• Anybody using the TI silent 703 terminal

    From Jay Cotton@21:1/5 to All on Fri Aug 20 18:41:27 2021
    I got on on ebay, and of course it's dead.

    It will not post. I looked inside at the main board and there is an empty socket U16. I think this is an eprom. If any one can confirm my suspicion that would be very helpful.

    I did find a complete schematic for the printer, and am planning a debug session this weekend.

    tnx
    jc

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Scott Dorsey@21:1/5 to lbmgmusic@gmail.com on Sat Aug 21 14:50:09 2021
    In article <e722dd86-3b20-47a9-bc96-782e4b42f5efn@googlegroups.com>,
    Jay Cotton <lbmgmusic@gmail.com> wrote:
    I got on on ebay, and of course it's dead.

    What does dead mean? No power light?

    It will not post. I looked inside at the main board and there is an empty >socket U16. I think this is an eprom. If any one can confirm my suspicion >that would be very helpful.

    It was not at all unusual for people to design in extra eprom sockets in case later revisions of the firmware bloated. If it looks like memory and the pinout is right, it's likely an eprom but that doesn't mean anything is missing.

    I did find a complete schematic for the printer, and am planning a debug >session this weekend.

    Then you should have no problem seeing with U16 is. The first thing is to check power supply rails. It has been 40 years since I have worked on one
    of these things but as I recall they have a weird rail for the thermal head. But if you have no 5V, nothing will work.

    As always with equipment of this vintage, suspect electrolytics and tantalum capacitors before anything else. Beep out the supply rails before applying power because shorted tantalums will cause collateral damage.
    --scott

    --
    "C'est un Nagra. C'est suisse, et tres, tres precis."

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jay Cotton@21:1/5 to All on Sun Aug 22 11:59:01 2021
    Here is the situation now.

    The power supplies all check out.
    The clock circuit is working. (8.08mhz)
    The reset circuit seems to be working.

    The processor is dead. No activity at all.

    I tried to find documentation for the TMS 7041 chip.
    No luck there.

    Kinda stuck at this point.

    jc

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Scott Dorsey@21:1/5 to lbmgmusic@gmail.com on Sun Aug 22 19:22:08 2021
    Jay Cotton <lbmgmusic@gmail.com> wrote:
    Here is the situation now.

    The power supplies all check out.
    The clock circuit is working. (8.08mhz)
    The reset circuit seems to be working.

    The processor is dead. No activity at all.

    So the buss isn't cycling? Do you see anything moving on the buss at all?

    I tried to find documentation for the TMS 7041 chip.
    No luck there.

    There should be a TMS7000 databook online somewhere. There are various different versions of the IC... I don't remember the difference between
    the TMS7041 and TMS7040 offhand but the databook will tell you. It was
    not one of the more popular microcontrollers but they were certainly out
    there.

    Do you have a logic analyzer?
    --scott
    --
    "C'est un Nagra. C'est suisse, et tres, tres precis."

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Dennis Boone@21:1/5 to All on Sun Aug 22 15:13:10 2021
    I tried to find documentation for the TMS 7041 chip.

    http://bitsavers.org/components/ti/TMS7000/1983_SPND001A_TI_TMS7000_Family_Data_Manual.pdf

    De

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Three Jeeps@21:1/5 to lbmg...@gmail.com on Sun Aug 22 19:03:43 2021
    On Sunday, August 22, 2021 at 2:59:03 PM UTC-4, lbmg...@gmail.com wrote:
    Here is the situation now.

    The power supplies all check out.
    The clock circuit is working. (8.08mhz)
    The reset circuit seems to be working.

    The processor is dead. No activity at all.

    I tried to find documentation for the TMS 7041 chip.
    No luck there.

    Kinda stuck at this point.

    jc
    A quick read of the databook indicates that the CPU clock is suppose to be 2.5 MHz across pins 17,18. If the 8 Mhz is at the CPU clock pins, that is high. For processors of that era I would do o/c experiments and they usually fell apart around 105% of
    spec. Perhaps the reset line is frozen? Any activity on the address, data or r/w lines? Processor socketed? As asked earlier, got a logic analyzer?
    J

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jay Cotton@21:1/5 to All on Sun Aug 22 21:46:16 2021
    Yes I have a logic analyzer (8 channels).

    All of the chips are directly soldered to the board. With exception of U16.

    No action on the address/data or r/w lines. But then the chip may be running internal code that is polling at I/O port
    somewhere.

    jc

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jay Cotton@21:1/5 to All on Sun Aug 22 21:42:54 2021
    Several things pop out as I read through the data book.

    1. 4k rom on board, and 128 bytes of ram on board the chip.
    2. It is possible that the chip is looping on its internal code/ram polling a status
    bit/byte before moving on to the next phase.
    3. The 8Mhz clock is way high, and 2.5mhz is more likely. Need to drill in on that.
    A side note here, things like clock into the chip come from a gate array, where the 8mhz crystal
    is connected.

    4. need to revisit the clock pins and document each one.
    5. need to take a better look at the reset pin. wonder if reset is dodgy.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Scott Dorsey@21:1/5 to jjhudak4@gmail.com on Mon Aug 23 18:17:09 2021
    Three Jeeps <jjhudak4@gmail.com> wrote:
    A quick read of the databook indicates that the CPU clock is suppose to be = >2.5 MHz across pins 17,18. If the 8 Mhz is at the CPU clock pins, that is = >high.

