SWTPC Progress...

Hi, Some time ago, I asked for some help locating documents for the Percom LFD-400 disk system. A number of people kindly helped out so I thought I would give a brief progress report to the list. 1) I removed various hacks on the MP-C and MP-S boards to return them to "factory" spec. 2) I implemented a switch on the MP-A CPU board to turn the onboard RAM on and off. This made debugging much easier. At this point, I was able to boot into SWTBUG with the MP-A RAM turned on. 3) I debugged several problems with the GIMIX 16K static RAM board. The original owner reported that this card always caused problems. No schematics seem to be available for it so the troubleshooting was slow going. The most interesting problem was that one of the 5 volt regulators (there are 4 on the board) was only providing 1.8 volts. As well as powering one of the 4K banks, this regulator was supplying power to some of the address decoding chips. With the out of spec supply, these chips were interfering with the correct operation of the address bus, preventing the whole system from operating. Other problems included some dead RAM chips, some bent pins on socketed chips and 39 undocumented dip switches. If somebody turns them up, I would love to get the docs for this board. At present, I have determined that 16 of the switches control the address of the 4 4K banks of RAM. I have left the other 23 switches as they were set when I acquired the system. Experiments as simple as turning them all on or off result in a non- functioning system. 4) I implemented the MP-B motherboard enhancement documented in the September 1978 issue of Interface Age. Fortunately, my MP-B was built with sockets, so the modifications can be easily removed if it should be necessary to restore the system to "factory" behavior. 5) I tied the MP-B mod into the address decoding on the 32K GIMIX card and the system now has 48K less 32 bytes of RAM and boots into SWTBUG with the MP-A RAM turned off. 6) I then proceeded to the Percom floppy interface board and determined to my great pleasure that it was functioning perfectly. I was able to jump into MINI-DOS+ with the SWTBUG 'Z' command, format a disk and create some files. I did this initially with a different 5 1/4 inch floppy drive as the original Percom (Shugart) drive had some problems. 7) I corrected two issues with the Percom drive and it now functions correctly. The first issue was shorted caps on the logic board. These were easily identified as they were smoking. The second problem puzzled me for a while. The drive worked fine as long as I did not put the cover on it. MINI-DOS is, unfortunately, pretty cryptic with its error reporting, only giving integer error codes. The magazine articles that I got don't document them. I finally figured out the problem when I noticed that the drive stopped working when I leaned over it. Turned out that there was some dust partially obscuring the index sensor. With the lid off and the overhead light shining on the drive, there was enough index signal for the drive to work. Put the cover on, or even shade the drive, and the internal LED was not producing enough light to detect the index holes. Interestingly, canned air was not enough to dislodge the dust. I had to reach in with a small, soft, unused paint brush and poke around to get it to the state where it works reliably with the cover on. One neat thing about the drive is that it has two sets of index and write-protect sensors. It is "flippy". It only has one read/write head, but you can insert a double sided diskette in either side up and use both sides. So, now I need some software. For those visually inclined, I put some pictures here: http://wsudbrink.dyndns.org:8080/swtpc6800/ Pictures 1 to 11 are of the AC-30 cassette interface and CT-64 terminal that I will be working on next. Pictures 16 and 17 are of the 6800 cover (I'm missing the screws). Pictures 18 to 21 show the output and listing of an SWTPC BASIC program I wrote to solve last week's Car Talk puzzler. Pictures 22 to 25 show SWTBUG working and using both serial interfaces. Pictures 27 and 28 show booting from SWTBUG into MINI-DOS+, getting a file listing and loading BASIC from a diskette. Pictures 30 to 35 show the beast itself. The cards from front to back are 1) The GIMIX 32K RAM 2) The GIMIX 16K RAM 3) The Percom floppy interface 4) The SWTPC MP-A CPU 5) The I/O card to the left (slot 0) is the MP-C and the card to the right (slot 1) is the MP-S. Picture 34 also shows the Percom floppy drive. Picture 35 is from the back. You can see the switch to control the MP-A RAM and also the small brown card that is the MP-B addressing enhancement. Bill