On Thu, 14 Nov 2002, Mark Tapley wrote:
> > > Most DEC machines have at least one power-OK line from the PSU.
> >
> > I want to figure out which that one is.
>
> I think there is no such line. On the mainboard connector to the PSU,
> there are Red (3), Black (4), Orange (1), and Blue (1) wires.
>
> Red are all connected together, and all go to appx. 5 V when the system
> is on. Black are all connected together and to ground and are always at
> 0 V. Orange is one wire, and goes to appx. 12 V when the system is on.
> Blue is one wire, and goes to appx. 12 V. when the system is on.
Black is nearly always ground, red is commonly used for +5V. This wouldn't
be the first supply I've heard of that didn't have a PG connection.
> The red and black cross-connections are true on the mainboard and on the
> PSU. They are not all adjacent on the connector, which makes not much
> sense to me, but that's how it is. Red and Black on the mainboard
> connect to the corresponding colors on the Disk power connector.
It may have made it easier for the engineers who were laying out the power
busses on the mainboard.
> As for voltage: I really need a good DVM. Calibrated the 12V setting at
> work against a good 4-digit fluke DVM, so I think my reading of 12.1V
> for orange and -12.0 for blue is close. Did not have time to check at
> 5V, but afterward measured 3 Ray-o-Vac Renewal D cells in series at 4.7
> V, so I may not be way far off. If so, my reading of 5.3 to 5.4 for the
> Red group is worrisome. This reading is consistent even at the Disk
> power connector. Anyone else have a good VOM, a working VLC, and 3
> Renewal cells? Or just the VOM and either of the other two?
It sounds like your DMM is close enough. Since I'm not familiar with these
supplies, I can't say if 5.3-5.4 is too far beyond the norm, but I usually
don't get too worried unless it reaches ~5.5 and beyond for +5V.
> I know more about it, now. Turned it on cold, it ran for a while, then
> failed, then started to try to boot again. Blew the hair dryer on it, it
> went into hard fail almost immediately and did not try to boot. Blew a
> fan on it, it went OK and actually ran far enough to give me a login
> prompt. In both cases I was trying to aim at the part of the mainboard
> near the power connector, but the system is so small I was probably
> affecting most of the mainboard and possibly a lot of the PSU as well.
It certainly sounds like something is overheating. You might want to try
using some cardboard to carefully direct the airflow from your hair dryer.
Do these systems have an internal fan? I haven't been inside one enough to
remember, but if they do, is the one in your system running properly?
> Near the power connector are two power-looking components (as in they
> have big leads, and in one case a heat-sink attach point, whereas nearly
> everything else is microscopic surface-mount stuff). One is a
> power-transistor looking thing with a heat-sink attach point (but no
> heat-sink) that has 3 pins and says
>
> LT1086CT
> 9151
Linear Technology 1.5A adjustable voltage regulator, TO-220 package.
Datasheets for this part can be found here:
http://www.linear.com/prod/datasheet.html?datasheet=233
> The right leg is connected to 5.3 V (Red-wires).
> The center leg stays around 3.4 V, system running or failed.
> The left leg stays around 2.15 V, running or hung (as far as I can tell).
Pins from left to right:
1 ADJ (ground for fixed voltage regulators)
2 Vout (also connected to the mounting tab)
3 Vin
It seems like the board uses the regulator to regulate +5V down to 3.3V.
> The other is an 8-pin DIP that says
>
> M9124
> LM393N
> QST
Low Power Low Offset Voltage Dual Comparator
Datasheets can be found here:
http://www.national.com/pf/LM/LM393.html
> Are these two part of a crowbar circuit, or reset circuit, or some such?
> Should I be able to see some clue about why the system hangs by looking
> at their pins?
I expect to find a crowbar would be inside the PSU, and not on the
mainboard. I don't think these components would be used for a crowbar
anyway. It really does sound like your system has a thermal problem. I'd
suggest some careful directing of the air from your hair dryer to help
pinpoint the problem area.
-Toth
Received on Thu Nov 14 2002 - 09:46:01 GMT