Image Map Image Map
Page 7 of 12 FirstFirst ... 34567891011 ... LastLast
Results 61 to 70 of 111

Thread: REMEX Paper Tape reader

  1. #61

    Default

    Quote Originally Posted by bobaboba View Post
    As Dave suggests, the only foolproof way is to 'single-step' the reader so that it can pause while the program is off doing other things. If you don't do that you will be restricted to tape lengths <= buffer length which will vary from program to program.
    We ran into this at the RICM when we make a 20mA<->RS-232 adapter and tried to read images of Focal source files. Periodically Focal would do some housekeeping, and ignore the input. We added the RTS/CTS signal to the adapter and turned on flow control in the the TTY emulator, and everything worked OK.

    Some devices, like tape readers, will stream if you send it a read command within a short time of it finishing the last command. Maybe yours can support this, and doing individual character reads will still keep the speed up.
    Member of the Rhode Island Computer Museum
    http://www.ricomputermuseum.org

  2. #62
    Join Date
    Dec 2013
    Location
    Near Milwaukee Wisconsin
    Posts
    1,048

    Default

    Well, I suppose if I want it to be right, I'll follow your advise (everyones) and investigate how to make this Device Handler a line by line reader. Maybe I start with a sample program and a rewire of the M863. It's time to review the documents again and start fresh. Mike

  3. #63
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    2,071

    Default

    I don't think there is much rewiring required - I think a few additional wires at the most. From what I can see you would have to wire up:

    /SYSRDY (an input to the DR8).

    /HISPD (an output from the DR8).

    /DR (an output from the DR8).

    DATA READY (an input to the DR8) - you should already have this connected up.

    DATA MODE SELECT (an output from the DR8).

    /WL (an output from the DR8).

    /WDG (an input to the DR8).

    My assumption here (that may be incorrect) is that the direction RIGHT is for normal reading and LEFT is for (re)-winding.

    You may (of course) consider wiring up all of the signals in one go!

    You should then be able to write a simple test program to test high speed synchronous read mode - and incorporate a long software delay between each character to ensure the overall system works as expected.

    Using /SYSRDY before a read should ensure that the reader is ready and waiting - with tape loaded - and is ready to go before you start the read and wait for a DATA READY pulse.

    One thing you may need to check is what happens to the signals from the DR8 on a power-up or reset condition. If the default signal state drives the REMEX 'mad', then you may need to incorporate some external inverters somewhere. I have had a quick look - but can't see anything on a firts pass that would give me casue for concern.

    I think the 'key' timing diagram (for high speed synchronous transfer) is the second half of Figure 3-1 on page 3-3 of document http://bitsavers.trailing-edge.com/p...S7300_1979.pdf.

    If I read the timing diagram correctly, you would set DATA MODE SELECT to a '1' - indicating MODE 6 operation followed by /HISPD to '0' - indicat6ing high speed mode of operation.

    Setting /DR to '0' will then start the reader moving 'right'. The data will appear followed by the sprocket hole indication (the DATA READY signal transitioning from '1' to '0'). At this point you can read the data but you must then set /DR to a '1' within 75 milliseconds to stop the reader from moving on to the next character.

    When the driver is entered for the next character in the sequence, you would set /DR to '0' again and the whole process repeats itself.

    By wiring up the rewind signals - after the tape has been read, you could get it to rewind the tape automatically for you!

    Dave
    Last edited by daver2; November 7th, 2018 at 08:29 AM.

  4. #64

    Default

    Happen to see this REMEX paper tape board on ebay. Never seen a Dual Omnibus board.

  5. #65
    Join Date
    Dec 2013
    Location
    Near Milwaukee Wisconsin
    Posts
    1,048

    Default

    Actually my PDP8E has a regular omnibus for the first half and a special second half mother board in the back. The back board has single, double and full slots in it. This back half held all the Kearney and Trecker machine tool interface cards of various sizes. The REMEX interfaced with my PDP8E thru a card like this. The card was just a buffer which grabbed characters and then signaled an interrupt, when the PDP8E would retrieve the character. There was some special code in the machine tool routine that would handle it all. So far I have not thought of how to use the back half. Mike

  6. #66
    Join Date
    Dec 2013
    Location
    Near Milwaukee Wisconsin
    Posts
    1,048

    Default

    Winter has arrived last night. The temperatures have dropped into the 20's and everything is covered with snow. Even the dog was not impressed. On the brighter side, old computer season, at least at my house has moved into full swing. To that end, I rewired my M863 so that IN00 now uses the latch. IN00 will be connected to the REMEX data ready pulse. I also wrote a simple program to test moving the REMEX only one character at a time.

