Can MOSFETS be used as switch in the feedback network of opamp?

[Van Grays]

here goes an explanation of my project:

The existing circuit uses 2 op-amps The High gain range to measure current from 0.010 A to 1.0 A is 27 and The High gain range to measure current from 1.0 A to 5.0 A is 3.44. The op-amp has GBW of 10MHz with wide input range.

Now, my objective is to reduce the usage to only single op-amp mainly to reduce PCB footprint. The gain range to measure current should be same and It should correctly amplify frequency up to 4khz.

I was suggested to use MOSFETS in feedback path to trigger multiple gain stages. Is it reliable to use MOSFET as a switch to achieve this? If so, how can I stabilize my op-amp with a MOSFET in the feedback loop??

What might be the advantages and disadvantages of doing so?

Are MOSFETS reliable? If not, can you please suggest me some alternate methods

additional to this, someone suggested me to use this circuit
 But I am finding issues with the connections part, please guide me on building a proper circuit.

[andrewtaylor]

Depends on the accauracy you expect from the final circuit.
It's a tradeoff between the n-FET you choose, and the low resostance for the higher amplification range.

n-Channel FET with much less than 0.1 Ohm are available, so this will work well.
Make shure the FET is full conducting (== Vgs high enough for full conduct)
If possible, choose highr R than 845 to relax the situation.

[moffy]

Try to choose a low capacitance D-S, D-G MOSFET as that will be a parasitic when the device is off. When on it should be fine as the MOSFET will have minimal signal drop across it as it should be low value compared to the 845R resistor. There is one problem though, the body diode will start to conduct if the voltage gets too negative when the MOSFET is off, this limits the maximum input signal during its off state.

[mikeselectricstuff]

That will work fine for a single supply system, but for bipolar signals the body diode may be an issue.
Another approach, eliminating the issue of switch resistance,  would be to use a SPDT analogue switch, to select the -ve input between 2 taps on a resistor chain

[PCB.Wiz]

Are MOSFETS reliable? If not, can you please suggest me some alternate methods

Of course. You could look for ESD protected MOSFETs if you are worried.
Run this in spice, and check bandwidth.

As above, you need a low resistance FET to give low errors and low C fet to not give too much peaking when OFF.
A feedback C from output may be needed.
Rds and Cin are somewhat of a trade off, but you can sort in Digikey by both to find a suitable part.
A 800mOhm fet will be ~ 0.1% and a 8 ohm fet ~ 1% adder to your resistance.

There are also Analog Switch parts, that are made using MOSFETS, that give you other RDS and Coff choices.
If the voltage swing on the input pin is 0 < Vi < 5V, you have many to choose from.

[ajb]

If reducing board size is the primary goal, you can get single op amps in SOT363 packages, so a second op amp isn't much bigger than one op amp and transistor.  It still takes up a bit more space with routing the extra power lines and feedback path, but is also more flexible, for example if you wanted to use both ranges simultaneously in the crossover region, or have different filtering or something.

[sourcecharge]

That will work fine for a single supply system, but for bipolar signals the body diode may be an issue.
Another approach, eliminating the issue of switch resistance,  would be to use a SPDT analogue switch, to select the -ve input between 2 taps on a resistor chain

You could use back to back mosfets for the ac signal if you want.

[David Hess]

A MOSFET works fine to switch feedback network elements.  A JFET or bipolar transistor can also be used.

Things can get tricky if bipolar operation is required.

[Van Grays]

Will this input of mosfet or opamp?