Re: 7904 Mainframe production dates

Denis Cobley <denis.cobley@...>

Hi all
B010100 is always the first saleable unit serial number.
All below that were non saleable demo units (they were often upgraded for
sale at a later date and sold with full warranty as ex demo units).
Major serial number breaks were used to denote significant changes in
Tek often had a number of ongoing mods as late field / customer identified
problems were fixed or redesign took place,.
These were incorporated into new board layouts and circuits (this is why you
have 2465B service manuals for below B050000 - discrete components and
B050000 and above - same scope but now with several boards surface mount
parts only).
As has already been noted, Tek recycle the same serial numbers on most
products and use the first letter/number to designate the plant / country of
A couple that were not mentioned are "N" which denotes manufacture by
Hinditron - this was Tek India and "C" which you will find on more products
as China. All The current TDS1000/2000 range is only made in China (as were
the later TDS200 series).

----- Original Message -----
From: "Fred Olsen" <>
To: "Miroslav Pokorni" <>; "washesmelon"
Cc: <>
Sent: Friday, July 11, 2003 3:41 PM
Subject: Re: [TekScopes] 7904 Mainframe production dates

I had said:
>>Jeff, the only one I can address at the moment is the early 7904. It
>>would be reasonable to assume that early production exceeded the 250k
>>point, hence the restart at 260k rather than 250.
>>That said, I still would find it difficult to
>>comprehend that Tek built over a quarter-million 7904s!

to which
Miroslav Pokorni wrote:
> When you expressed doubt that Tektronix built over a quarter million
> 7904, I did agree, but did not think it was worth commenting.

I didn't mean to imply that Tek had built anywhere near a
quarter-million 7904s in the first series. I can't believe that.
First, in the later years many (all?) units' S/Ns started with B010100
rather than the traditional 00101 of the old days. Second, I suspect
that many of the runs were discontinuous. Earlier on Tek seemed to put
a clean break at a major rev, a redesign, such as "1A1 >S/N 20000" -
which was when the FET front end started IIRC. Later on the "greater
than"s seem to more often be in succession with lesser revisions which
were handled with a numeric break. So do I think that there were 19999
Nuvistor 1A1s? No, but close to it.

As a hypothetical I grabbed a late-production 7A26 manual from the file.
First Printing was Oct '72 at model start, with a Revised Feb '87. A
good example in later gear (all right, I know that the "new Tektronix"
doesn't consider a 7k-anything to be "later") - a good example is board
revisions. The amp board P/N went through a bunch of revs: -00 through
-08, -10, -13, -15, -17, and -21. All of those are listed with
consecutive S/N breaks. Bear with me, it's easier to see in a little

-00 B010100 ~ B049999
-01 B050000 ~ B069999
-02 B070000 ~ B083789
-03 B083790 ~ B089999
-17 B245520 ~ B251089
-21 B251090

I would suggest that it's reasonable to assume that revisions such as
the -02 to -03 were running changes and were indeed made at that S/N
break. But the ones at an even ten thousand, such as -00 to -01 or -01
to -02, would seem to simply be available blocks of assigned numbers.
There is no reason to think that all of those blocks were filled. There
might have been as few as a hundred or less used out of any 10000. It
all depended on sales demand, and what was at the time their traditional
engineering-driven continuous implementation of design improvements.

The only way to know for certain would be either to have a large pile of
all of the change sheets, which perhaps only Tucker might have been able
to put together outside of Tek; or to gain access to Tek's production
numbers, which isn't likely to happen even if they still have them.

So, do I think that they built a quarter-million 7A26s? No way. Or
that many 7904s? Even less likely. When I told Jeff that the
early-series 7904 production exceeded the 250k point, it was meant only
as a number break and not as an actual quantity. Granted that this
doesn't take into account the non-U.S. production using different S/N
series, but in most cases those numbers were small compared to

If pressed for a SWAG of the number of 'early' 7904s, the ones below
260k, I wouldn't think it could exceed four figures. Much of this is of
course my opinion, and I'll readily concede that I'm no expert. So,
grain of salt, YMMV. Anyone with actual knowledge of this please weigh

Best to all,
Outgoing checked by Norton AV

Your use of Yahoo! Groups is subject to

Join to automatically receive all group messages.