Robotron hell!!!! (1-2-1)

Kid Raster

Well-known member
Joined
Sep 25, 2006
Messages
762
Reaction score
405
Location
Tulsa, Oklahoma
Hey everyone-
I'm looking to you William's experts out there.
Let me start by saying that I have even tried sitting in a chair in front of this machine
and stared at it for hours, hoping to will it to work. ;)

Here's the story....
this Robotron used to work great.
One day we plugged it in, and it did not.

Work log summary:

Error code '1-2-1'

Suggestion:
Possible bad RAM chip at location '2-1'
Purchased new chips, replaced all RAM chips.
Turned machine on.

Error '1-2-1'

Next, it was suggested the voltages were incorrect.
Purchased new switching PSU from Bob Roberts.
Installed switch, verified 12.0v and +5.0v at output of switch and at board.
Turned machine on.

Error '1-2-1'

Next, it was suggested that all cables/terminals on the board may need to be
cleaned and/or reseated.
Removed all cables and connectors from board.
Cleaned all connectors and pins.
Reconnected all cables and connectors to board
Turned machine on.

Error '1-2-1'

Next, it was suggested the socket at location '2-1' had cold solder.
Removed RAM chip from socket at location '2-1'.
Re-flowed solder on all pins on socket at location '2-1'.
Replaced RAM chip into socket.
Turned machine on.

Error '1-2-1'.

Next, the battle-axe was removed from the attic.....

Anyway...
The machine is dedicated original other than the switching PSU.
Please help if you can!

The survival of the last human family depends on you!!!
 
Could be an addressing problem. You'd need to take a look at the schematics and have a prod around with a logic probe. 1-2-1 is the first one its failing on but it could be a whole bank that it can't read.
 
The first thing I do on any Williams board is pull the board and remove all the old solder from the pins (of the connector) that supplies power to it. Usually the pins have broken solder connections.

Then I inspect the pins both on the top part and bottom part for pitting or signs of overheating. If they look good, I re-solder them to the board and try it again. Loss of voltage can show up as 121 error.

I have even had to rotate the pins one quarter of a turn each and then replace the spring connector on each wire on the harness, so that they make good, solid contact. The last time I used Bob Roberts Williams OEM power supply connector and repair kit. Works great.
 
Normally power supply problems manifest as a 1-3-1 error because that is the first chip the RAM test tries to read.

The most common issue is with the addressing chips (74153s) at 4E, 3D, 3E and 4F. You need a logic probe to test them or a logic comparator. Next most common are the 74LS373s at 1H, 3H and 4H. Again you wil need a logic probe to test them.

Other possibilities include the address and data buffer chips at 1D, 2D, 1G and 2G.

Quirky memory addressing issues are always fun to track down and they are usually the last chip you would suspect of being the problem.

ken
 
I think after reading a lot of the replies, most of this goes a bit over my head as far as technical expertise. I do not own a logic probe, and at the very least know how to operate one...lol- does anyone know of someone who would be willing to take a look at this board and get it running for me? I'd be willing to pay to get it working again. The funny thing is this machine worked fine at one point and then just stopped.

Anyway, if anyone could reccommend a good repair person I would be quite appreciative. Nothing worse than having a great game like this staring you in the face for months and not wanting to work....:p
 
what ? you gonna just give up like that? Dude..buy a probe a soldering iron and some solder and learn to track down the problem. It gets expensive to send boards out for repair every time it breaks..and it will break again.
 
Normally power supply problems manifest as a 1-3-1 error because that is the first chip the RAM test tries to read.

The most common issue is with the addressing chips (74153s) at 4E, 3D, 3E and 4F. You need a logic probe to test them or a logic comparator. Next most common are the 74LS373s at 1H, 3H and 4H. Again you will need a logic probe to test them.

Other possibilities include the address and data buffer chips at 1D, 2D, 1G and 2G.

Quirky memory addressing issues are always fun to track down and they are usually the last chip you would suspect of being the problem.

ken

You're right, its been a couple of years since I ran into this problem and I think the last time it was a 74153 that was bad.
 
what ? you gonna just give up like that?

Yeah.


I don't want to buy and have to learn how to use all of that shit, I'd rather just pay an experienced and reputable person to fix it for me. There are tons of people out there that do it, and I'm not hurt about giving them some business... ;)
 
Yeah.


I don't want to buy and have to learn how to use all of that shit, I'd rather just pay an experienced and reputable person to fix it for me. There are tons of people out there that do it, and I'm not hurt about giving them some business... ;)

Shall we start the bidding at $65 plus shipping?
 
Yeah.


I don't want to buy and have to learn how to use all of that shit, I'd rather just pay an experienced and reputable person to fix it for me. There are tons of people out there that do it, and I'm not hurt about giving them some business... ;)

My Joust did the 1-3-1 thing last year...its still sitting waiting for the issue to be fixed/resolved. I even sent it out for repair. Probably something simple that any Williams nut could fix but I have given up. Some day, ill come across a solution.
 
Last edited:
I usually do not hijack or resurrect an old thread but I am having very simmilar isues with my Robotron and have followed the exact same steps, only mine is a RAM 11 error...
Replaced
Reseated
Resoldered (header pins and sockets at ram 11)

It does play for a minute but only for about 5 waves and then it freezes, locks up and I (punch it) I mean turn it off...
 
Before fishing in the board cesspool that is eBay, try pressing on the connector at the MPU end of the ROM ribbon cable. Because the connectors don't have the little cams on the sides it is not uncommon to pull the wires loose when unplugging the connector onto the MPU board. I normally use a flat bladed screwdriver to press on the top of the connector and lift it on one side then the other until it lifts free. That way I don't stress the cable.

If you see a difference in the behavior while pressing on the connector, you may need to recrimp the connector.

ken
 
Before fishing in the board cesspool that is eBay, try pressing on the connector at the MPU end of the ROM ribbon cable. Because the connectors don't have the little cams on the sides it is not uncommon to pull the wires loose when unplugging the connector onto the MPU board. I normally use a flat bladed screwdriver to press on the top of the connector and lift it on one side then the other until it lifts free. That way I don't stress the cable.

If you see a difference in the behavior while pressing on the connector, you may need to recrimp the connector.

ken

Thanks for the tip!
I reflowed the pins on the board and I must have managed to wiggle a wire loose. Good catch!
So far its working ok but not saving highscore or keeping my settings...
 
New batteries and make sure everything looks good around the battery contacts and solder points on the board. Use your meter and make sure the 4.5 volts is making it to where it should.
 
The most common problem there is corrosion on the battery posts. Either buy a battery case from Radio Shack and mount it off the board, or do the lithium battery mod.

ken
 
Ram 5 error.
Plays for a level or two then goes wiggity whack.

Actually in a same spot at sgtsmackaho.
Prior to the Ram error It played great but would never save the settings. would have to set free play at every power up.

I have rebuilt the PSU and upgraded to the lith battery when I got the game. its minty and never saw route use.
 
Last edited:
You may want to play 24 RAM pickup. Pull all of the RAM chips, dump them into a big bowl, stir them around and then replace them them. The rug test is based on a pseudo random number gernerator that is seeded with 2 different values (one for each pass). The problem with this is that there are over 100 bits in RAM that never get tested. The odds are that the RAM test will find the faulty RAM chips, but if there is a single bit failure in the right place....

By pulling the RAM and replacing it in a truely random fashion, you increase the odds that every bit will get tested. That way the RAM test finds the bad bit, not the game.

ken
 
Back
Top Bottom