Neotec NT-500DX Repair

[Rocketeer2001]

There wasn't too many threads on repairing on of these monitors, so I thought I'd start one. Maybe that means these are so reliable that they don't break often?

Anyhoo, I just picked up a cabinet that has the 27" Neotec NT-500DX Trisync monitor in it. I need it to put into my Big Buck Hunter 2 with a failed WG D9200 EGA monitor in it.

I tried powering on the Neotec and it makes that recognizable 'toomph' sound and one single click, but then not much of anything else. No OSD or anything; just black deadness. There is a super high pitched whine, but I think that's just your normal power supply kind of sound.

After reading over some old threads and watching this guys videos, I've compiled a list of possible things to check:

http://forum.arcadecontrols.com/index.php?topic=99464.0
https://forums.arcade-museum.com/threads/neotech-nt-550dx-manual-and-schematic-needed.515427/
https://forums.arcade-museum.com/threads/neotec-nt-500-woes.408148/



Here's my list:

1. Check/replace the SMPS (power supply) driver chip IC501:  KA3842A
2. Check/replace the 5 caps in the power section: C525 (up its rated voltage to 25v or 35v),  C527, C519, C501, C502.
    The caps blow and the SMPS starts loading down till the FET blows and the SMPS control chip goes with it. Suggests “upcapping” the voltage rating on these caps in the power section.
3. Check/replace the Vertical IC 301 goes bad and takes out R623A & R623 (5.6 ohm, 2W, 5%). Replacing all ensures reliability.
4. The transistor that controls the power to the electronics Q501 (HSB772P) tends to heat up till it burns a hole in its place.
    Replacing it with an NTE185 and attaching a small aluminum heat sink to it to help dissipate heat really helps.
    This is a transistor that simply switches the monitor 15 and 12 volt supplies off/on using the monitors CPU.
5. Q504 (mosfet 2SK956) might be a blown transistor. Check if legs are shorted to ground.
6. When the SMPS is ticking, check R520 a 68 ohm resistor, and R526 a 0.15 ohm 2 watt 5% resistor.
7. Check R539

I've already written down all the caps and their values and will be replacing all of them. I've heard some success stories after simply replacing all the caps, so I'll start there.

Things I did find / test:
I found one lifted leg on C2A8 on the neck board. I reattached that bought I doubt that was the cause of it not working. Aside from that I've cleaned a bunch of gross black dust off everything.
I tested Q504, which I believe is the HOT, and it seems fine. (see pic)
Tested the big filter cap C504 and it's still very healthy. Tested C521 and it's still good too.
Tested R R623A & R623 and they're both good.


A couple helpful tips for the next NT-500DX lost soul:

1. You need to remove the metal cage around the neck board in order to get to the back of the board to desolder caps. You just need to desolder the 4 corner tabs and one big on in the middle and then the board will wiggle out.

2. I wanted to remove the high voltage wire from the neck socket so that I didn't have that dangling about when I'm flipping the chassis board around replacing caps. I couldn't figure out how to get the wire out of the socket until after I really cleaned it and desoldered it from the neck board to inspect closely.
There's a little tab you can pry away from the wire ever so slightly, and while holding that you push up on a little nubbin. It slides up about 1/4" and then you can pull the wire out. (see pic)

[lilshawn]

looks good visually. nothing stands out off the hop.

don't forget that some of those larger transistors and FETS used to switch the horizontal yoke or flyback will sometimes test good even though they are actually bad. It's due to the low voltages that testers use. the small 2 or 3 volts used to test won't be enough to flashover the gate junction... but when you apply full circuit voltage to it, it does.

so, I would say if it tests bad, its obviously bad... but if it tests okay, it could still be a 50:50 chance... only real way to tell for sure is to swap with a known good one, or devise a testing circuit that tests it at the circuit voltage (140v or whatever)

sounds like the monitor is kind of on, so a good place to start is by checking all the power rails and making sure they are all there. basically ground your negative lead and probe a device that is on the "out" side of the diode coming out of the SMPS transformer. there is a lot of rails in this chassis, but they all do something. hit up the "power" sheet and test them all. you may find one sagged down.

something else is do is use a different highlighter for each rail voltage... and go through the schematics and highlight any of the volt markers on other pages with the same color. makes isolating issues a little easier down the way. like, if the 8v rail is down to 6 you can easily find all the stuff running on 8v by finding all the pink highlighted volt markers.... and that way you don't have to read every single one, every time you look at the schematic.

