WG K5515 intermittent HV Shutdown

[gwarble]

Hey all,
I've had a few monitors over the years exhibit this behavior, sometimes a cap kit fixes it sometimes not, but I'd like to learn the proper way to test and find the real problem...

In this case its a 25" Wells Gardner K5515 (medium res from a 720°) that I think is going into HV shutdown, intermittently and only when "cold". Sometimes its fine but sometimes I have to power cycle it a few times before the screen will stay on. Usually its "shutdown" before any raster is visible, but on the third or fourth try you might see raster/picture collapsing when it shutsdown. I had this problem with this monitor a few years ago, haven't noticed it recently and forgot about it but in testing to loan out the machine its apparently still doing it. Once it stays on for the first 15 seconds or so without shutting down it will stay on no problem for as long as I keep it on or power cycles, the problem only returns after it sits off for a while.

I know some monitors have an HV Shutdown cutoff pot, usually glued in place, that might need a little tweak, but I don't see that on this monitor. I will start by checking B+ while running normal and while in shutdown, but looking for tips for what part of the circuit could be causing this.

Thanks,
Joel

edit: should be str380
B+ measures 123vdc when running, but after shutdown happens it jumps to 151vdc
i guess this pot pictured is the hv shutdown cutoff... not sure if i should adjust hopefully some experts can advise

 

[bakerhillpins]

Reading through your problem description I'd say it sounds like the chassis has cold solder joints and needs an inspection and reflow. If you haven't capped the chassis I suggest you do that at the same time.

 

[gwarble]

thanks, good advice in general but I'm trying to understand better what would cause what i think is hv shutdown, or even know for sure thats what I'm seeing.
next time i pull the chassis i will reflow everything, but the fact that it works fine after a few fire ups and them never fails has me a little skeptical. Its also something I almost surely would have done last time I pulled it but its been years

not many threads on this chassis, so i stumbled on my own thread from when i got this game in 2011 and the monitor was completely dead... turns out once i got it running i never followed through with the voltage regulator... but although the B+ is correct at 123vdc, the VR installed is an STR381 (what the 19" uses, which is 130vdc) but this monitor is supposed to have an STR380 (123vdc)... I still have the one I bought for it but never installed, so I will swap that in next and see what happens to the B+. I would expect it to drop by 7v but then maybe I can adjust the horizontal width coil to bring it back up to 123v?

dead WG 25K5515 (medium res) in my 720°

monitor gets 120, no glow fuse is blown (big one on the left side of chassis) but HOT does not test shorted... so that would lead me to the flyback? should i start by just changing the fuse (or putting in a breaker for testing) and see if it was a fluke/environmental? or does that fuse being...

forums.arcade-museum.com

edit: swapped in str380, first attempt came up 129.5v with a picture, powered down and adjusted horizontal width coil thinking it was related from that old thread... now its shutting down again (151v)

 

[gwarble]

hmm... fact checking a little bit and both the WG 25K5515 and 19K4915 (medium res) manuals i found online reference STR380 (page 12), which should be 123v (and STR381=130v)... I was under the impression that B+ was supposed to be 123v for the 25" and 130v for the 19"

with the STR381 in place, despite my shutdown issue, when working B+ is spot on 123v
with the STR380 in place, I was only able to run it once and it was at 129.5v

seems backwards, but I don't know, about to put the STR381 back in

 

[gwarble]

str381 back in, b+ back to 123v, monitor working... i don't know what to make of that discrepency

edit: very few posts about this chassis, but this one i think confirms what i'm seeing but don't know what he searched:
"OK, from a quick search, I'm seeing two visual differences between the 25k5515 & 19k4915 chassis: the 19k4915 uses STR380 VR, the 25k5515 uses STR381 VR. Looks like the 19k4915 chassis, has a metal plate that attaches to the top of the flyback, from the chassis wall. Can someone verify that this is accurate please?"

