2 Dec
1999
2 Dec
'99
7:22 a.m.
Argh!! You've got an intermittant connection, and
they're the _worst_ to
trace...
My reaction exactly. The connections in question _look_ OK, though
obviously aren't. There are some others on the front panel that look even
worse than what I can do with what I've got right now, and my questionable
soldering skills.
It could be bad solder connections (although that's
not that common in
machines as 'new' as the 8/m), it could be bad connections on the
backplane connections, it could even be cracked tracks on the board.
Is there any way to check for either of these problems?
The nasty bit is that disturbing _anything_ will often
'fix' it. So you
think the last thing you did (resolded the LEDs, cleaned the backplane,
whatever) was the fix. And as soon as you get the case on, it's gone again.
The thing to do is to see if you can find some action (moving a
component, bending the board, etc) that makes the LEDs come and go. If
you can do that you're getting close. Just make sure that you're not also
disturbing something else, though.
Well, once I got the face plate off yesterday I was able to wiggle the LED
itself in its 'socket' and it would come on.
When 2 instruments disagree you really should
investigate further. Yes,
it may be that one instrument can't see the pulse (it's just too narrow
for it). Or you might really be seeing noise on a ground connection or
something.
Oh, wonderful. I just took both the scope and the logic probe to both E46
and E39. They both gave the same results, the chip looks to be good.
Something tells me this system is going to drive me crazy! But it means
that what the scope told me a couple days ago is correct.
This machine sounds as though it's full of bad
connections, alas. Have
you cleaned _all_ the edge fingers on all the boards? Have you cleaned
(and tightened if necessary) the PSU connectors. Have you cleaned the
backplane slots?
I cleaned the edge fingers before even turning it on, though I think I'll
clean them again. I'll have to take a look at the PSU connectors. I blew
out the backplane slots with canned air, how can you do a better job of
cleaning them?
Has it been stored under poor conditions?
Not since I got it a couple weeks ago, before that I've no idea. I suspect
it might have spent a short amount of time outside as I found a couple pine
needles in it.
Personally, I use a Weller TCP iron. It's one of
the Magnastat
temperature-controlled ones. It's a lot simpler (== less to go wrong)
than an electronically controlled iron, and it's Good Enough for
everything I'm ever likely to want to do with it. I find the #7 tip too
cold for many things and generally have a #8 in it.
This iron is very popular with serious hobbyists and research/development
labs in the UK, which is probably a good recomendation.
Sounds about like the one I was going to buy last weekend, but they were
out of ALL of their Weller's except one really cheap looking model at the
local electronics store. I know Fry's has a whole cabinet of different
models, but I didn't fell like going down there twice in one week. Sounds
like I need to plan on that Saturday.
On a possitive note, either Saturday or Monday I'll have a 4k Core Stack
and a 8k Core Stack. So I can see what happens with different a different
core stack in place. I think I'll also be getting the spare M8330 at the
same time, the pair of CPU boards will be coming later, probably in a week
or to.
Zane
| Zane H. Healy | UNIX Systems Adminstrator |
| healyzh(a)aracnet.com (primary) | Linux Enthusiast |
| healyzh(a)holonet.net (alternate) | Classic Computer Collector |
+----------------------------------+----------------------------+
| Empire of the Petal Throne and Traveller Role Playing, |
| and Zane's Computer Museum. |
| http://www.aracnet.com/~healyzh/ |