AIM 865A Refer to Antenna facility

AIMuhf

I used an AIMuhf for Measuring balun common mode impedance – #2 using the OSL calibration facility.

AIM also claims to have a means of backing out a known transmission line between reference plane and DUT. This article discusses use of AIM’s Refer to Antenna facility.

AIM’s developer recently said of AIM’s Refer to Antenna facility:

Version 882 does have a problem with the Refer to Antenna function. Version 865A can be used for this function.

This function does have it’s limits though. It should only be used for good quality coax. The impedance and velocity factor of coax is not constant over the whole length and this limits the accuracy. Also the impedance may not be equal to the “nominal” impedance in the catalog. The impedance of 50 ohm cable can vary quite a bit. AC6LA.com has some interesting data showing how coax parameters vary with frequency.

Custom cal is much better when it is possible to put the cal loads at the far end of the transmission line. This takes into account variations in impedance, velocity factor, and loss and it can be used when there is coax and ladder line in one transmission line system.

This article looks at use of AIM’s Refer to Antenna facility in AIM 865A to measure a choke at the end of 0.93m of RG58.

Screenshot - 16_02_2015 , 09_43_31

Above is the result of a scan with the AIM calibrated at its coax socket.

Screenshot - 16_02_2015 , 08_18_06

Above, the same DUT with Refer to Antenna turned on prior to the scan. The results are plainly nonsense, the negative resistance values are not physically possible on such a DUT.

Screenshot - 16_02_2015 , 08_19_10

Above, the result of saving and reloading the graph. Now the resistance is positive, but the reactance below choke self resonance is negative whereas it should be positive, nonsense again. Then there are the glitches at 18 and 28MHz. The Refer to Antenna facility is seriously broken and appears to have been released without having been adequately tested.

To the latter part of the developer’s comment:

This function does have it’s limits though. It should only be used for good quality coax. The impedance and velocity factor of coax is not constant over the whole length and this limits the accuracy. Also the impedance may not be equal to the “nominal” impedance in the catalog. The impedance of 50 ohm cable can vary quite a bit. AC6LA.com has some interesting data showing how coax parameters vary with frequency.

Custom cal is much better when it is possible to put the cal loads at the far end of the transmission line. This takes into account variations in impedance, velocity factor, and loss and it can be used when there is coax and ladder line in one transmission line system.

That is all true, but the question is whether AIM 865A achieves what can be achieved within those limitations.

Screenshot - 16_02_2015 , 09_53_26

Above is the data entry screen for Refer to Antenna. On it, the loss at 1MHz is specified, and AIM extrapolates that to the measurement frequency. The algorithm is not exposed, but with only one loss parameter, it is likely that it calculates loss(f)=loss@1MHz*fMHz^0.5. So, for the above case, it would extrapolate loss/100m of RG58 to be 1.332*1000^0.5=42.1dB whereas the loss of RG58 at 1000MHz is more like 70.8dB, almost 30dB higher!

Conclusions

  • AIMuhf and AIM865A as a measurement system produces inconsistent and unreliable results, the Refer to Antenna facility is severely broken.
  • The model for line loss is probably over simplified and not very accurate at higher frequencies, it could be more accurate with a better model.