[HECnet] No connectivity to arsgea 4

[Dave McGuire]

On 1/22/21 12:36 PM, Johnny Billquist wrote:
> Not really trying to claim there is anything particular about DEC 
> equipment here. While they were built properly, I don't think there 
> necessarily is anything exceptional about them.

   I can tell you from experience that there definitely is.  I've worked 
on computer equipment a very wide variety of manufacturers.  By far the 
highest-quality by most measures has been IBM, with DEC and HP following 
close behind.  Nothing else even comes close.  Note that I'm only 
talking about minis and mainframes here.

> But no, I do not believe any components have been replaced.

   That would be very unusual.  Field service guys carry boards for a 
reason.  When I was running VAXen commercially, we had several boards 
fail in our production machines, and they were replaced by DEC field 
service personnel.  The machines were VAX-11/750s, MicroVAX-IIs, and 
MicroVAX-3s.

> One of the 
> biggest issue is when equipment are left turned off for longer times. 
> That usually seems to be the time when problems appear.

   When that happens, the problem usually turns out to be dried-out 
electrolytic capacitors in power supplies, and occasional popcorning of 
small dry tantalum capacitors.

> By the way, we're talking 11/23 machines here. So BA23 box. Not that 
> much loaded in them, so no big strain on the power supply.

   Load, or lack thereof, makes no difference with the most common 
failure mode of the BA23 power supply.  The safety capacitors across the 
line always see the same line voltage.  Their flaws are aggravated by 
humidity in the air, not load.

   In fact, running the power supply under heavier load actually 
improves the safety capacitor situation due to "bake-out" effects on 
those capacitors.  Of course such elevated temperatures then shorten the 
life of the other capacitors, so..

   FYI, the safety capacitors in the BA23 power supply are:

Refdes  App           C       Class   Lead Spacing
--------------------------------------------------
C101    line-line     0.47uF  X       25mm
C102    line-ground   0.22uF  Y       20mm
C201    line-line     0.1uF   X2      20mm

> My experience with 11/70 machines are the same. If they are kept 
> running, they usually just continue to work. If you have them off for a 
> couple of years, you'll have a lot of work before they are stable again.

   Yup, and this is something we deal with all the time at the museum. 
We obviously cannot keep all of that hardware running all the time. 
During times of sparse tours (pandemic..), we rotate through the 
machines and run them for a little while every few weeks.

> But disk drives get into trouble with age, even if continually running. 
> And some are worse than others. The RD53 is sortof infamous here...

   Yeah, they started dying 20+ years ago.  Awful.

            -Dave

-- 
Dave McGuire, AK4HZ
New Kensington, PA