We occasionally receive requests for a mixer that will operate above our highest frequency mixer, the MM1-2567LS. The truth is that, contrary to the datasheet, the MM1-2567 actually operates above 67 GHz. It was designed to operate up to a frequency of 80 GHz on the RF and LO sides. The issue is that our test equipment can only measure conversion losses up to 67 GHz directly. To solve this problem, I devised an experiment to prove whether the mixer frequency actually extended to it’s simulation limit or not. Here is the experimental setup:
First we generate a 50 GHz LO using an A-2050, which provides the highest power, highest frequency LO we can provide. We use this to upconvert a 5-30 GHz input signal up to two sidebands: 45-20 and 55-80. We filter out the LO feedthrough and the lower sideband using an FH-5500. Next we downconvert the resultant single sideband signal using a second MM1-2567. The result is a high conversion loss, and an ugly signal:
However this includes two conversions, as well as the loss of the filter. If we divide this by two and make some assumptions about the loss of the filter (remember we can’t measure that directly above 67 GHz either), we get this:
So there you go, mixer operation up to 80 GHz. The conversion loss appears pretty high (20 dB) at these frequencies, but remember that we don’t know exactly what the insertion loss of the filter is, the connectors are all only rated to 67 GHz, and this is in a connectorized microstrip package that has its own losses. With all that, the performance of the chip doesn’t look too bad. The predicted CL is actually the measured loss of the MM1-2567LS from the datasheet, so it is there to confirm that our experimental method makes some sense. So feel free to buy the MM1-2567LS and try to use it above 67 GHz, just keep in mind that we can’t guarantee the performance since we can’t measure it, so use at your own risk!