Noise figure measurement - Noise generator vs ENR source?

Hi,

I am trying to determine if it's possible or not to do a NF measurement using a Noisecom UFX7110 noise generator.

I am familiar with doing NF tests using an ENR noise source (Y-Factor method), but unfortunately I do not have a ENR source available, but I do have this noisecom generator.

Looking at the block diagram of the UFX110A (see attached jpg)
it appears that all the noise generator is, is a noise source with a switch and attenuator in series with it's output. So this should be suitable for a NF measurement. Only problem is there is no ENR table for the noise source.

So I'll probably need to measure the excess noise ratio (ENR) values of the noise generator at different frequencies, but how would one do this accurately using a spectrum analyser?

If anyone has done something similar or has any advice it would be appreciated.

Mark

vfone said:

https://mwrf.com/test-amp-measurement-analyzers/measure-noise-without-calibrated-source

Click to expand...

Thanks Vfone

At first the method seems pretty good. But one of the comments below it suggest that the author has realized (and admitted) that the new method is in fact the same as the Y-Factor method? In other words you would still need a calibrated ENR source.

Any thoughts?

myRF said:

Thanks Vfone

At first the method seems pretty good. But one of the comments below it suggest that the author has realized (and admitted) that the new method is in fact the same as the Y-Factor method? In other words you would still need a calibrated ENR source.

Any thoughts?

Click to expand...

As far as I know, NoiseCom noise generators come with an ENR table showing ENR over a frequency range. With an attenuator inserted to the output, you have to include corrections of ENR as a function of loss and attenuator temperature.

Measuring ENR is possible with a spectrum analyzer but it must be well calibrated, best to a good termination at a known temperature.
Find Agilent document on noise measurement with all details.

- - - Updated - - -

Better than a spectrum analyzer is to use a good power meter which usually has a better accuracy and a better resolution of a power level.

Can you indicate what your task is? Frequency range, bandwidth, noise figure of a tested device...

jiripolivka said:

As far as I know, NoiseCom noise generators come with an ENR table showing ENR over a frequency range. With an attenuator inserted to the output, you have to include corrections of ENR as a function of loss and attenuator temperature.

Measuring ENR is possible with a spectrum analyzer but it must be well calibrated, best to a good termination at a known temperature.
Find Agilent document on noise measurement with all details.

- - - Updated - - -

Better than a spectrum analyzer is to use a good power meter which usually has a better accuracy and a better resolution of a power level.

Can you indicate what your task is? Frequency range, bandwidth, noise figure of a tested device...

Click to expand...

Thanks jiripolivka,

As far as I know there is no ENR table for this generator. I have looked in all the accompanying docs with no luck. I might try getting in touch with noisecom to see if they can help.

Unfortunately we don't have a power meter here. Basically the task is to measure the NF of a high gain receiver module (around 30dB gain), dc-2GHz. I have previously measured the NF of these and it is about 13dB.

myRF