[comp.sys.dec] More MVII Hardware Help

jcd@pitt.UUCP (Jim Durham) (10/09/90)

Thanks to all who replied to my inquiries concerning the TK50/RQDX2
problems I have been having. That conversation has taken off in
a slightly different direction, but is still good reading.

Here's my latest dilemma with my "used" MVII 8-) : I have
a need for a SLIP link to another computer. I have two AVIV SDZV11
8 port serial cards, which are supposed to emulate the Dec cards.
WHen running SLIP or just Kermiting to the other machine, the
whole serial system just dies at random times. It seems to be
that the incoming data is lost. If I ping the other machine, the
Microvax sends, but does not receive the reply. This also happens
when running Kermit, so it doesn't appear to be a SLIP driver
problem.

I have tried both cards, and the results are identical. Sometimes
I can move a 100k file and sometimes I can't get 5 characters between
the machines. All 8 ports die together. A terminal on tty00 will die
also if SLIP on tty02 dies.

This sounds like an input buffer problem to me. By the way, this
is Ultrix 3.1 . Oh yes, the only way to restore the serial system
is to reboot. Any ideas?

Incidentally, for those of you to whom old MVII's are starting to
filter down, let me reiterate my first two hassles with this machine.
First...keep the VR260 monitor *cool*. The flyback in this monitor
is capable of blowing with any kind of heat. I know. Second...if
the MVII is dusty, check the voltage on the bus. The connectors from
the power supply to the bus are cheesy and you can get over a volt
of drop in the +5 supply. TTL doesn't work well on 4 volts. I've
been through this one, too. If you have an extender card, pull
some module, put it on the extender, and measure with a digital
voltmeter from ground to +5 on some chip. If you don't have
at least 4.9 volts, better clean up the power suppply/bus connectors.
I hope this helps someone to not lose as much cranial hair as I did.

-Jim Durham

terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr) (10/10/90)

In article <8810@pitt.UUCP>, jcd@pitt.UUCP (Jim Durham) writes:
> Second...if
> the MVII is dusty, check the voltage on the bus. The connectors from
> the power supply to the bus are cheesy and you can get over a volt
> of drop in the +5 supply. TTL doesn't work well on 4 volts. I've
> been through this one, too. If you have an extender card, pull
> some module, put it on the extender, and measure with a digital
> voltmeter from ground to +5 on some chip. If you don't have
> at least 4.9 volts, better clean up the power suppply/bus connectors.
> I hope this helps someone to not lose as much cranial hair as I did.

  If you're getting 4V there, it isn't dust 8-(.

  Seriously, if you have a BA23 master box (with front panel halt and restart
buttons), you should have the power harness checked. The old harness can catch
on fire. The way to tell if you have the proper harness is to open the BA23
up and look in the drive bays. The good harness is DEC part number 17-01311-01
and uses white connectors and various colored wires. If you have yellow con-
nectors or a webbed cable (rather than individual wires) you have a potential
safety hazard. Again, this *only* applies to BA23 boxes. If you're not comfort-
able with doing this yourself, you can get DEC to look at it (which will cost 
you if you don't have a contract).

	Terry Kennedy		Operations Manager, Academic Computing
	terry@spcvxa.bitnet	St. Peter's College, US
	terry@spcvxa.spc.edu	(201) 915-9381

don@zl2tnm.gp.govt.nz (Don Stokes) (10/10/90)

terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr) writes:

> In article <8810@pitt.UUCP>, jcd@pitt.UUCP (Jim Durham) writes:
> 
>   Seriously, if you have a BA23 master box (with front panel halt and restart
> buttons), you should have the power harness checked. The old harness can catc
> on fire.

I can second this -- we had intermittent problems on a MicroVAX I a
while ago.  The problem was a cooked power harness.  The cable had been
pretty hot -- the connectors had cracked and the cable was quite an
unhealthy shade of toasted brown.


Don Stokes, ZL2TNM  /  /                            Home: don@zl2tnm.gp.govt.nz
Systems Programmer /GP/ Government Printing Office  Work:        don@gp.govt.nz
__________________/  /__Wellington, New Zealand_____or:_PSI%(5301)47000028::DON

cdl@chiton.ucsd.edu (Carl Lowenstein) (10/10/90)

In article <8810@pitt.UUCP> jcd@pitt.UUCP (Jim Durham) writes:
> Second...if
>the MVII is dusty, check the voltage on the bus. The connectors from
>the power supply to the bus are cheesy and you can get over a volt
>of drop in the +5 supply. TTL doesn't work well on 4 volts.
> If you have an extender card, pull
>some module, put it on the extender, and measure with a digital
>voltmeter from ground to +5 on some chip.

