs a separate high voltage probe. Follow the directions
EXACTLY as stated with the literature that comes with the probe.

so I don't think the R905 pot would have any effect on the 90V supply voltage.

again, just documenting this for others who may have a similar problem in the future.
 
Thanks for the help, I look forward to getting this one going 100%. I'll check back in when I get some time to get back to it.
 
19v

I think you have hit on the problem. You can check for the 90v on pin 5 and at the same time you adjust that voltage you can measure the beam current and it will increase or decrease relative to the 90v return voltage adjustment on the same pot. That would case the issue you described.
 
put the HV probe on it & it was running a bit high...15K- 15.2K. It was a bit of a pain getting it tuned in to 14.5K, but got it set at around 14.6K & it ran a lot cooler & for about 25 mins until smoke started pouring out of the cap at C900 like a pothead on payday.
I think the problem is with the pot at D905, since it was a bit of a bear to get dialed in. I've seen the pots on the HV section of a Bally pinball wreak havock if they're starting to go. I hope that's what it is, because the next thing would presumably be the circuitry that regulates supply voltage.
Good thing I ordered pots, I hope that does it.
 
Sorry for necropost BUT
This was a fix I did a while ago and before I forget I'd like to post it here. R900 was getting hot and I had no HV(at least no spark when discharging)/no image.
After looking over the HV for a while, I saw that the insulator that goes through the hole and against the nut was missing on Q900 (one of the transistors on the side of the case).
In the pics you can see that my Q903 has an insulator and Q900 didn't, and so Q900 was shorted and breaking the regulating circuit.
Fixed it by quickly measuring and 3D printing a spacer.
I was surprised at how quickly I fixed the HV problem, and hopefully this helps somebody.
 

Attachments

  • PXL_20210201_191957091.jpg
    PXL_20210201_191957091.jpg
    1.5 MB · Views: 10
  • PXL_20210201_191953180.jpg
    PXL_20210201_191953180.jpg
    1.6 MB · Views: 10
Last edited:
That's great, but you should buy a proper TO-220 silpad (or use a mica + grease, but the silpad is better). You want that insulator material to be highly thermally conductive, which silpads are designed to be.

Not that it's super critical in this case, as those transistors don't get scorching hot, but in general, you want to use the right parts.
 
That's great, but you should buy a proper TO-220 silpad (or use a mica + grease, but the silpad is better). You want that insulator material to be highly thermally conductive, which silpads are designed to be.

Not that it's super critical in this case, as those transistors don't get scorching hot, but in general, you want to use the right parts.
Sorry, I meant the through-hole ones. Forgot to upload the pics, will edit
I would never use plastic instead of the mica pads.
 
Back
Top Bottom