[Rocketeer2001]

That's a good point about the low voltages of the testers. I could test the voltages coming off those transistors in-situ when the board is powered up though. I should be able to read if the correct voltages are coming out each leg then.

I'm going to do the voltage tests tomorrow, and I took your advice and highlighted all the voltage locations on the diagrams. That was extremely helpful! Had to go over the sheets a few times as I kept missing some.

[lilshawn]

I took your advice and highlighted all the voltage locations on the diagrams. That was extremely helpful! Had to go over the sheets a few times as I kept missing some.

nothing more frustrating that trying to find an issue and having to search all over hells half acre for a tiny circles only to find out the issue was on one you missed along the way.

i've started doing this to all my schematics that are broken upp like this. makes troubleshooting so much easier.

[Rocketeer2001]

Been some slow progress the last few weeks, but I'm back at it!

One improvement I made was at the connection at P903, which is for the degause coil wire that comes from around the yoke. The connector was just shoved onto some pins on the board. I think it should have had a JST socket for it to snap into. Maybe mine just fell off somewhere along the way. I salvaged one off a random board and added that in.

I tested the HOT with my multi-meter instead of my component tester and it might be half working? I put my meter in diode mode, put the negative on the center pin, the put the positive lead on either outer leg. To my knowledge, I should get around 0.4-0.5v on either of them, but I only got around 0.4v on one leg and nothing on the other. I tried juggling the leads around but got the same result.

Decided I should test this again when it's in the board and powered on, so I soldered it back in and reinstalled everything back into the cabinet. When I turned it on I was getting some strange results. I had the neg. probe of my meter grounded to the chassis and I was probing everything else with the pos. lead to check the voltages.
For the Q504 HOT I was getting:
Right leg (leads to L512) =  71VAC or -78VDC
Left Leg (leads to L513) = 71VAC or -74VDC
Center Leg (leads to L511) = inconclusive. Meter was going nuts beeping at me and auto changing rapidly between AC and DC.

Per the recommendation, I checked all the other voltage rails. I tried metering right at the pins coming off the main power transformer T501, but I was either getting a scrambled readings or none at all. If I probed farther down the circuit it was better.
On the AC side at pins 1,2,3,7 I get 71VAC. Pin 5 freaks out.

On the DC side I should have
this voltage @ this location = but I get:
51v @ R544 = 60v
15v @ C501 = 14.7v
12v @ C505 = 11.9v
7v @ C527  = 3.25v
5v @ C507 = 3.7v

I also measured some voltages at the neck board and I got:
7v = 0v
12v = 11.95v
140v (105v) = 146.6v

To me, it looks like I have an issue with my 7v and 5v supply. Also not sure if my AC volts should be 71v or if they should be higher, but it does appear to match the reading at the HOT.
Thoughts? Next things to check?

[princess prin prin]

I tested the HOT with my multi-meter instead of my component tester and it might be half working? I put my meter in diode mode, put the negative on the center pin, the put the positive lead on either outer leg. To my knowledge, I should get around 0.4-0.5v on either of them, but I only got around 0.4v on one leg and nothing on the other. I tried juggling the leads around but got the same result.

That's not the correct way to test an NPN transistor (the HOT here is a 2SC5411 so it falls into this category). NPN means it has two PN junctions (which behave like diodes): one is BC, the other is BE. So you put the red lead on B and the black on C and you get a reading (0.5V or close) and that's what you did. Then you keep the red lead on B and you move the black lead on C and you get a reading a little higher than the other. And that's it. Of course no readings if you reverse the leads and no reading on CE since it is not a junction.

[Rocketeer2001]

Ah, ok good to know. Went out and tried that just now and I got 0.5v and 0.68v. This is with the HOT still in the board, but not turned on.

Given that result, and the voltage readings I got when it was running, does that mean it's ok?

[Rocketeer2001]

Did some more testing tonight.