    Could be bad crystal. More likely bad decoupling capacitors on supply rails. --scott
    --
    "C'est un Nagra. C'est suisse, et tres, tres precis."

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jay Cotton@21:1/5 to All on Mon Aug 23 15:14:05 2021
    Page 4-27 lists the max clockin to be 10.2 mhz. The CPU has an internal divide by
    4 circuit. I guess the 8mhz clock is o.k.

    Probing with a scope.

    I see pin13 has (INT1) 76hz. It seems to conform to the description in the terminal manual.

    No data buss or address buss action at all. The MC bit is 0, as documented in the terminal
    manual, for full expansion mode.

    Reset is a 1. I blipped it with a ground, and did not see any changes at all.

    Here is an interesting point. The CLOCKOUT pin 2 is a . No clock. This may be set by the boot
    ROM.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Three Jeeps@21:1/5 to lbmg...@gmail.com on Mon Aug 23 18:44:41 2021
    On Monday, August 23, 2021 at 6:14:06 PM UTC-4, lbmg...@gmail.com wrote:
    Page 4-27 lists the max clockin to be 10.2 mhz. The CPU has an internal divide by
    4 circuit. I guess the 8mhz clock is o.k.

    Probing with a scope.

    I see pin13 has (INT1) 76hz. It seems to conform to the description in the terminal manual.

    No data buss or address buss action at all. The MC bit is 0, as documented in the terminal
    manual, for full expansion mode.

    Reset is a 1. I blipped it with a ground, and did not see any changes at all.

    Here is an interesting point. The CLOCKOUT pin 2 is a . No clock. This may be set by the boot
    ROM.
    I looked in the data book to see if the clk had a divider ckt but didn't find any. If you found one then ya, 8MHz would put it into specs. U16 is the processor? or the ROM? I am not at all familar with the thing you are working on.
    I quickly gleaned that on startup the CPU does a data fetch at FFFF,FFFE for boot sequence and yet you dont see any bus activity? I'd have to read more about how the CPU operates to offer help.
    I did a quick google search and sources of that chip seem to be long gone. I did come across a ref to Toshiba as a second source but had no luck finding any.
    J

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Andy Valencia@21:1/5 to Three Jeeps on Mon Aug 23 21:30:33 2021
    Three Jeeps <jjhudak4@gmail.com> writes:
    I quickly gleaned that on startup the CPU does a data fetch at FFFF,FFFE
    for boot sequence and yet you dont see any bus activity?

    I feel like he should at least see the reset line blip on powerup.
    Without that, it seems like something basic upstream has failed. As
    somebody said, caps on the power rails, something like that.

    Andy Valencia
    Home page: https://www.vsta.org/andy/
    To contact me: https://www.vsta.org/contact/andy.html

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jay Cotton@21:1/5 to All on Tue Aug 24 11:50:03 2021
    My next official step will be to replace the power supply caps.
    I doubt that this is the primary problem, but its not that hard to do.

    WRT reset. The reset goes to a 1 after the power is applied. I have
    not measured the delay. Should be about 500 usec after the power
    rails are stable. Controlled by a custom package (why oh why) lm3905 timer.

    WRT memory read at init. I will setup the analyzer and try to catch that. Guess I will trigger on the rising edge of reset.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jay Cotton@21:1/5 to All on Wed Aug 25 19:09:56 2021
    Well, I am seeing a bit of racket on the +5 rail. So, a few bucks for caps seems
    in order.

    Bypass caps eh? I'll have a go at them also, the the power supply caps come in.

    jc

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Scott Dorsey@21:1/5 to lbmgmusic@gmail.com on Thu Aug 26 01:49:22 2021
    Jay Cotton <lbmgmusic@gmail.com> wrote:
    My next official step will be to replace the power supply caps.
    I doubt that this is the primary problem, but its not that hard to do.

    That's not what I said. Check the decoupling capacitors on the logic.

    It probably won't hurt to do the power supply caps in the process, mind you, since they are almost certainly bad. But it's the logic decoupling that
    may be keeping the processor from running right.
    --scott


    --
    "C'est un Nagra. C'est suisse, et tres, tres precis."

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jay Cotton@21:1/5 to Jay Cotton on Wed Dec 8 18:33:41 2021
    On Wednesday, August 25, 2021 at 7:09:57 PM UTC-7, Jay Cotton wrote:
    Well, I am seeing a bit of racket on the +5 rail. So, a few bucks for caps seems
    in order.

    Bypass caps eh? I'll have a go at them also, the the power supply caps come in.

    jc
    Closing out this discussion.

    I did not get the 703 running, changing caps was not the solution. I suspect it may have had a
    voltage spike or the wrong power supply attached or something like that.

    However, this turns out not to be a complete disaster. I got another 703 (that works) and now I have
    a source of parts for the running terminal.

    For those interested, https://www.youtube.com/c/CuriousMarc has a series on the silent 703 that is
    very interesting. Not to mention the resurrection of the Apollo guidance computer etc.


    jc

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)