More testing of 5806223 replacement card and redesign to eliminate donor card requirement

CONVERTING THIS TO A DIRECT SLT CARD CONNECTOR ON THE PCB

I found some spring contact parts that would allow me to make a compatible end on my PCB that would slide into the card compartment and mate with the pins in the socket there. This did require some very careful measurement of IBM SLT cards to understand all the dimensions 

I implemented the pads onto which I will solder the contacts. The result will look very much like the original IBM end of an SLT card. I then need a 3D printed cover similar to IBM's black molded plastic part that protects the contacts and guides the card into the slot when it is inserted. 

The spacing of SLT pins in the card slot is .125" or 3.175mm. The connectors will fit easily in that spacing and are about the same width, length and height as the ones IBM uses. 

The plastic cover on the end of a card has grooves to slide over guides in the SLT board (backplane) at each card slot. It has notches to ensure the card can't be inserted rotated 180 degrees. It has .1"x.1" square openings that the SLT board pins slide into. 

I will need to design these covers and print them. I have time since the parts and PCBs will take a week or so to arrive. 

TESTING TO UNDERSTAND WHY INTERRUPT LEVEL 3 WAS ON INITIALLY

I did find that often the 1130 would power up with a level 3 request active. My design should initialize correctly and only drive the request after a plotter command has been executed. I took the card out and put it on the bench, reproducing the erratic IL3 behavior. 

The big breakthrough was while I was redesigning the card to feature an SLT connector. I spotted an error in part of the reset logic for the flipflop that drives the IL3 request. 

A gate will reset the flipflop either when the 1130 system reset signal is active or when an XIO Sense Device with bit 15 set is issued to area 5 (the plotter address). The signal for the latter case passes through an inverter. I realized that the inverter was a 74LS05, an open collector inverter, thus it should have had a pullup resistor on its output but it did not. 

I did a bodge change to the board on the bench and verified the correct functionality. I placed the card back into the 1130 and indeed the IL3 does not come on spuriously. I don't want to spend too much time working on the old design, when I will soon have the integrated SLT connector on the updated board.