Everywhere on the board where there should be 7v, there is 0v, and everywhere it should have 5v, it has 3.6v.

I went back to pin 15 of the main transformer which is supposed to be outputting 7.2v and working my way along that circuit. IC 503 (AMC-76285) might be the culprit, but I have no idea what that is...maybe a voltage regulator? Maybe Q506? I tested it while on the board and I do get results instead of a short or zero's, so I'm inclined to think it's ok.

More than likely it could just be bad caps. There's only 3 and when I test them on the board with my ESR meter I get:
C525 - 16v, 1000uf = ESR 9.5 ohm (should be around 0.09 ohm) 
C527 - 16v, 100uf = ESR 2.5 ohm (should be around 0.7 ohm)
C507 - 16v, 100uf = ESR 0.63 ohm (should be around 0.7 ohm)

Guess I should just order some caps and recap everything and see what happens, unless anyone has other suggestions of what I should look at next.

[princess prin prin]

IC503 is a voltage regulator for sure albeit an obscure one (compatible with 7805??). It gets its input voltage from the cathode of D502 and outputs regulated 5V. The input must be at least a couple of volts higher than 5V and lower than the 16V rating of the smoothing cap C526. If you don't have a good voltage at D502, you could desolder the regulator, run the monitor without it and check if the input voltage is now OK. The regulator could be tested with an external voltage source. You could do the same with Q505. 7V is used as the heater voltage so no picture until you have that back.

[Rocketeer2001]

I'm thinking the IC503 voltage regulator is probably working correctly. I metered the voltage at the input side and it was 3.9V, and on the output side it was 3.6v. It is stepping down the voltage, I just don't have enough of it to begin with.

The voltage after D502 is also 3.9v, but the voltage before is unknown. My meter just goes nuts when I try to probe right at the transformer pins. Perhaps too much noise and it confuses the meter?

Kinda thinking that L509 or D502 might be my issue. Or perhaps even the whole transformer. Is there anything on the AC side of the transformer that would be missing up just the 7VDC side?

[lilshawn]

linear regulators need at least 2 volts higher (some 12+ volt regulators need 2.5 or even 3 volts) than their output voltage in order to work properly...

yes that part is a regular bog standard LM7805. it's output voltage is 5 volts, so you need at least 7 volts to get 5 out. (6 would be pushing it and it would technically work, but would be very prone to dropping out) since this is on a 7 volt rail and it's looking like it
s low... you might want to get the 7 volt rail a good looking at. something is dragging it down.

you cannot probe the output of the transformer directly, as it's output a very high frequency pulsed output that is smoothed out by the inductors (L5xx parts) and the diodes immediately after. the average layman should probe the cathode side of the diode for proper readings.

toasted caps can certainly cause rails to be low, and as these (and every other CRT monitor in existence), are all over 20 years old and can purchase their own alcohol now, should be replaced.

[Rocketeer2001]

Thanks for the suggestions and info, guys! Looks like I have some caps to order and then we'll see what happens. I'll probably get a spare voltage regulator and HOT while I'm at it, just in case. Stay tuned for an update in a week or two after I get the parts!

[Rocketeer2001]

I've come across a new hurdle; obsolete parts!

I found a LM7805 easy enough to replace AMC-76285 (IC503), but I can't find a replacement for the HOT (Q504) 2SK2648 (or 2SK956). I know I can use alltransistors.com to find something compatible, but I don't know which specs I can vary by whatever amount to choose something else.

Also can't find a 250v, 1uf NP cap. It's an oddball size and being bi-polar doesn't help. I could find a 1.8uf on aliexpress, but I don't want to use a cap from aliexpress, haha. I should just go test the ESR of those caps (there's 3 of them) and see if they're still ok. If they are, Ill just reuse them.

I was going to pickup some other spare parts too (for just in case), but similar issue with these:
IRF654B
HSB772P (or B772-Y). Shawn, I can't find a NTE185 that you recommended in an old Neotec post either.
KA7812
IRFU214A
KA3842A
TDA8172
C023M-15 (or CD023M or ES023M)
2SC5411 (or J6815 409)

[Rocketeer2001]

Decided to check all the major components coming off that pin 15 of the transformer. I removed them and put them in the component tester. These are good:
Q505 (mistakenly mentioned this previously as Q506)
D502
C552

I think IC503 is good, but it looks like a little PNP transistor rather than a big blocky mosfet like a LM7805. I don't think a LM7805 will fit in this space (see pics).

