(SOLVED) WG-K7400 Intermittent Loss Of Blue Emissions...

[MKFan4Life]

Recently got a WG-K7400 25" from a fellow a couple hours away in Georgia.  He showed it running and looking sweet at his house.  After the ride here (and getting a temporary AC connection going since I am awaiting an adapter for the chassis AC connector), I first noticed a little static coming and going at random times in the display.  Then sometimes it would get a little dark and then come back in just like nothing was wrong (I think this was when it was losing a color emission, not sure which at that time).

I tapped lightly around on the neckboard and sometimes it seemed to affect these problems.  I noticed one solder joint that looked cold and like it needed attention on the neckboard, so I powered her down.  Then I disconnected the neckboard, and reflowed that joint (it was for a post that a red wire coming from the flyback plugged onto) and several others that looked iffy to me including all the points for the RGB adjustment pots.  (--BINGO! Either that, or I was attempting to say "before" but it was too many letters to type-- doing this, the Green Drive pot seemed like when adjusted it wanted to vary up and down a little unless you had it "just so").  After doing this, the monitor ran flawlessly for about 2-3 days where I would turn it on and either play the game or let it run for a few hours to make sure the problem was gone.

Then this afternoon, I wanted to show it off to my brother-in-law since he saw the old one and it's weak output.  I fired it up and he was impressed... for about 2 minutes.  Then it suddenly went a bit dark, but still clear and well-focused.  I was like crap.  After he left I went into the Mortal Kombat 1 monitor test screens and checked the color bars.  Blue was GONE!  No blue output at all.  The red and green were just fine.  At first I was like, "Not another Blue Gun failure!" (last tube had bad blue gun)  So I let it run about 10-15 minutes... no change.  I cut it off and unplugged it.

About 10 minutes later I fired it back up.  Same scenario.  No blue output.  I let it run about 5-10 minutes... checked it again, and the blue had come back like nothing was wrong.  I readjusted the flyback's Screen and Focus controls (I fiddled with them when blue first went out) and it has been running fine for about 45 minutes now.

Since the reflowing of the joints on the neckboard a few days back, there has been absolutely NO color OR static problems and it has looked GREAT until now.  I have a rejuvenator, but don't want to use it unless it's really needed since this display seems to have fine emmisions when blue is there.

The guy I bought it from was an electronics engineer and told me he never rejuvenated because tubes usually died soon after and that he wouldn't cap this chassis because it wasn't that old (thought they were over 10 years old?).  I can tell there have been caps replaced, because some have either old glue residue or, (I HOPE NOT!), electrolyte from old leaking caps near the newly-mounted ones on the pcb.

What could cause this?  Help me oh "masters of all that is arcade monitor-related"!  

[smartbomb2084]

Whenever I have had color problems with these monitors it is usually the color drive transistors on the neckboard, Q503 for Blue in this case, or it is cold solder on the large resistors connected to them. They just cook themselves until done.

These transistors 'leak' when heated and often saturate the color associated with it. Although this is not your particular problem, it should be investigated and eliminated. If you don't have a new transistor to replace it with, try switching red with blue for diagnostics sake. Heed a bit of warning though, if your neckboard is severely discolored from heat in this area, it is best to wait until you have a new replacement to use or the traces on the board will just lift and melt from any kind of desoldering/soldering heat.

[MKFan4Life]

Whenever I have had color problems with these monitors it is usually the color drive transistors on the neckboard, Q503 for Blue in this case, or it is cold solder on the large resistors connected to them. They just cook themselves until done.

These transistors 'leak' when heated and often saturate the color associated with it. Although this is not your particular problem, it should be investigated and eliminated. If you don't have a new transistor to replace it with, try switching red with blue for diagnostics sake. Heed a bit of warning though, if your neckboard is severely discolored from heat in this area, it is best to wait until you have a new replacement to use or the traces on the board will just lift and melt from any kind of desoldering/soldering heat.

