Hi,

To calculate Composit Second Order Distortion I use formula:

CSO(dB)=(Pi-Ps)dB + 10Log(NoDistProd)dB

where

NoDistProd(Lo)=N(1-(f/(fH-fL+d))) or

NoDistProd(Hi)=(N-1)(f-2fL-d)/(fH-fL+d).

Based on these formulas NoDistProd is a real number or double.

To measure CSO I have to set Number of Distortion Products,

but imd.CSONumDistortionProducts is an integer value.

I am not able to set the trace because I am getting an error when send this command!

Please explain.

Thanks,

Dragan.

To calculate Composit Second Order Distortion I use formula:

CSO(dB)=(Pi-Ps)dB + 10Log(NoDistProd)dB

where

NoDistProd(Lo)=N(1-(f/(fH-fL+d))) or

NoDistProd(Hi)=(N-1)(f-2fL-d)/(fH-fL+d).

Based on these formulas NoDistProd is a real number or double.

To measure CSO I have to set Number of Distortion Products,

but imd.CSONumDistortionProducts is an integer value.

I am not able to set the trace because I am getting an error when send this command!

Please explain.

Thanks,

Dragan.

CSONumDistortionProducts Property is written:

VB Syntax - imd.CSONumDistortionProducts=value, where

value - (Double) Number of Distortion Products

Return Type - Double

Now I am confused.

Dragan.

1. Do we use next formula to calculate Composit Second Order Distortion:

CSO(dB)=(Pi-Ps)dB + 10Log(Number of Distortion Products)dB,

where

Number of Distortion Products(Below carrier) = N(1-(f/(fH - fL + d)))

Number of Distortion Products(Above carrier) = (N-1)(f - 2fL - d)/(fH - fL - d)

Pi = Power level at the second order intercept point (dBm)

Ps = Power level of each carrier (dBm)

N = Number of carriers

f = Frequency of distortion product in Mhz

fH = Frequency of highest channel in MHz

fL = Frequency of lowest channel in MHz

d = Frequency separation between channels in MHz.

2. To measure Composit Second Order Distortion using PNA-X Network Analyzer we have to send

Number of Distortion Products to the equipment using command:

imd.CSONumDistortionProducts = Number of Distortion Products.

To get Number of Distortion Products do we use next:

Number of Distortion Products(Below carrier) = N(1-(f/(fH - fL + d)))

Number of Distortion Products(Above carrier) = (N-1)(f - 2fL - d)/(fH - fL - d).

3. And if we do,

do we need to round result (Number of Distortion Products = Round(Number of Distortion Products))?

Please advise,

Dragan.

I should first say that, of course, we never stimulate the DUT with more than 2 tones, so that this is really a model of what the CSO would be if the DUT is non-linear with a normal distortion curve (2nd order goes up 2x with pwr, 3rd order goes up 3x with power, etc).

So, for CSO, we measure the IM products (second order for CSO) and we know the tone power in the main tones: then we need to compute how that tone power would distribute over the channel. If you say 2, then it is just the two tone result, if you say 10, then the power of each tone is presumed to be reduce by 5x ,and the implied change to the 2 order term is computed, and the result is reported as CSO. I'm not sure I know if it matters if they are above or below the carrier.

Of course, CSO is measured on the second order and so it must fall into the band of the amplifier or we won't see it. I have seen setups where CSO lower is measured sweeping the spacing of the two tones for Fl (say 55 MHz) to Fh-Fl (say 805 MHz) so the second order low (F1-F2) always falls in the BW (in this case 55-860 MHz); then a second test for the high side second order where the tones are swept with a fixed spacing of 6 MHz from 55 MHz to about 430 MHz, so the second order sweeps from around 110 MHz to around 860 MHz (you work out the exact numbers). In all cases, you need the main tones and the second order result to be in-band with the amplifier.

And then the number of carriers that this is supposed to represent is the number of distortion products used.