Installed prototype enclosure and attempted testing of Virtual 2315 Cartridge Facility

INSTALLED THE POWER COMPONENTS IN THE PROTOTYPE ENCLOSURE

The enclosure has an incorrect opening size for the Virtual 2315 Cartridge Facility (V2315CF) main unit, but I used it to work out the mounting locations for the various power components that fit inside. These are the power distribution pcb, the power supply, and the timer. I routed all the wiring through the side holes and the two ribbon cables through the rear hole of the enclosure.

This is the view that the operator sees when they tilt the top right cover of the IBM 1130 up (you can just see the hinge on the right of the picture), with the tabletop in front and the enclosure sitting atop the internal disk drive of the IBM 1130. The 3 3/4" diameter mini versions of the 2315 disk cartridge that are used with the 1130's internal disk drive (2310) are inserted into the opening on the right, which is a scaled down model of what the opening of the 2310 looks like. 

ATTEMPTED TO COLLECT DATA FOR THE WRITE ISSUE

I had brought out a signal on one of the unused output lines of the main unit of the V2315CF so that I could capture it along with the -Write Gate,  -Write Clock Phase B and -Write Clock and Data signals that the 1130 sends to the 2310 disk when it is writing a sector of data onto the 2315 disk cartridge. The V2315CF intercepts this data and stores it in RAM so that it can be written out to the mini 2315 cartridge which serves as the disk cartridge for our system with the V2315CF installed. 

The testing is simple. I first issue a read command from the 1130 that requests the data from cylinder 0, head 0, sector 0 be brought into memory at the location specified in the I/O instruction. The data on the mini cartridge should appear there - it has been reading correctly up until today - but instead of the valid data I found the filled with random garbage, mostly 16 bit words of 0xAAAA. 

If it had read correctly, I would then modify a few words in the memory buffer and issue a write command to store the buffer back in cylinder 0, head 0, sector 0. When I unloaded the cartridge at the end of the session, what had been written back to the mini cartridge was just the random junk, not the changes I made. Further, I found random junk written through many other sectors on other cylinders. 

It is acting as if the RAM is not working properly now. When I load a mini cartridge, the V2315CF reads the contents from the mini cartridge and stores it in RAM, then turns on the File Ready status allowing the 1130 to issue I/O instructions to the disk. However, the data is either not loaded properly into RAM or corrupted somehow while the testing is underway. Another possibility would have been that the rewrite to the mini cartridge when we turn off the disk drive is what is failing, except that the read command is finding the garbage so it is definitely in RAM. 

I may have introduced this error during the recent code changes I made for the diagnostic output needed to conduct this test. I will go home and use simulation to examine the behavior of my logic to discover what is going wrong.