gbuce@pixar.UUCP (George Buce) (01/21/89)
[eat me]
Probably a rehash of older LUCAS board conversations, but I just got on
the net about a month ago and have only seen mention of the proposed memory
board design...
I'm putting together a LUCAS board for a friend (mine will soon follow)
and I'm having difficulty with external memory. I'm using a 12 Meg '020 and a
16 Meg '881, replaced all the PALs with D series PALs. All other parts are the
74F series parts called for (except U9, of course).
With the B series PALs, the only U9 I could get to work was a 74HC74
(well, it's not TTL compatible, but it works). With the D series PALs, I can
get two 'HC parts to run and one each of 'LS and 'ALS. I'm clocking the system
at 12 MHz (hopefully to be at 16 MHz or 20 MHz). The system will work fine
with no external memory attached, but with a Micron 2 Meg expansion board (and
two slot box) or an Alphanetics 2 Meg board, the system hangs.
Did I miss any news on the net about a fix for the board that was a lot
more tolerant of changes in the system and/or didn't need the 'tuning'? I've
got to think that with a little rewiring or a PAL change, the board would be a
lot more solid. I'm going to put the board on a logic analyzer soon to try to
fathom the problem, but I'm hoping that someone out there has already solved
the problem and I just missed the posting.
One tip for those about to assemble the LUCAS board: The ribbon cable to
the disk drive is a real mechanical problem, with the folding of the cable
pressing up against the bottom of the LUCAS board. I found that by seperating
the ribbon into 14 pairs (every other wire is a GND wire, so each pair has a
ground), the cable folds much more cooperatively. A little care and an Xacto
knife makes splittling the cable a straightforward task. I actually made a new,
shorter cable. The parts are rather cheap, about $7 total.
Thanks...
_______________________________________________________________________________
George Buce || Not representing my company, but boy!
(8{> || do they make neat toys! (Can anyone
ubcvax!pixar!gbuce || say 48 bit planes?) (How about $30K?)
_______________________________________________________________________________c60a-1fy@web-2a.berkeley.edu (Anon) (02/04/89)
It seems I was a bit premature when I posted to comp.sys.amiga that I had the lucas board working correctly with my starboard. Now, however, I have been going for 3 hours without a crash, so here's my fix. Build the terminator, but use 3.3 k pull-downs instead of 1k's for /as, /lds, and /uds. These signals looked a little low on the scope. Build a little piggy back board to split u9 into two chips, one for each gate. This is easy to do using 3 sockets. One is connected to pins 1-6 and the other to 8-13, and both are connected to 7 and 14. For me I got the thing to work using an S part for U8, and ALS part for the first gate in U9, and a 7474 for the second half. This works like a charm. I hope this helps. -Vince Lee
c60a-1fy@web-2b.berkeley.edu (Anon) (02/05/89)
On my last message, I swapped the two chips used as U9. I have a 7474
connected to pins 1-6 and a 74ALS74 connected to pins 8-14. It has now
been running 6 hours with 0 (zero!) crashes!!!
I never did like the idea of trying to find "the right" U9 to satisfy
both gates. Sure, you might be able to do it with an empty amiga, but
when you start stacking on them expansion devices, Lucas gets real
unhappy. If you're lucky, you might get something that works, but it
might me flakier than if you separated the gates.
My guess is that spitting the gates in U9 and possibly terminating
the bus if you have a starboard should get any system to work, since that
is really the only "flaw" in the lucas board.
I make the piggyback assy with 3 14 pin sockets as follows. It's pretty
neat, although it will guarantee that you'll never get that emission shield
back on (awww... gee, big loss).
//- -\\ <- Two sockets w/ one edge in base socket
// \\
======= <- Socket plugging into U9 socket
| |
Then, just add two stiff wires so both chips are connected to 7 and 14 (pwr),
and also to keep the unconnected edges of the side sockets from touching.
-Vince Leeewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) (02/06/89)
In article <19949@agate.BERKELEY.EDU> c60a-1fy@web-2b.berkeley.edu (Anon) writes: >On my last message, I swapped the two chips used as U9. I have a 7474 >connected to pins 1-6 and a 74ALS74 connected to pins 8-14. It has now >been running 6 hours with 0 (zero!) crashes!!! > I hope you remembered to ground all the unused inputs on the two chips to keep the unused input gates from going linear and sucking current. Of course, it's been a long time since I hacked hardware, so design details may have changed... _-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_ Leo L. Schwab -- The Guy in The Cape INET: well!ewhac@ucbvax.Berkeley.EDU \_ -_ Recumbent Bikes: UUCP: pacbell > !{well,unicom}!ewhac O----^o The Only Way To Fly. hplabs / (pronounced "AE-wack") "Work FOR? I don't work FOR anybody! I'm just having fun." -- The Doctor