Andrea C

The appendix B code is configuring the instrument to use binary data format (format.data = format.REAL32).<br> Compared to ASCII, this allows fewer characters on the bus for better throughput.<br> <br> To interpret the data, checkout some IEEE754 compliant techniques.<br> I think numpy and fromfile might be useful.

Model 2450 uses integrating A/D converter type.<br> It is custom implementation, but is reasonable to think of it as dual slope integrating A/D of about 22 bits at 1 NPLC.<br> The NPLC setting is affecting the duration of the integration or how long you let your signal “charge up” the cap.<br> Regardless of NPLC value, a reported measurement is based on sample size of one.<br> <br> In contrast, some of our products have both high accuracy integrating A/D and fast digitizing A/D.<br> The digitizer always runs at 1MHz; &#160;requested sampling rates slower than 1MHz can make use of auto aperture based averaging from the 1MHz data. &#160;SMU models such as 2461, 2651A have digitizers in addition to integrator.

That sort of performance info would be specified by the OEM of the relay that is mounted into the S46 chassis.

Assuming Radiall relay types, their document here may be of help:
Radiall RF Relay Info

On pg 14 of that document they give their recommendation about hot switching (don't).
The estimated amount of degradation in the relay life/number of operations is a function of the average power of the signal.

On pg. 9 of that document they give the Average Power Rating vs. Operating Frequency.

Check your gpib-32.dll properties.  Is it from NI or Keithley or possibly CEC?

My guess is that something installed the NI driver for GPIB.  And it over-wrote the required version of that DLL needed by the KPCI-488A card.
Can you reinstall the driver for KPCI-488A?

Did you close the channel?
For the reading that is placed in the buffer, what channel tag does it have?

My .NET code from prior post does cause the front panel to update the watch channel measured value.

On the instrument, in the event log settings, you can enable command logging.  This will let you see exactly what the IVI driver commands are sending to the DAQ.

If still not working, check your firmware level.  I'm at 1.7.12b

 

Check that you have the most recent version of Test Script Builder (J05).
Also check the firmware on your 2612B (3.4.2 is most recent).

Test Script Builder is making use of NI-VISA for communication with the instrument.
From error screen, it is not clear what the issue is.

If you have NI-MAX installed, try verifying basic VISA IO from that.

If you still need help, tell us the VISA resource name that you use for the 2612B.
 

What is the state of interlock?
If interlock not satisfied, the max voltage is clipped/only a portion of 100V range can be used.

Closing the loop on this one.

Suggest updating firmware to latest (1.7.12b) ;  you are at 1.7.7b.

Even so, when using the integrating A/D, we are able to duplicate the issue with DMM7510 but not with DMM6500.
A bug has been logged for DMM7510 behavior.

Workarounds if using DMM7510:

First:  your trigger model operation is essentially sampling until told to stop;  You could get to essentially same outcome by making use of abort() command to stop an infinite count acquisition into a recycled finite sized buffer.

Second:  your trigger model operation will work and *TRG can be detected/used to stop it if the digitizing A/D is used instead of the integrating A/D.

Here is a code sample:

my_instr.write("reset()")
my_instr.write("errorqueue.clear()")

#comment out the use of integrating A/D
#my_instr.write("dmm.measure.func = dmm.FUNC_DC_VOLTAGE")
#my_instr.write("dmm.measure.range = 0.1")

​​​​​​​#use the digitizing A/D
my_instr.write("dmm.digitize.func = dmm.FUNC_DIGITIZE_VOLTAGE")  
my_instr.write("dmm.digitize.range = 0.1")
my_instr.write("dmm.digitize.samplerate = 10e3")
my_instr.write("dmm.digitize.aperture = dmm.APERTURE_AUTO")

my_instr.write("measure_buffer = buffer.make(12)")
my_instr.write("measure_buffer.fillmode = buffer.FILL_CONTINUOUS")

my_instr.write("trigger.model.setblock(1, trigger.BLOCK_BUFFER_CLEAR, measure_buffer)")
my_instr.write("trigger.model.setblock(2, trigger.BLOCK_MEASURE_DIGITIZE, measure_buffer, trigger.COUNT_INFINITE)")
my_instr.write("trigger.model.setblock(3, trigger.BLOCK_WAIT, trigger.EVENT_COMMAND, \
                                              trigger.CLEAR_ENTER, \
                                              trigger.WAIT_OR, \
                                              trigger.EVENT_DISPLAY)")
my_instr.write("trigger.model.setblock(4, trigger.BLOCK_MEASURE_DIGITIZE, measure_buffer, trigger.COUNT_STOP)")

Stands to reason that the SMU that is sinking the current needs to be lower potential than the external voltage source that the SMU is loading.

If need be, the SMU will transition to quadrant 3 (negative current and negative voltage) to allow the source current setpoint to be achieved.

See if the attached PDF helps.

What are you proposing to connect the 2651A to?

Can I ask:  what do you need the operating voltage to be?

I've uploaded a block diagram level of the SMU.

For your question about using it as a load, you will operate it as a current source with a voltage compliance limit.
To operate in quadrant 4 as a load, you would have the SMU source negative current.

The V-meter provides input to the voltage compliance limiting.