At Measuring balun common mode impedance – #2 I mentioned a glitch on the AIMuhf scan that appears to be a defect of the instrument / client software and that it undermines confidence in the system.
The article documents a test of a known load to attempt to prove the measurement system good.
Note that AIM 865A is not the current version, but problems with the current version were described at AIM 882 produces internally inconsistent results.
The scenario is:
- AIMuhf with a NM to SMA-F adapter, and 1m of RG58 OSL calibrated;
- test load comprising 25+j0Ω at the end of 1.14m of RG58C/U.
I note though that attempts to OSL calibrate to this test fixture produced a warning no matter how many times the process was tried, connectors cleaned and fastened to specification torque.
Above is the scan using AIM865A. Note the glitches around 46 and 140MHz, these are artifacts of the measurement system, they are not expected and do not appear on a measurement of the test load using a Rigexpert AA-600.
Exploring the glitches reveals another defect of the AIM software in the magnified view of the cursor in the previous screen shot. Rs,Xs at the cursor is given as 82.686,-0.254 in the right hand side of the display, but the magnified view of the curve indicates more like 60,5… there appears to be some misalignment of the blue cursor line and ‘logical’ cursor.
Above is a measurement of the same test load connected to the instrument using a N-M to BNC-F adapter and the instrument OSL calibrated at its N connector. No glitches evident. This scan is consistent with expectation, and with the Rigexpert AA-600 measurement.
The “Refer to antenna” function does not work, though presumably it estimates line loss on an assumption that it is entirely conductor loss and the result of well developed skin effect at 1MHz… both dubious assumptions, and so you might not want to use it far from 1MHz anyway.
- AIMuhf and AIM865A as a measurement system produces inconsistent and unreliable results on a simple known load