1) There are test points behind the front panel cover (on a BA23)
that let you measure the +5 and +12 where it goes into the backplane.

2) The cable from the power supply to the backplane has been identified
by DEC as a known fire hazard, a mandatory ECO.  Field service will
replace it for you at no charge.  (at least they did for me)  The
new design of cable & connector seems much more reliable.  Last week
I had to shut down my MVII after 140 days of uptime, just because the
building was being re-wired.

-- 
        carl lowenstein         marine physical lab     u.c. san diego
        {decvax|ucbvax} !ucsd!mpl!cdl                 cdl@mpl.ucsd.edu
                                                  clowenstein@ucsd.edu

amichiel@rodan.acs.syr.edu (Allen J Michielsen) (10/13/90)

In article <8810@pitt.UUCP> jcd@pitt.UUCP (Jim Durham) writes:
>a need for a SLIP link to another computer. I have two AVIV SDZV11
>8 port serial cards, which are supposed to emulate the Dec cards.
>WHen running SLIP or just Kermiting to the other machine, the
>whole serial system just dies at random times.

The AVIV cards are quite old I would wager.  Given that assumption,
DEC had a problem with a batch of defective propriotary q-bus interface
chip sets.  As I recall they had the same problem, and it probably gets
worse with age.  Dec didn't seem to have any of their own cards affected
but sent notices to all licensed developers of cards.
You may find help from the code info on the bus interface chips and your
local dec office.  The digital unlisted spares program MAY also have info
of this type.  I'd say your chances of getting the replaced for free are just 
about zero, and the chip set costs about $100, if you can buy it from that
same spares program.
al


--
Al. Michielsen, Mechanical & Aerospace Engineering, Syracuse University
 InterNet: amichiel@rodan.acs.syr.edu  amichiel@sunrise.acs.syr.edu
 Bitnet: AMICHIEL@SUNRISE 

terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr) (10/13/90)

In article <Bw7Zq1w163w@zl2tnm.gp.govt.nz>, don@zl2tnm.gp.govt.nz (Don Stokes) writes:
> I just pulled my MicroPDP-11 (11/73) to bits, and located the power
> harness (under a *lot* of dust).  It's part number 70-20450-01, with
> yellow connectors, but separate coloured wires.

  Winner! Winner! You have one of the flaming power harnesses 8-)

> What was the nature of the overheating problem?  The part I have in my
> hand looks OK, no sign of any heating.	I flexed the conectors to check
> if they had stiffened or become brittle from heat, and they're fine.

  You'll get blackening of the +5V (red wires) at the connector body. You may
also get discoloration of the connector pins.

  I had a '73 that ran for several years with no problem, and then one day
went *POOF*. No new boards, but it flamed out anyway. It's an accident wait-
ing to happen.

	Terry Kennedy		Operations Manager, Academic Computing
	terry@spcvxa.bitnet	St. Peter's College, US
	terry@spcvxa.spc.edu	(201) 915-9381

don@zl2tnm.gp.govt.nz (Don Stokes) (10/13/90)

terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr) writes:

>   If you're getting 4V there, it isn't dust 8-(.
> 
>   Seriously, if you have a BA23 master box (with front panel halt and restart
> buttons), you should have the power harness checked. The old harness can catc
> on fire. The way to tell if you have the proper harness is to open the BA23
> up and look in the drive bays. The good harness is DEC part number 17-01311-0
> and uses white connectors and various colored wires. If you have yellow con-
> nectors or a webbed cable (rather than individual wires) you have a potential
> safety hazard.

Drat.

I just pulled my MicroPDP-11 (11/73) to bits, and located the power
harness (under a *lot* of dust).  It's part number 70-20450-01, with
yellow connectors, but separate coloured wires.

What was the nature of the overheating problem?  The part I have in my
hand looks OK, no sign of any heating.	I flexed the conectors to check
if they had stiffened or become brittle from heat, and they're fine.

I can't remember what type of cable the MicroVAX I that I saw cook a
power harness had.  Can you (or anyone else) remember the details before
I go rummaging through the "museum" looking for the offending part or
hassling DEC?


