With the performance I'm looking for, it means
detecting distortion products down more than 100dB from the inputs, but
that's something I'm up for. I do need to work on a little better
combiner for my sources, though, so they don't intermodulate before
going into the D.U.T.. I can get fairly close to the sort of
performance I asked about with some modern op amps, but not across the
full HF band.

Cheers,
Tom

Hi Tom. Look for Anzac combiners, we used them to measure +55 dBm Ip3 on the
VMP-4 amplifiers and didn't get measureable IMD from them at +6 dBm in.

John Thorpe (AOR 7030 fame) assures us that the simple MCL PSC 2-1 is
blameless to 0 dBm.

http://www.aoruk.com/comments.htm

I had been having what appeared to be hybrid IMD using some Merrimac
combiners, which totally cleared up by using DDS sources instead of low
noise HP 8640's modified to remove the levelling crosstalk. The DDS's do NOT
talk to each other and less attenuation for isolation is required.

Gimmee a hint on your "modern" op-amps. Split PSU's I presume?

Regards
W4ZCB

Since I really need wider bandwidth than the super-low-distortion ones
will give me, I'm not remembering those off the top of my head.
Well-at AUDIO, you can get distortion down in the -140dBc range for
outputs at a couple volts RMS with either the OPA627/OPA637 or the
AD797 at low gains, and the '797 also has about a 4dB noise figure for
a 50 ohm source. But for wider-band RF stuff, the OPA847 is one
candidate, if used carefully. It's not quite as good as I'd like to
see, but close.

Cheers,
Tom