Some questions about a Nubus card in an Apple II

[Eeyore]

I've got another request for help with some 'antique' kit.

The equipment in question uses a Nubus interface card to communicate with the
software on an Apple II.

Thse 'interface cards' have a fair bit of local intelligence, in fact there's a
whole 68020 @ 16MHz on it with a shedload of I/O.

The problem that's typically experienced is that the host platform / software
fails to recognise / talk to this 'interface'. There seems to be no special
rhyme nor reason to it, aside from the fact that they gradually fail, with
resets becoming more and more often required to get things working until
eventually there is no response at all.

One thing that caught my attention was the use of 74ACT651 interface
transceivers on this card. I was puzzled about (a) the use of 'A' series - it
seems that Nubus rarely ran faster than 10 MHz and (b) the use of 'T' parts
since almost all the other generic logic on the card is 74HC.

I have schematics btw but no other accomanying documentation, test procedures or
test software.

Initially I plan on checking for obvious stuff like the presence of suitable
power-on reset signals and any obvious bus conflicts.

The company whose product it is supplies a $5000 'refurb' board on receipt of
the old one but this would only make sense if the equipment had commensurate
value today which it hasn't at those prices.

Any comments ?

Graham

 

[Donald]

I've got another request for help with some 'antique' kit.

The equipment in question uses a Nubus interface card to communicate with the
software on an Apple II.

Thse 'interface cards' have a fair bit of local intelligence, in fact there's a
whole 68020 @ 16MHz on it with a shedload of I/O.

The problem that's typically experienced is that the host platform / software
fails to recognise / talk to this 'interface'. There seems to be no special
rhyme nor reason to it, aside from the fact that they gradually fail, with
resets becoming more and more often required to get things working until
eventually there is no response at all.

One thing that caught my attention was the use of 74ACT651 interface
transceivers on this card. I was puzzled about (a) the use of 'A' series - it
seems that Nubus rarely ran faster than 10 MHz and (b) the use of 'T' parts
since almost all the other generic logic on the card is 74HC.

I have schematics btw but no other accomanying documentation, test procedures or
test software.

Initially I plan on checking for obvious stuff like the presence of suitable
power-on reset signals and any obvious bus conflicts.

The company whose product it is supplies a $5000 'refurb' board on receipt of
the old one but this would only make sense if the equipment had commensurate
value today which it hasn't at those prices.

Any comments ?

Graham

I don't think the Nubus was used on the Apple II.

http://en.wikipedia.org/wiki/NuBus

 

[Eeyore]

I don't think the Nubus was used on the Apple II.

http://en.wikipedia.org/wiki/NuBus

Silly me. I meant the Mac II of course. I can just remember the Apple II as well. II-e
in fact..

Graham

 

[Lionel]

I don't think the Nubus was used on the Apple II.

It wasn't. The old Apple II had an extremely simple 8 bit bus.

http://en.wikipedia.org/wiki/NuBus

The Wikipedia page has a link to the official Nubus spec's, which
presumably include data on the correct logic levels on the bus, &
maybe the recommended logic family for the bus drivers:
<http://www.bitsavers.org/pdf/ti/2242825-0001_NuBus_Spec1983.pdf>

I suspect that Graham's on the right track WRT to his comments about
74ACT vs 74HC logic, as I've seen many similar problems in mixed logic
family circuits over the years. Random failures that increase with the
age of the system are the classic symptom of that particular design
(or manufacture) error.
I'd first scope out the logic levels between the ACT drivers & the HC
main logic inputs, as that's the point of biggest mismatch between C &
TTL spec's.

--
\___ Proud Cog #1 in the AUK Hate Machine
_(AUK)====:: Do *you* think that you have the Right Stuff?
/='='='='-, Apply TODAY by addressing a gratuitously cruel
(O+O+O+O+O) flame to: "Uncle Fester", C/O soc.singles & AUK.
~^^^^^^^^^~~~^~^^~'~~^'^~~~"~~'"~^~'"~~^~"~'~^'^~^~^^~^~"~^~"'~'"~^~~

 

[joseph2k]

I've got another request for help with some 'antique' kit.

The equipment in question uses a Nubus interface card to communicate with
the software on an Apple II.

Thse 'interface cards' have a fair bit of local intelligence, in fact
there's a whole 68020 @ 16MHz on it with a shedload of I/O.

The problem that's typically experienced is that the host platform /
software fails to recognise / talk to this 'interface'. There seems to be
no special rhyme nor reason to it, aside from the fact that they gradually
fail, with resets becoming more and more often required to get things
working until eventually there is no response at all.

One thing that caught my attention was the use of 74ACT651 interface
transceivers on this card. I was puzzled about (a) the use of 'A' series -
it
seems that Nubus rarely ran faster than 10 MHz and (b) the use of 'T'
parts since almost all the other generic logic on the card is 74HC.

I have schematics btw but no other accomanying documentation, test
procedures or test software.

Initially I plan on checking for obvious stuff like the presence of
suitable power-on reset signals and any obvious bus conflicts.

The company whose product it is supplies a $5000 'refurb' board on receipt
of the old one but this would only make sense if the equipment had
commensurate value today which it hasn't at those prices.

Any comments ?

Graham

It has to do with drive levels to match nubus. Try swapping them out for
AHCT parts. They have better endurance properties.

 

[Eeyore]

It has to do with drive levels to match nubus.

That's what I was wondering. The Nubus drive level is only guranteed to be TTL
level I suppose ?

Try swapping them out for AHCT parts.

AHCT ? Hmm, I've never some across those before - or have I ? Hmmm... Looks
pensive and wonders where that memory card reader/writer I designed 15 years ago
got to. BCT rings a bell too.

They have better endurance properties.

Endurance ? In what way ?

Many thanks, Graham

 

[whit3rd]

I've got another request for help with some 'antique' kit.

The equipment in question uses a Nubus interface card ....
The problem that's typically experienced is that the host platform / software
fails to recognise / talk to this 'interface'.

Nubus was made to be self-configuring, and each addressable
slot has unique ID. You can use utilities to find out what cards are
in what sockets (and this would be a first step if 'talking' isn't
happening).

Also, each card has its own ROM for ID and autoload of drivers; if
your
card's EPROM is losing data (not unlikely at this date, Nubus was
used 1986-1995), that could be messing things up the way
you describe. If you can, try to locate/remove/read the EPROM
and program a new one.

Heck, if it's in a socket, just yank it and reinsert (to clean the
contact
points).

 

[Eeyore]

Nubus was made to be self-configuring, and each addressable
slot has unique ID. You can use utilities to find out what cards are
in what sockets (and this would be a first step if 'talking' isn't
happening).

Utilities ?

The Mac world seem devoid of anyone technically competent in such matters.

Also, each card has its own ROM for ID and autoload of drivers; if
your
card's EPROM is losing data (not unlikely at this date, Nubus was
used 1986-1995), that could be messing things up the way
you describe. If you can, try to locate/remove/read the EPROM
and program a new one.

I have 2 of these boards and the code EPROMs match. I can't see any ID ROM or
EEPROM on the schematics.

Heck, if it's in a socket, just yank it and reinsert (to clean the
contact points).

Done that. No help sadly.

Graham

 

[joseph2k]

That's what I was wondering. The Nubus drive level is only guranteed to be
TTL level I suppose ?




AHCT ? Hmm, I've never some across those before - or have I ? Hmmm...
Looks pensive and wonders where that memory card reader/writer I designed
15 years ago got to. BCT rings a bell too.



Endurance ? In what way ?

Many thanks, Graham

Better ability to survive voltage or current excursions on the I/O pins
outside normal operting conditions.