Just went to take some detailed pics of this neckboard and guess what I noticed?  The Blue Q503 appears to have a VERY loose heatsink.  Aren't these supposed to have some type of thermal compound on them?  It's wobbling all around.  Wouldn't that cause it to overheat after a period of time if it's not making good contact with the transistor?  The other two for red and green are firmly in place on the transistors.  I will upload detailed pics of both sides of neckboard soon... wife calling me to dinner. 

[newmanfamilyvlogs]

http://forum.arcadecontrols.com/index.php?topic=103165.0
http://cotmm.org/?p=54

i had the identical problem

[grantspain]

ay,change the drive transistor as a matter of course

[smartbomb2084]

Aye

[MKFan4Life]

http://forum.arcadecontrols.com/index.php?topic=103165.0
http://cotmm.org/?p=54

i had the identical problem

Thanks for your links to your similar problem.  The traces look pretty good (there is some darkening from the heat of those transistors though).  I don't have a spare transistor guys, or I would do the swap immediately.  I guess I could, however, go ahead and try swapping it with one from either red or green like was stated earlier in the replies.  Then I imagine if blue stayed steady and say, red for instance, started crapping out the same way (if I swapped in blue's), I'd be sure it was the transistor.

Do any of you think, though, that the heatsink being so loose could simply be causing it to overheat and exhibit the problem even though the transistor might still be in good condition?  Or would the overheating already have caused damage to it?  I just don't have a spare on hand to swap in.  I do have a known good K7197 chassis I COULD rob one from if they are compatible.  I just don't like butchering known good equipment, especially if there's a chance I might kill the robbed transistor troubleshooting the K7400, lol!

Here's some pics of the neckboard:









The last pic is a close-up of the solder joints for the color transistors.

THANKS!

[newmanfamilyvlogs]

Don't have the link here on my phone but I got those transistors for .63 each on mouser, I think.
I think the highest price I saw was like 1.30 each. Might as well do them all. I had a spare neckboard board did a working swap, to ensure itwas the problem but then the green went out so I opted for all new.

[MKFan4Life]

Don't have the link here on my phone but I got those transistors for .63 each on mouser, I think.
I think the highest price I saw was like 1.30 each. Might as well do them all. I had a spare neckboard board did a working swap, to ensure itwas the problem but then the green went out so I opted for all new.

I'm new to this part of monitor repair, so I guess that means you can't use the transistors (in general) from just any old standard res chassis, right?  I will definitely order some.  Just hate waiting don't you?  Lol!  Especially on a weekend when I'm all ready to work on something like this.  But all good things come to those who are willing to wait, eh?  Thanks!

[MKFan4Life]

When I get the new transistors, if I have to reuse the old heatsinks off the others, what kind of thermal compound will work?  Can I use the type computer cpu's require, or will that not work?  I usually use a generic silver paste I imagine similar to "Arctic Silver" on pc cpu's.

cotmm68030, I hope the Mouser Part#512-KSC1507YTSTU is what you ordered.  I found it at this link on Mouser:

http://www.mouser.com/_/?Keyword=KSC1507&FS=True

It is the top one on that list (only one showing in stock).  Here is a link to that exact one on Mouser:

http://www.mouser.com/ProductDetail/Fairchild-Semiconductor/KSC1507YTSTU/?qs=sGAEpiMZZMtn7beagP2hIdb0BRj3v8d8vtakSiZO1QI%3d

I am ordering 3 of them to head off any future failure.

[newmanfamilyvlogs]

None of mine had any sort of thermal compound on them. I would imagine if they were going to have something like that it would be more like those thermal 'pads' but even then its probably overkill:

Maximum Operating Temperature:   + 150 C

[lilshawn]

yeah, this heat sinks are terrible.

i usually take the heat sink off and put a little heat sink grease on the transistors, and use some pliers to bend the metal to get a better grip on the transistor.

BTW those heat sinks are terrible.