    Code:
    START,	CLA		/ Send out START TAPE bit
    		TAD SSBIT
    		DBSO
    
    RDY,		DBSK		/ Wait for DATA READY, loop until ready
    		JMP RDY
    		CLA		/Clear data ready latch
    		TAD SSBIT
    		DBCI
    
    WAIT,		CLA		/ Wait 1/2 mSec
    		TAD CNT
    		DCA CNT1
    DELAY,	NOP
    		NOP
    		ISZ CNT1
    		JMP DELAY
    
    STOP,		CLA		/ Send out STOP TAPE bit
    		TAD SSBIT
    		DBCO
    
    KEY,		KSF		/ Wait for a key flag from the keyboard
    		JMP KEY
    		KCF		/Clear Flag
    		JMP START
    I tested this program with jumpers rather than the REMEX and the program seems to work. Then I connected the REMEX to see how well it actually works and something must have failed in the REMEX. Now it doesn't do anything. Another side step to repair the REMEX. Mike

  7. #67
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    2,071

    Default

    The 20's - that's warm! Oh, you mean F not C...

    Why are you assuming that it is the REMEX that is at fault when the last thing you did was some rewiring and a software change?

    Just a thought...

    How have you got the REMEX wired up now?

    Dave

  8. #68
    Join Date
    Dec 2013
    Location
    Near Milwaukee Wisconsin
    Posts
    1,048

    Default

    Gee... I thought you were English, doesn't that mean English units? The reason that I suspect the REMEX is that it will not move even in manual. The Spooler motors work but the stepper is not. I think that it is the drive transistors. Did some voltage checks, but now I have to fire up the Scope and look at square waves. Mike

  9. #69
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    2,071

    Default

    I am afraid we left "English" units behind many years ago! It turns out it is much easier counting in decimal ...

    If it doesn't drive in manual then that will be a good indicator...

    I have managed to secure a couple of paper tape punches that I am going to pick up next weekend. They come from some equipment that I used to look after in the early eighties that is about to be replaced. Right place at the right time for a change! Unfortunately I missed out on the LA120's that were skipped - but I have managed to secure the five brand new replacement print heads that are surplus to requirements though. I know where there are some more LA120s that will be surplus to requirements. They also have some ASR33 type 'clones' that I am trying to extract. Getting hold of the right person is proving elusive though.

    Dave

  10. #70
    Join Date
    Dec 2013
    Location
    Near Milwaukee Wisconsin
    Posts
    1,048

    Default

    Well, here in the States, we usually do things rather oddly. Most companies want to be international and for that reason they use the metric system. Yet, most people still use feet and inches, pounds, Fahrenheit, seconds. So in my garage my tools and I suffer between the two. Good thing the hours and seconds were not changed. Imagine 100 seconds in a minute, 100 minutes in an hour or whatever hours in a day. No one would know what is going on or when. For this reason I dread most changes. Software especially. Most times those changes just end up making more problems than they solve.......................................

    The REMEX suffered another chip failure. Z9, the 7400 NAND chip that takes the DRIVE GATE signal and applies it to the clock pulse chip, went bad. The output of the 2nd gate would not rise above 1.2 volts. Apparently the top transistor of the TTL output gate must have burnt open. After replacing this chip the REMEX once again ran in manual. I was anxious to see if the step program I wrote would work on the REMEX. At first it didn't, but after adjusting the wait delay between start and stop, the program worked. Next I added a read data and display the tape data to the program.

    Code:
    START,	CLA		/ Send out START TAPE bit
    		TAD SSBIT
    		DBSO
    
    RDY,		DBSK		/ Wait for DATA READY, loop until ready
    		JMP RDY
    		CLA		/Clear data ready latch
    		TAD SSBIT
    		DBCI
    
    WAIT,		CLA		/ Wait 1/2 mSec
    		TAD CNT
    		DCA CNT1
    DELAY,	NOP
    		NOP
    		ISZ CNT1
    		JMP DELAY
    
    STOP,		CLA		/ Send out STOP TAPE bit
    		TAD SSBIT
    		DBCO
    
    /************************
    RDATA,		DBRI		/Get data
    		TLS		/display data on terminal
    		TSF
    		JMP .-1
    /************************
    
    KEY,		KSF		/ Wait for a key flag from the keyboard
    		JMP KEY
    		KCF		/Clear Flag
    		JMP START
    Each time a key is pressed on the terminal, the paper tape advances one character, the new line is read and displayed on the terminal. The program then waits for the next key. I'm using KERMIT on my IBM XT as the terminal. So holding a key down results in a 30 character per second repeat. The REMEX will keep up with this. I read more than a 1000 characters and saw only a couple of errors. Maybe the time delay needs a little more adjustment. Seems the paper tape does not run smoothly. There is an audio rhythm you can hear. Sounds like it reads a few characters then pauses slightly and then reads again. Yet there are no skipped characters, just maybe a few tenth's of a percent errors.

    Next I need to see how this new program can fit into the PT8E.PA program. The amount of code is more, so more of the punch code will have to be removed to make space. There is other issues that I'll have to address. 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
  •