Sunday, October 12, 2014

More progress restoring the 1053 console printer, plus continued coding of keypunch interface


After studying the mechanisms last night, I was armed with a good understanding of the parts involved in triggering and resetting the index cycle. I went out to the printer, discovered the errant pivot and freed it up. I ran the machine with power on, watching it cleanly execute carrier return and index operations exactly once.

The next issue is lack of activation of the second operational cam, which drives space, backspace and tab actions. When I activate the interposer links for space or tab, the cam is not activating and turning. I took the time for a bit more studying, to understand all the moving parts involved with this activity, before freeing it up.

There is a plate that rotates relative to the operational cam, either pivoting a pawl into the ratchet when moving the cam or retracting it when the cycle is done. This was so gummed up that it wasn't moving, but I worked the old lubrication out and got it moving. At first, it would latch but not release, until it was free enough to work reliably both ways.

Next, I found the escapement which controls spacing and backspacing was not working smoothly. It seemed that the pivot mechanism for the escapement and backspace pawls has those pawls slide left and right as well as rotate, but due to a gummy film between the layers, they were not sliding back and forth. Some attention to that cleared up the problem giving me a crisp spacing action.

Tab works smoothly as well. The problem is that the three actions driven by the left operational cam, space, backspace and tab, are can stick on taking repetitive cycles. The fault has to be in the restoring mechanism, guiding me to the next spot for special attention.

I also found a writeup in the theory of operations, sans any diagrams, that explains a bit about those bails for the operational cams which were configured two different ways. Based on the writeup, I should be able to sort out which is wrong and correct things. These are meant to block re-triggering, so the left one could be causing my problem with the tab, space and backspace.

The left operational cam is still repeating cycles. When I hand cycle the machine slowly it almost always rests properly. Hand cycling rapidly will sometimes give a few cycles before resetting, other times it resets correctly after just one cycle. Under power, however, it never resets, striding across to the right with continuous spacing or continuous tabbing or racing backwards in continuous backspaces.

I will keep at this until I find the sticky pivot and free it up. I had triggered some character cycles which seem to work properly, although I didn't have the platen installed or paper and ribbon in the machine. If I can get the operational cam restoring well I should be in position to start testing printing and then reinstall this in the 1130.

After more work, the tab and backspace take just one cycle, as they should, but the space function continues to be pesky. Under motor power, it never stops, taking constant cycles until I pull the plug, but when I cycle it slowly the clutch works properly. I will look into the adjustments for the restore, although it seems that this machine was working properly when retired so its adjustments should be correct.


I structured the program module with separate tabs for each major module and began more detailed coding. I completed the function which extracts a command line from the input serial port stream, then moved on to coding the parsing routine that breaks apart and validates the command line.

I did discover one flaw in my protocol, which is built around a line of no more than 81 characters, but has a command to load a user translation table requiring 516 characters. I had to modify the protocol to handle this given a max of 81 characters per command. The _LOAD command now handles one position of the translation table per line, specifying a decimal value for the character to be updated and then four hex digits as the keypunch actions for that position.

No comments:

Post a Comment