Lowest Mixer IF Frequency

[hem213]

Hello everyone! sorry if these is a basic question

I am having a problem with figuring the lowest possible IF frequency that can be achieved with a specific RF mixer that I need for a project.

https://www.digikey.com/en/products/detail/analog-devices-inc-maxim-integrated/MAX2685EEE/1867170

the LO and Mixer-IN frequencies for my project are going extremely close, like something like around 991.5 Mhz and 991.6 Mhz for example. On the datasheet, no matter where I looked it did not seem to include any information about the lowest possible IF freq. but on the "Mixer Typical S-Parameters" section, it seems like the lowest tested IF frequency they did was at 10mhz. I'm not sure if this is the limit or just the constraints of their testing. Other mixer ICs I've shopped around for always seems to include the IF range so it is confusing why this does not

thank you in advance!

[radiolistener]

The mixer don't know about IF, it just do frequency shift by multiplying input RF signal with LO signal.

[hem213]

so there is no limit? could it go all the way to down to the DC range? maybe I might be misunderstanding

I only ask because I when I see the datasheet of an mixer IC like the one below, it specifically gives an IF range

https://www.minicircuits.com/pdfs/ADE-2+.pdf

https://www.digikey.com/en/products/detail/mini-circuits/ADE-2/13927551

[Zero999]

It has a lower frequency limit. The IC is only specified over 800MHz to 1000MHz. The low noise amplifier is AC coupled via an internal capacitor, so it can't work down to DC.

[hem213]

It has a lower frequency limit. The IC is only specified over 800MHz to 1000MHz. The low noise amplifier is AC coupled via an internal capacitor, so it can't work down to DC.

yes I think I misspoke, I'm aware of the frequency limits for the inputs of the mixer, I was just asking about how low the downconverter output IF frequency can reach.

For example, when mixing 991.500 MHz and 991.600 MHz signals, am I going to get an 100Khz IF frequency or is there something on the datasheet that I'm missing that specifies the lowest absolute limit of said IF output frequency?

[gf]

The schematic in the datasheet shows a diode ring mixer with transformer-coupled RF and LO inputs, but IF output is DC coupled. So basically, why should there be a low limit for the IF?

(Edit: I'm referring to the linked ADE-2+.pdf)

[RFDx]

so there is no limit? could it go all the way to down to the DC range? maybe I might be misunderstanding

Yes, you can go down to DC. The IF-outputs are open collectors. Use 50...100 Ohm resistors instead of inductors as pull-ups to Vcc.

[newbrain]

so there is no limit? could it go all the way to down to the DC range? maybe I might be misunderstanding

Yes, you can go down to DC. The IF-outputs are open collectors. Use 50...100 Ohm resistors instead of inductors as pull-ups to Vcc.

It might go down to DC, but note the "Mixer Gain vs. IF Frequency" graph on page 5.
It seems to be optimized for ~80 MHz, 6 dB gain, and already at 60 MHz the gain drops to about -12 dB.
All AC specification are given for an 80 MHz IF (960 MHz LO, 880 MHz RF), so I would really not rely on working (well) at a low IF frequency.

[PA0PBZ]

It might go down to DC, but note the "Mixer Gain vs. IF Frequency" graph on page 5.
It seems to be optimized for ~80 MHz, 6 dB gain, and already at 60 MHz the gain drops to about -12 dB.
All AC specification are given for an 80 MHz IF (960 MHz LO, 880 MHz RF), so I would really not rely on working (well) at a low IF frequency.

That graph is for the evaluation kit, where among other components there's a 680nH coil connecting ifout+ and ifout- which certainly limits the lower IF performance. The mixer (and certainly the EV kit) seems to be designed for the 900MHz band/80MHz IF but that doesn't mean it will not work for a lower IF with adjusted external components, time for some testing I guess.

[radiolistener]

there can be a problem with higher noise floor at low frequency part of spectrum on the output.

[DimitriP]

If it doesn't have to be cheap,small or not need an IF amp,  it can do everything else :  SYM-63LH+  ,   ZX05-43-S+

https://www.minicircuits.com/WebStore/dashboard.html?model=ZX05-43-S%2B
https://www.minicircuits.com/WebStore/dashboard.html?model=SYM-63LH%2B

[CaptDon]

Very low I.F. frequencies can be a problem with image rejection. At 900MHz the front end can't roll off image frequencies that are as close as your 100KHz I.F. so you run into a problem. With the L.O. so close to the receive frequency any L.O. leakage including its noise components may actually cause de-sensing of your front end amplifiers if you are using them. How did you end up with R.F. / L.O. frequencies so close together?? Standard off the shelf I.F.'s tend to be 10, 30 or 60MHz and you can see many offerings from RHG listed on Ebay. Many of the radar systems I worked on were full of Avantek and RHG products from the mixers down to the log/lin video outputs.

[iMo]

Yep, "images" and "mirrors" - at the output of a "single sideband mixer" you will see every possible combination of

n*LO+m*RF and n*LO-m*RF and m*RF-n*LO

where m, n are the harmonics of the RF and LO signals.

With an I/Q mixer the situation looks a bit better.

[ftg]

Jason Milldrum of Etherkit has done some experimenting with MAX2681 in direct conversion receivers.
Scroll down to "MAX268x Experimental Direct Conversion Receiver" on https://nt7s.substack.com/p/summer-etherkit-update

If he gets usable performance from it with nominal 300Hz - 3400Hz IF, then your 100kHz IF also sounds feasible.
But it is still outside the nominal spec of it.

For guaranteed performance I'd echo the suggestions of others about going with some nice double balanced diode ring mixer.
I would have suggested Mini-Circuits RMS-2, but their MCA1-24+ seems to be cheaper on Mouser and seems just as usable.

Edit: Just noticed that Jason Milldrum has a whole complete post on evaluating MAX268x-series mixers.
https://nt7s.substack.com/p/max268x-active-mixer-evaluation

Edit2: Fixed some typos.

[hem213]

THANK YOU ftg! this is so much more than I expected, I really appreciate it. and thank you everyone for detailed explanations! This forum is a Godsend