P299 Filter Capacitors Question (Battlezone)

murphy112358

Member

Donor 2018
Joined
Jan 22, 2018
Messages
45
Reaction score
12
Location
Kent, Washington
Hi all. I've recently acquired a Battlezone cabaret (newbie here) and just discovered the big filter caps on the deflection board are not 2x 6700 uF but instead 1x 4700 uF and 2x 2200 uF. I suspect the latter caps are wired in parallel, so I figure I have the equivalent 2x 4500 uF.

The display is very playable, but the vectors are a bit crinkly. I suspect this "under capping" would be the cause of this. (Too much AC ripple?) Could anybody confirm or deny? I wonder if this is a well-known substitution trick and not anything to be concerned about. I don't have access to an oscilloscope to measure.

Also, but perhaps unrelated, the HV, focus, brightness, and contrast knobs are all set relatively high. I am unable to make the display show spiderwebs (not that I really want to, but this may be indicating something.) All of the monitor voltages check OK, perhaps a little higher than what is in the schematics but plausible because of the high settings. (Did NOT check HV anode voltage.) I am guessing possible leakage or tired transistors somewhere in the Z-amp?

Any insight welcome, including where to buy high quality new 6700 uFs, thanks!
 
The P299 board normally came with 6800uF's, but I've used 4700's before in a pinch (as this is what the 6100 deflection boards used, and they're a similar design).

They are not wired in parallel. There is a separate cap for the + and - voltages, so in your case you're under-filtering on the 2200uF side.

Replacement caps here:
https://www.digikey.com/products/en?keywords=338-1649-ND

I'm sure arcadepartsandrepair.com has them as well, though I don't know his part number offhand.


Also, you should do the resistor mod, if it hasn't already been done. See post #33, here, regarding it, and brightness issues:

https://forums.arcade-museum.com/showthread.php?t=356350&page=4
 
Thanks for the insight, but do not I understand the "not wired in parallel" remark. I have yet to take the board out and inspect it, but I see three capacitors zip-tied together in an area where there should only be two capacitors. The three cap values are 4700, 2200, 2200. I can only imagine the 2200s were combined to make a 4400, otherwise, it would be in real sad shape.

I do still have the 15W resistors on the board, but don't have any degrading z output when left on for several hours (FWIW).
 
Ah, I misread that there were two 2200's. My bad. (I read it as one.)

You want to get rid of the resistors anyway, as if they haven't already caused damage to the transistors, they will. (It could just be that they've fried one or more of the transistors in that area, as that's where the brightness control circuitry is, and I've seen cases where it's just dim from the start, like yours.)

I'd replace the filter caps, remove and bypass the power resistors, replace and Q500/501/502, and see where you stand, as those are things you'll want to do anyway.

Also, reflow the solder on the pins of all of the connector headers on the deflection board, as that's another common cause of issues with these. If they have never been reflowed, they are almost certainly cracked. Reflow each one with a drop of fresh solder.
 
The P299 board normally came with 6800uF's, but I've used 4700's before in a pinch (as this is what the 6100 deflection boards used, and they're a similar design).

They are not wired in parallel. There is a separate cap for the + and - voltages, so in your case you're under-filtering on the 2200uF side.

Replacement caps here:
https://www.digikey.com/products/en?keywords=338-1649-ND

I'm sure arcadepartsandrepair.com has them as well, though I don't know his part number offhand.


Also, you should do the resistor mod, if it hasn't already been done. See post #33, here, regarding it, and brightness issues:

https://forums.arcade-museum.com/showthread.php?t=356350&page=4


yes we have the 105c caps, transistors and the zero ohm jumper wire in stock. if OP has trouble finding anything let me know as i would be happy to help.
 
Is the "105C" like a temperature rating? Higher the better? The Digikey link had theirs at 85C.

I might replace the Q503 too, while I am in there. I'll speculate that my cathode voltage has drifted too positive--I think that results in a dimmer spot.

For fun, I'll report the outcome here in a couple of weeks after I've collected all of the parts needed.
 
Yes, it's the temperature rating. Generally the higher the better.

In this case those filter caps don't get very hot, so a high temp cap isn't critical. Any quality cap will work.
 
Regarding the transistors in the spot killer and Z-amp, are the part designations still commonly available? I've dug out the part numbers from the monitor manual:

- NPN TPS 98 (MPSA06 is a substitute?) (Q500/Q501) (spot killer)
- PNP MPSA56 (Q502) (spot killer)
- NPN 2N3904 (Q503) (spot killer)
- NPN D40P3 (Q504) (z-amp)

I ask this because that thread you referenced mentioned a common replacement for one of the transistors. Thanks again!
 
Those are basically correct.

