Which instruments can be used to measure the s-parameter, IIIP3, noise figure of LNA?

Hi,

i had design a CMOS LNA operates at 2.4GHz. the input and output impedence is match to 50ohm at that frequency. can somebody tell me which instruments can be used to measure the s-parameter, IIIP3, and noise figure of my LNA at that frequency?

What the devices i need to add if i wish to measure a differential LNA?

Thank you for your help.

Hi

About instruments: try visit www.agilent.com and find noise figure analyzer, network analyzer (PNA series i think)

Regards

To do these measurements you need a Network Analyzer, Two Signal Generators and a Spectrum Analyzer:
http://www.home.agilent.com/USeng/nav/-11882.0/pc.html
For balanced measurements (or multi-port) you can use one of the ENA from Agilent:
http://www.home.agilent.com/USeng/na...881709/pd.html

you may follow many concepts, i.e. minimize the number of instruments.

For IP3 you need a SA and two signal generators. If the frequency separation may be choosen the same as SA IF freq, so the internal SA LO out may be used as signal generator. The other one, may be the Network Analyzer's generator.

To measure spar, you need a VNA, but ask yourself are spar really needed, or |s11| and |s21| are enough? In the latter case a more economical SNA may be used instead.

Also NFA require (or has built-in) generator that may be shared with previosly described measurements.

A note about factories.
Agilent is an important supplier but others exists.
Also look for www.rohde-schwarz.com and www.anritsu.com

Ehh... for usable using of S11 and S21 (in simulator etc.) you need vector information in degree or cartesian form, scalar network analyzer can only give power (magnitude) of signal, not phase relation between ouput and input signal.

Phase and magnitude information needs if working with smithchart or polar chart and scalar network analyzer is more or less useless for more advanced work above tuning cavity filters... IMHO.


[little offtopic]

Very simple standing-wave measure can built with directional coupler (or bridge) and SA with sweep generator and using good RF-switch between reflection measure and transmission measure and you make poor mans scalar network analyzer - good enough to tuning not so high performents cavity filter in field. (generator impedance is importet to precision of measure)

(if everybody have seeing small, ligth weigth and _fast_ VNA in same size as R&S FSH6 - yell! - I need this! - I don't understud why NVA is so fragile, big and so heavy in day modern technology...)

R&S FSH6 can measure with phase information and present smitchart, but is so slow (1 sek sweep in best case) and complicate mounting and demounting of reflection bridge between reflection measure and transmissions measure (lack of embedded RF-switches) and always need calibrate process between shift measure mode - is practical inpossible measure and tuning filters on the field... take so much time and irritate on struggling of bridge...

I not tested if possible to take out value from FSH6 to make S-paramters.

FSH6 is very good idea to make portable NVA in combination a SA, but bad software implementation (only make of software guy and no practical testing in field ?) and miss embedded RF-switches in reflection bridge, make FSH6 useless for field using more than just SA - using.

don't buy network capability option on FSH6, is not worth a price IMHO.

hp8594 with sweep generator and directional coupler and RF-switches work so much faster and easier for simple retuning filter and other testing (as gain and power etc.) on field. I travel around with it defiance as equipmet weight 20-times more compare to FSH6.

---

[ok, little more offtopic, but...]


FSH3(6) has also more programs idioti as sweep generator only working in transmissions measure mode, and close down if going back to SA-mode

Is impossible to make -70 dBm sweep generator level with external attenuator and measure in SA-mode +30 dBm (with externa power attenuator) to check amplifier a 100 dB gain and rigth output power.

ie impossible to kick on sweep generator - measure and adjust generator levels (with extra attenuator if need) in dBm with SA-mode to wanted level on DUT, after this mounting protection attenuator (if strong amplifier) to input, setting refrence offset level to compesate away protector attenuator and levels view in dBm, and coupling to DUT and measure - this simple basic measure is impossible to make on R/S FSH3(6) handheld SA depend of stupid software programmers and measure kit-making......


In transmission measure you can only measure relative dB, not absolute, and max 25 dB gain in DUT - how measure amplifier with 90 dB gain (using external attenuator) and testing compression point at 30 dBm in this case ? GHAAAAAA

(Software guy forget I want measure amplifier, not only filters..)

etc. etc. stupid basic things...


----

I want using know absolute level, high dynamic adjustable range of sweep generator (FSH3(6) only 20 dB and relative value (-27 dBm - 7 dBm measured level), hp8594 handle 68.5 dB and give value in reliable dBm from -66 to +2.5 dBm) and can using refrence offset level both on generator side and SA-side for compensate extra attenuator or extra LNA to give value in dBm! - not only relative value as FSH3(6)

hp8594:s have sweep generator totaly free running (but synced to sweep) without incluence to other used SA-mode. FSH3(6) is only runing on transmissions measure (specific application) and not usable to other measure - Idiots!.

---

FSH3(6) and other modern SA, also from HP etc. have tendency to make solution of few specific measure problem (standard SA measure wit difference of analytic filtera and demodulation), transmission, reflection of filter, (but not amplifier with high gain...)) etc. in separate mode-button, possibly made by non RF expirence programmers with object thinking and this pilosophy give smaller and smaller space to make custom solution of measure problem depend of idiot:s limit of scale, not independence function and coupled to other functions and 'kit'-thinking of measure problem and want sell it on high licens cost.

Every RF-guy knows not exist 'stardard measure' in development phase and maintance of old equipment and needs very very flexible equipment. - but is hard to find to days...

---

So, if thinking to buy new equipment, testing very carefully before and try near 'impossible' measure as a simplest problem to find programmers idiot limits or locked funktion - or you feeling blast six month later with expensive equipment cannot or heavy work to measure 'simple' problem, problems easy measure with older (possible already sold out) equipment...