specturm analyzer reading on cascaded amplifers

[FrankTCS]

I have cascaded two LNA's and when i apply voltage to both amplifers the floor on my spectrum analyzer increases 20dB to about (-50dB). When i apply voltage to just one amplifer, the specturm analyzer's noise floor(?) stays at (-75dB). Is that normal? or a product of the cascaded amplifer? Also will this increased floor reduce the ability to receive singals lower than the floor?
thx.

Frank

 

[logbook]

Frank
the key facts omitted here are the gains of the LNAs.

I don’t understand what you mean when you say that you applied voltage to just one amplifier. Do you mean
a) you just used one amplifier, correctly powered
b) you had both amplifiers wired into circuit, passing the signal though both, but applied power to only one.

b) would be a bizarre thing to do and the result has no useful meaning.

Consider just one amplifier. Suppose it has 20dB gain. If the noise floor on the spectrum analyser only goes up by 10dB, then the noise on the spectrum analyser, referred to the amplifier’s input, has dropped by 10dB. You should expect this sort of effect because spectrum analysers have lousy noise figures (say 28dB) and you would have to try really hard to get an amplifier that bad!

 

[FrankTCS]

Logbook,
sorry for the confusion. I applied voltage to both amplifiers (30dB gain each) for a total of 60 db (with an input signal of -80dBm). Its then when the noise floor jumps up.
I was told that this may be a function of the video resolution BW on my signal generator also, and i'm going to try to reduce the BW to 5kHz.

So from your response i should expect the increased noise floor to be normal - correct?

If i could afford a noise figure meter i could probably confirm this, but im hoping some actual real world testing would have to do.

 

[logbook]

You have 60dB gain, but the noise floor only went up 20dB. That means that the noise figure of the amplifier, referred to its input, is 40dB better than the spectrum analyser. Notice that one amplifier did not produce significantly more noise on the spectrum analyser display because the spectrum analyser is so noisy.

Looks like you have a really poor spectrum analyser, since the noise figure of the amplifier has to be perhaps 1dB or 2 dB, and the noise figure of the spectrum analyser is therefore more like 42dB, >10dB more than a good spectrum analyser.

 

[FrankTCS]

logbook, thanks for help really appreaciate it. I did call HP about the analyzer and they let me know its one of thier lower-end models. Again much appreaciated.

best regards,

Frank