[time-nuts] Fine print on HP 5334B
richard at karlquist.com
Thu Feb 22 18:39:45 EST 2007
I don't have much more of an answer to give you other than
just reiterating that Ken the software man did more than we
had a right to expect, as it is, given what he had to work with.
At HP, what we liked to do was to run a synthesizer from
the frequency standard and set it to put out an offset
frequency, like 10.0005 MHz. (We had the luxury of having
a lot of 8662's around :-) We would then use the HP
mixer box from the 5390 system (K59-53??? or something
like that) to mix with the DUT and produce a 500 Hz
beat note. This could then be counted by any old counter to high
resolution. We didn't give this up until the 53181 came
out. It can do an honest 12 digits thanks to our FPGA
wizard Eric. It would be unreasonable to expect the '34 to
be able to match that.
Rick Karlquist N6RK
SAIDJACK at aol.com wrote:
> In a message dated 2/22/2007 23:15:49 W. Europe Standard Time,
> richard at karlquist.com writes:
> 3. The interpolator has somewhat questionable accuracy when you
> get down to 12 digits. This is hidden somewhat by the firmware,
> but if you monitor the bits the interpolator puts out, you can
> see what is going on. To me, it is kind of amazing it works as
> well as it does considering the design.
> So please lower your expectations!
> If you want a really good HP counter, move up to the 53181 which
> is much more modern, even being 15 years old.
> Rick Karlquist N6RK
> R&D Project Manager of 5334B (circa 1986)
> Hello Rick,
> I had a couple of Q's on these units, glad you are on this mailing list!!
> I have an Rb/Cs feeding the 10MHz ref ext. I feed 10MHz to the front panel
> from my DUT.
> Then I subtract 10.0MHz from the display using math, and set a 99s gate
> time. This gives me down to 100uHz (1oo -6 indicated) resolution (11
> digits) of
> the frequency difference.
> Q1: Now when I press the 100 average button, I would expect the unit to
> calculate the averages of 100 (albeit very long) measurements and give me
> least one, if not two more digits of resolution.
> But the unit just comes out of this with some sort of overflow error.
> Q2: How come I cannot enter more than 99s measurement intervalls? It would
> be great if the unit supported 1K and 10K intervalls to improve resolution
> or two digits.
> Is there any other trick I can use to increase the resolution, say by
> feeding the Rb/Cs also into channel B etc?
> Thanks in advance,
> <BR><BR><BR>**************************************<BR> AOL now offers free
> email to everyone. Find out more about what's free from AOL at
More information about the time-nuts