[time-nuts] 53131A or 5370B

[John Miles]

I've heard this recommendation before, but I'll admit I'm not sure I
understand it.  My 5370B has no problem reading frequencies out to 1E-10 at
1-second gate times.  10 digits/second, just like the 53131A.  Sure, the two
LSDs (1E-11 and 1E-12 at 10 MHz) are effectively tied to a random-number
function, but if it bothered me that much I'd put a piece of electrical tape
over that part of the display.

I've seen JohnA's page at
http://www.febo.com/time-freq/hardware/5370B/index.html but am not sure how
to reproduce the graph at the bottom that shows random noise at the +/- 1E-9
level.  That's at least 10x worse than what I observe on the display,
counting a 10 MHz signal at 1 second gate time.

An automated application with the 5370B can (and should) use its "computer
dump" mode to move the processing to the host PC, so I don't see the 5370B's
slow CPU as a real drawback.  I haven't used a 53131A before.  What,
specifically, is better about it?

-- john, KE5FX

> >
> > >Here  is the question, I only have room for one counter,
> should it be a
> > 5370B or a  53131A?  Which >is more useful to measuring
> frequency  variations?
> >
> > >Jeff
> >
> >
> > Hi Jeff,
> >
> > In my opinion the 5370B is better for phase offset
> measurements, and  the
> > 53131A better for frequency measurements.
> >
> > The 5370B is not good at all for direct frequency measurements.
> >
> > The 53131A is not that good for phase measurements.
>
> I agree with Said here. The problem with the 5370B is really in
> the processing
> it does, not the hardware, which is excelent for its time.
>
> I would use the 5370B for more static setups doing longer logging
> sessions, but
> for most bench-measures the 53131A is probably more versatile.
>
> Personally I use my 5335A for versatile bench-counter.
>
> Cheers,
> Magnus