Don Stokes, ZL2TNM  /  /                            Home: don@zl2tnm.gp.govt.nz
Systems Programmer /GP/ Government Printing Office  Work:        don@gp.govt.nz
__________________/  /__Wellington, New Zealand_____or:_PSI%(5301)47000028::DON

don@zl2tnm.gp.govt.nz (Don Stokes) (10/14/90)

terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr) writes:

> In article <Bw7Zq1w163w@zl2tnm.gp.govt.nz>, don@zl2tnm.gp.govt.nz (Don Stokes
> > I just pulled my MicroPDP-11 (11/73) to bits, and located the power
> > harness (under a *lot* of dust).  It's part number 70-20450-01, with
> > yellow connectors, but separate coloured wires.
> 
>   Winner! Winner! You have one of the flaming power harnesses 8-)

Oh wonderful.  I'll see if I can hassle a new part out of DEC.  What's
the story on getting fire hazards replaced outside warranty?  Service
contracts?  You gotta be kidding!

>   You'll get blackening of the +5V (red wires) at the connector body. You may
> also get discoloration of the connector pins.

She seems ok at the moment, no discoloration of connector or cable
attachment (I took off the black cap) -- I even threw caution to the
wind and stuck a finger on the +5V wires after running the thing for
a while.  I also took off the black cap over the attachement of the
wires to the connectors.  Can't hurt to get it changed though -- *if*
it can be done on the cheap!

Don Stokes, ZL2TNM  /  /                            Home: don@zl2tnm.gp.govt.nz
Systems Programmer /GP/ Government Printing Office  Work:        don@gp.govt.nz
__________________/  /__Wellington, New Zealand_____or:_PSI%(5301)47000028::DON

jcd@pitt.UUCP (Jim Durham) (10/16/90)

In article <cNy2q3w163w@zl2tnm.gp.govt.nz> don@zl2tnm.gp.govt.nz (Don Stokes) writes:
>terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr) writes:
>
>> In article <Bw7Zq1w163w@zl2tnm.gp.govt.nz>, don@zl2tnm.gp.govt.nz (Don Stokes
>> > I just pulled my MicroPDP-11 (11/73) to bits, and located the power
>> > harness...(stuff deleted)
>
>>   Winner! 

>Oh wonderful..(stuff deleted)
>
>>   You'll get blackening of the +5V (red wires) at the connector body. You may
>> also get discoloration of the connector pins.
>

This is pretty simple physics..Power (read: Heat) = Volts times Amps.
In my case, 1 volt drop times , lets say, 20 amps of current = 20 w
of heat.

The trick is to keep the voltage drop down. My experience with these
"Molex" style connectors is that they are not good for high-current
applications. Even if DEC has a new whiz-bang style replacement, the
pins on the power supply and bus are still the same old tinned junk.
My original comments still apply. Keep 'em clean!
>
>Don Stokes, ZL2TNM  /  /                            Home: don@zl2tnm.gp.govt.nz
Hi Don, I'm W2XO (that's ham radio talk, for you bit-pusher-only types 8-) ).

-Jim Durham   jcd@vax.cs.pitt.edu | durham@w2xo.pgh.pa.us

terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr) (10/16/90)

In article <cNy2q3w163w@zl2tnm.gp.govt.nz>, don@zl2tnm.gp.govt.nz (Don Stokes) writes:
> Oh wonderful.  I'll see if I can hassle a new part out of DEC.  What's
> the story on getting fire hazards replaced outside warranty?  Service
> contracts?  You gotta be kidding!

Ok. Here's the complete scoop (and by the way, BA123's *are* also affected):

Box         Bad cable    Good cable   FCO Kit p/n  Status  Cost USMLP  Date
----------  -----------  -----------  -----------  ------  ----------  -----
BA23-AX,BX  70-20450-01  17-01311-01  EQ-01427-01  I, N/A  N/A         12/86
BA23-C/D    70-20450-02  17-01312-01  EQ-01467-01  I, N/A  $34.00      07/87
BA123       17-00865-01  17-01311-02  EQ-01468-01  I, N/A  N/A         01/88

  This should give you all you need to know. The "Status" means that it's an
"Improvement" call FCO, and is chargable to non-contract customers and free
upon request to contract customers. The "N/A" in "Status" means that as far
as I can tell, the EQ kit is no longer available and you have to order the
loose piece part. The "Cost USMLP" is the cost to buy the loose piece part.
If it's "N/A", it means it is an unlisted part and you have to get it by spec-
ial order.