Be careful with the transistors on those deflection boards. Some of them were shipped with TIS98, not TPS98, and some of the boards were silkscreened with the part outlines for TIS98.

MPSA06 is a direct sub for TPS98 (same orientation). But if the original part was TIS98, you need to flip the MPSA06 around, so it has the opposite orientation when installing.

Also, the D40P3 rarely fails. (I've never seen one go bad.) I'd leave that one till last.
 
Regarding the transistors in the spot killer and Z-amp, are the part designations still commonly available? I've dug out the part numbers from the monitor manual:

- NPN TPS 98 (MPSA06 is a substitute?) (Q500/Q501) (spot killer)
- PNP MPSA56 (Q502) (spot killer)
- NPN 2N3904 (Q503) (spot killer)
- NPN D40P3 (Q504) (z-amp)

I ask this because that thread you referenced mentioned a common replacement for one of the transistors. Thanks again!

if you look on our site they all are there except the D40P3. you can search by the original number and all subs will come up.
 
If you do need the Q504, you can sub Digikey part number 2SC4793FM-ND. Or also an MPSU10, or 2N6558. But again, it's more likely one of the other transistors.
 
FWIW, here is a top/bottom photo of my P299. All of the header pins appear reflowed, nice and shiny.
 

Attachments

  • P299 Top.jpg
    P299 Top.jpg
    1.1 MB · Views: 18
  • P299 Bottom.jpg
    P299 Bottom.jpg
    1.1 MB · Views: 12
Given the thread's title, I figure a photo of the deflection board with the new, per spec, filter capacitors was in order. (6800uFs)

Funniest thing is, there was no impact. Which means two things: (1) I did it right (my second repair), (2) It wasn't a problem. Indeed, it seems using smaller filter caps is just fine.

The bw vector monitor faq mentions the Spot Killer LED (D904) could be temporary removed to quickly verify whether its causing dimness. Could it be possible this LED is leaking current?
 

Attachments

  • P299 Top New 6800 Caps.jpg
    P299 Top New 6800 Caps.jpg
    1 MB · Views: 10
I would remove the two big ceramic resistors R100 and R101 (as andrewb mentioned) and install wire jumpers in their place. Here is one way to do so, by scraping off some solder mask and installing them on the bottom of the board:
1aaea5e8928715406547d30f09e3cf80.jpg
 
What douglasgb said regarding R100/101.


Did you replace the transistors that were previously discussed (Q500-503)? If not, that should be next.
 
No, I haven't [yet.] I want to try to avoid shotgunning it. (Doing this all as a learning exercise, expedience not important.) It seems removal of the diode would be enough to implicate the xy spot killer circuit block as the culprit (whether it is just the diode, just the transistors, or both.) If there was no change after removal of the LED, then it would seem the transistors would be good.
 
I don't think the LED is the issue, nor will it tell you much if you remove it.

Those four transistors are heat-worn and tired anyway. Plus they are cheap and simple to replace. In this case, shotgunning them is fine, as they're a likely culprit anyway. And even if they don't fix it, replacing them isn't a bad thing, given the heat they've taken over 30 years from the R100/101 resistors. Consider it cheap insurance.
 
I would remove the two big ceramic resistors R100 and R101 (as andrewb mentioned) and install wire jumpers in their place. Here is one way to do so, by scraping off some solder mask and installing them on the bottom of the board:
1aaea5e8928715406547d30f09e3cf80.jpg

Yeah, the bw vector monitor faq reports there was an Atari service bulletin instructing operators to remove these. However, I suspect the context at the time was the game only had 2-3 years left before getting discarded.

Those big filter caps cause a huge current spike on power up, punching your power transistors (and other stuff) in the mouth each time. It seems replacing/upgrading R100/R101 would be better for increased longevity.

Nice bridging technique, BTW.
 
Atari themselves removed R100/101 on the later revisions of the deflection boards. They aren't needed.

I've removed them on literally 50+ boards I've refurbished for folks. It's never been an issue, and the heat savings is better for all of the parts around them.
 
I don't think the LED is the issue, nor will it tell you much if you remove it.

Those four transistors are heat-worn and tired anyway. Plus they are cheap and simple to replace. In this case, shotgunning them is fine, as they're a likely culprit anyway. And even if they don't fix it, replacing them isn't a bad thing, given the heat they've taken over 30 years from the R100/101 resistors. Consider it cheap insurance.

Seems temporary removal of the D504 LED would quickly confirm/deny whether the XY spot killer circuit is causing problems or not. This would have to be a common troubleshooting technique. On the schematic, this LED is on the rail from the B+ to the emitter of Q504. If this path is open, then the XY spot killer can no longer be exerting any influence.
 
Back
Top Bottom