What am I seeing?

[k7buc]

Ok, using Pomona 1296 bnc to terminal adapter...
When I calibrate, I finish the cal procedure, then without disturbing the resistor, I run a scan. There is a significant amount of reactance showing down at 100khz to about 900khz where it comes to zero. Then it crosses, and becomes capacitive thereafter.
Shouldn't the calibration routine cancel out all that reactance? I'm seeing about 25 ohms inductive at 100khz.
From 1mhz to 170mhz it's straight-line Capacitive reactance to about -3 ohms.
Using the supplied calibration connectors directly connected to the AIM4170 I get much the same result, though the inductive reactance starts at only about 12 ohms and drops from there to zero at 695 khz. From 1 to 170 mhz it only drops capacitive about 1 ohm, so thats pretty close.
The 200 ohm resistor is actually 199.468 ohms according to my HP34401A, and when I plug in that value during calibration instead of 200 ohms the curve is flattest.
Any ideas on what I may be looking at here?
-Del

[Bob]

k7buc said:

Ok, using Pomona 1296 bnc to terminal adapter...
When I calibrate, I finish the cal procedure, then without disturbing the resistor, I run a scan. There is a significant amount of reactance showing down at 100khz to about 900khz where it comes to zero. Then it crosses, and becomes capacitive thereafter.
Shouldn't the calibration routine cancel out all that reactance? I'm seeing about 25 ohms inductive at 100khz.
From 1mhz to 170mhz it's straight-line Capacitive reactance to about -3 ohms.
Using the supplied calibration connectors directly connected to the AIM4170 I get much the same result, though the inductive reactance starts at only about 12 ohms and drops from there to zero at 695 khz. From 1 to 170 mhz it only drops capacitive about 1 ohm, so thats pretty close.
The 200 ohm resistor is actually 199.468 ohms according to my HP34401A, and when I plug in that value during calibration instead of 200 ohms the curve is flattest.
Any ideas on what I may be looking at here?
-Del

Hi Del,
I tried a similar test using a banana adapter and one of the same 200 ohm resistors used in the AIM cal devices.
I used one lead of the same resistor for the short circuit.

There is some reactance at the low end and theta is greater than 0.5 degree below 300KHz. This can be improved significantly by using the Custom Cal procedure (new in version 650). I ran it both ways and I've uploaded screen shots showing the results.
w5big.com/Cal200_1.gif -- Custom Cal
w5big.com/Cal200_2.gif -- Standard Cal

For this test, I set the custom cal sample interval to 0.01MHz.

Note theta is 10 degrees full scale in these shots.

-- 73/ Bob

[k7buc]

It appears the custom cal cannot begin at 0.100 mhz. How did you get around that? It complains that "Delta is too large for this start freq...".
I tried using 0.122 as start freq and 0.01 as delta. It still complained. It will allow starting at 1mhz with 0.01 as delta. That should have made (Start - (2*Delta)) > 0.1 I would think.
-Del

[k7buc]

Hmmm. There seems to be more to this formula than is stated in the error message. If I reduce the "range" so that the upper end is below about 20 mhz, then it will calibrate. Still, though, I can't start at 100khz.
The pgm doesn't allow scanning out of the calibrated range so the lowest freq (100 khz area) cannot be used when custom cal is used.
At least thats how it appears.
-Del

[k7buc]

Very interesting. Using the "custom cal" I get better than 1/10 % accuracy! This is for a range of 120 khz through 10 mhz.
Whatever range I tried, the result was better than 1/10 % using the custom cal procedure. Custom calibrate looks like a winner... if it could only work over the whole range of the instrument at once....wow.
-Del