Subject: Re: help - 3max (DS5k/200) will not start
To: port-pmax list NetBSD <port-pmax@netbsd.org>
From: Eyal Lebedinsky <eyal@eyal.emu.id.au>
List: port-pmax
Date: 12/31/2002 11:38:04
"Peter C. Wallace" wrote:
> Could be anything but my first guess would be a bad power supply, I've seen
> several of those in my DS5000s. DCOK is active high AFAIK, you might check the
> +5V and +-12...
Well, I did check the voltages but I do not have a
reference for the mobo that tells me what it correct.
Here is what I measured (view from top).
----------------------------------- front -------+
+--+--+ |
+-----+ +-----+ |1 |2 | +-------------+
| | | | +--+--+ | | |
| B | | R | |3 |4 | | | |
| | | | +--+--+ | | |
+-----+ +-----+ |5 |6 | | power |
+--+--+ | supply |
| |
| |
| |
| |
| |
| |
| |
/ /
/ /
| |
| |
| |
| |
| |
| |
| |
| |
a b c d e f | |
+ + + + + + | |
------- | |
| |
/ /
/ /
The voltages at the power feed are:
B 0
R 5.15
1 12.24
2 -12.14
3 0
4 0
5 4.94
6 0.20
and at the pin headers I have
a 0
b 5.13
c 0.02
d 4.94
e 0.02
f 4.94
All voltages measured against the chasis.
So, it looks as if the voltages are fine, but I am not sure that the
power OK signal is properly delivered. I assume this is one of 5/6?
Now some speculations:
I read somwhere that e-f is a jumper that clears the nvram. This does
not match the fact that the chip is cleared by shorting pins 12 (gnd)
and 21 (/rclr) and I verified that [f] is not directly connected
to [pin 21] or any other pin on the clock/vnram chip.
In other places I read that the 4-pin brown header on other DS5k models
has the pinout:
a TRIG
b GND
c MFG
d DCOK
and that
shorting MFG-DCOK will reset the machine
shorting GND-MFG will enable some manufacturing tests
--
Eyal Lebedinsky (eyal@eyal.emu.id.au) <http://samba.org/eyal/>