Price Check - 25k5515 Chassis Working

I just removed and replaced solder all over this chassis and neck board, cleaned up the chassis a bit, replaced the missing AC connector. This has not had any other components replaced. I tested this in my Hard Drivin cockpit, dialed in the colors, adjusted the focus and brightness. Great...

forums.arcade-museum.com

edit: just to clarify for future visitors, the metal plate is also in place on my original 25K5515, but another visual indicator might be the extra ceramic resistor on the outside of the left chassis wall on the 19K4915 which isn't on the 25", and the extra heatsink around the VR only on the 25"

does anyone have a working 25k5515 they can check the vr (bottle cap on the left of the chassis)?


edit: https://forums.arcade-museum.com/th...edium-resolution-monitors.530450/post-4884476
edit: thanks for chiming in bakerphilips, i see your posts in that thread and you seem familiar with this chassis

edit: is the manual and a lot of threads wrong and the 25K5515 supposed to use a STR381 (rated 130v) with B+ at 123vdc (measured at R503 end farthest from the tube)?

 

[bakerhillpins]

thanks, good advice in general but I'm trying to understand better what would cause what i think is hv shutdown, or even know for sure thats what I'm seeing.
next time i pull the chassis i will reflow everything, but the fact that it works fine after a few fire ups and them never fails has me a little skeptical. Its also something I almost surely would have done last time I pulled it but its been years

So the problems you are describing are most often caused by bad solder joints. It's not just knee jerk, that most likely IS the issue here. It just happens to also be good advice in general because it's a frequent issue - these chassis ran long hours for many years and the solder around elements that got warm/hot (e.g. power resistors among other things) fatigue and need reflowed. Cap kits help with restoring a poor picture and reduce stress on other components with older operating units.

hmm... fact checking a little bit and both the WG 25K5515 and 19K4915 (medium res) manuals i found online reference STR380 (page 12), which should be 123v (and STR381=130v)... I was under the impression that B+ was supposed to be 123v for the 25" and 130v for the 19"

Both of these monitors are essentially a Frankenstein to fit a niche that seemed to be opening up when "graphics" was becoming the hot button. If you look them over it's all mods on a STD resolution model to make them Med Res.

Both of these chassis should run on a STR380 at 123vdc.
There are MANY errors in the manuals as a result of the mods to make Med Res

Let's document the K4900 series medium resolution monitors

Running with the theme started here, I'm going to add some info that I've found while working on my k4915s. But lets not limit it to 19" models. I'm midway through rebuilding 2 chassis and in the process I'm pulling and testing the power resistors since these can tend to be issues on the...

forums.arcade-museum.com

with the STR381 in place, despite my shutdown issue, when working B+ is spot on 123v
with the STR380 in place, I was only able to run it once and it was at 129.5v

str381 back in, b+ back to 123v, monitor working... i don't know what to make of that discrepency

I've got a VR marked as an STR380 (green printing) in the basement that regulates at 130vdc on any chassis I plug it into. It looks to be OEM so I suppose it could be a factory misprint. Threw me for a loop for a bit.

Every time you pull the chassis and manipulate it you flex the board and change the problem around. You should stop what you are doing and inspect/reflow before you contine to chase ghosts.

 

[gwarble]

So the problems you are describing are most often caused by bad solder joints. It's not just knee jerk, that most likely IS the issue here. It just happens to also be good advice in general because it's a frequent issue - these chassis ran long hours for many years and the solder around elements that got warm/hot (e.g. power resistors among other things) fatigue and need reflowed. Cap kits help with restoring a poor picture and reduce stress on other components with older operating units.