[MKFan4Life]

I did order the correct transistors didn't I?  Had to do a little cross-referencing to find what appeared to be the right ones.

Also, would the heat sink being loose cause this problem in the first place and the transistor still be good?

Thanks guys!

[newmanfamilyvlogs]

just replace them. my 7400 is only 4 years old and they were already failing.

[MKFan4Life]

Guess I'll find out when they get here and I install them if they're the right ones...

[newmanfamilyvlogs]

Finally dug through my history to find the component I ordered:
http://www.mouser.com/ProductDetail/Fairchild-Semiconductor/KSC1507YTSTU/?qs=sGAEpiMZZMtn7beagP2hIdb0BRj3v8d8vtakSiZO1QI%3D

Looks like that's the exact same C1507 you ordered.

One small thing I noted with mine was that the leads are a little shorter than what was on my neckboard, so they sat a little snug up against the PCB.

Also if you're really that concerned with the heatsinks on them, why not find a little low RPM fan just to move some air across them?

[lilshawn]

Also if you're really that concerned with the heatsinks on them, why not find a little low RPM fan just to move some air across them?

i've been doing that with some of the chassis that are chronic transistor burners. it works quite well actually... maybe i should be finding the root of the problem that's causing the transistors to overdrive in the first place...

*wanders off to ponder*

[newmanfamilyvlogs]

Also if you're really that concerned with the heatsinks on them, why not find a little low RPM fan just to move some air across them?

i've been doing that with some of the chassis that are chronic transistor burners. it works quite well actually... maybe i should be finding the root of the problem that's causing the transistors to overdrive in the first place...

*wanders off to ponder*

Are there any display benefits to cooler transistors or is it purely a protective measure?

[lilshawn]

Also if you're really that concerned with the heatsinks on them, why not find a little low RPM fan just to move some air across them?

i've been doing that with some of the chassis that are chronic transistor burners. it works quite well actually... maybe i should be finding the root of the problem that's causing the transistors to overdrive in the first place...

*wanders off to ponder*

Are there any display benefits to cooler transistors or is it purely a protective measure?

nothing really, longevity only.... they often run the transistors right at their max (or close)... often if it's rated at 2 amps they run a 2 amp transistor instead of going with a 3 amp. without knowing why a particular transistor was chosen, it's difficult to sub something better, as often the engineers used that one for some intrinsic value to optimize the circuit based on it.

cooling the transistors often helps them last longer, as the air in the cabinet is often quite warm - hot even.... the max temp degree quote often supplied my the manufaturer is at 25 degrees Celsius ambient temperature... boost that up to 50 degrees Celsius and the reliability at maximum load drops dramatically.

[MKFan4Life]

Finally dug through my history...

Looks like that's the exact same C1507 you ordered.

One small thing I noted with mine was that the leads are a little shorter than what was on my neckboard, so they sat a little snug up against the PCB.

Also if you're really that concerned with the heatsinks on them, why not find a little low RPM fan just to move some air across them?

Yep!  That's the ones!  Appears these are going to be gone before long since all the other compatible ones they had were not available and these stated by the description "End of Life: Scheduled for obsolescence and will be discontinued by the supplier."

Yeah, I did notice in the description they said they had short leads.  Mine are really snug to the pcb already, so I guess mine were short lead transistors to begin with.  On my K7197 chassis, the transistors have REALLY long leads and sit very high with NO heatsinks at all.

The fan sounds like an excellent idea, and fun project.  Maybe I can find a way to mount it and tap into some power from the chassis in a way that will make it look factory... to a newb anyways.  Lol!

[smartbomb2084]

Sorry, I forgot about this.

http://www.electronix.com/advanced_search_result.php?search_in_description=1&keywords=2sc3782&osCsid=htjbtkcn3h7usd45r3ccd18f16&x=9&y=10

[MKFan4Life]

Is that one compatible with a SEC C1507-0?

[smartbomb2084]

According to the drawing these are what is supposed to be on the neck board.

