Hi,

Currently I am trying to simulate MOSFET operates in triode/linear region. When the Vds is less than Vgs, it suppose to operate in linear region as shown in IV Curve.png. But from the simulation using Multisim, I am getting quite large resistance as shown in Multimeter - XMM4 in the attached figure. Anything wrong with my simulation or the expected result is correct? Please comment, thanks.

 

Hi,

Currently I am trying to simulate MOSFET operates in triode/linear region. When the Vds is less than Vgs, it suppose to operate in linear region as shown in IV Curve.png. But from the simulation using Multisim, I am getting quite large resistance as shown in Multimeter - XMM4 in the attached figure. Anything wrong with my simulation or the expected result is correct? Please comment, thanks.

I am not certain that the way you have connected the meters in your simulation is valid. This will depend on the rules of your software, but certainly in the real world an ohm-meter could not be expected to give a meaningful reading when connected to a powered circuit in this way. In a real circuit, the ohmmeter could affect the supply voltage, and it might be damaged.

That said, in your model the FET seems to be passing only about half a milliamp for 10V Vgs and nearly 3V Vds. The FET seems barely turned on. What device are you using?

 

If we could not measure when the supply voltage is turned on, then how are we going to characterize the resistance of MOSFET for triode region operation? Is it just base on calculations? Any other ways of connection or method to do real measurement?

The FET is virtual model in the simulator.

 

I don't know about your virtual ohmmeter, but with a real one, you would get rid of V1 and R1.
You can keep XMM3, if you want to see how much voltage the ohmmeter is applying, and XMM1, to see how much current is flowing, but you need to put the XMM4 red probe on the other side of XMM1. This wouldn't work with a real ammeter, but your virtual one should have zero resistance. A real one has finite series resistance.

 

I don't know about your virtual ohmmeter, but with a real one, you would get rid of V1 and R1.
You can keep XMM3, if you want to see how much voltage the ohmmeter is applying, and XMM1, to see how much current is flowing, but you need to put the XMM4 red probe on the other side of XMM1. This wouldn't work with a real ammeter, but your virtual one should have zero resistance. A real one has finite series resistance.

Hi Ron H,

I have put the XMM4 red probe on the other side of XMM1 as shown in attach, is that what you mean? I am still getting the same amount of current and measured resistance. Can I directly measure the resistance in that direct manner by just probing D-S terminals?

 

Try a look at a data sheet. This links will get one for an IRF3703 - https://ec.irf.com/v6/en/US/adirect/ir?cmd=catSearchFrame&domSendTo=byID&domProductQueryName=IRF3703

While not absolutely ordinary, the FET is reasonably typical. The FET has an Rds (on) of 2.8 milliohoms. Apply Ohm's law and see what your current might be. How can you have a nearly 3 volt drop across the FET when the fixed resistor is 100 ohms?

Even an old IRF510 has an Rds (on) of .54 ohms

 

Hi Ron H,

I have put the XMM4 red probe on the other side of XMM1 as shown in attach, is that what you mean? I am still getting the same amount of current and measured resistance. Can I directly measure the resistance in that direct manner by just probing D-S terminals?

You didn't read my post very carefully. I also said you had to get rid of V1 and R1. Take them completely out of the circuit. You will treat the MOSFET as a voltage-controlled resistor, with the gate voltage as the control.

 

Try a look at a data sheet. This links will get one for an IRF3703 - https://ec.irf.com/v6/en/US/adirect/ir?cmd=catSearchFrame&domSendTo=byID&domProductQueryName=IRF3703

While not absolutely ordinary, the FET is reasonably typical. The FET has an Rds (on) of 2.8 milliohoms. Apply Ohm's law and see what your current might be. How can you have a nearly 3 volt drop across the FET when the fixed resistor is 100 ohms?

Even an old IRF510 has an Rds (on) of .54 ohms

He said that the FET is a virtual model in the simulator. MOSFETs can be tiny, as well as large.

 

To get a better idea of what is going on, why not try different values of Vgs, or even better do a do a DC sweep of Rds versus Vg? This should give you a clearer picture, for instance whether your FET has a large threshold voltage or perhaps a very high Rds(on).