  *PLEASE NOTE*. I don't work for DEC. The above is not an official statement
of DEC's position. Contact your local DEC representative for official infor-
mation. [The preceding sentences paid for by the "Avoid the Lawyers" campaign.]

> She seems ok at the moment, no discoloration of connector or cable
> attachment (I took off the black cap) -- I even threw caution to the
> wind and stuck a finger on the +5V wires after running the thing for
> a while.  I also took off the black cap over the attachement of the
> wires to the connectors.  Can't hurt to get it changed though -- *if*
> it can be done on the cheap!

  The problem with the old cables is that the yellow/black connector depends
on IDC for the connection. Imagine sticking a piece of insulated wire into
the tines of a dinner fork. Get a good connection? I thought not 8-). That's
what is being changed - to a conventional stripped+crimped connection.

	Terry Kennedy		Operations Manager, Academic Computing
	terry@spcvxa.bitnet	St. Peter's College, US
	terry@spcvxa.spc.edu	(201) 915-9381

johnd@physiol.su.oz.au (John Dodson) (10/21/90)

In <cNy2q3w163w@zl2tnm.gp.govt.nz> don@zl2tnm.gp.govt.nz (Don Stokes) writes:

>terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr) writes:

>> In article <Bw7Zq1w163w@zl2tnm.gp.govt.nz>, don@zl2tnm.gp.govt.nz (Don Stokes
>> > I just pulled my MicroPDP-11 (11/73) to bits, and located the power
>> > harness (under a *lot* of dust).  It's part number 70-20450-01, with
>> > yellow connectors, but separate coloured wires.
>> 
>>   Winner! Winner! You have one of the flaming power harnesses 8-)

>Oh wonderful.  I'll see if I can hassle a new part out of DEC.  What's
>the story on getting fire hazards replaced outside warranty?  Service
>contracts?  You gotta be kidding!

Oh dear !
First time this happened to me, I just bought the connectors locally (nice
white ones ;-) & crimped (& soldered to be safe ;-) the new cables myself.
This has worked for 7 PDP11/73's & 4 uvax II's, if anyone in Australia
need's  these connectors mail me, I have about 20 left ! ;-) I've used a lot
in other departments at this Uni, having bought 50 originally ;-)
You can have them cheap, say 50c each & $50 for the advice ;-)
I could even mail "ready made" to the US if you are desperate & would like to
make a donation to medical science ;-)

Please guys (& gals) consult your technical staff or local connector supplier &
save $$$$$$. Don't hassle DEC field service they are too busy with their fire
extinguishers ;-)

John Dodson,					Dept of Physiology,
johnd@physiol.su.oz.au				University of Sydney,
Phone	+61-2-692-3277				NSW 2006
Fax	+61-2-692-2058				Australia.

don@zl2tnm.gp.govt.nz (Don Stokes) (10/24/90)

terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr) writes:

[Flaming power harnesses on BA23s]

> Ok. Here's the complete scoop (and by the way, BA123's *are* also affected):
> 
> Box         Bad cable    Good cable   FCO Kit p/n  Status  Cost USMLP  Date
> ----------  -----------  -----------  -----------  ------  ----------  -----
> BA23-AX,BX  70-20450-01  17-01311-01  EQ-01427-01  I, N/A  N/A         12/86
> BA23-C/D    70-20450-02  17-01312-01  EQ-01467-01  I, N/A  $34.00      07/87
> BA123       17-00865-01  17-01311-02  EQ-01468-01  I, N/A  N/A         01/88

The good news is that DEC presented me with a new power harness cable
today, free of charge (or freight -- it hitched a lift with our friendly
DEC Field Service Person who was attending to a slightly malfunctioning
8550).

The bad news is that the part they gave me is P/N 17-01311-02 (ie the
part for a BA123), not -01 (the BA23 part), and no matter what contortions
I try, it physically will not reach between the connectors on the power
supply and the BA23's backplane.  I didn't think it looked quite right,
and questioned it on taking delivery (but didn't have part numbers on me
at the time, and was more concerned about the ailing VAX).

Sigh.


Don Stokes, ZL2TNM  /  /                            Home: don@zl2tnm.gp.govt.nz
Systems Programmer /GP/ Government Printing Office  Work:        don@gp.govt.nz
__________________/  /__Wellington, New Zealand_____or:_PSI%(5301)47000028::DON