In message BLU170-W99A6289D3FD1B2B5AEA2BACEBB0@phx.gbl, Mark Sims writes:

That's a smaller catastrophe than it sounds:  The display has a
pretty trivial parallel interface, and it will be pretty trivial
to connect something else instead.

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

I did have a display that was going dim on one digit so I ordered a new VFD
display from Agilent, not the whole board just the actual display.  When I
compared it to the original I found that the top to bottom distance between
the pins was larger than the original display.  So I had to carefully bend
the pins inwards towards the center until the distance was correct.  Imagine
a DIP IC with a 0.6 inch distance being reduced to a 0.5 inch distance.
Fortunately the pitch from pin to pin in the rows was the same.

There are 56 pins on this sucker and bending them inwards and then getting
them into the 56 holes on the PC Board was a real trick but it can be done
with a great deal of patience.  I never did question Agilent if they had
changed the PCB in later models to accomodate this change.  The new VFD was
a Noritake just like the old one.  The distance on the old one was 1.375"
and the new one 1.5".  I don't remember if the part number was different.
Since this is a very custom display made expressely for Agilent you would
have thought that they would have kept the dimensions the same on the pins.
Once I got done the display worked perfectly which was a big relief.

In looking at the Agilent site it looks like these might have already become
unobtainium but there is a repair service available and a PCB exchange.  My
guess would be that because of the size difference they want to do the
repair work themselves.

----- Original Message -----
From: "Mark Sims" holrum@hotmail.com
To: volt-nuts@febo.com
Sent: Saturday, January 18, 2014 9:55 AM
Subject: [volt-nuts] What made a HP3458A so expensive

fluorescent display.  These have a rather limited lifetime and leaving them
on 24/7 puts a lot of hours on the tube.  Someday those displays will become
unobtainium...

https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts

On 18 January 2014 21:15, Bill Gold wpgold3637@att.net wrote:

HP / Agilent seem to have a habbit of changing parts, without changing
the part numbers. This is quite a recent experience of this for me.

1. I have a couple of HP verification kits for vector network
   analyzers (85055A type-N and 85051B APC7). These consists of a couple
   of airlines, a couple of attenuators, in a wooden box, with foam to
   hold pieces in place. The foam degrades over time, so I ordered new
   foam for two kits. While the foam fitted in the bottom of the boxes,
   it was too small to fit the lid. So I had to make up my own.

2. I contact Agilent and ask if they can selll me the foam to fit the
   a third verification kit I have - HP 85053B 3.5 mm. The part numbers
   for the foam are still available, but I think they concluded they
   would not fit my box.

3. Agilent said I could buy a new box, with the foam that fits. The
   cost of this was around £350 (GBP) which seemed a bit excessive for a
   small wooden box with a bit of foam inside. There was nothing wrong
   with my box, so I declined.

4. Much to my delight, Agilent offer me a box free of charge to keep
   good customer relations.

5. Box arrives, but the airlines from my verification kit don't fit,
   as they are too long.

So basically Agilent have changed

• Size of airlines
• Design of box
• Size of foam to fit in lid of box

all without any change in part numbers!!

I've also got a 18 GHz type-N 85054B calibration kit for a VNA. The
manual on the Agilent web site, is supposed to cover the serial number
of the kit I have, yet the parts list printed on the bottom of the kit
is different to the Agilent manual. In this instance they have changed

• Part numbers of the sliding loads, although physically they look to
  be similar or identical.
• Part number of the connector gages, which have changed the layout of
  these a bit, so they don't fit new foam perfectly.

It does seem they can make significant changes to parts, without
giving them a new part number.

Dave