[time-nuts] HP 5071A Electron Multiplier of Cesium Beam Tube

John Miles jmiles at pop.net
Mon Sep 7 05:30:18 UTC 2009

> Out of curiosity, what do the Datum tubes sell for? Are they any good?

I've never seen a price list anywhere, just the brochure (
http://www.gigatest.net/datum/cesium_tubes.pdf ).  No doubt the replacement
A9 and A11 modules increase the package price substantially, if they still
sell them at all.

> Wait a minute.  Datum owns Symmetricom which bought HP/Agilent's
> CBT production line and admits that their tubes were never as
> good as HP/Agilent's tubes.  The 5071, as opposed to the 5061, has
> a ROM with a password in the CBT so that only OEM tubes will work.
> The password consists of a poem written by Len Cutler.  The poem is
> copyrighted so that it would be illegal for a competitor to simply
> reproduce it, according to Len.

Heh, that's awesome!  But with inkjet case rulings coming down on the side
of the aftermarket vendors (e.g.
http://www.techdirt.com/article.php?sid=20041026/103208 ) I suspect the
judge would chuckle, give the plaintiff an E for effort, and then find for
the defendant.  Of course the 5071A predates the DMCA by a few years, as it
did the Symmetricom purchase, so who knows what would have happened in

My 'new' eBay-special 5061A has one of the Datum tubes and a 10811-60109
upgrade.  I don't know how old the tube is, or how it performed relative to
HP's specs when new, but the short-term noise on the 5 MHz output is within
specs despite the tube's measured (Vp-Vv)/Vb value being only about 0.8
instead of >1 as called for in the 5061A manual.

Peak beam current with this tube is about 25 nA (-980mv pk into 40 Mohms),
compared to the 80-200 nA typical range that the manual calls for, and the
Cs resonance valley isn't much higher than the noise floor observed with no
RF applied (-550 mv versus -520 mv).  But the output phase noise, at least,
is well within specs (-88 dBc/Hz at 1 Hz in short time constant mode,
spec=-82 dBc/Hz, and -105 in LTC mode, which is actually better than the -96
dBc/Hz spec for the option-004 tube.)

One thing I find confusing is how HP's published ADEV and PN specs didn't
change between the 5061A->5061B generations.  I'd think the short-term
stability would have gotten better when they went to the 10811-60109 for the
'B' model -- hence my better-than-expected PN reading at 1 Hz -- but the
5061B manual still has the same spec charts in it.  Obviously most were much
better than spec, but the numbers are so conservative that it's hard to say
how well a particular unit is really performing unless you have a whole
warehouse full of them for comparison.

-- john, KE5FX

More information about the time-nuts mailing list