[Top] [All Lists]

Re: [TowerTalk] Third Order Intercept

To: "'Tony'" <>, <>
Subject: Re: [TowerTalk] Third Order Intercept
From: "Gary Schafer" <>
Date: Thu, 25 Jan 2007 23:46:09 -0500
List-post: <>

> -----Original Message-----
> From: [mailto:towertalk-
>] On Behalf Of Tony
> Sent: Thursday, January 25, 2007 8:18 PM
> To:
> Subject: [TowerTalk] Third Order Intercept
> Greetings all:
> Have a question about third order intercept point. I understand the
> measurement takes sensitivity into account at the point were IMD products
> start to show at the noise floor; the greater the sensitivity is before
> distortion shows up the better.
> If that's the case, then a good TOI figure alone doesn't tell the whole
> story. If the sensitivity isn't in the useable range for most HF work (-
> 110
> ~ 130dbm or so) then the receiver will have good IMD rejection, but would
> be
> somewhat deaf compared to one with greater sensitivity.
> It would be similar to adding attenuation to improve the TOI by making the
> rig less sensitive; a 10db attenuator might cause the TOI improve by 10db.
> Would this be an accurate interpretation of the TOI measurement?
> Tony KT2Q

Let me see if I can remember some of this. The 3rd order intercept point is
a fictitious number that can never really be reached. It is an extrapolated
number from a set of measurements at lower signal levels. You will reach the
"compression" point in an amplifier and the amplifier will give little gain
after the compression point before you can reach the 3rd order intercept
point. The 3rd order intercept point is usually many db above that point.

The 3rd order intercept point is where the wanted signal and the 3rd order
product levels would be equal in amplitude "IF" the amplifier was able to
amplify both without going into compression. The intercept point is plotted
from wanted signal levels and 3rd order signal levels measured at some point
well below the compression level where the changes are predictable.

3rd order intercept is a useful term in evaluating how well (how big of a
signal) an amplifier will do with more than one signal present at the same
time (two tone test). Compression level tells you how strong a single signal
can be before the amplifier goes into compression (will not amplify any

You are right about adding an attenuator in front of a receiver improving
the intercept point (and also the compression point) but it will reduce the
usable sensitivity by the same amount. But sensitivity has no bearing on the
intercept point by itself. In other words we could lower the noise floor of
the receiver thus making it more sensitive (by using better transistors) and
the intercept point may remain the same or it could be better or worse.
There is no direct relationship between sensitivity and the intercept point.
However, you really need to know both the sensitivity (or noise figure) and
the intercept point to evaluate an amplifier or receiver more fully. How
many db above the noise floor is the intercept point.

One thing to watch out for when looking at 3rd order intercept specs is
whether the 3rd order intercept point is referenced to the input or the
output of the amplifier. A lot of pre-amplifier manufacturers like to
reference to the output because it looks better as the figure given includes
the gain of the amplifier. To properly evaluate those figures you need to
subtract the gain of the amplifier so you have what the number is at the
input of the amplifier so you can properly compare it with another amplifier
that may have a different amount of gain. Of course some of those that do
reference to the output don't tell you that is what they are doing, for a

Gary  K4FMX


TowerTalk mailing list

<Prev in Thread] Current Thread [Next in Thread>