These caps are definitely bad:
C525 - ESR= 7.1 ohm, 574uf [should be 1000uf and ESR 0.09 ohm]
C507 - ESR= 4.2 ohm, 81.2uf [should be 100uf and ESR 0.7 ohm]
C527 - ESR 27pf [should be 100uf] (this one is so busted it doesn't even tell me the ohms or vloss %. THIS IS A REALLY BAD ONE)

I may have these caps kicking around in a parts bin somewhere. I could try swapping out just those caps and see if it fires up.

[lilshawn]

IC503 won't test with one of those testers since it's a power regulator. you'd have to feed power in excess of it's rated output (in this case 7v as we have discussed previously) and load the output to see if it's working properly.

i'd throw a new cap on anything 100uf and up. (they are the most susceptible to degradation that the smaller value caps.) and anything high voltage excluding the main DC B+ cap.

[Rocketeer2001]

Ah, yes that makes sense. I'm feeling a little lazy now, so I'll skip that IC test and just hope for the best with the cap replacements.

I ordered the caps and now we wait.
For anyone else that needs to assemble a cap kit for one of these monitors, here's a complete list I compiled:

CHASSIS BOARD
C504   400v, 330uf   
C510   50v, 4.7uf   
C545   35v, 68uf   *
C582   50v, 10uf   
C951   50v, 1uf   
C953   50v, 1uf   
C948   50v, 1uf   
C935   50v, 1uf   
C952   50v, 1uf   
C901    50v, 4.7uf   
C907   50v, 10uf   
C922   16v, 22uf   
C585   16v, 100uf   
C523   25v, 100uf   
C505   16v, 100uf   
C949   50v, 1uf   
C910   50v, 10uf   
C922   16v, 22uf   
C519   25v, 470uf   
C527   16v, 100uf   
C507   16v, 100uf   
C525   16v, 1000uf   *
C501   25v, 470uf   
C530   25v, 1000uf   *
C520   160v, 22uf   *
C521   100v, 330uf   *
C305   25v, 470uf   
C325   25v, 100uf   
C323   25v, 1000uf   
C744   100v, 47uf   *
C745   16v, 22uf   
C722A   200v, 22uf   
C722   200v, 22uf   
C623   16v, 22uf   
C601   16v, 100uf   
C607   16v, 47uf   
C612   50v, 4.7uf   
C613   50v, 4.7uf   
C615   16v, 47uf   
C715   50v, 4.7uf   
C773   50v, 4.7uf   
C774   50v, 4.7uf   
C772   50v, 4.7uf   *
C704   50v, 10uf   
C703   35v, 100uf   


NECK BOARD
C276   200v, 22uf   *
C263   250v, 1uf, NP   *can be 400v 1 uf
C203   250v, 1uf, NP   *
C233   250v, 1uf, NP   
C274   16v, 47uf   
C272   16v, 47uf   
C270   16v, 47uf   
C290   16v, 47uf   
C292   16v, 47uf   *
C280   16v, 47uf   
C279   16v, 47uf   *
C244   25v, 22uf   
C288   25v, 22uf   
C296   16v, 220uf   *


INPUT SIGNAL BOARD
C105   16v, 100uf


* = caps that cannot be physically larger due to space limitations


TOTAL OF EACH SIZE
(1) 400v, 330uf
   
(3) 250v, 1uf, NP
   
(3) 200v, 22uf
   
(1) 100v, 47uf
(1) 100v, 330uf
   
(1) 160v, 22uf
   
(4) 50v, 10uf
[8] 50v, 4.7uf
(6) 50v, 1uf
   
(1) 35v, 68uf
(1) 35v, 100uf

(2) 25v, 1000uf
(3) 25v, 470uf
(2) 25v, 100uf
(2) 25v, 22uf
   
(1) 16v, 1000uf
(1) 16v, 220uf
(6) 16v, 100uf
(4) 16v, 22uf
(9) 16v, 47uf

TOTAL = 60

[lilshawn]

if you need to replace high voltage non-polarized caps, and are unable to source them (often 50v is the most you can get these days) you can take 2 of an acceptable voltage rating.... but DOUBLE the uF rating...

for instance if you need a non-polarized 1uf cap with a 200v rating, then get two 2.2uf 200v caps which will be close enough

then take the 2 negative leads and solder them together and insulate the lead (solder mask, glue, paint, nail polish, whatever). this will give you a capacitor with 2 "positive" leads that you can use in place of the non-polarized cap.

