# Gate Resistance (RG) Auto Impedance Method

Discussion in 'General Electronics Chat' started by Engr, Dec 22, 2010.

1. ### Engr Thread Starter Member

Mar 17, 2010
114
5
Hi everyone. I am searching for a way to measure the Gate Resistance of the MOSFET with a value that is less than 1ohm. I saw a method called the Auto-Impedance method but the methodology was not explained. Does someone know how to do the Auto-Impedance Method to measure the gate resistance (Rg) of a MOSFET?

Nov 29, 2010
199
16
3. ### Engr Thread Starter Member

Mar 17, 2010
114
5
Yes, it is correct. The MOSFET has a very high input impedance, but the high input impedance is between the Gate-to-Source junction of the MOSFET. If only the Gate Pin of the MOSFET is considered, it has an impedance (called Rg) that is very small, sometimes less than 1ohm. It can be seen on the datasheet of the MOSFET.

4. ### timrobbins Active Member

Aug 29, 2009
318
16
Engr - the series gate resistance is a very difficult measurement to make because it feeds into quite small capacitances. Do you really need to measure it ? That is usually the domain of device scientists.

You can have a go by a few methods. One relatively simple method is to assume the gate is a capacitance with ESR. You need a step source voltage with low effective series resistance, and connect a known series resistance between your step source and gate, and measure step voltage waveforms with different known series resistances. Measurement of rise time is probably in the ns range so the setup has to be pretty careful. Not for the faint hearted or inexperienced I suggest.

Ciao, Tim

5. ### Engr Thread Starter Member

Mar 17, 2010
114
5
Yes, we really need to measure it since it is one of the parameters in the datasheet of a MOSFET. The small capacitance you are talking about, I think it is called Ciss. The method you mentioned above is what we are doing right now but it has limitations in measuring the gate resistance that is why we are looking for other ways to measure it accurately.

6. ### tyblu Member

Nov 29, 2010
199
16
Interesting - what are you measuring it for? Is it only to verify the datasheet specifications, or is it for a specific circuit? There are ways to perform the above-mentioned measurement at all ports (gate, drain, source) to obtain a system of equations, in order to isolate the gate. Transconductance will skew the measurements if taken at the gate alone. 'Suppose you could repeat the measurements at the gate alone, but in the full range of modes, then use MOSFET modeling theory to isolate the gate variable. What are you doing now?

7. ### solis365 Member

Nov 5, 2008
21
0
by gate resistance do you mean the contact resistance of the package to the actual IC?

if you can get your hands on a semiconductor parameter analyzer i bet that will do it real quick and without much hassle.

but those are in the \$50,000 range, so unless your university has one...

8. ### timrobbins Active Member

Aug 29, 2009
318
16
Apart from reiterating tyblu's first few pertinent questions, what limitations were you experiencing with testing - that may assist us in appreciating how you are approaching the topic.

9. ### Engr Thread Starter Member

Mar 17, 2010
114
5
Yes, it is only to verify the datasheet specifications. We are making a test solution that can measure Rg less than 1ohm.

10. ### Engr Thread Starter Member

Mar 17, 2010
114
5
The limitations that we are experiencing right now is that when we measure an Rg that is below 1ohm, we cannot measure accurately the value of the Rg.....

11. ### timrobbins Active Member

Aug 29, 2009
318
16
Engr - can you summarise why you can't measure accurately?

12. ### Engr Thread Starter Member

Mar 17, 2010
114
5
If it is below 1ohm, the path resistance, stray capacitances and inductances will start affecting our measurements. That is the reason why we are looking for a way to measure an Rg that is less than 1ohm. A measurement that is not affected by the path resistance, stray capacitances and inductances....

13. ### timrobbins Active Member

Aug 29, 2009
318
16
I think you better to go back to square one, and rethink why you actually have to do a measurement, given that you seem to think your capabilities are limited in preparing a suitable measurement setup for what is one type of relatively simple test.

You said you wanted to measure the parameter because it was on the datasheet, and you were verifying datasheet parameters.

Are you verifying datasheet parameters for fun, or because you have a technical problem with that fet in a circuit, or ???

14. ### Engr Thread Starter Member

Mar 17, 2010
114
5
I am not verifying datasheet parameters for fun and it is also not because of a technical problem with a fet. I am a junior test engineer and my job is to develop test solution to make sure that the device parameters and functionality is within the specs specified on its datasheet. It just happened that the device assigned to me is the MOSFET and the Rg is one of the parameters needed to be tested for a MOSFET.

15. ### timrobbins Active Member

Aug 29, 2009
318
16
Engr, sorry to be a pain - nothing meant by the questions other than it's a surprise to find somebody trying to do what you are doing from my myopic perspective. What do you think you will try and do? Would I be second guessing that there are quite a few parameters on the fet datasheet where you have the same outcome - ie. that from the equipment and practices you know, you aren't in a position to be able to validate the specified level of that particular parameter.

Ciao, Tim

16. ### Engr Thread Starter Member

Mar 17, 2010
114
5
Yes, its correct. Another parameter where we are also having a problem on accuracy is on the measurement of the Ciss parameter of the MOSFET. The Rg parameter is measured using the Ciss parameter but we are also having problems measuring the Ciss parameter when it is below 1nF.