How much IF power can I pump into this diode mixer...

After doing a bit of reading...

The "rule of thumb" is that P1dB is around 10dB (mid- and high-frequencies) to

15dB below IP3.

From Mini-Circuit's data sheet, the (input, presumably!?) IP3 is specified as

12dBm. Hence, at mid- to high-frequencies, P1dB should be around 2dBm (referenced to the input).

Anyone care to concur or refute this?

Thanks,

---Joel

I'll concur. This app note explains it:

It'll probably be a tad under 0dBm.

--
Regards, Joerg

http://www.analogconsultants.com/

One more remark: If you really have to push the limits here I'd measure it. Hook up a generator plus two step attenuators, 1dB steps and 10dB steps. Then measure the output with an RF meter or with a good spectrum analyzer and see where the output lags 1dB.

--
Regards, Joerg

http://www.analogconsultants.com/

Thanks for the pointer Joerg.

I'm a little red-faced now to notice that in tiny print (at the bottom of the electrical specifications box) the data sheets lists, "1dB Comp: +1dBm typ."

Good to know we're all in agreement, plus or minus a dB or so. :-)

I once presented a cost analysis in front of "suits". Doctors, attorneys, bankers and one EE (the CEO). Everyone was in a good mood and the occasional joke was cracked. Then a banker asked me what my confidence levels in those numbers was. "Plus minus 3dB."

The CEO later told me "My heart almost stopped. Don't ever do that again".

--
Regards, Joerg

http://www.analogconsultants.com/

The rules of thumb are largely jokes except for establishing broad orders of magnitude of signal levels. That 40mA maximum input "current" specification means DC current, not RF signal current, and is somewhat important for certain applications such as using the mixer as a current controlled RF attenuator. As for IP3, there are two types, output IP3 denoted OIP3, and input IP3 denoted IIP3. The default is nearly always taken to mean OIP3 unless stated otherwise, making that midband 12dBm an OIP3 and the IIP3 about 12dBm+6.5dB (conversion loss)=~20dBm at the input. And that would be about right for that little part. Of course all of these numbers are strongly dependent on the LO level, and unless stated otherwise are taken to be at an LO level of +7dBm. The intercepts drop off rapidly with LO power. Another rule of thumb is that in order to realize the maximum dynamic range available from the mixer, the signal levels should be no greater than 20-30dB below LO, meaning -23dBm to -13dBm range, note that this puts the intermodulation down 2x the difference between these levels and the IP3, or 2x (+12-(-13)=50dB down from the fundamental worst case, and this is all measured at the output.

LOL

Another appnote that may help to compare is this one: HP Application Note 995: "The Schottky Diode Mixer"

Joop