[Rocketeer2001]

Shawn, that's a handy tip in the event I need to replace those non-polar caps; I didn't know that could be done, so thank you very much for that info!
For now, it seems I don't have to worry about those because I replaced all the caps with the exception of those 3 non-polar ones on the neck board, and I re-used the big filter cap (C504) and.... it turns on!!!!!   

These caps were definitely bad:
C272
C525
C276
C585, C527, C507, C505

These ones were surprisingly ok and sometimes even better than my new ones:
C296
C325
C744
C521
C922, C623
C907, C582, C910, C704

All other caps were low on their uf rating or had a higher ESR reading, and was just time for them to be replaced.

Initially the screen was mostly white with some raster lines, so I turned down the 'screen' brightness on the flyback and there was the good ol' "No Signal" message! I then tried to fiddle with the focus on the flyback because the image looked a bit fuzzy. Didn't seem to matter which way I turned it, it either got fuzzier or less fuzzy. Then I remembered reading a post where someone else had a similar issue and the solution was to remove the glass and clean it and the monitor. It worked! Glass and screen were both gross.

A couple final issues:
1. Looks like the image has barrel distortion (curling in at the corners). Not sure if this will go away after it has a proper signal and not just the "no signal" message.
2. Image is rotated a bit. On the D9200 it had a pot on the chassis board for adjusting this, but I don't see one on this board. Is this something I can adjust in the OSD menu?
3. The menu doesn't pop up when I push the 'menu' button on the switch board. I'm going to go test the switches to make sure they work. Perhaps they only function when a signal is present?

[lilshawn]

great, glad to know you got it sorted out!

yes, the menu only displays when the signal is present... as the adjustments you make are specific to that resolution. and each resolution has its own settings.

[Rocketeer2001]

I can't tell you how happy I was that it just fired up and I don't have to mess around hunting down transistors, resistors, or mosfets. So far, I'm a fan of the Neotec over the D9200 and I haven't even played a game on it yet!

I rewatched that first video from youtube I posted. Near the end he goes through the OSD menu and it has all the adjustments I need to correct the issues I'm seeing.

I'll wheel this machine over beside the other one tomorrow to hook up the game board and see how it looks before I do the big swap. Any tips or tricks I should know about removing/handling a monitor?

[lilshawn]

  i just unbolt them and throw my monitors on a cart and wheel them around to service/test them. got an old school AV cart with a plug and cord set on it and everything. makes transporting and testing easy.

i guess don't go poking your fingies in there while it's turned on is a good start.

while the main glass envelope of the monitor is quite thick and strong, (some parts of the front and corners are 1/2" thick) the neck of the monitor is the single most fragile thing on a monitor. (often only 1/16" thick) so if you can get a person on each side while you slide it out of it's old cabinet you lessen the chance of knocking it. most of the time the frame around the monitor sticks out enough to protect it, but sometimes when you go tilting and pushing and pulling on it, you can get it hung up on wires or something in the cabinet and snap the neck right off, and then it's toast. no fixing that.

[Rocketeer2001]

So the monitor works, but it has some image issues I can't remedy with the OSD menu.

1. General blurriness. I adjusted the focus on the flyback but I still can't get the image crisp. Seems like if I get one portion of the screen mostly focused, another part gets out of focus. Perhaps the yoke needs shimming?

2. Big Buck Hunter comes with built in test images, one of which is a grid. I could see in the top left corner that the white line looks like it's diverging, but in the top right corner it's fine. Convergence issue perhaps? It also has solid color test screens, and in Red there is a dip in the top left, in the Green there's a slight dip, and Blue there is no dip. The convergence rings on the neck still have a white paint stripe across them indicating it's been factory calibrated, so those obviously haven't been fiddled with. Perhaps it's just time to re-calibrate them, but also sounds like a giant time-suck once you start playing with those rings.

