[SOLVED] Ad8302 based VSWR measurement

hi ,
Im trying to use ad8302 for vswr measurement along with a dual bidirectional coupler.

Ad8302 is basically a logarithmic detector which would amplify the difference between two inputs.
The incident and reflected power is coupled using a directional coupler and given to ad8302.Here one input (incident power coupled)is constant and other (reflected power , coupled) is varying according to load.

But the datasheet says , the minimum reflection coefficient you would be able to measure in this method will be limited by directivity of the coupler..

| ΓMIN(dB)|<|D(dB)|

the below table depicts various load impedances and corresponding reflection coefficient in dB calculated by (Zl-Zo)/(Zl+Zo)



Zo=50Ω

Zl
Ω Reflection coefficient in dB
0 0
10 -3.521825
15 -5.376906
20 -7.359536
25 -9.542425
30 -12.0412
35 -15.06655
40 -19.08485
45 -25.57507
48 -33.80392
49.9 -59.99131
50 ∞



Say i ve got a dual bidirectional coupler with 35dB directivity and coupling 20dB



So is the data sheet saying that with ad8302 i wont be able to measure the reflected coefficient beyond 48 Ohms load in this method??

can somebody help me to understand this?
**broken link removed**

Directivity of the reverse coupler is very important when measuring the return loss (in dB).
When return loss (in dB) equals the coupler directivity (in dB) the measurement error is very high.
For the forward coupler the directivity is not such important, because in this case the coupled signal level generally is high.

ALL VSWR measurement systems have the exact same problem. When you get a well matched load, there is a very little reflected signal, and it can get swamped out by bigger system leakage signals. That does not necessarily mean that you can not make the measurement! What you want to do is to get the best 50 ohm load you can find, hook it up to the system, and make your measurement as you step or sweep the frequency. The load might have 45 dB return loss, but your system will measure it like it varies between 30 to 50 dB return loss. Record those measurements at each frequency, save them in a computer file, and in future measurements just subtract those numbers from the new measured data. You are "calibrating" your measurement system.

The load might have 45 dB return loss, but your system will measure it like it varies between 30 to 50 dB return loss.

Click to expand...

this 45 dB return loss which maybe the ideal return loss fo a load,,, is it to be calculated by connecting that load on to a vector network analyzer or some standard equipment? or do you meant it in some other way?

Record those measurements at each frequency, save them in a computer file, and in future measurements just subtract those numbers from the new measured data. You are "calibrating" your measurement system.

Click to expand...

So if the ideal value is 45dB ,, and my measured value is say 35dB then my error is 45-35=10dB,,,and i need to add(or subtract?) 10dB to every measurement i take with this system right?

Thanks
Steeve

---------- Post added at 21:01 ---------- Previous post was at 20:46 ----------

By the way i came across an article depiciting the effect of directivity in vswr measurements...This gives a graph..as shown below.Does anyone know if this is a standard graph and i can apply it to my condition??And any theory behind developing this chart??

Link:-h**p://www.evaluationengineering.com/index.php/solutions/instrumentation/directional-power-measurements-and-the-effects-of-directivity.html

If you really are trying to measure 45 dB return losses, you are going to have to spend considerable time learning about "error correction models".

In the old days, one would hook up a "moveable" 50 ohm load, and slide it to different positions. At each position you would measure S11 vs frequency. Then you used simple algorithms to compute the "center" of the S11 points at a single frequency, and that was your calibration vector.

In more modern times, you take a know reflective load, as in offset short circuits, or an E-cal unit, make the S11 measurements, and knowing what the calibration stadards actual data should be, you can calculate the correction factors. There are a number of patents held by Vahe Adamian (sp?) of ATN that explain the process.