Every time you pull the chassis and manipulate it you flex the board and change the problem around. You should stop what you are doing and inspect/reflow before you contine to chase ghosts.

i totally agree with you, and when I do pull the chassis I will of course reflow all header pins and large component legs, I always do and probably already did many years ago when I pulled it.



and this is a potentially different issue than the VR... but when I saw it still had the str381 i figured B+ would be higher than it should be (by 7v, meaning closer to the HV shutdown point, which seemed related). I'm trying to understand why it was at 123v, but swapping in (you can do this without pulling the chassis) an str380 brought the B+ UP to 129.5v before going into shutdown. That thread also has a lot of 25K5515s with 381, so I'm a little skeptical at the moment that maybe the manual was copied from the 19" and maybe should have the 381?

If the voltage regulator was mislabeled by the factory or counterfeit, this would mean they would both have to be wrong? My notes say the STR381 in there now was either factory or put in during a rebuild in '96 (before i owned it), and the STR380 i swapped in was bought in '11. I did buy some extras of each so i might test them all and see what results i get.

 

[bakerhillpins]

I think some things are getting lost in translation here.

19K4915 and 25K5515 use the STR380 (123v) voltage regulators. IIRC it will actually RUN with either because the spec on the HV shutdown is sufficiently over normal operation range to avoid false triggers.

I have a mislabeled/counterfeit part in my basement. Was only used as a point of reference to illustrate a point about parts and weird readings.

. I'm trying to understand why it was at 123v, but swapping in (you can do this without pulling the chassis) an str380 brought the B+ UP to 129.5v before going into shutdown.

If the VR socket is bad or solder joints are cold you will see LOTS of oddball problems.




Inspect and reflow, return to your questions.

 

[mecha]

I think some things are getting lost in translation here.

19K4915 and 25K5515 use the STR380 (123v) voltage regulators. IIRC it will actually RUN with either because the spec on the HV shutdown is sufficiently over normal operation range to avoid false triggers.

I have a mislabeled/counterfeit part in my basement. Was only used as a point of reference to illustrate a point about parts and weird readings.



If the VR socket is bad or solder joints are cold you will see LOTS of oddball problems.




Inspect and reflow, return to your questions.

I have seen the solder to the legs of the voltage regulator socket crack. I would do both that and the HOT one.

also overlooked is the B+ jumpers along the edge. probably this weekend I'll get around to my new tutorial video.

 

[gwarble]

seems like every thread (which is few) has contradictory info on the str381 vs str380 on this monitor. edit: I know the manual states str380, but i have now seen more instances of actually being populated with str381 than not. I'm under the impression B+ is supposed to be 123v, but can't find anywhere this is conclusive either

would love if anyone with a working 720 could check whats in their chassis (and their b+)

yes the VR socket legs and connections would probably be something i didn't reflow, will inspect and do so... the 4915 flowchart does mention those b+ jumpers up by the tube and they are on my list to check when i pull the chassis

 

[gwarble]

of course the most fun part of intermittent problems... swapped the VR back and forth and no failure yet, but I have to leave it off a while to be sure
or maybe moving the VR holder in the process adjusted the compromised solder joints


although the 5515 and 4915 manuals both list the str380 while the std res 4901 lists str381... many examples across 15 years of threads have confusion and or reference to this but actual chassis' more likely having str381 from what i can tell

and i can't find ic501 on the schematic, but this is interesting:

4900 is pretty well documented as 130v B+ using an str381 (rated 130v)
4915 is less well documented as 130v B+ but using an str380 (rated 123v)
5515 is poorly documented as 123v B+ using: manual says str380, klov threads conflict, most examples have str381 installed i think

should B+ directly match the VR rated output on every monitor or does more on the circuit affect the B+?
if so, are 4915 and 5515 supposed to have B+ of 123v?

i think i will try reading the output (pin 2) of the VR directly and see if i can get in there (edit: tried, same as B+, 123v in my case even with a VR labeled str381)
edit: after testing this (touching the vr socket leads) i got shutdown again, maybe confirming those cold solder joints on the vr socket. when in shutdown even the vr output leg measured the 151v same as b+ in shutdown)

unrelated but check: 19" K7000 chassis use a 123v STR3123 regulator, while 25" use a 130v STR3130