Hello, i want to have a control circuit like this:

i want to control the voltage drop across the R, for Q1, what should i use? BJT or MOSFET which is better? if use MOSEFT, how should i chose a mosfet, it will work in it's linear region, but what i know is the linear region of MOSFET is very nerrow... what parameter should i pay attention to?i used mosfet before as a switch, in that case, i will choose a low Ron and low Qg/Cds, in linear mode, should i choose high Qg one so the linear range will be wider ?


thanks and regards

 

Without telling us the intended voltage, current, power, or frequency of the intend circuit, how can you expect an answer? Also, the linear region of a MOSFET is not very narrow.

ak

 

Do you want to adjust the voltage drop across R (voltage control) or the current through R (current control)?

Either way there are techniques to linearize the control, such as using an op amp in a feedback loop with a transistor (either MOSFET or BJT).
Without feedback it's difficult to obtain linear control in that circuit.

 

Hi, yes I want to control (adjust ) the current pass through R,2-5A, using feedback to opamp，would you please give me some example circuits？

Thank you very much for help

 

You can use a low Ron, but remember that in the linear region, you won't be there. Try putting a small sense resistor between the source and ground and use and op amp to compare it with a reference and drive the gate. Remember that you will be dissipating power in the MOSFET (Vds times current), which will show up as heat in the MOSFET or Transistor.

 

Here's a schematic that can help. The current through resistor R2 is simple V1/R1. Since 5 amps is a lot of current you must use a power transistor, like the good old 2n3055. However, that BJT has a low gain and, for that reason, a darlington configuration must be employed. The voltage source V1 can be any voltage source, for example a simple voltage divider. Hope it helped.

 

Hello, i want to have a control circuit like this:
View attachment 86866
i want to control the voltage drop across the R, for Q1, what should i use? BJT or MOSFET which is better? if use MOSEFT, how should i chose a mosfet, it will work in it's linear region, but what i know is the linear region of MOSFET is very nerrow... what parameter should i pay attention to?i used mosfet before as a switch, in that case, i will choose a low Ron and low Qg/Cds, in linear mode, should i choose high Qg one so the linear range will be wider ?


thanks and regards

Put a resistor to sense current in the source lead and use the voltage developed across it to bias a bipolar transistor in common emitter. The collector terminates the bottom of a voltage divider that provides gate bias for the MOSFET. You can bootstrap the gate bias with a high value resistor to maintain a good input resistance.

 

Hello All:
Thank you very much for all the suggestions, I will try first and let you know the result.

best regards

 

hi
I did a simulation using LTSpice, like this :


but when I try to change the C1 to 200nf, the simulator give a warning of :


anybody know why?

 

I didn't have your op amp and transistor, but it works ok with some I have.

 

Hi, Ronv:
thank you for your reply, I attached my opamp and transistor library in my last post, I upload them again, would you please help me try again? my problem happen when I change C1 to 200nf...

thank you for help.

 

Sorry Ronv:
or, would you please share with me your simulation circuit?

thanks

 

Put a resistor to sense current in the source lead and use the voltage developed across it to bias a bipolar transistor in common emitter. The collector terminates the bottom of a voltage divider that provides gate bias for the MOSFET. You can bootstrap the gate bias with a high value resistor to maintain a good input resistance.

Hi, Ian:
thank you very much for your reply,i am very interested on your idea, is it possible you can do a simple drawing for me?

 

Sorry Ronv:
or, would you please share with me your simulation circuit?

thanks

sorry Ronv:
i didn't see your circuit this morning
i had tried your circuit, yes, it works fine with changing C1, and the problem is the opamp: OPA2251 vs LM358.

Thanks and regards

 

Hi, Ian:
thank you very much for your reply,i am very interested on your idea, is it possible you can do a simple drawing for me?

I'm a little busy to look for a schematic app that will output "universal" files.

The idea is too simple to need a diagram - decide what mid-point drain current you want to flow and calculate a resistor to drop 0.7V at that current. That resistor now becomes the current sense between source and GND. Connect a common emitter NPN transistor, E to GND and B to source of the MOSFET, its best to put a resistor between base and source to soak up any spikes - about 15R is sufficient. An actual voltage divider probably isn't necessary to bias the gate - just connect at least 47k from collector to Vcc, current increase in the source resistor will now tend to bias on the bipolar transistor so the collector voltage falls - that is your nfb that stabilises the operating point. You need to string a high resistor between the bipolar collector and the MOSFET gate otherwise the collector will shunt the input.

So far with just the transistor and a few resistors, you have both AC & DC nfb so not much gain - you need an AC shunt (small electrolytic) on the nfb path. You could try from collector to GND, but that may not be fully effective, put a few k from the collector to the cap, then a high resistor from the cap to gate.

The transistor, "few k resistor" and cap develop your average bias to hold the MOSFET drain at its mid-point, the high resistor from cap to gate gives you a high resistance input.