It remains to be seen however whether your FET model is representative of anything you might use in practice. This "virtual" model may be for a really tiny FET which would normally exist only as an integrated circuit element. If you want to model something that you could use for an experiment, you may need something more specific.

 

You didn't read my post very carefully. I also said you had to get rid of V1 and R1. Take them completely out of the circuit. You will treat the MOSFET as a voltage-controlled resistor, with the gate voltage as the control.

Sorry I have missed that out. The attached Simulation_circuit_3.jpg still gives the same results. Perhaps there are some FET modeling issues with the Multisim. Whenever I tried to use a company's model for Eg IRF244, there's netlist generation error as shown in Simulation_circuit_4.jpg.

I tried to use ADS to simulate the I-V characteristics and successfully shows that the current is quite high and resulting a low resistance between VDS, shown in figure ADS_FET_Simulation. The turned on in triode region is pretty low, less than 4Ω!!! So if I would want to design a resistance, say 1kΩ using NMOS, how could I do that? In IC design, they can design a large value of resistance using MOSFET, any idea how's the architecture?

 

Did you simply type in IRF244, or did you select a MOSFET from the existing library?
If the device doesn't exist in the library, you'll have to obtain it from somewhere and add it to the library, add the model or subcircuit to the schematic, or use the .include libraryname.ext Spice statement.

 

Sorry I have missed that out. The attached Simulation_circuit_3.jpg still gives the same results. Perhaps there are some FET modeling issues with the Multisim. Whenever I tried to use a company's model for Eg IRF244, there's netlist generation error as shown in Simulation_circuit_4.jpg.

I tried to use ADS to simulate the I-V characteristics and successfully shows that the current is quite high and resulting a low resistance between VDS, shown in figure ADS_FET_Simulation. The turned on in triode region is pretty low, less than 4Ω!!! So if I would want to design a resistance, say 1kΩ using NMOS, how could I do that? In IC design, they can design a large value of resistance using MOSFET, any idea how's the architecture?

You have to also omit V1!



When you measure a resistor with an ohmmeter, the ohmmeter must be the sole source of current. Otherwise, you will get an erroneous measurement.

Device resistance is a function of the device geometry. Resistance is proportional to channel length, and inversely proportional to channel width.

 

Did you simply type in IRF244, or did you select a MOSFET from the existing library?
If the device doesn't exist in the library, you'll have to obtain it from somewhere and add it to the library, add the model or subcircuit to the schematic, or use the .include libraryname.ext Spice statement.

I select the MOSFET from the existing library as shown in the attach. I have checked the library folder, those components in the list are not there. The folder only has one DEFAULT.LIB file, that's why it still gives the simulation error as shown in the figure previously. Where can I get those models? From vendor's website?

 

You have to also omit V1!



When you measure a resistor with an ohmmeter, the ohmmeter must be the sole source of current. Otherwise, you will get an erroneous measurement.

Device resistance is a function of the device geometry. Resistance is proportional to channel length, and inversely proportional to channel width.

Yeah, you are right, I should eliminate the VDS voltage source. I did not know the reason of removing V1 and did not notice that it will cause erroneous in the measurement, thanks for the reminder and hint . Now I got more realistic results from the simulation as shown in attach. Few questions came to my mind from this simulation:

1) From the figure, we can see that the a 0.5V bias voltage difference will cause quite large variation in the resistance (in this case is 263Ω). Besides reducing the step size of the bias voltage, is there any way we can increase the tuning sensitivity, say 100Ω per 0.5V step size? I have tried connecting a resistor series with the VGS voltage source, but it doesn't make a difference.

2) By varying the VGS and without VDS, how is the MOSFET to operate in triode region? From the I-V characteristics, aren't suppose we need some small amount of VDS to fix the load line?

3) How to determine the resistance range for different devices from the datasheets? Which plot we can analyze to get the resistance value before we purchase a particular MOSFET? We can use the IRF3703 datasheet (link is given by beenthere in previous post) for this example.

Sorry for asking so many questions, they just appeared in my mind .Appreciate if anyone of you could share your expertise and experience.