3. There's a white ghosted vertical line that waves around with a frequency of 1 second along the left half of the screen. I tried to film this but it just doesn't show up well. It's visible in person though. I have no idea what could cause this. I did a doodle of about where it is and looks like. Pretend it's flapping about like a wave.

I did notice the degauss coil had come loose at some point in its life and was drooping down on the upper right corner of the screen, so you'd think there would be issues there, but there isn't. Not sure if this relates to any of these issues.

The ghost line is annoying, but is mostly only noticeable on the high score screen. The blurriness is also annoying, but is mostly prominent when trying to read text and the deer don't hold up a lot of signs so...
Not sure if I should start mucking about trying to fix these issues or just be glad I have a working monitor and just play the game.

[lilshawn]

So the monitor works, but it has some image issues I can't remedy with the OSD menu.

1. General blurriness. I adjusted the focus on the flyback but I still can't get the image crisp. Seems like if I get one portion of the screen mostly focused, another part gets out of focus. Perhaps the yoke needs shimming?

it's an issue with single focus flybacks/tubes. just get the center part as sharp as you can...it's where the action is... believe me, you won't notice it.

2. Big Buck Hunter comes with built in test images, one of which is a grid. I could see in the top left corner that the white line looks like it's diverging, but in the top right corner it's fine. Convergence issue perhaps? It also has solid color test screens, and in Red there is a dip in the top left, in the Green there's a slight dip, and Blue there is no dip. The convergence rings on the neck still have a white paint stripe across them indicating it's been factory calibrated, so those obviously haven't been fiddled with. Perhaps it's just time to re-calibrate them, but also sounds like a giant time-suck once you start playing with those rings.

 unless you've done convergence adjustments before... i'd probably leave it. convergence is usually off in the corners. it's really hard to get it good in high deflection angles like these. again, as long as the middle is good... you won't notice it when you are playing a game. you aren't going to have white squares you need to follow around. once your sage brush and trees are displayed... you won't notice.

3. There's a white ghosted vertical line that waves around with a frequency of 1 second along the left half of the screen. I tried to film this but it just doesn't show up well. It's visible in person though. I have no idea what could cause this. I did a doodle of about where it is and looks like. Pretend it's flapping about like a wave.
I did notice the degauss coil had come loose at some point in its life and was drooping down on the upper right corner of the screen, so you'd think there would be issues there, but there isn't. Not sure if this relates to any of these issues.
The ghost line is annoying, but is mostly only noticeable on the high score screen. The blurriness is also annoying, but is mostly prominent when trying to read text and the deer don't hold up a lot of signs so...
Not sure if I should start mucking about trying to fix these issues or just be glad I have a working monitor and just play the game.

it could just be your brightness is up too high or your width too wide. the beam can deflect off the side of the tube causing weird ghosting.

you could also try unplugging the degauss coil and see if it goes away. sometimes an old or worn out PTC switch for the coil that is normally supposed to whittle down to nothing will allow excess current to flow causing screen to do some weird things as it stays partially energized.

if it makes no difference... could be noise in the video. try rerouting the cable from the board to the monitor differently. it could be picking up noise from other cables it's being run next to.  also try moving the monitors cables around to different places. i've had issues with the focus/g2 leads being bundled near video cables and causing all sorts of weird video issues. you could also remove the video signal so the no video signal message is displayed and see if there is any evidence of the wave there. it might be super hard to see, but if it is there, there might still be a cap somewhere that is needing replacing causing it.

[Rocketeer2001]

I had similar thoughts to issue 1 and 2. I think I've played with convergence rings before, but it's been a while. I do remember watching videos and reading posts about people trying to get a clearer image by messing with those and it was a 50/50 mix of "I got it way better!" and "I think I made it worse".

All good suggestions regarding my ghost line, thanks shawn!
I'll go play around with cable placement and stuff and see what kind of results I can get. You may be onto something about that degauss coil because I believe there's one running down that area.
would those non-polar caps have anything to do with this? They're the only ones I didn't replace.

