Repairing the video card for CAVEMAN
Video problems
Identify the origin of frequents failures of the video block.
Presentation of the video block
It comes in the form of a rack placed at the bottom of the cabinet of the pinball. The block is composed of 6 cards plugged on a backplane bus.
Functions of cards and references from left to right:
- A23 - MA-253 - I/O RAM
- A22 - MA-254 - EPROM
- A21 - MA-252 - CPU
- A20 - MA-255 - VIDEO 1
- A19 - MA-257 - VIDEO 3
- A18 - MA-256 - VIDEO 2
The video bloc is a stand-alone device built around a 8086 CPU:
Schema:
The VPTF
To troubleshoot video block of the Caveman, ideally to have VPTF (Video Prototype Test Fixture). This device allows individual card test and made an accurate diagnosis.
Thanks to Jean-René Karr, who kindly gave us the photos and allow to share the documentation.
VPTF techbook:
The problem is that this kind of material is very rare... a single copy known in France!
Diagnosis
In case of failure, he must first identify the card in question. Fail to have the VPTF, must have a second functional block or not having the same kind of failure. In our case, we had two blocks where on the first, the joystick was not working and on the second, the video was down (squares of color display).
It is then possible to perform a cross test to determine the defective card. Exchange the cards one by one with the second block and look at if the behavior changes. In our case, we identified the problems came from the I/O RAM card. Often it is this board (the first in the rack) that is in question.
Repair
Once the board is isolated, it remains only to read the schema and to arm themselves with a multimeter and an osciloscope to try to find the failure.
For the failure of the joystick, the U1 and U3 circuits were involved. Once replaced, the board works again.
For the failure of the video, the RAM 2114 were defectives. As a precaution, we have all replaced them (U13 to U16).
Another troubleshooting
On another pinball, game works properly but the video showed yellow vertical strips, of a width of 4 pixels. It was obviously a problem of video memory, either at RAM 6116, either at the decoding access or generating the video signal.
A cross test with a functional module has enough to identify the faulty card: it was A18. On this board, the video signal is generated from two buffers 8282 (U3/U4) and two shift registers 74LS165 (U5/U6). Analysis of the schemes, the cause of the failure was determined to be U3 or U4. The U4 circuit has been replaced and the card works again.
