Keithley 6221 & 2182A Waveform logging

[Pluscrafter]

The setup is as follows the 2182A is connected to the 6221 via Trigger cable and the 6221 is connected to the network.

My plan is that the 6221 should create a square wave (1Hz to ca. 50 Hz, 50% DCycle) and the 2182A log the data on 90degree. The Trigger out of the 6221 is set to tlink #1 and the Trigger In to tlink #2. The 2128A is set to external trigger. The Filter of the 2182A is on and the rate is set to slow.

When the frequency is =< 1Hz it is working and the 2128A read the right values. But on a higher frequencies it seems that the 2128A is averaging the values. My understanding is that, if the phase angle reaches 90degree an trigger impulse is given to the 2182A the device reads the voltage. 

[Kleinstein]

AFAIK the K2182 has both some analog and digital filtering. The analog filtering is fixed frequency and normal exponential settling. So it take a relatively long time to fully settle and it may thus not be a good idea to use it with modulation so fast. The digital filter is usually a simple running average filter. If the data are send to a PC anyway, one could as well do the averaging later on the PC.

With > 1 Hz modulation, one may consider to use the non AZ mode. This gives faster readings. Depending on the frequency the noise can be lower: the AZ mode wastes about 1/2 the time for reading a zero, but ideally suppresses 1/f noise this way. However with many Keithly meters the 1/f noise suppression does not work so well in all cases. So using non AZ mode may be attractive if one is interested in the modulated part. In this case one would not mind a slow offset drive.

[dietert1]

Looks like the K2182A implements two autozero methods in two different input stages.
One is called "Autozero" and the other one "Front Autozero". About "Autozero" the manual mentions a possible drift when disabling it for extended periods of time. About "Front Autozero" the manual mentions polarity reversal and that this method can be disabled when using the "Delta" method. The "Delta" method appears to be a third method of nulling the instrument.

Anybody has more details? For example: Running the instrument with a low thermal short in CH1 in 10 mV range with "Autozero" enabled and "Front Autozero" disabled gives about 80 +/- 10 uV (!). Is this instrument damaged? With "Front Autozero" enabled it reads -80 +/- 10 nV.

Regards, Dieter

Edit: The "Delta" method appears to be what the OP wanted to use.