AnsweredAssumed Answered

Noise power with 029 option

Question asked by noise029&028 on Apr 1, 2019
Latest reply on Apr 26, 2019 by noise029&028

Like • Show 0 Likes0
Comment • 2

Hello everyone, I use the noise option '029' to make noise power measurements for a measurement chain composed of an LNA and an impedance tuner just at the input of the LNA. I am trying to determine the output noise power (from the chain only) for each tuner position. Basically i have two questions to ask

1- What is the appropriate parameter that allows to do this measurement because there are two parameters that are "DUTNPDI and DUTNPD" knowing that my noise calibration is done at the end of cable of port 2 with a noise source and also at the level of reference plan of DUT?

2- can I say that these two parameters are the total noise power available at the output and which is the power of noise available at the output due to the sources (KTB Ga) + the internal noise power of the chain ?

Ga: available gain of the measurement chain

T: ambient temperature

K : Boltzmann constant

B: bandwidth (1Hz)

Thank you so much

Cordially

Dr_joel

Apr 12, 2019 5:32 PM

Correct Answer

There are two noise power measures: Available noise power density (DUTNPD, power to a conjugately matched load) and Incident Noise power (DUTNPDI, power delivered to the a non-radiating non reflective load, e.g. 50 ohm). Normally you want to use DUTRNPI. But in some cases you might want available noise power, for example, to get the equivalent temperature of the noise. A resistor at a temperature T, will have available noise power of kTB, but will deliver Incident noise power to a 50 ohm load of the value kTB*(1-|GammaLoad|l^2). For a 50 ohm resistor, gamma=0 and DUTNPDI=DUTNPD. We measure S22 of the DUT to determine gamma of the load to make the computation (and for some other stuff too).

Sorry, I'm a little confused by the question.

can I say that these two parameters are the total noise power available (only DUTNPD is available, DUTNPDI is incident from the DUT). They both measure the total noise power received by receiver, including kTB*G plus and added noise. They exclude the noise generated in the PNA receiver (that's the difference between DUT and SYSNPD). There is no distinction from the input generated noise, as it is a total noise power measurement.

In noise figure, we do account for the incident noise caused by the input termination, and this portion of the noise (times the gain) is in-essence subtracted from the total noise to determine the Te of the DUT. But only done as a NF measurement and not as a Noise Power measurement.

See the reply in context

No one else had this question

Outcomes

Dr_joel

Apr 12, 2019 5:32 PM

Sorry, I'm a little confused by the question.

1 person found this helpful

Actions

noise029&028 @ Dr_joel on Apr 26, 2019 3:23 AM

Hello, thank you for all your answers, which are very clear and precise.

I would like to ask you four more questions, if that's possible, please.

The option (029) of the PNA-X uses the cold method for noise measurements , so the ncident noise power at the output of the DUT is written as follows normally:

DUTNPDI = K B Ga (Tcold + Te)

With Te: equivalent noise temperature of the DUT

Question 1:

Ga: is the available gain of the DUT or DUT + receiver?

In general, the total incident noise power (of the system) measured on the noise receiver is:

SYSNPDI = K B G * (Tcold + Te) * G recei + N recei

With G is the gain of the DUT, G recei is the gain of the noise receiver, N receiver is the power of noise generated by the noise receiver. While the DUTNPDI exclude the power generated by the noise receiver (N recei = 0).

Therefore DUTNPDI = K B G * (Tcold + Te) * G recei ?

Question 2:

In the help of this option, I did not understand how the Ga depends on S11, S22, Gamma S (the figure below), while the source is adapted to 50 ohms (RF source of the PNA-x adapted to 50 ohms) , so the Ga must depend only on S21!

Question 3:

In which case , we can estimate that T cold = T0 = 290°k, and how to estimate the value of Tcold (knowing that my measurements are made in an environment of 23°C -> 296.15°K )? . Is it necessary to configure the ambient temperature parameter during the noise calibration phase at the temperature of 269.15 ?

Question 4:

In my noise power measurements, I use a scalar noise measurement ( Internal E-CAL Module is disabled) in the calibration phase, it is assumed that the RF source provides a perfect combination of 50 ohms, and counterpart I use an external impedance tuner to measure the noise parameters of the DUT in the noise measurement phase. So do I have the right to use the tuner just in the measurement phase, or do I have to use it in the calibration procedure (calibrate the noise receiver to the same impedances as I use them for noise power measurements at the output of the DUT) ?