[smartbomb2084]

Or... In case you are rich you can buy these. Who knows, they might work better. HA

http://www.happcontrols.com/electrical_supplies/67040700.htm

[MKFan4Life]

Yep... I think you're correct.  I should have looked at the K7501 schematic that the guy who sold it to me gave me 'cause he said they were almost identical (the K7501 & K7400).  I just figured a picture was worth a thousand words, and got the SEC C1507-0 identification straight from a pic of one of the currently installed transistors on the neckboard.

After some Googling of the specs, they look compatible, but the ones I ordered are rated at 300V and the originals you listed are rated at 200V.  Would that be a problem or benefit?

Also, lucky me, came across an exact PDF copy of the parts list and diagram I am looking at now in hard copy, but it's labeled for the K7400, and they are completely identical.  (at a glance, anyways)

Hope I don't end up having to order 3 of the "originals" if these end up causing problems, lol!

[MKFan4Life]

Or... In case you are rich you can buy these. Who knows, they might work better. HA

http://www.happcontrols.com/electrical_supplies/67040700.htm

Your first link is for 2SC3782 and the next (the $4.75 one) is a 2CS3782.  Any difference?

[smartbomb2084]

The 300V ones are able to stand a higher operating voltage obviuosly, which I would think would be a benefit being that they are going to be operated at 114V.

But you know what they say:  Assumption is the mother of all screwups!

Tha Happ link is just a typo.

[newmanfamilyvlogs]

On the 300V vs 200V issue... I don't know enough about transistors and CRT components to form a reasonable opinion on this so here's my question:

After getting mine up and running after replacing those transistors, it somehow feels like the picture isn't as 'punchy' as it could be. As though there could be a higher contrast ratio, as compared to my memory of arcade screens, as well as various CRT televisions I've seen. Obviously this is a pretty subjective observation. The convergence is also a bit off still, which I suspect can affect perception of the screen as well.

Could the difference in rated voltage effect this? I'm open to the suggestion that I've either got other settings non-optimal, that I have too high an expectation for what the K7400 is capable of, or that there is some other maintenance that needs to be performed. The manufacture date on mine is Jan 2006; to me that seems young enough for something like a cap kit, but again I don't have enough experience to justify my opinion.


MKFan4Life: If you know, what's the date of your monitor?

[lilshawn]

Replacing drive transistors with ones not specified by the manufactuere can result in all sorts of video problems.

The transistors where specifically chosen for a reason. If you replace them with a "better" higher voltage transistor, it may switch slower, resulting in a slightly blurry image due to different rise and fall times of the different transistor.

In theses cases, milliseconds mean the difference between the electron beam hitting one phosphor dot and three because it takes longer to turn on the beam. Even the secondary drive transistors being changed with "better" inferior parts can cause issues.

Unless you scrutinize the specs on your sub choice and the original, you're looking for trouble.

[newmanfamilyvlogs]

Thanks for that info! Now I'm curious just how far off the original spec these transistors might be.
I'm going to assume that my transistors were not stock then, as the C1507 that I replaced them with were what were printed on mine.

They certainly work, but I'll have to replace them one day with the correct ones and see what a difference that makes. At 0.63c each I think I'll deal with it not being perfect (:

Edit:

Scanning the data sheets what really jumps out at me is the difference in the transition frequency:
C1507 Typical: VCE = 30V, IC = 10mA 80Mhz
2SC3782 Typical: VCE=(–)30V, IC=(–)50mA 400Mhz

What can we infer from these specifications? The C1507 doesn't list a "Max" range on the sheet I looked at, but neither does the C3782.

[MKFan4Life]

From cotmm68030:  MKFan4Life: If you know, what's the date of your monitor?

Mine appears to have same tube and chassis that were mated from factory with matching serial #'s on the tube and chassis frame (that's not saying the pcb couldn't have been swapped since I don't see a serial # on it anywhere).  But anyways, the tube has a date of September 1996 and the frame says manufactured August 1997.  The seller told me he thought they went out of production in 1995 and it was the last American made Wells Gardner STD resolution monitor.  Now about that, I have no idea.  But I didn't know they manufactured them all the way past 2000!  Wow.

