dynamic range definition microwave

I'm used to that the dynamic range of amplifier is definied as the range between the 1 dB compression point and the noise floor. But in Lee's book "The design of radio-frequency integrated circuits" it is definied as the range between the intercept point and the noise floor. Is this a common definitions also? The intercept point is usually about 10 dB higher than the compression point so the latter definition gives a "non-linear" dynamic range then.

Lee's definition is correct. Usually you are not able to use the amplifier very close to its compression point, because intermod products and distortion will corrupt your signal.
The definition of SFDR (spur free dynamic range) is more significative, being the range between the lowest signal to noise ratio that your system needs (on left size) and the point in which the intermod products are higher than noise on other side.
I hope it can help.
Mazz

But the range between the noise floor and the compression point is more common, right? No definitions could ever be incorrect but Lee's definition doesn't make common sense. To me, an amplifier is certainly not linear between the noise floor and the intercept point and a better definition would be the range between the noise floor and the compression point? As you said the "real" linear range is the spurious free dynamic range. Maybe I shouldn't associate an amplifiers linear range with dynamic range then...?

In my opinion a general or common sense idea has no meaning.
In RF for example the SFDR is more common.
IP3 and P1dB are two different metrics and should be considered both in the evaluation of an amplifier.
But this is just my opinion.
Mazz

If you're working with an (RF) amplifier in the lab it might be interesting to know the maximum input signal you can apply before the amp goes into saturation. To know this, the definition of the dynamic range with the intercept point is of no use (unless you calculate the compression point from the intercept point). And btw the intercept point doesn't exist while the compression point "almost" does. This is why I think the intercept point is a bad definition for the dynamic range. To me, dynamic range means the usable range, not the usable plus some unusable range.

Yes, right.
I was thinking to any real application in which you don't have clean CW signals, but wanted modulated signals + interferers.
As I said, these are two different metrics and we've got the simple conclusion that each one of them describes different situations.
Regards
Mazz

If you says dynamic range, then it goes to the compression point; it says SFDR, then intermodulation point. I think it's quite clear