Measuring IP3 with cheapskate equipment

Discussion in 'Homebrew and Kit Projects' started by N5HXR, Jul 19, 2021.

ad: L-HROutlet
ad: l-rl
ad: abrind-2
ad: MessiPaoloni-1
ad: Left-2
ad: Left-3
ad: L-MFJ
ad: L-Geochron
ad: HRDLLC-2
  1. N5HXR

    N5HXR Ham Member QRZ Page

    So, since my last big question thread turned into some wonderfully patient explanations of IP3 and why I might care, I thought I'd play around with measuring it. I took some of the amplifiers I've made, and measured them to the best of my ability. A question arose today that I can't really find an answer for...

    I know that I don't have the best equipment, so I suppose any measurement I make is a "worst case" situation :). But anyway, I noticed that if I measure the signal from my generator, and then put the amplifier in, I see the following results on my oscilloscope's FFT (left is siggen -> scope -> dummy load; right is siggen -> amp -> scope -> dummy load):


    My question is, since I see intermodulation products without the amplifier, does that somehow spoil my measurement when I put the amplifier inline? Is there any problem with these measurements that means I can't draw useful conclusions about the amplifier?

    Thanks for any help!
    VK4HAT likes this.
  2. W7HV

    W7HV Ham Member QRZ Page

    Interesting. Is that intermod coming from the generator or the scope...or both. If from either, amplitude gain settings can affect them. Thinking out loud: If the intermod goes way down after adding 3dB of attenuation between the sig generator and scope (keeping sig gen amplitude and scope vertical scale the same), it's the scope. You can try simply turning down the sig gen but that could affect intermod in either the sig gen or scope.

    I haven't played much with the FFT on my Rigol scope.
  3. N5HXR

    N5HXR Ham Member QRZ Page

    OK, here's the result with a 10dB attenuator inline (don't have a 3dB):


    The purple trace is without the attenuator, white trace is with. So the two injected signals go down by 10dB, but the intermodulation products don't change. Does that suggest that it's in the signal generator?

    Attached Files:

    Last edited: Jul 19, 2021
  4. N5HXR

    N5HXR Ham Member QRZ Page

    Actually, check that -- it seems to change substantially depending on the vertical resolution of the signal trace. The above graphs are at 50mV/division. Here's the same test at 10mV/division:


    Here, peaks 1 and 4 are much lower to start with, and they go up when I add the attenuator.
  5. K7JEM

    K7JEM Ham Member QRZ Page

    N5HXR likes this.
  6. N5HXR

    N5HXR Ham Member QRZ Page

    I'd be happy measuring that too, but one thing at a time, guys :).
    K7JEM likes this.
  7. N5HXR

    N5HXR Ham Member QRZ Page

    Or, I guess put another way, K7JEM, does measuring P1dB tell me that the BGA2866 in front of an ADE-1+ will suck as much as measuring its IP3? And will a P1dB measurement tell me if the JFET preamplifier I plan to build will be any better?
  8. SM0AOM

    SM0AOM Ham Member QRZ Page

    What kind of two-tone signal generator are you using?

    N5HXR likes this.
  9. N5HXR

    N5HXR Ham Member QRZ Page

    Presently it's my Siglent SDG 1032X signal generator. I've played with it in two configurations. It seems to be somewhat better if I use its wave combination feature to sum channels 1 and 2 together into one output. I also have an external attenuating combiner I put together awhile back, but I'm pretty sure it affords less isolation than the wave combination feature...
  10. K7JEM

    K7JEM Ham Member QRZ Page

    Well, either number reflects the ability of the amp to perform at high levels. The P1dB is a number that can actually be measured, the IP3 is a calculated figure, and purely imaginary. Both give an indication of what the maximum signal output can be, although (IMO) the P1dB is a better indicator.

Share This Page