[sebhc] H89 Power-up problems.

dwight elvey dkelvey at hotmail.com
Sat Jul 8 12:45:17 CDT 2006

I recommend that once you expose the contact
put a film of silicon grease, such as DowCorning #4,
on the contacts. This will improve the contact and
stop the multiple keys.

>From: "Mark Garlanger" <garlanger at gmail.com>
>Reply-To: sebhc at sebhc.org
>To: sebhc at sebhc.org
>Subject: Re: [sebhc] H89 Power-up problems.
>Date: Fri, 7 Jul 2006 16:57:10 -0500
>Hi Guys,
>  Thanks for all the suggestions. When I last turned off that H89
>(about 3-4 days ago), it was still having the problem. Today, the
>keyboard is working (although not as good as on my other H89).
>Several of the keys seem to have problems with repeated characters (as
>if the debouncing circuitry is not working well). Several of the keys
>will intermittently generate 2 (or more) characters when I press it
>(more likely if I press it slowly). Also, the '4' key will
>intermittently not register on the key press, but will on the key
>release ( and sometimes on both). It doesn't seem like all the keys
>have the problem (not even all the keys in a given row). On my other
>H89, I am unable to get any of the keys to act this way.
>Is there any trick to popping off the keys? Maybe they need cleaning 
>BTW: It still usually only beeps once even though the computer part is
>working.  It could be the timings that Lee mentioned below, since once
>(out of about 10), it did beep twice.
>  Mark
>On 7/5/06, Lee Hart <leeahart at earthlink.net> wrote:
>>Mark Garlanger wrote:
>> > Does anyone know what the power up beeps mean?
>>The TLB and CPU boards each independently generate a "beep" when they
>>are powered up. The speaker plugs into the TLB, so you get one "beep"
>>for an H19 trminal, or for an H89 with the OFF-LINE key depressed.
>>This first "beep" tells you that the TLB has power, and that its Z80 is
>>running and executing software (even if incorrectly). It will "beep"
>>again each time the TLB is reset.
>>The CPU board in an H88/89/90 generates a separate "beep" when it is
>>powered up or reset. To hear this as a second "beep", the TLB needs to
>>be working, and on-line (i.e. the OFF-LINE key is up). This "beep" is
>>sent to the TLB serially as a software command at the same time as the
>>"H:" prompt. It indicates that the CPU board's Z80 is running the
>>monitor program.
>>Note that the timing of these "beeps" is set independently by the R-C
>>networks on each board that control the power-on Reset. Old TLB and CPU
>>boards tended to have long delays; later ones had shorter delays. It's
>>possible to swap boards so you have a long delay on the TLB and a short
>>delay on the CPU -- when this happens, the CPU board sends its "beep"
>>before the TLB has started running, so you only get one "beep" even
>>though nothing is wrong.
>> > working fine, but now the keyboard no longer works (except for the
>> > 'shift-reset' combo), all other keys don't even keyclick (they had
>> > initially).
>>The keyboard encoder chip is an old PMOS part, and pretty sensitive to
>>static damage. Touching the bare keyboard connector pins or traces to
>>the keyboard can kill it. If the problem isn't something simple like the
>>ribbon cable to the keyboard being unplugged, you may have a bad
>>keyboard encoder IC.
>> > Between the time it was working and now, during a few power-ups the
>> > screen the screen went a little crazy, for example, the "H:" was
>> > shaking and showing up in two places, another time various characters
>> > showed up on the screen. I'm wondering if a CAP or other component
>> > failed.
>>These symptoms both point to problems on the TLB. I'd check the supply
>>voltages. The TLB needs +8v, +16v, and -16v to work. It has onboard
>>regulators to convert these to +5v, -5v, +12v, and -12v.
>> > BTW: The memory is installed strangely in the system. The top row of
>> > memory ICs have a second memory chip soldiered on top of each one,
>> > along with a wire that connects them all to one of the lines on the
>> > slot that you would normally install the extra memory card to bump it
>> > up to 64k. This must have been a cheaper way to get to 64k than adding
>> > the board.
>>Right. This is a 'quick and dirty' way to upgrade to 64k of RAM without
>>any extra PC boards.
>>Ring the bells that you can ring
>>Forget the perfect offering
>>There is a crack in everything
>>That's how the light gets in
>>         -- Leonard Cohen
>>Lee A. Hart   814 8th Ave N   Sartell MN 56377
>>leeahart_ at _earthlink.net
>>Delivered by the SEBHC Mailing List
>Delivered by the SEBHC Mailing List

Delivered by the SEBHC Mailing List

More information about the Sebhc mailing list