Star Wars Cockpit Vector Issues

Well.. What has been done to the game since you got it? Have you done a cap kit? Worked on the PS?
 
I am getting, under MATHBOX TEST "Divider Errors"...

Hmmm...

Here is the error screen:

SWERRORS.jpg


Both the AMP boards and the PS have been recently rebuilt and should not be the problem...
 
Last edited:
Have you gone through that board at all ? Has everything been reseated ? Checked voltages etc..

I've done nothing thus far. I have no idea really where to begin. The game was working fine when it was powered up this afternoon and by the time Jeff and family left it was acting up. Any idea where the chips are that deal with the mathbox functions?
 
First I would check the voltages , second I would pull off that edge connector that connects all 3 boards and make sure the edges are clean and put that connector back on, you could try reseating the roms but i doubt it would help , if none of that fixes it then you have a bad IC and will probably have to send it out for repair unless someone else has an idea.
 
Welp, according to Spaeth the chips in question aren't socketed... and would require some work to determine what is bad and what isn't. He's suggesting I pick up a replacement boardset then sell the non-working one as of right now. He's also not in a position to do any work right now (with the move)...

...anyone else here work on Star Wars PCB's?
 
I thought elektron forge worked on these as well. And I thought I heard he had fairly quick turn around. I would check with him first to see if he had an idea of the repair costs. The board itself looks to be working fairly well. But it may just need one part...Or twenty.
 
This is out of the "STAR WARS TROUBLESHOOTING GUIDE"

Divider Tests

The divider circuit on the Main PCB is an unsigned 15-bit fractional divider which assumes that the dividend (numerator) is less than twice the divisor (denominator). The hardware consists of a dividend latch, a dividend shift register, a divisor inverting latch, an adder, a difference latch, a quotient shift register with tri-state buffer, and a divider control which includes a divide cycle counter. Refer to the detailed block diagram on Sheet 5B to aid in understanding the circuit operation.
The dividend latch is written into by the microprocessor and has two parts: the dividend latch high byte (DVDDH) and the dividend latch low byte (DVDDL). The dividend latch is not altered by the operation of the divider so it may be left at its previous value, if desired.
The divisor latch is also written into by the microprocessor and has two parts: the divisor high byte (DVSRH) and the divisor low byte (DVSRL). The DVSRH line loads the divisor high byte, clears the quotient shift register, and loads the dividend shift register from the dividend latch. The DVSRL line loads the divisor low byte and starts the divider.
The divider subtracts the divisor from the dividend in the dividend shift register and puts the result in the difference latch. If the result of this subtraction is positive, the carry (C16*) will be a "1." The "1" is shifted into the quotient shift register, and the value from the difference latch is stored into the dividend shift register. (In a conventional divider algorithm, the dividend shift register would then be shifted to the left, but here the output of the adder is wired to the difference latch which is already shifted to the left.) If the result of the subtraction is negative, the carry (C16*) will be a "0." The "0" is shifted into the quotient shift register and the dividend shift register shifts once to the left. The value in the difference latch is otherwise ignored.
The following Test 21 through 25 determines the condition of the divider circuit. The test number will be shown on the display (assuming that the display and the AVG PCB are working). Each test is retriggered every 50 to 60 microseconds for as long as the test is selected.
Perform the Preliminary Procedure under Hardware Diagnostic Tests and set the option switch at location 10D on the Main PCB as indicated in the test.
The Mathbox Tests self-test display performs Hardware Diagnostic Test 21 through 25. The Mathbox Tests display shows the Option switch setting of any test that fails to produce the correct results.

Test 21

Tests for 15 pulses at pin 10 of divide cycle counter at location 8P.

Option Switch Setting:

87654321ONONONONOFFOFFOFFOFF
4000 (Dividend)/4000(Divisor) = 4000 (Quotient)

Test 22

Tests for shorted bits in the dividend and/or quotient.

Option Switch Setting:

87654321ONONONONOFFOFFOFFON
5555 (Dividend)/4000(Divisor) = 5555 (Quotient)

Test 23

Tests for shorted bits in the dividend and/or quotient.

Option Switch Setting:

87654321ONONONONOFFOFFONOFF
2AAA (Dividend)/4000(Divisor) = 2AAA (Quotient)

Test 24

Tests for shorter bits in the divisor. The data is inverted and appears at the outputs of the divisor inverting latch at location 4P, 5P, 6P, and 6L.

Option Switch Setting:

87654321ONONONONOFFOFFONON
2AAA (Dividend)/2AAA(Divisor) = 4000 (Quotient)

Test 25

Tests for shorter bits in the divisor. The data is inverted and appears at the outputs of the divisor inverting latch at location 4P, 5P, 6P, and 6L.

Option Switch Setting:

87654321ONONONONOFFONOFFOFF
5555 (Dividend)/5555(Divisor) = 4000 (Quotient)

Does that mean I could change the DIP switch settings to those indicated to correct the divider??
 
I thought those instructions referred to a piece of equipment that Atari techs had to test the board set with.
 
Do not loose your faith in the Force. Your game will get fixed. Lets see..though. Have tried any of the suggestions so far in this thread?
 
Naw...not losing my faith in the force. I've pulled and reseated the edge connector. After talking with Spaeth about the issue he made it clear it wouldn't be a socketed chip needing re-seating. I still need to pull out the boardset and re-seat all the connectors between the boards.
 
There is a section on the AVG board that implements a hardware divider with discrete logic chips.

You can look at the schematic package page 5B for a block diagram of what's involved and pages 9A and 9B for the actual chips involved.

The N's and F's in the first column refer to positions you can set the DIP switches on the mathboard to to run some hard-coded instructions. The manual shows some LED states when you do this along with some waveforms that you could use an oscilloscope to view.

There are 20 or so chips in this section. I don't think any of them are socketed. You could use the schematics to eliminate some of them if your clock output is good or if you're sure some data is getting latched in. I don't know if this is beyond your skills or not.

At a minimum, you could use the chip numbers from the schematics to do the 'touch-test'. See if any of them are burning hot when the game is malfunctioning.

With a logic probe or oscilloscope, you can look at the input/output pins to see if signals are getting passed.

Good luck.

Kerry
 
Short of the "touch test", the rest is beyond my capabilities and/or tools on hand.

It looks like I am pretty much looking at sending these boards out for repair or purchasing a working set and keeping these for backup "parts" purposes.
 
Short of the "touch test", the rest is beyond my capabilities and/or tools on hand.

It looks like I am pretty much looking at sending these boards out for repair or purchasing a working set and keeping these for backup "parts" purposes.


Aw, don't part them, get them fixed. Even if you buy a new set make sure these ones stay intact and alive.

Or if you sell them, I'll take them, and get them fixed. I might need a SW set for my project anyway depending on how damaged the ones I have are.
 
Have you even bothered to check the voltages ? no offense Frizz, but it amazes me when people dismiss the simple things like the voltages and edge connector and will put in a new boardset. If your voltages are off then the new board could be damaged too.
 
Back
Top Bottom