From lilshawn:  Replacing drive transistors with ones not specified by the manufactuere can result in all sorts of video problems...  Unless you scrutinize the specs on your sub choice and the original, you're looking for trouble.

I would agree totally.  But I wanted to order them quickly (I don't mind ordering the proper ones now also.), so I just got the specs off the ones already installed since the fellow claimed he had used it without problems for awhile, and I (at that time) had no reason to believe they had been changed to some with the wrong specs.  Funny thing is, what are the chances that me and cotmm68030 both would acquire a WG-K7400 which had the transistors all improperly swapped with C1507's if they were known to be improper?  Guess it COULD happen, just odd.  Definitely NOT what the Parts List and Schematic calls for, though.

Mouser doesn't have the 2SC3782.  Since that site lists more data for the C1507, I figured we could compare specs better, but no dice.

From cotmm68030:  Scanning the data sheets what really jumps out at me is the difference in the transition frequency:
C1507 Typical: VCE = 30V, IC = 10mA 80Mhz
2SC3782 Typical: VCE=(–)30V, IC=(–)50mA 400Mhz

It's all Japanese to me friend, lol!  I saw some different "collector currents" when first comparing them, but then noticed I was comparing "Continuous Collector Current" with "Maximum DC Collector Current"... apparently not the same thing AT ALL!  One day I will understand all this stuff.

EDIT:  Sure wished Mouser would make up their mind what they're charging me for these little things.  At first, they only gave me a total without shipping, $1.89 for all three.  (I used their little chart to figure smallest weight possible at 1lb. shipped via USPS Priority @ $6.95 just to make a note for the old checkbook register since I used my check-card and didn't want to forget to deduct total later when they notified me of items shipped and full total.  That would be $8.84, just as an estimate.)  Well, then they charged my account $9.89 plain as day in my online banking account history showing from Mouser.  After updating my register with this amount, now I see that charge is gone, and I have a new charge of $7.99 on my account... weird!  It's like that little $1 charge they tack temporarily on your account when using a check-card at certain gas stations.  Irritating, lol!

[MonMotha]

EDIT:  Sure wished Mouser would make up their mind what they're charging me for these little things.  At first, they only gave me a total without shipping, $1.89 for all three.  (I used their little chart to figure smallest weight possible at 1lb. shipped via USPS Priority @ $6.95 just to make a note for the old checkbook register since I used my check-card and didn't want to forget to deduct total later when they notified me of items shipped and full total.  That would be $8.84, just as an estimate.)  Well, then they charged my account $9.89 plain as day in my online banking account history showing from Mouser.  After updating my register with this amount, now I see that charge is gone, and I have a new charge of $7.99 on my account... weird!  It's like that little $1 charge they tack temporarily on your account when using a check-card at certain gas stations.  Irritating, lol!

It's probably an artifact of using a debit/check card rather than a credit card.  Chances are, they're not actually posting that extra money as a charge, just putting it on as a "hold" until they know the actual shipping amount.  Restaurants sometimes do this in order to reserve funds for tips; hotels will do this when you put the card down as the "deposit" for a room; rental car companies do the same; etc.  If you don't see it as a "charge" then a "credit", then it was probably just a hold.  It shouldn't have actually posted at the higher amount, but it may show up in your "pending transactions" list that way.

If you want the actual 2SC3782, MCM Electronics has them in stock.

[MKFan4Life]

Yep.  I am sure you are correct on the hold.  I was just wondering where they pulled that fictional $9.89 amount out from?  Lol!  It just really looked more like a real charge than a hold.  If I had seen $1 or $10 or $20, I would have waited longer before recording it in my register.  Lol.  No biggie, just ranting.

And thanks for the link for the original transistors friend!