I'll try to get a better video of it in the event I'm unsuccessful narrowing down the issue.

[lilshawn]

would those non-polar caps have anything to do with this? They're the only ones I didn't replace.

I'll try to get a better video of it in the event I'm unsuccessful narrowing down the issue.

it's possible. the bi-polar caps i normally have to deal with are either:

1: on the neckboard for the gun drive. (usually 1uf 200v NP) if you have streaking where if you have a bright spot... and it's good for a bit along the scanline... but further down the scanline it seems to darken or change color. these caps need to be changed out. (again, two 2.2uf in a positive positive config will net you 1.1 uf bi-polar which is close enough)

2: on the main deflection board. (usually 4.7uf sometimes up to 10uf 50v NP (a big fat cap, not a small cap... there seems to be 2 variations and i have no idea what the difference is but this cap is usually a big fat one) this is a cap used for horizontal width timing. so if you have an inability to squeeze down the width because it's too wide, or adjusting the hsize does little or nothing, this is the cap that needs to be changed. (i don't know specifically why there is a difference but i always had piles of the proper large 4.7 non-polar caps. not sure if swapping for the "small" caps would fetch any additional issues.)

see talk about non-polar/bi-polar 10 years ago here: https://forum.arcadecontrols.com/index.php?topic=143822.0 with good ol' ken coming in clutch with the making of bi-polar caps.

"wavy gravy" stuff is usually indicative of power supply. especially down the left side (which is the start of the scanline). What happens is the scanline starts...the cap supplying power, supplies its power, but "runs out of juice" before the scanline can finish, so the scanline gets darker or changes... but then when the scanline ends and goes back to the start again... the cap gets a chance to recharge again. so you end up with brighter image on one side, than the other. sometimes this is static (not moving) sometimes it shows up as a lighter or darker wave in the picture.

not sure if it's just a line you are talking about... or a section of picture that is lighter while the other section is darker.

just lines, i tend to look towards video interference.
whole sections, i tend to look towards power supply.

but i wouldn't discount either.

[Rocketeer2001]

The 3 bi-polar caps are on the neck board, but I don't have a bright to dark streaking like you mentioned, so that's probably not the issue.

There's this pot on the chassis board called VR 601 H.V ADJ. I don't know if that's for 'horizontal / vertical' or if it's for 'high voltage'. I don't know what this pot adjusts, but perhaps it has something to do with my ghost wave. After I get everything hooked back up in the cabinet I'll take a video of the issue.

Speaking of which, I did the big swap taking the Neotec out of the 'new' cabinet and taking the D9200 out of the 'old' cabinet. These SOB's are heavy! After removing the Neotec out the one cabinet and almost breaking our fingers and wrists (needed a helper, I don't see how anyone could remove these things solo), we decided to lay the other cabinet on it's back. It was a lot easier to just lift a monitor straight up and out rather than trying to lift it with outstretched arms while trying to jiggle bolts loose.

[lilshawn]

that is a high voltage adjust. it's normally set at the factory and then never changed.

running the HV (AKA "B+") at too high voltage can cause excess x-rays to be produced and can cause an x-ray protection circuit to kick in and shut the monitor down.

same as stuff like transistors and FET's, sometimes HV caps (over 100v) can measure okay with the low voltage a meter/tester uses to test, but fail when you apply the full beans voltage to them. it's recommended if you haven't changed all the caps, to do so. sometimes they will cram a small one in behind the flyback out of sight. i don't remember if this chassis does that, but it's worth a look.

[Rocketeer2001]

Yup, changed all the caps except those 3 bi-polar ones and the big filter cap in the power section. There was a few caps hidden between the flyback and the heatsink and I replaced those.

But if VR601 H.V ADJ. is the high voltage adjust, then what's this other one (VR501 B+ ADJ)  that's in the power section?

