[time-nuts] HP 5372A vs. 5370A

[John Miles]

> Here is a simple question: Why should I try to get an HP 5372A
> (or 5371A)? What
> are the benefits over the 5370A? Worth spending?

The 5370A/B is nice because its one-shot resolution is better than any other
HP/Agilent counter, prior to the release of the 532xxA models a few months
ago.  Also, you can hit the power switch and be making measurements five
seconds later -- no CRT to warm up, no menus to navigate, no persistent
modes to enable/disable.  Likewise its input coupling, level, and offset
controls are easier to deal with.  Its fan is much quieter than the 5372A's,
so you can leave it running for weeks if necessary without getting tired of
listening to it.

The 5371/5372 has a lot of bells and whistles that the 5370 lacks, but these
features belong on a host PC anyway, for the most part.

In short, I like the 5370 a lot.  You should get a 5371A/5372A if you need
it to do something specific that you can't accomplish with a 5370.
Otherwise I don't think it rises to the level of "You gotta have one of
these!"

> The biggest limitation of the 5372A is that you can only make 8191
> frequency measurements or 4095 time interval measurements with each
> measurement lasting no longer than 8 seconds.  If you want to use it for
> making more or longer measurements (e.g. long term Allan Deviation
> measurements) you have to start doing some GPIB programming, but I think
> you lose the no-dead-time advantage.

The 537x counters are best operated in TI mode, not frequency mode.  None of
them has any dead time in single-shot TI mode and (at least in the 5370's
case) the TI resolution is also somewhat better.  This often requires an
external frequency divider for the START input, of course.

> We really do need a simple app to "extend" the 5371 and 5372 for
> longer time
> ranges. If there's one out there, I've certainly missed it as I've looked
> around.

The current TimeLab beta has some basic 5371/5372 acquisition support
(http://www.ke5fx.com/timelab/setup.exe) in addition to the 5370 driver
that's always been there.  The 537x driver source code is included in case
it's useful to anyone else writing custom software.

The main limitations of the 537x TimeLab drivers are the lack of support for
FFT and histogram views of data acquired from TI counters in general, which
is a limitation of the program itself that may be addressed at some point,
and lack of support for fast binary transfers and host-based TI averaging.
I don't plan to spend much more time on the 537x-specific drivers, but
(well-tested) patches are always welcome.

-- john, KE5FX