EMRFD Message Archive 3166

Message Date From Subject
3166 2009-06-09 07:16:35 Kenneth Stringham Hp 8640 Generator
As I recall from using this generator, this a very clean generator. The major problem will be that it can drift which can be mitigated by using the display lock feature. It will tend to jump frequency slighty and microphonic should you be using it for precise measurements and very narrow filtering, but otherwise it was/is a great signal source.

Ken - AE1X
3167 2009-06-09 14:06:47 WB6TPU Re: Hp 8640 Generator
The HP8640 is one of the spectrally cleanest generators available.

I've never had any problems with mine, however one classic failure mode on units that have been in storage for a long while with the front panel facing the ceiling and back panel facing the floor is that the lubricant in the cavity oscillator can cold-flow out of it's proper place in the cavity and cause intermittent tuning as the cavity tuning mechanism moves. If possible, be sure to check for constant RF output across the tuning range (i.e. no dropouts).

The other main thing to beware of is that the nyl
3169 2009-06-10 03:13:46 drmail377 Re: Hp 8640 Generator
How does the HP 8654B compare with the 8640B? It is 10MHz to 520MHz and lots cheaper than the 8640B which is something like 1MHz to 500MHz.

73's David

3170 2009-06-10 05:33:10 Mike Czuhajewski Re: Hp 8640 Generator
> ...the nylon gears that perform the bandswitching can
> deteriorate over time and break. Mechanical replacement parts
> have become scarce as hen's teeth.

That need not be a significant problem, as many have reported
success in using epoxy to glue them back together. (That's "real"
epoxy, the type requiring a 24 hour cure, not 5 minute epoxy.)
One person even went so far as to wrap a turn or two of bare wire
around the hub for reinforcement before applying the epoxy, for
reinforcement. (He sent me some photos of it, which I can pass
along to anyone who wants to see them.)
3173 2009-06-10 10:57:07 Wes Hayward Re: Hp 8640 Generator
Hi David, et al,

This question is one that I'm probably better able to answer than most folks, for I have both generators -- a HP8640B and a HP8654A. There is no comparison -- the 8640 is the real thing. It has a wider frequency coverage (0.5 to 500 MHz), great tuning resolution, good stability (although it is an analog generator, so this is not a millihertz stable thing), and absolutely stellar low phase noise. There are some high end synthesized generators today that may reach or even exceed the phase noise performance of the 8640, but they are not the kind of thing that those of us reading this group will be purchasing, I suspect.

In contrast, the 8654 is a relatively small and light weight box that has no digital readout and suffers from really poor phase noise. I recall an experiment in an earlier lifetime when I was doing some IM testing while using an 8640 and a 8654 at the same time. This was dramatic, for one could see the phase noise on the 8654 carrier, even with a routine spectrum analyzer. Generally, if the phase noise is so poor that it is easily observed with a spectrum analyzer, the generator is considered unsuitable for communications testing. The phase noise is worse than routine appliances.

It's interesting to put the specs aside and just describe the results. Consider a simple experiment: Build a modest superhet, such as the EZ90 or the S7C from EMRFD, but use the signal generators in place of the LO. The phase noise performance of the 8654 is so poor that the receiver would sound like junk with severe reciprocal mixing. In contrast, the 8640 is so good that the folks building high end receivers such as the Triad (10 dB NF with IIP3=+45) use this generator as the LO as a mechanism to test the design.

Both of my generators have been tested for amplitude accuracy with a HP power meter. Both are great in this regard with accuracy commensurate with the resoluti
3178 2009-06-10 22:28:24 Gary Chatters Re: Hp 8640 Generator
drmail377 wrote:
> How does the HP 8654B compare with the 8640B? It is 10MHz to 520MHz and
> lots cheaper than the 8640B which is something like 1MHz to 500MHz.

A couple more 8640B notes:

The frequency range spec is 0.5 - 512 MHz, but it will underrun the low
end down to about 450 KHz, so is useful for 455 kHz IF work.

Options are:
1 - Variable AF modulation oscillator (standard is fixed 400Hz and 1000Hz)
2 - Doubler. Extends top range to 1024 MHz. The range switch goes to
1024, but it doesn't output that range without this option.
3 - Reverse power protection to 50w.
4 - Avionics test version. Several variations from standard. I do not
know how useful one of these would be for general purpose use.
323 - The military version AN/USM-323. Built in to a weatherproof box,
but missing some features: phase lock feature and external counter input.

If price is a concern and you do not mind nearly twice the mass and
volume and can get along with the missing features, consider getting the
AN/USM-323. There is one on eBay now, item 300321044094, but it has
been taken out of its big yellow box.

3180 2009-06-11 01:00:43 Ian White GM3SEK Re: Hp 8640 Generator
I have an HP8640A (repeat... "A") which has many of the same features
but doesn't have the digital readout or locking capability.

Does anyone have comments or experience of the 8640A that they'd like to

Also looking for a downloadable manual...


73 from Ian GM3SEK
3181 2009-06-11 11:18:35 WB6TPU Re: Hp 8640 Generator

Manuals for the HP8640A seem to be a rare commodity these days. I don't know where to find a free downloadable one, but if you are desperate, the following vendor (Global Test Equipment) seems to have a hardcopy available for purchase: http://www.4gte.com/Manuals7.htm

There may be other sources for the 8640A manual, but I'm not aware of them.

Usual disclaimer: I have no financial interest in or other knowledge of the company other than what is