[time-nuts] Question concerning failure and value of HP 5371A

[Magnus Danielson]

From: "Arnold Tibus" <Arnold.Tibus at gmx.de>
Subject: Re: [time-nuts] Question concerning failure and value of HP 5371A
Date: Sat, 15 Oct 2005 19:46:41 +0200
Message-ID: <E1EQq8V-0003QR-6Y at febo.com>

Hi Arnold,

> Hi Magnus and the others in the group,
> thank you very much for your comments and the help!
> 
> I had a look to the counter at the seller, propably really ok, but its at ebay 
> already above 1000 $, and he believes to get much more. 
> (We will see soon where it ends).
> see #http://cgi.ebay.de/1-St-HP-5371A-Frequency-and-Time-Interval-Analyzer_W0QQitemZ5817342262QQcategoryZ31337QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

That is not all that supprising.

> In Germany such devices are rare and very expensive, as there are some 
> dealers buying all the good equipment and reselling them to companies with 
> multiplicators....
> Even for a really defective 5371A, showing fatal errors on RAM,  
> mathunit and others, they ask for at least around 600 $. 
> That's not my collar size....

Right. A failed battery is expected and really part of standard maintenance
IMHO. It does not take much effort to change either, just some basic care as to
not heat up the battery too much in the soldering process.

When I got my HP5372A (w option 30 and 40 if you care to know) I had to replace
the battery. After the setup-process it has been flawless. I think my 1500 USD
was a good price considering the options and state.

> This forces me to try to develop a very high speed gating device with 
> d-flip-flops as ripple counter, capable to work below few ns, best would 
> be to count up to 500 MHz pulses (1ns), but unfortunately I am already 
> an old codger who liked to work with analogue rf signals, not liking all 
> the digital devices.
> The point is, I do not find gates capable to work well beyond 30, 40 MHz.
> I there somebody having a helpful idea how this could be solved?

Look at one of those spiffy Xilinx Spartan III FPGAs. It is prooved in battle
for commercial lowcost counters.

> PS: I had a look to the internal h/w of the hp 5371A, the ocxo is a
> normal hp 10811-60111 in, I thought to find a higher qualified type in 
> such devices. 

You are thinking backwards. For normal spectrum analysers and counters you had
to ask for option 010 to get an 10811-60111 installed. Here you had it by
default. If you need better, you have the 10 MHz external reference input in
the back. You are arguing from a "spoiled" level. The 10811-60111 is a very
good oscillator compared to alot of the things we see normally in counters from
that time. You can *always* do better, but will the customers usually gain from
it as much as it costs? How much gain in typical measurement performance will
the additional cost be? Looking at the suite of applications notes for the
5371, I'd say that they are usually short-timed enought for it to be not a
major issue.

> No, I am not watching all that illnesses about jamming gps etc., 
> because I think that this is good calming down people, but this would 
> not stop informed guys at all navigating to the target...there are other 
> ways. So what.

Well, if you can fly through a superpowers defenseline with a small airplane
and land on the square in the middle of the capital, next to the center of
political powers, then you can ram your airplane into a nuclear power plans
without functional GPS. So, the effect is highly debateable, while the hurt is
certainly not good, definite and documented.

Cheers,
Magnus