Stargate Power Supply - C11 reversed before recapping, should it be?

[WhereEaglesDare]

C11 was reversed and the board had already been recapped once before by someone else, but a few were swollen.  Here is a pic.

you will see the stripe next to the plus. Please help.

[WhereEaglesDare]

Also, does anyone have access to a schematic?  R16 is broken in half and can't figure out where the 2N3055 is suppose to go on the heat sink. Please help?  I need the value for r16 and I need to know which transistor on the heatsink to change out. Please help. I have to reinstall this board for a customer tomorrow.

[smartbomb2084]

C11 is wrongly reversed.

You can get a schematic here along with a lot of other good stuff.

http://www.robotron-2084.co.uk/manualsstargate.html

[Ken Layton]

Schematics here:

http://arcarc.xmission.com/PDF_Arcade_Williams/Stargate_Drawing_Set_(16P-3002-103)_Dec_1981.pdf

[lilshawn]

well according to the schematic, c11 should be a non polar cap. (.1uf 50volt as evidenced by it's straight plates) likely a ceramic cap.

check what C8 is, it's doing the same job in the 5 volt rail.

the silkscreen could just be wrong and someone could have just plopped that one in there.

[smartbomb2084]

The power supply shown in the pic above is the smaller board power supply with the detached heatsink like found in ROBOTRON. C11 on that supply is a 22uF/10V electrolytic and is wrongly reversed in the pic.

The power supply shown in the STARGATE drawing is like the one found in DEFENDER and should have a .1uF/50V non-polar cap at C11.

You can get a ROBOTRON schematic from the link I posted above.....

[lilshawn]

see, i look at the robotron schematic you linked and the power supply board schematic shows the same .1uf 50 volt cap. 



I'm just thinking because this is a -5 volt rail the cap could actually be properly installed, but the silk screening could be incorrect.

[smartbomb2084]

Sorry...that is a STARGATE Schematic link above...here's the ROBOTRON one.


http://www.robotron-2084.co.uk/manuals/robotron/robotron_upright_drawing_set_mar_82.pdf

[lilshawn]

right, that's better. thought i was going crazy there.

Still part of the -5 volt system though. (LM7905 is a -5 volt regulator) I still get that feeling the cap is proper, just the silkscreen is wrong.

the + side of the cap in the picture IS near the edge and as such, the + lead of the cap could be tied to ground (as often ground plane goes around the outside of the board.)

OP should follow the trace and see if the + side of the cap in question is indeed tied to the ground (as it should in a (-) volt rail) and the negative of the cap follows back to the plug that feeds the -5 out to the rest of the system.

[smartbomb2084]

You are right about the ground often being on the edge of a board.

I'll check my game when I get home to see if the silkscreening is wrong....

Can't stand an easily solvable mystery going unsolved...

[smartbomb2084]

Silkscreened '+' is correct on my board.

[lilshawn]

but is the orientation of your cap match the orientation on the board in OP's picture...

IE, have you a revision where they've fixed the silkscreen? ir is the cap in OP's board indeed reversed?

[jennifer]

     A reversed screen?.... Now that's a new one for Jennifer, I would be thinking that that would be rare, like finding a two headed penny or something.  (Maybe not just never actually seen one).

[smartbomb2084]

but is the orientation of your cap match the orientation on the board in OP's picture...

No.... C11 is correct on my board oriented with the screening....The OP's board has it reversed.

[lilshawn]

huh, i wonder how it hasn't blown up. 

incorrect silkscreens are more common than you might think. For instance, some of the wellsgardner k7000 monitors have the silkscreen the wrong way around on c204 on the neckboard. some bipolar(nonpolar) caps have polarity markings too.

I've always just done them one at a time, noted the orientation of the cap when I removes them, and then replaces it the same ways I takes it out.

I think what happens is they do the schematic design, plunk it into some program that autoroutes and optimizes the board and they manually go in and change some stuff (cause auto-routing usually sucks) but the silkscreen layer doesn't get changed and slips by unnoticed until 10000 board show up and they start assembly and testing and they all fail. rather than throw out 10000 boards, they fix the issue (reversed the cap) and ship 'em.

[smartbomb2084]

You're right man...why didn't it blow up?

 Maybe the board never saw power after it was previously recapped?

The  '+' mark for C11 is actually a GROUND connection...Do you think that could be why?