I need to include overcurrent protection in my circuit. Previously, I used current sense amplifiers to achieve that, but the this application needs current sensing in the microamp region and most of ics I used before did not have a appropriate resolution. I need to limit the load to 150 microamps. I was thinking on using something like MAX4373 but its leakage current is too big. The Vrs+ in my case is 0.08-1.2V and I don't mind having bigger voltage drop at Rsense since the load is a thermistor.

Any suggestions?

 

Any schematic to hand, why do you need current limit in a thermistor?

 

Would a constant current source be a better approach?

 

The thermistor cannot handle more than 0.2 mA (200 micro) and this component cannot be changed.

Currently, I am using a psoc and its internal 8-bit current source which can be set to 100 micro. However, the firmware can be accessed by anyone so if someone makes changes by mistake, the element is no longer protected.

Hence, I needed a hardware solution (similar to View attachment 70446 ). I could not find something accurate enough for micro amp range.

 

your link dont work, if you post a circuit then we could see what to do.

 

Load is the thermistor. Reset is done with psoc. Basically, I need a way to protect the thermistor that can only handle 200 micro amps with hardware maybe similar to current sense amplifier solution as attached.

Another solution I was looking into is to insert a current regulative diode that works similar to clamping diodes by regulating the voltage drop.

 

What is the thermistor doing, is it measuring temp change, if so where is the rest of the circuit?

 

Yes, it is for temperature measurement.

As for the circuit:

I have a current source (inside the Cypress PSOC 5 LP chip) going to the thermistor (no part number). I am feeding the voltage drop across the thermistor to the ADC of the chip. That is it. The circuit I included is only an example of what I was thinking to do to protect the thermistor. The reason I needed a additional protection is that the firmware controlling the current source can be easily changed by mistake.

 

ok so why not use a constant current source using an op amp.

 

How about this?

Edit: The second circuit enables a ground-referenced thermistor to be used.

 

Thanks Alec.