The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Measurements >> RF Measurements >> Noise figure definitions
https://designers-guide.org/forum/YaBB.pl?num=1283769005

Message started by tromeros on Sep 6^th, 2010, 3:30am

Title: Noise figure definitions
Post by tromeros on Sep 6^th, 2010, 3:30am

Hi everybody,
Lately I designed and measured a direct conversion downconverter comprising of LNA (single ended input and output), mixer (single ended input – differential output) and a differential to single converter for the measurement.
Since the architecture is direct conversion the overall noise is selected in the spectre simulation to be NFdsb. According to textbooks and the simulation comparing single and double sideband NFs, the following expression occurs:
NFssb = NFdsb + 3

However during the measurement the agilent spectrum analyzer that I used has the ssb and dsb options but they have the opposite dependence. According to the textbook of the Y-factor method (http://cp.literature.agilent.com/litweb/pdf/5952-3706E.pdf) page 23:
“In the simplest case, the SSB noise figure will be a factor of 2 (3.0dB) lower than the DSB measurement.”

This fact causes confusion to me. Can anyone share the reason why the textbook definition is opposite from the instrument feature?
Which option should I take to compare simulation with measurement?
Thanks in advance for the help.

Title: Re: Noise figure definitions
Post by pancho_hideboo on Sep 6^th, 2010, 6:57am

tromeros wrote on Sep 6^th, 2010, 3:30am:

According to textbooks and the simulation comparing single and double sideband NFs, the following expression occurs:
NFssb = NFdsb + 3

There are many cases where this relation is not satisfied in actual measurement.

Measurement using actual instruments give NFasb instead of NFdsb.
http://www.designers-guide.org/Forum/YaBB.pl?num=1162399761
http://www.designers-guide.org/Forum/YaBB.pl?num=1239925339/4#4

However your confusion is not related to this.
Your confusion is no more than very simple, easy and basic issue about output noise quantity.

tromeros wrote on Sep 6^th, 2010, 3:30am:

(http://cp.literature.agilent.com/litweb/pdf/5952-3706E.pdf) page 23:
“In the simplest case, the SSB noise figure will be a factor of 2 (3.0dB) lower than the DSB measurement.”

This does not mean NFdsb.
This means DSB measurement. So this sentence is very correct and never contradict any textbook definition about NF.
Consider SSB measurement and DSB measurement with numerator of NFssb.
http://www.designers-guide.org/Forum/YaBB.pl?num=1258675545/3#3

Difference between NFssb and NFdsb is due to difference of denominator of NF.

Also surely read http://cp.literature.agilent.com/litweb/pdf/5952-3706E.pdf.

NF which actual instruments gives is NFasb not NFssb.
This has no relation to whether DUT is SSB or DSB.

Title: Re: Noise figure definitions
Post by tromeros on Sep 7^th, 2010, 3:36am

Thanks pancho_hideboo, you were very clear.

The question that arises for me, is how do I compare the simulated and the measured NFs since their definition is different?

Thanks!

Title: Re: Noise figure definitions
Post by pancho_hideboo on Sep 7^th, 2010, 3:55am

tromeros wrote on Sep 7^th, 2010, 3:36am:

The question that arises for me, is how do I compare the simulated and the measured NFs since their definition is different?

No, they are same essentially.
Simply you can't understand a concept of NF correctly.

pancho_hideboo wrote on Sep 6^th, 2010, 6:57am:

NF which actual instruments gives is NFasb not NFssb.
This has no relation to whether DUT is SSB or DSB.