Image Map Image Map
Page 2 of 11 FirstFirst 123456 ... LastLast
Results 11 to 20 of 106

Thread: REMEX Paper Tape reader

  1. #11

    Default

    I agree, the 7474s seem to be the most common to fail. Of course, if you have Fairchild parts, you should just replace them before even bringing out the 'scope. It would be interesting to have a failure analysis done.
    Dwight

  2. #12
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    2,069

    Default

    There are various websites describing failure modes of ICs. The JEDEC website is pretty much the definitive guide.

    Manufacturing processess have been getting better with time - which basically means that the older the device is the more likely it is to fail (as you have found).

    Various manufacturers - and even specific batches of devices - have their problem areas. For example, some batches of ICs I know about used a bonding agent that attacked the IC itself over time (the aluminium mask if I remember correctly). As others have said, moisture ingress is a big killer for devices of the age of your PDP8.

    Welcome to the world of owning a vintage computer. Fix it first then use it for a while... Repeat for as long as you own the machine. Eventually, though, you may weedle out the 'weak' ICs and it may stay reliable for a while. Who knows?

    Dave

  3. #13
    Join Date
    Dec 2013
    Location
    Near Milwaukee Wisconsin
    Posts
    1,040

    Default

    Interesting, my equipment is in my basement. The only place that I can use. I have a dehumidifier running, but it does get a little cool during winter. I have measured low 50's on the floor. My feet can testify to that. Maybe I should run a little heat in my computer cabinet. At the power plant I worked at, years ago, all the large motors and electric cabinets had heaters in them. I suspect it was an attempt to keep the moisture out. I'd probably would have to close up my computer cabinet some. It is missing a side and the front door. Mike

  4. #14
    Join Date
    Jan 2014
    Location
    Freedom City
    Posts
    6,253
    Blog Entries
    2

    Default

    Maybe a Dampp-Chaser is in order.

  5. #15

    Default

    All of my toys are in my cellar. The dehumidifier keeps it 40%-60% during the summer, and it will go below 40% in the winter. In the summer it rarely goes above 70F, and in the winter it will get in the 50s.
    Everything seems to be pretty reliable in that environment. The spindle bearings in the RP06 complain when they are cold, so I have to run the drive for about 30 minutes before they get quiet and I can load the heads.

    At the RICM, the systems that are in the Lab (office) space are in a nice climate controlled environment. Only one flipchip has failed in the PDP-8/I in the last several years. The RX8-E or the RX02 just died in the PDP-12, so I need to fix that.
    Member of the Rhode Island Computer Museum
    http://www.ricomputermuseum.org

  6. #16

    Default

    Temperature per se isn't really much of an issue especially non-operational aside from people comfort. As mentioned earlier its moisture that kills things like 'plastic' 70's ICs. My little workshop in the back of the garage generally gets up to around 80F in summer and I have a heater on a thermostat which keeps it from going below a nominal 40F in winter as a concession to the age of the kit. My own operating range isn't as wide though, I prefer around 65F +/-5. Some of my kit was rescued from a pretty wet environment and after many years moisture ingress is still taking a steady toll, not so bad now though - maybe only a single fail per year or so.

  7. #17
    Join Date
    Dec 2013
    Location
    Near Milwaukee Wisconsin
    Posts
    1,040

    Default

    Well..... I only lost one chip this time, two last season and one the season prior. So maybe it's not a big deal and I'll live with it, especially if maybe I'm sculling the herd. The machine seems to be working fine at the moment, and the previous years also worked fine after the initial repairs.

    I still want to get the REMEX to work and add it to my equipment that is useful, rather than taking up shelf space. I think the next item I want to accomplish is to make a good tape that I know works. I figure that I'll make a tape that has LBAA, the bin loader on it and the DEC-8-5-u-SYM, the binary punch on it. See if I can load both programs from my ASR33 punch/reader and be sure that they run. Then I should have a reliable tape to use for testing the REMEX.

