0
\$\begingroup\$

What are the most important characteristics to look for in a MOSFET for switch a precision 50nA load?

The MOSFET should be able to switch the load on or off within 1us. The overshoot/undershoot while switching should be minimised.

Edit:

Here's a simplified schematic of the circuit I'm describing.

circuit

The load is at 3.3V and the 66M Ohm resistor gives very close to 50nA current when M1 is switched on. This is affected by the Rds_on of the MOSFET but with a 66M Ohm load resistor it's going to be negligible for pretty much any reasonable Rds_on.

Overshoot/undershoot is fine as long as the voltage at the load (non-grounded terminal of R1) does not go above 3.3V or below 0V.

Here's an example simulation:

Sim Circuit

The transistor is a BSS84, the En pin is fed the following waveform:

Enable pin signal

The rise/fall time of this signal is 10ns.

Here is the current through R1:

Current through R1

The turn-on response is quite fast, the rise time from 0nA to 50nA is 14ns. There is some undershoot of 3.7nA. The turn-off response is terrible, the fall time is 18ms from 50nA to 50pA. There is also 6nA of overshoot.

During turn-off, the current/voltage drops very quickly until the load voltage is 2.1V then it decreases slowly like a capacitor is discharging.

I want the load current to drop to 0A within 1us ideally. I presume this could be accomplished by finding some way to drain the drain-source capacitance of M1 quickly once M1 has been shut off. I tried using another MOSFET as per the circuit below but it didn't achieve what I wanted.

Alternate Circuit

Edit 2:

I replaced the P-channel M2 with an N-channel and it seems to work well:

N-chan

The current through R1: I(R1)

I chose an M2 with very low Rds_on. The current through R1 now drops to ~0A in 12.5ns.

Why was everyone saying this would be so difficult? Am i missing anything here?

\$\endgroup\$
  • 4
    \$\begingroup\$ Supply voltages to mosfet please. What are you driving it with? What precisely is the load? Have you found a mosfet with low leakage? What leakage would be regarded as good? What would the overshoot/undershoot come from - is your load reactive? Let's NOT make this a guessing game dude. \$\endgroup\$ – Andy aka Apr 20 '14 at 13:16
  • 2
    \$\begingroup\$ At 50nA, 0.05pF of capacitance will only allow 1V change in 1usec, this is not going to be easy in the general case. It might be better to describe what you're trying to do. \$\endgroup\$ – Spehro Pefhany Apr 20 '14 at 13:42
  • \$\begingroup\$ Most of the characteristics are unspecified in the question, such as source and load impedance, capacitance , rise time, voltage , path length. Or... The geometry of the current path. \$\endgroup\$ – user38637 Apr 20 '14 at 14:30
  • \$\begingroup\$ I don't think it is possible to use a mosfet to switch a 50nA load. The mosfet leakage with Vgs=0 is in the micro Ampere (uA) range so you can't really turn off the load. \$\endgroup\$ – alexan_e Apr 20 '14 at 16:32
  • 1
    \$\begingroup\$ Too broad. Please narrow down what you want to achieve, per these comments, by including the relevant requested information in your original question. \$\endgroup\$ – Adam Lawrence Apr 20 '14 at 16:38
3
\$\begingroup\$

"Why was everyone saying this would be so difficult? Am i missing anything here?"

Does it bother you at all that the BSS84 used in your model has a leakage spec that is about 2X the 50nA you are trying to switch?

Single MOSFETs are designed for low Rds for die size, not so much for switching extremely low currents (although tcrosley found one that will almost do what you want). To switch low currents like this, one would use a low leakage analog switch. For example something like DG2012 might work.

Remember that all models are wrong, and it is up to you to figure out if it is right enough to be useful.

\$\endgroup\$
2
\$\begingroup\$

The SD5000 quad N-channel MOSFETs have extremely low drain-source leakage -- typically under 1 nA (although the maximum is 10 nA). Typical Vgs is under a volt. Turn on and off times are typically under 1 ns. It is rated to handle ±10v signals.

They are available in a DIP package at Future Electronics.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.