如何缩小Agilent 4338B中的扩散

我使用Agilent 4338B毫欧表来测量电池的交流阻抗。我按照说明进行了短暂的校准，并将设置更改为适合我测量的电池。我遇到的问题是
在读数中仍然存在（看似）巨大的差异。我在不改变任何东西的情况下测量了相同的单元10次，在取下数据后移除引线，并且每次将引线放置在完全相同的位置。保持这种控制措施
我仍然有0.20mΩ的扩散，因为阻抗大约介于.395 - .431mΩ之间，如何缩小这种扩散的想法吗？同样短的校准似乎完全不可靠，将引线连接在一起并缩短它
至+/-.0009mΩ断开，然后重新连接，它将读取+/-.0700mΩ范围内的任何位置（虽然数字是一致的，就像它将转到.0009或.0500，只有那些2，
不只是随机数）这个
仪表刚刚在2周前进行了校准，我使用的另一个具有相同的问题，并在6个月前进行了校准



     以下为原文

  I use an Agilent 4338B Milliohmmeter with my work to measure the AC impedance of batteries.

I have followed the instructions for short calibrations and have changed the settings to what is appropriate for the batteries that I measure.

The issue that I am having though is that there is still a (seemingly) huge discrepancy in the readings.

I measured the same cell 10 times without changing anything, removing the leads after taking down the data, and placing the leads in the exact same spot every time.
Keeping this measure of control I still got a spread of .02 mΩ  which is a lot seeing as the impedance is somewhere between .395 - .431 mΩ

any idea on how to narrow this spread?

also short calibration seems totally unreliable as well, connect the leads together and short it to +/- .0009 mΩ disconnect, then reconnect and it will read anywhere in the range of +/- .0700 mΩ (though the number is consistent, like it will either go to .0009 or .0500, and only those 2, not just any random number)

This meter was just calibrated 2 weeks ago, the other that I use has the same issues and was calibrated 6 months ago