Because I use the tuner before the DUT to change the reflection coefficient at the input of the DUT, in order to measure the noise power at the output of the DUT for each impedance, for this , I performed a scalar noise receiver calibration (no tuner in the measurement bench).

Very Best regards.

Actions

2 Replies

Dr_joel Apr 12, 2019 5:32 PM
Unmark Correct
Correct Answer
1- What is the appropriate parameter that allows to do this measurement because there are two parameters that are "DUTNPDI and DUTNPD" knowing that my noise calibration is done at the end of cable of port 2 with a noise source and also at the level of reference plan of DUT?
There are two noise power measures: Available noise power density (DUTNPD, power to a conjugately matched load) and Incident Noise power (DUTNPDI, power delivered to the a non-radiating non reflective load, e.g. 50 ohm). Normally you want to use DUTRNPI. But in some cases you might want available noise power, for example, to get the equivalent temperature of the noise. A resistor at a temperature T, will have available noise power of kTB, but will deliver Incident noise power to a 50 ohm load of the value kTB*(1-|GammaLoad|l^2). For a 50 ohm resistor, gamma=0 and DUTNPDI=DUTNPD. We measure S22 of the DUT to determine gamma of the load to make the computation (and for some other stuff too).

2- can I say that these two parameters are the total noise power available at the output and which is the power of noise available at the output due to the sources (KTB Ga) + the internal noise power of the chain ?
Sorry, I'm a little confused by the question.
can I say that these two parameters are the total noise power available (only DUTNPD is available, DUTNPDI is incident from the DUT). They both measure the total noise power received by receiver, including kTB*G plus and added noise. They exclude the noise generated in the PNA receiver (that's the difference between DUT and SYSNPD). There is no distinction from the input generated noise, as it is a total noise power measurement.

In noise figure, we do account for the incident noise caused by the input termination, and this portion of the noise (times the gain) is in-essence subtracted from the total noise to determine the Te of the DUT. But only done as a NF measurement and not as a Noise Power measurement.
1 person found this helpful
Like • Show 1 Like1
Actions
- noise029&028 @ Dr_joel on Apr 26, 2019 3:23 AM
  Mark Correct
  Correct Answer
  Hello, thank you for all your answers, which are very clear and precise.
  I would like to ask you four more questions, if that's possible, please.
  
  The option (029) of the PNA-X uses the cold method for noise measurements , so the ncident noise power at the output of the DUT is written as follows normally:
  DUTNPDI = K B Ga (Tcold + Te)
  With Te: equivalent noise temperature of the DUT
  
  Question 1:
  Ga: is the available gain of the DUT or DUT + receiver?
  In general, the total incident noise power (of the system) measured on the noise receiver is:
  SYSNPDI = K B G * (Tcold + Te) * G recei + N recei
  With G is the gain of the DUT, G recei is the gain of the noise receiver, N receiver is the power of noise generated by the noise receiver. While the DUTNPDI exclude the power generated by the noise receiver (N recei = 0).
  Therefore DUTNPDI = K B G * (Tcold + Te) * G recei ?
  
  Question 2:
  In the help of this option, I did not understand how the Ga depends on S11, S22, Gamma S (the figure below), while the source is adapted to 50 ohms (RF source of the PNA-x adapted to 50 ohms) , so the Ga must depend only on S21!
  
  Question 3:
  In which case , we can estimate that T cold = T0 = 290°k, and how to estimate the value of Tcold (knowing that my measurements are made in an environment of 23°C -> 296.15°K )? . Is it necessary to configure the ambient temperature parameter during the noise calibration phase at the temperature of 269.15 ?
  
  Question 4:
  In my noise power measurements, I use a scalar noise measurement ( Internal E-CAL Module is disabled) in the calibration phase, it is assumed that the RF source provides a perfect combination of 50 ohms, and counterpart I use an external impedance tuner to measure the noise parameters of the DUT in the noise measurement phase. So do I have the right to use the tuner just in the measurement phase, or do I have to use it in the calibration procedure (calibrate the noise receiver to the same impedances as I use them for noise power measurements at the output of the DUT) ?
  Because I use the tuner before the DUT to change the reflection coefficient at the input of the DUT, in order to measure the noise power at the output of the DUT for each impedance, for this , I performed a scalar noise receiver calibration (no tuner in the measurement bench).
  
  Very Best regards.
  Like • Show 0 Likes0
  Actions

Noise power with 029 option

Outcomes

Related Content

Recommended Content