 

Yeah, you are right, I should eliminate the VDS voltage source. I did not know the reason of removing V1 and did not notice that it will cause erroneous in the measurement, thanks for the reminder and hint . Now I got more realistic results from the simulation as shown in attach. Few questions came to my mind from this simulation:

1) From the figure, we can see that the a 0.5V bias voltage difference will cause quite large variation in the resistance (in this case is 263Ω). Besides reducing the step size of the bias voltage, is there any way we can increase the tuning sensitivity, say 100Ω per 0.5V step size? I have tried connecting a resistor series with the VGS voltage source, but it doesn't make a difference.

What are you really trying to do? Is this just a lab experiment, or are you trying to build working hardware?
Your model should allow you to change device parameters, such as L and W (channel length and width). In LTspice, if I place an NMOS4 symbol on my schematic, then right-click on it, a window opens which allows me to edit the geometry of the device (see attachment). However, if you know nothing about MOSFET design, this may not be useful to you.

2) By varying the VGS and without VDS, how is the MOSFET to operate in triode region? From the I-V characteristics, aren't suppose we need some small amount of VDS to fix the load line?

A real multimeter applies a small drain current (and voltage, by Ohms law).

3) How to determine the resistance range for different devices from the datasheets? Which plot we can analyze to get the resistance value before we purchase a particular MOSFET? We can use the IRF3703 datasheet (link is given by beenthere in previous post) for this example.

As you say, go to the datasheet. Rds is specified, but generally only at one or two values of Vgs.

Sorry for asking so many questions, they just appeared in my mind .Appreciate if anyone of you could share your expertise and experience.

 

What are you really trying to do? Is this just a lab experiment, or are you trying to build working hardware?
Your model should allow you to change device parameters, such as L and W (channel length and width). In LTspice, if I place an NMOS4 symbol on my schematic, then right-click on it, a window opens which allows me to edit the geometry of the device (see attachment). However, if you know nothing about MOSFET design, this may not be useful to you.

I am trying to simulate and build in a filter application after knowing the resistance characteristics. Ya, I do know about MOSFET design, learned in university. Well, if I would to purchase the MOSFET, the geometry is already fixed. What I mean is there might be external circuitry that I can implement in order to increase the tuning sensitivity. Anyway, it's just a thought .

A real multimeter applies a small drain current (and voltage, by Ohms law).

In this case if I would to use it for filter application, the resistance would change accordingly to the current? It would not perform like a fixed passive resistor?

As you say, go to the datasheet. Rds is specified, but generally only at one or two values of Vgs.

I only see Fig 4 (IRF3703 datasheet) shows the Rds(on), but it is normalized and vs temperature. How can I denormalized it, or am I referring to the wrong figure?

 

Does the circuit REQUIRE that you use a MOSFET?

Other silicon devices can operate as voltage controlled resistances.

 

You change the resistance by changing Vgs. For low resistances, the resistance would not be a function of current, except to for the nonlinearity of the FET (there will always be some nonlinearity, even in the "linear" region).
You can use a FET as a VCR, but each unit will require a different Vgs to get a given value of resistance, so it would not be suitable for production.
Some people use JFETs in this application, but they are generally in a feedback loop, where the resistance value is servoed to accomplish a function such as automatic level vontrol.
JFETs are small-geometry devices, so their resistance is higher, which I think might be better for your application. If you use a switching MOSFET such as IRF3703, I think your resistance will be extremely nonlinear if you try to attain resistances on the order of thousands of ohms. If you are looking for milliohms, you should be OK with the power MOSFETs.

 

Does the circuit REQUIRE that you use a MOSFET?

Other silicon devices can operate as voltage controlled resistances.

My circuit does not require MOSFET, any approach to function as a voltage controlled resistor is fine for me. Any good suggestions besides JFET as suggested by Ron H that we are not aware of?

 

My circuit does not require MOSFET, any approach to function as a voltage controlled resistor is fine for me. Any good suggestions besides JFET as suggested by Ron H that we are not aware of?

If you Google "voltage-controlled resistor", you will find information about JFETs and other techniques.
VCR2N4N7N is a JFET which is specifically designed to be used as a VCR.

Have you searched for "voltage controlled filter"? You might find something useful.
Here is an example:
http://www.musicfromouterspace.com/analogsynth/AAA_VCFilterIndex.php

Don't stop there. Be creative in your searches.