[time-nuts] Re: Frequency counter recommendation wanted

[Bob Camp]

Hi

One “interesting” point about the 1992: Coming up with an accurate answer to the question 
“how good is it?” is not at all easy. 

One *could* spend the day with the Racal sales guy asking that question. At the end of the 
day, the result would be confusion. Spend a day with the spec sheets an confusion also is 
the result. 

About the only conclusion you can come to is that they *really* don’t like to commit to a number.
One might wonder ….

Next you could grab one of the ones sitting around the lab and start looking at the results. 
Considering this is done in 1970’s, the gear to do so isn’t quite what you have today. Best
guess at the time was something around 4 ns for a “single shot” sort of resolution. 

On the same basis, the 5334 comes in at 2 ns and the 5335 comes in at 1 ns. The 53131
is around 0.5 ns and the 53132 is around 0.2 ns. 

The 5345 is also a 2 ns single shot device. Like the 5334, (and others) it can do averaging
to get lower.“Single shot” and “averaging” aren’t quite the same thing ….

The the 5370 (went it works) can get to 0.02 ns “RMS” and the SR620 can do about the 
same. The 53230 can get to 0.01 ns with a good input. Since these are one sigma rather than
a fixed LSB, the comparison isn’t quite perfect …..

Does any of this matter? It depends a lot on just what you want to / need to do. That’s always
the gotcha with these questions. Things are rarely tightly defined ….

If I’m looking at the delta between two 1 pps pulses, then I likely *do* care about single shot
sort of stuff. If I want to know if my transmitter is 2 Hz off at 10 GHz, that’s a different question.

Bob



> On Mar 15, 2023, at 4:07 PM, John Vendely via time-nuts <time-nuts at lists.febo.com> wrote:
> 
> I concur.  I've also experienced keypad problems with the 1992, and know of someone else who has, as well.  Otherwise, they're good instruments...
> 
> 73,
> 
> John K9WT
> 
> On 3/15/2023 3:08 PM, Ken Winterling via time-nuts wrote:
>> John,
>> 
>> I've owned three of the Racal-Dana 1992 counters for years. The one
>> weakness I have noticed is that the keypad tactile switches.fail at a very
>> high rate. I got my counters very cheaply because they were advertised as
>> defective, for parts only.  If you encounter a switch problem I can send
>> you information on what I have done to fix the problem.
>> 
>> 73,
>> 
>> *Ken*
>> *WA2LBI*
>> 
>> 
>> 
>> 
>> On Wed, Mar 15, 2023 at 12:15 PM John Miller via time-nuts <
>> time-nuts at lists.febo.com> wrote:
>> 
>>> Right now I think I have a line on some Racal-Dana 1992s from another
>>> member, but I have heard so much good about the 5334B I think I want to get
>>> one just on principle. Especially once I have these 1992s, I could pick up
>>> a faulty 5334B and repair it! I think that could be a lot of fun and save a
>>> good bit of money.
>>> 
>>> I really appreciate all the feedback, folks, and hopefully in another 5-10
>>> years those I'll be able to get my hands on an  SR 620 or a multi-GHz model
>>> like the HP 53132A. I have an Agilent E4402B that needs some work....
>>> 
>>> Thanks all,
>>> John
>>> 
>>>> On Mar 14, 2023, at 4:39 PM, Stewart Cobb via time-nuts <
>>> time-nuts at lists.febo.com> wrote:
>>>> I've bought perhaps a dozen HP 5334B counters off ePay for my company
>>> labs
>>>> over the last decade, at prices from about $75 to $225. I've never had a
>>>> problem with one. Rick built a good box. For various reasons, the B model
>>>> is a better bet than the A model.
>>>> 
>>>> If it does everything you need it to do, the 5334B is the counter to get.
>>>> It's blessedly quiet because it doesn't have or need a fan.
>>>> 
>>>> Pro tip: you will occasionally see one listed "as is" with the display
>>>> showing "no osc". Those counters are generally fine. All that message
>>> means
>>>> is that the timebase switch on the back is set to "external" rather than
>>>> "internal", and the seller doesn't know to flip it back. If you can see
>>>> that on the photos, don't be afraid.
>>>> 
>>>> If it's an OCXO model (option 010), it's possible that someone harvested
>>>> the 10811 OCXO before selling the remainder. I've never seen that, but
>>> it's
>>>> possible. If so, it's a bit tedious to change back to the default
>>> internal
>>>> crystal, but these counters will always work with an external 10 MHz
>>> input.
>>>> Cheers!
>>>> --Stu
>>>> _______________________________________________
>>>> time-nuts mailing list -- time-nuts at lists.febo.com
>>>> To unsubscribe send an email to time-nuts-leave at lists.febo.com
>>> _______________________________________________
>>> time-nuts mailing list -- time-nuts at lists.febo.com
>>> To unsubscribe send an email to time-nuts-leave at lists.febo.com
>>> 
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at lists.febo.com
>> To unsubscribe send an email to time-nuts-leave at lists.febo.com
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe send an email to time-nuts-leave at lists.febo.com