    Thinking ahead, will I have to make a device handler for the REMEX once I get a program to work? I was wondering exactly how other programs would access it. Mike

  8. #18
    Join Date
    Dec 2013
    Location
    Near Milwaukee Wisconsin
    Posts
    1,040

    Default

    Today I successfully made a paper tape of the BIN LOADER LBAA and the Binary Pinch program DEC-8-5-U. I had both programs on my IBM XT, Then connected the ASR33 to the COM1 port of the XT. I then tested the paper tape by reading it in from the ASR33 to the PDP8E. Both programs loaded fine. I then took some time to play around with the Binary Punch program. You have to enter the number of blocks. I found this a little confusing and still do not know exactly what this is. I entered SR Bit 11 = 0. This turned out to use block zero. I wonder what it would mean if you entered SR = 1? Would that be the first two blocks of memory, I suspect so. I then entered the initial address as 0000 and the finial address as 0010. The program punched the leader and then punched 13 (decimal) punches followed by the trailer. Turns out the first two punches is the initial address, 0000. This program punches a BIN file format. Then the values of 0000 to 0010 were punched next, followed by the CheckSUM. After the leader, the high byte of the initial address (bit 0-5) is punched with bit 7 = 1. This is the address marker. The next punch is the low address is punched (bit 6-11). I didn't know exactly how the CheckSUM worked. I seem to remember that each punch is added, the overflow is ignored and then the 2's compliment is punched as the CheckSUM. Apparently DEC does this, but doesn't do the 2's compliment. I added the data and could not obtain the same number as on the paper tape. The Punch on the paper tape was 0622 and I would only get 0522 when I added the list up. I later realized that you have to also add in the address marker, but 7, to the CheckSUM, that was the difference. The paper tape data agreed with what was in memory. I then tested the tape by reading it back into the PDP8 and that worked also. I did this entire process a couple to times to be sure it all works. So, my programs are good, the ASR33 reader/punch is good and my tape is good. Now I have a good tape to test the REMEX. Mike

  9. #19
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    2,069

    Default

    Good going Mike.

    The number of blocks is just the number of contiguous blocks of core you wish to punch to the paper tape.

    If you only have 1 contiguous memory block (say 0000 to 0100) then the block count is 1.

    If you have 3 contiguous areas of memory you wish to save (e.g. 0000 to 0100, 0200 to 0230 and 0300 to 0345) then the block count will be 3.

    For each block, you will then be asked for the corresponding start and end address.

    Dave

  10. #20
    Join Date
    Dec 2013
    Location
    Near Milwaukee Wisconsin
    Posts
    1,040

    Default

    Today, I wanted to find and use the program I wrote last year that read data from the REMEX to the PDP8. I had forgotten that I wrote it in PAL8 on the SIMH, assembled it and keyed in the assembly code. So I thought that maybe I could create a PDP8 file from the source code file by reading it in from My IBM XT COM1 port. First I tried PIP.

    .R PIP
    *RIMTST.RA<TTY:/B

    The PDP8 seemed to stop, so I started to send the source file.

    C:/>copy RIMTST>RA COM1:/B

    The PDP8 lite up and after a few minutes of waiting, the XT said file copied, indicating the end of the transmission. But I did not get the * prompt back from PIP. I tried Escape and Ctrl L, I could not get PIP's attention. Ctrl C brought back OS/8, but no file was saved. I tried a few other configurations of options, but nothing worked.

    Next I tried to use EDIT.

    .R EDIT
    *RIMTST.RA<TTY:/B
    #R

    Here again the PDP* seemed to stop, the source file was started.

    C:/>copy RIMTST>RA COM1:/B

    The PDP8 lite up and after a few minutes of waiting, the XT said file copied, indicating the end of the transmission. But again I did not get the attention of EDIT, but CTRL L followed by ESCAPE and ENTER, brought back the * prompt. Listing in EDIT revealed that some data was copied, but there were many mistakes, character replaces and omissions. Later, I changed the EDIT command to use an ASCII rather than BINARY file.

    .R EDIT
    *RIMTST.RA<TTY:/A
    #R

    This repaired all the mistakes and character replacements, but some text was missing.

    After some reading and thinking, I found that EDIT reads until it encounters a form feed. I looked at the file I got from SIMH and it ended in a '$', no Form Feed. So I changed it to a Form Feed and tried again. This time the source file seemed to appear in the EDIT buffer. But there were lines that were missing. The first two lines, then 12 line were missed, then it correct read the next 37 lines, missed the next 10 lines and correctly read the reminder of 24 lines. Watching the front panel display of the PDP8, there seemed to be a pattern in the blinking, but doesn't seem to match the good versus bad data. I was thinking that maybe EDIT filled up it's buffer and then had to do something, where the data was missed, and then came back to receive more. I read that the EDIT buffer is some 5000 characters, my source file was 4681 bytes. The #E command did save the file. So, I can just edit the source file with EDIT to get it right, but I thought I'd have a way of communication between the PDP8 and SIMH via floppy, but as of now it is not reliable. I'll try some experimentation with EDIT later, and I also wonder why PIP didn't work. Mike

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •