Strange SVMI behaviour ALM1000

When going through the StudentZone May 2018 material, I found that I could not match the waveforms measured and went to investigate. I found the following strange behaviour, with no leads connected:

- ALM1000 Oscilloscope running, AWG's configured according to the training material, curves CA-V and CA-I visible:

Looks what you would expect, except maybe for the "not very straight Square wave".

- Next, when I turn on curve CB-V as well, the trace of CA-I changes dramatically:

My ALM1000 is Rev. F, so I would expect it to be calibrated. It seems like some stray capacitance is introduced by switching on the CB-V curve?

Can someone please advise, is this normal behaviour, does my unit need (re)calibration? Please help.

Small typo
[edited by: MartijnSc at 11:36 AM (GMT -4) on 31 Mar 2021]
  • Ok, there are a couple of details about how the software tries to automatically configure the hardware when the time base (mSec/Div) is set less than 0.2 mS. The software tries to switch from 100KSPS to 200KSPS if just two of the 4 possible curves (trace waveforms) are selected. The switch to the higher sample rate is to provide the better time resolution at higher frequencies. The hardware can only sample two signals at a time at the higher sample rate vs 4 at the slower rate. Just CA-V and CB-V for example. When it sees that say three of the curves it tries to switche to what is called "alternate sweep" mode where it alternates between two of the three (or four) signals each time it sweeps.

    It looks like in your examples the hardware multiplexer is not properly syncing to the proper two signals. In your first screen it looks like you were getting CA-V and CB-I rather than CA-I. In the second case it looks like you were getting the correct signals. I was able to reproduce similar results. I was able to properly set the 2X sample rate / alternate sweep manually by opening the Set Sample Rate controls from the Options drop down menu.

    EZ seems to be broken for me and I can't insert an image or attach a file to this response to show you what to do. Also, unfortunately there isn't any figures in any of the basic User Guide pages showing that set of controls either. Figure 7, Base Sample Rate control,s in the advanced pages on Equivalent Time Sampling does at least show what it looks like. If you select the Alternate Sweep Mode check box and also have the Double Sample Rate box checked it should alternate between sampling VA and VB / IA and IB and you should get the right signal traces.

    It looks like how the software automatic chooses to configure the hardware when the user happens to select three or four traces when the time base is set to 0.1 mS/Div needs to be looked at to see if it is working as expected.

    Thanks for pointing this out.

  • So, trying your suggestion, when I use the same settings as before and change the Sample Rate, again strange things happen. For example, when selecting the Alternate Sweep Mode, the IA jumps to -200mA (with nothing connected to CHA or AIN), disabling and re-enabling the Alternate Sweep Mode may make it jump back to 0 mA, where it may also stay at -200mA. In short, it's behaviour is unpredictable.

    I can see the selector indicator flash between "VA and VB" and "IA and IB" when I have both Double Sample Rate and Alternate Sweep Mode selected, but the Oscilloscope screen shows all kinds of "transient (?)" signals flashing around!

    I further found that there is a difference in the behaviour, depending on the selected ADC MUX Mode, even with Alternate Sweep Mode unselected:

    1. With Double Sample Rate selected and Alternate Sweep Mode unselected, with VA and VB selected, I get a flat curve (blue in pic below) for CA-I (which I would expect, as noting is connected):

    2. With VA and IA selected, other settings same, I get that strange ripple:

    If I disable Double Sample Rate, I always get that ripple...

    I'm kinda lost, I do not know how to best measure (what settings to use) when measuring the voltages accross and the current through a LC-circuit, being fed with a Square wave at 2750 Hz, as described in that StudentZone topic.

    (by the way, running Windows 10 on a Intel I7 3770K, with 16 GB of mem, connected to USB3

  • The alternate sweep mode only "triggers" properly when the "Sync AWG" box is checked and more than 2 traces are selected. Sorry but that is a hardware limitation.

    The "automatic" configuration when crossing the 0.2 to 0.1 mS/Div time base boundary will be fixed in the next release of the software (tentatively scheduled for the end of this Academic semester May/June).


  • Ok, super!

    One final question about current flowing even with nothing connected. I finally tried with AWG setting of SVMI, DC, 5V on both channels A and B. I measure 0.13 mA on CHA, and -0.13mA on CHB.

    Is this to be expected (acceptable), or should I somehow calibrate?

  • Do you mean a small DC offset in the current or the spikes due to charging the channel output capacitance? ( 400 pF ).

    Small DC offset of less than 1 mA is to be expected. It can be zeroed out in the software.