I'll look at it later. I'm mainly trying to make a VGA, which I guess is a lot like a mixer.

There are quite a few CMOS VGA’s out there. bulk-driven circuits are popular.

Interesting! This is not something I had come across. Are you aware of any good papers to read? I found this design, which seems interesting. It appears to be designed for extremely low supply voltages. Not sure about the trade-offs compared to more "conventional" designs, such as these I found in some other paper. I'm currently designing with sky130_fd_pr__nfet_g5v0d10v5 so low supply voltage is not one of my requirements.

That’s one I saw too. If you give me a rough idea of the specs you are trying to meet I can dig into the literature. Low power and low voltage has been the recent driver for the obvious reasons (IoT) but there are probably lots of others. Do you need it to be differential? I know a way to do a VGA with inverters that might be interesting.

So I'm using it for possible the oldest use of VGAs: a scope frontend. So I'm sortof aiming for 500MHz bandwidth, 50 Ohm, 5Vpp input into the offset stage, which is currently a simple single-ended opamp design. Then I'm looking for a VGA design that will output into the differential ADC with a range of like a volt.

So the VGA is to be digitally controlled?

Well, yea there will be a DAC somewhere I guess, but not necessarily on-chip. In theory I guess it'd be possible to just switch a resistor bank. The designer of the scope said he'd like fine-grained gain control though, so a resistor bank is my backup plan, if linearity and range of the VGA isn't good enough.

In CMOS, I'd suggest a switched attenuator instead of a VGA whenever possible. I have done them in 0.35um for 390MHz IF (with 1dB steps) and also 0.18um at 2GHz, with 6dB steps. You can use R-2R ladder, or a R:75R ladder (really R/5:15R or R/15:5R made of unit resistors). You could drive a 50 Ohm attenuator ladder directly from the input port, and use NMOS swqitches (maybe a T-arrangement of 3 switches if you need high isolation, I modified the T-structure to have 8 inputs and one output, to reduce the number of fets connected to the output node) to select a tap to feed to a high-impedance input of the next stage. Otherwise, if you have a signal current available (e.g. from the drain of an amplifier device), you can switch the signal current into different taps on the ladder for different attenuation, and the end of the resistive ladder gives you a resistive output impedance that is about constant. If you are just driving a high-impedance stage from it then you can put a bunch of switched resistors across the termination of the ladder, to do fine gain steps. I think I did 1/64dB steps on one baseband attenuator like that. In one of the Jim Williams books I recall that there is a design for a scope frontend intended for IC manufacture, that somehow keeps the high voltages off the chip IIRC, I should look at that again one day soon. BTW, I am thinking of doing a sampling scope, need a trigger comparator, adjustable delay and sample-and-hold, probably using an off-chip ADC due to not having time do design one. Should be possible to achieve many-GHz bandwidth on this process.

Yeah, this process has good resistors and characterized parasitic capacitance. It may be tricky sorting out the impedance of the fancy pad structures but I suppose that is well characterized too.

Here is the book chapter that I mentioned on oscilloscope front-ends (only really of interest for a high impedance input, probably little help if only 50 Ohm input is wanted): Steve Roach chapter @Adrian Freed I think an oscilloscope frontend is one of the cases where it might be necessary to ask for a straight-through pad so that one could provide one's own ESD structures etc., optimized for analog performance not limited by having to support logic IO functions too.

Here is a 180nm VGA http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.391.7858&rep=rep1&type=pdf

@Chris Jones wait, a R:75R ladder?? Why? Can't find anything other than R:2R ladders on the web.

@Pepijn de Vos Yes, R-75R (but to use fewer unit resistors, better put some in parallel so it is a (R/15) : 5R, or (R/5) : 15R ladder). I don't know if anyone else ever used them, but it gives 1.0024dB steps, occasionally useful if you want finer steps than 6.02dB that you get from R:2R. One needs to do some calculations to figure out the terminating resistor at the end.