[newmanfamilyvlogs]

So are you going to put the C1507s in then swap for the others later? The traces ok my neck board are so haggard that im hesitant to just go swapping again if there is minimal improvement.

[MKFan4Life]

Yeah, most likely I will.  Don't want to be an idiot or anything (by using something other than the parts listed by WG on the sheets), but the specs looked comparable and that's what was in there when I got the monitor.

I know with monitor cap kits, most people would probably agree if you're not sure what to put in a cap position, you'd be relatively safe to replace an old cap with a new one of same value if the monitor had worked ok before that way.  But I know this is a different beast, and I DO know what is SUPPOSED to be there according to the schematic and parts list.

Now if we can just find the culprit who is going around by night and replacing all the K7400 transistors with C1507's!!!  Lol!  Bet he has a stash of them just waiting for stocks to be depleted so he can set his own price on replacements!  Muhahahaha! 

[smartbomb2084]

If you want to pay $2.09 each at MCM for them go right a head.   But you could pay $1.69 each at www.electronix.com instead.  Don't you remember I already posted this earlier in this thread?  Aren't I the guy that alerted you to the correct ones in the first place?

[newmanfamilyvlogs]

Now if we can just find the culprit who is going around by night and replacing all the K7400 transistors with C1507's!!!  Lol!  Bet he has a stash of them just waiting for stocks to be depleted so he can set his own price on replacements!  Muhahahaha!  

Doing some looking around it just looks like the C1507 is a more available transistor. Also there are some monitors that DO use it:
http://www.neo-geo.com/forums/showthread.php?t=183766
http://www.gamehourz.com/Transistor-info-monitor-Neck-ftopict71309.html

I'm just going to wager a guess and say that if it works well enough an operator would purchase the C1507s in bulk and use them since they can obviously be found much cheaper than the C3782.

For what it's worth, I found a second neckboard among the ruins at a arcade graveyard and it has C3782s on it. The traces on the board are equally as fragile as on my other one, and the blue was out on it too.

Suppose once I source get some C3782s I'll replace them on this board and just do a swap so I can compare the differences between them more easily.

[MKFan4Life]

If you want to pay $2.09 each at MCM for them go right a head.   But you could pay $1.69 each at www.electronix.com instead.  Don't you remember I already posted this earlier in this thread?  Aren't I the guy that alerted you to the correct ones in the first place?

Yes indeed you are.  Just trying to be polite to him for also being helpful.  Thank you, too, just as much smartbomb2084.  For all your help on this monitor. 

I would think I can search for even more price options on the net, but pretty much, the shipping amount is the deciding factor. 

It might be some time before I even order those original ones, 'cause I ordered the C1507's at 1:30AM on October 11th (before you directed me to the original ones).  They are sitting here in front of me as I type this (was in the box today when I got home from work).

So I am going to try these and see the results even though I likely WILL be getting the 2SC3782's before long.

[MKFan4Life]

UPDATE:

Installed all 3 new transistors and seems to be running just great.  All colors are showing in all the Monitor Test screens fine and the game looks very good.  Don't notice any real improvement in colors, but they were very good to begin with until the blue started going in and out.  Hope I have nipped this one in the bud.

Found out those "awful" heatsinks (as you guys stated) had no thermal grease or paste on them at all.  I simply pulled them off of the old transistors.  Haven't put them on the new ones yet, but I also have not run the monitor for an extended period of time.  Really want to go ahead and do the little fan mod that we talked about earlier.  That would be nice I think.

I did want to mention, although my traces LOOKED good to start with, I quickly started having them lift in certain spots when trying to desolder the transistors.  I had to be really careful not to damage them, especially since my old 25-watt irons are about shot and I was trying to GENTLY use a 40-watter (I know, this is considered by many to be way too powerful, but I didn't have access to anything lower at this time.)  I got through it ok, but I need to run out and buy a couple of new lower wattage irons and some fine tips, too.

Thanks for all of the guys in this thread for your hints, suggestions, and support.