[lilshawn]

sorry, i'll try to explain a little better.

in an SMPS supply, you take AC and turn it to DC. this DC voltage is B+ then you take the B+ and pulse it 10,000 times a second through a transformer and create all the voltages (190v 20v 12v 6v etc.)

but quite often in older monitors, you'd take AC, turn it to DC... then run it through a linear regulator module to create "B+" that you then use to drive the flyback and the horizontal youke winding... then also resistors and crap to tune the voltage down for the various other circuits.

so the nomenclature of "B+" is used for a couple different kinds of voltages.

in the case of the nt500, with an SMPS... it's technically named "B+" is created with a feedback loop to a control IC (IC501) it's peak is about 392VDC this is the B+ used to be pulsed through the transformer. the R501 adjustment allows more or less voltage through to the sense pin of the control IC to raise or lower this voltage.

while technically not "B+" the main high voltage (HV+) of 140VDC used to drive the flyback (and adjusted by R601) which would normally be "B+" in other older monitors.

so in a nutshell typically, the high voltage DC side voltage is just normally called B+

generally if I say B+, i'm usually referring to the high voltage used to drive the flyback and/or possibly the horizontal deflection yoke winding if it doesn't have it's own supply (like the WG 9200/9400 does). (sorry, it does share a supply, but each has its own drive is what i meant by that.)

in any case ramble ramble ramble.

the HV and B+ in this case, very likely don't need to be messed with. quite often the spec on B+ is something super dumb like maybe... 147vdc +/- 12vdc...  so honestly it could be anywhere from 135 to 159 volts and still be fine. hell i've read specs saying B+ should be like 113 and i measure it and it's 145... and then turning it down makes it collapse and die. so take any stats you see with a grain of salt, since really...much like cooking... it should be done by feel. if it feels okay... it's okay. maybe it's just cause i'm an old grey beard.

i'm going to bed now, i feel old.

[Rocketeer2001]

Shawn, thanks for the thorough explanation. Sounds like I don't need to make any adjustments to those since they have a wide operating voltage range.

I was able to take a video of my ghost wave. It doesn't play well with the camera and I get some moire going on, but you can still faintly see the issue. Look for a vertical wavy ghost line on the left side of the screen:


What's odd is yesterday it was a thin white wavy line and I had the screen's focus adjusted pretty well. Today when I turned on the arcade the wavy line was now fat and transparent, and my previously focused images are now out of focus. Not sure what that's all about. I adjusted focus on the flyback again to get it more crisp, but I can't seem to get it as good as I had it, and the fat wavy line doesn't revert back into a thin wavy line.

Edit:
I was just playing with adjustments again and this time I can get that wavy line to be almost invisible and get the image sharper by tuning down the brightness (now at 50%) and contrast (now at 65%) in the OSD. It's not as bright as I'd like, but it has less image issues this way, so maybe that's the only fix I can do.

[Rocketeer2001]

I'm going to stick a fork in this thread, because I think this was as successful as it could be, so it's done!

Still some weird image warping and blurriness, but hey, I can still play the game and that's more than I could say a year ago.

Thanks for all the help Shawn, you the best! Here's some poor photos of the screen working.

[lilshawn]

honestly when you play a game, you won't notice these things. you won't notice if the corner is a little fuzzy, or the side is not quite straight... you are looking at the action in the middle... where the image is like 90% perfect. not much happens in the 50 - 60 % regions and you aren't looking at them except for peripherally.

flat and near flat tubes are notoriously hard to get "perfect".... especially with how short they have to make the tubes to not make them a mile long to get perfect geometry easily.

you basically have to get it "good enough" or "as close as you can get"... or even "split the difference".

think about this... look at the back of your tube... imagine an electron traveling at about 300,000 kilometers per second out the color gun and have to make a hard 50 degree left turn to get the edge of the monitor and hit a target the size of one of the CRT's pixels... then another electron... but at a very slightly different angle to hit the pixel next to it..... then the next time, slightly different again for the one beside that one...do this 196,605 more times to make a picture.... then do all that again 59 more times... then do all that in 1 second. that is somewhere in the neighborhood of 11.8 million voltage adjustments to the deflection yoke... per second. That's a lot of switching and adjusting and compensation... just to get a straight line.

it's REALLY hard to get things perfect without sophisticated computer controlled switching... even then it's not "perfect".

our eyes have been sullied by the perfect squareness of LCD manufacturing and the super bright displays of plasma and LED backlit displays. everyone has forgotten what a CRT actually looks like when trying to draw something at the speed of light.