Hi,

We are doing a test on a transistor using PNA-L N5230A and Precision Semiconductor Parameter Analyzer 4156C.

My PNA's port 1 and port 2 are connected to 2 probes which connected to the DUT.

By looking at the S11 and S22, we can identify the reflection of each port and calculate the impedance.

At the same time, we are using the 4156C to measure the IV curve of the DUT. Can we actually compare the resistance (from the IV curve measured by 4156C and the resistance (the real portion of the impedance calculated from S11 at low frequency)? Or else, is there anyway we can compare the resistance measured by both equipments?

Really hope that someone can help in this matter.

Thank you.

We are doing a test on a transistor using PNA-L N5230A and Precision Semiconductor Parameter Analyzer 4156C.

My PNA's port 1 and port 2 are connected to 2 probes which connected to the DUT.

By looking at the S11 and S22, we can identify the reflection of each port and calculate the impedance.

At the same time, we are using the 4156C to measure the IV curve of the DUT. Can we actually compare the resistance (from the IV curve measured by 4156C and the resistance (the real portion of the impedance calculated from S11 at low frequency)? Or else, is there anyway we can compare the resistance measured by both equipments?

Really hope that someone can help in this matter.

Thank you.

Thanks for your reply.

The connection of the setup is through a bias tee externally.

I am doing a test on a transistor. When I am trying to compare the resistance of both terminal using the IV curve (the slope) and the resistance calculated from S parameters, there are differences.

May I know what will be the possible problems that create these differences?

Thank you.

S-parameters can only do a reasonable job for impedance ranges of about 10 ohms to 250 ohms (in a pinch you can extend them more with care), so what is the resistance that you are trying to measure?

Thanks for your reply.

For the PNA, I have calibrated out everything up to the probe. So i believed the S-parameter shows only the transistor.

In one of our test, the calculated resistance from the S parameter is about 80 ohm but the resistance calculated from the IV curve's slope is less than 10 ohm.

Thats why I am wondering which part of the setup or method goes wrong that caused this difference.

Thank you.

We are doing the resistance calculation at 1MHz. I have attached an example of our s2p file below for your reference.

S2P Example.s2p

## Attachments

A bit of a strange transistor...0 dB gain, -3 db S12, very near 50 ohm S22, and 12 dB rl on s11. Is this a common emitter construction?

## Attachments

Thanks again for your informative reply.

From our calculation, the resistance we got from S11 also about 80ohm.

The resistance that we have calculated based on the slope of the IV curve from 4156C is about 10-20ohm.

If that is the case and I have verified my calibration, what is the possible reasons of this difference of resistance?

Thank you.

Insufficient information. Can you post your IV data, and also can you post a diagram of your measurement (a transistor is a 3 port device and you have posted a 2 port meaurements; from that I infer tht you have grounded one of the terminals, but which one; and what is the input and output ports?). Please include the informatino on the bias point (DC bias voltage and current) that is used on each port for the S-parameter measurement.

My best guess here is that the IV data that you have taken is not for the same operating point as the S-parameters that you have taken. You must of course do your IV computation at the same bias point as the S-parameters are measured. That means if you are looking at the IV curve at the input, you must bias the output at the same level as the S-parameter measurement does, then sweep IF and look at the IV values for the input around the same point as the S-parameters are made.