WinGPIB - Windows GPIB App (multi-purpose)

[yuhar]

USB cable connection, and address everything confirms that it works; (Yes I copy the correct VISA address from "Keysight connection expert")
 - the timer based logging works great; with comma and with decimal point settings too.
In the csv-file always decimal point will be, what You see in response field...

(I've had problems with this - somewhere else in the past - always, so I changed it now everywhere to decimal point.)

- I tried to make a autocal during logging (sample period 30 sec, - right after I got one reading I send Async command *CAL?), then autocal succeeded and few seconds leater: "error caused by remote command" on meter visible once, and then the logging continue... (-410 Query INTERRUPTED - if I see the error log on meter)

- I tried to enable  interrupt  command: *CAL   (looks like; start with autocal, and autocal failed on meter...) no logging at all.

Istvan

[IanJ]

It looks like the CAL command on your meter, 34465A I think you said it was, is somehow interrupting/locking GPIB and it has trouble continuing.

I will do some tests here to make sure INTRPT CMD is working good, but I don’t have a 34465A so can’t full test your setup.

Can you send CAL from WinGPIB from Send Async but with logging stopped. Try sending multiple times?

Ian

[yuhar]

(I have USB-GPIB interface and GPIB connector on meter too, but cannot use it this moment - I do not Why...)

I can send *CAL? from WinGPIB with Send Async and the logging is not stopped, Yes I can send muliple times everytime succeded, but after last one few seconds later - I get error message on meter once ("error caused by remote command") and the logging is continue no more error.

Autocal is around 15 seconds in duration.

If the sample period of logging is 5 or 10 sec then the autocal usually fails


I made a new test:
WinGPIB logging not running this time, I send *CAL? (Send Async) and right after send :READ? (Send Async) - then autocal is succeded; not comming the "error caused by remote command" but on the error list on the meter visible two "-410 Query INTERRUPTED"





Istvan

[IanJ]

I tried INTRPT CMD with my 3458A (ACAL DCV) and it works ok. No problem.

I also looked up the manual for the 34465A, there are various CAL commands in the sub-set.
If you are logging VDC then you only need to autocal ADC, i.e.:
CAL:ADC?
This would be faster.

I have a 34461A and although it doesn't support autocal I tried sending:
CAL:ADC?
The meter displayed an error "Calibration step failed......", which is fine, but then the meter stopped updating it's display, and won't only resume if I hit a key on the meter. Not good!
So, with this meter it looks like if there is an issue, or you don't send the right command properly then the meter hangs till you press a key. Maybe there is a GPIB command to restart?

In saying all that, you mentioned you are using the command:
*CAL?
In the manual it specifies that this command performs autocal and returns the result. So there is your problem, *CAL? is not a SEND ASYNC command, it is a QUERY ASYNC command and so will not work with INTRPT CMD.
You need to find a CAL command that is SEND ASYNC, i.e. does not expect to send a reply......with the current version of WinGPIB.
Or, you can STOP logging, send a manual QUERY ASYNC command, wait till it finished, and then manually RUN logging again.

Ian.

EDIT: Corrections made to above.

[IanJ]

New version in Post #1 available for download. - PLEASE READ THE INFO IN POST #766 BEFORE UPGRADING TO THIS VERSION IF YOU ARE COMING FROM V3.177 OR EARLIER.
WinGPIB_V3_182.zip
[Virus scanned using BitDefender AntiVirus]
Need help installing, upgrading or moving to a new PC, then see instructions in Post #1

Small bug fix on the INTRPT CMD checkbox. If manually unchecked whilst logging the status is now fully cleared and logging is resumed.
PS. Please be careful doing this i.e. not whilst an interrupt command is actually in progress!

Ian.

[yuhar]

Yes, Your Right,

Query Async command.

Autocalibration can be used without any special consequences if the ambient temperature changes. - it has a separate specification, tempco

"Autocalibration compensates for drift caused by time and temperature change in DC voltage and resistance
measurements, and adjusts some aspects of other measurement functions."

The other calibrations commands are not like this. (a password is also required for modification)

Thank You

Istvan

[IanJ]

New version in Post #1 available for download. - PLEASE READ THE INFO IN POST #766 BEFORE UPGRADING TO THIS VERSION IF YOU ARE COMING FROM V3.177 OR EARLIER.
WinGPIB_V3_183.zip
[Virus scanned using BitDefender AntiVirus]
Need help installing, upgrading or moving to a new PC, then see instructions in Post #1

Added a new checkbox to the Interrupt functionality in order to select between a Send Async or Query Async type command for the Interrupt.
This is because some devices, such as the Keysight 34465A require to send back confirmation status of an AutoCal when run and so require a Query Async type command.

Note: AutoCal, ACAL DCV are the types of commands for various instruments that this interrupt functionality is aimed at.

I have not tested this fully, so will leave the previous version available for download.

Per below, check this box for Query Async, otherwise it defaults to Send Async:



Ian.

[yuhar]

Hi Ian,

one the test is done with version 1.183

logging start normally... at the sixth reading, logging is stopt (it should be at 12, with my set conditions)
the meter not due Autocal, and the logging is stop till I push "abort retry" button; then all pending calls are completed one after the other,
logging continue until the next interruption is due. (no errors message in meter)


I don't think it's necessary to enter a duration with this Autocal. A successful result ( "+0" response) a condition for continue; - or duration is equal to timeout?
(- if Autocal is not successful, it may be possible to repeat it, but it would not be good if it stopped the logging.)

Thank You

Istvan

[IanJ]

Hi Ian,

one the test is done with version 1.183

logging start normally... at the sixth reading, logging is stopt (it should be at 12, with my set conditions)
the meter not due Autocal, and the logging is stop till I push "abort retry" button; then all pending calls are completed one after the other,
logging continue until the next interruption is due. (no errors message in meter)


I don't think it's necessary to enter a duration with this Autocal. A successful result ( "+0" response) a condition for continue; - or duration is equal to timeout?
(- if Autocal is not successful, it may be possible to repeat it, but it would not be good if it stopped the logging.)

Thank You

Istvan

It looks like your USB GPIB is not reliable enough on your PC, it seems to hang after a few READs. The 34465A has a Lan port, it would be better to use it.
My 34461A has the same settings as your 34465A and when I use USB it is very reliable, but of course I have a different USB card in my PC.

If you want to stay with USB then try a test:
Set up logging but without Interrupts.
Everytime it does a READ then check the length of time it takes (status window) and compare to your PERIOD in secs.
You should be able to run for hours/days without a problem.
Make sure READ is reliable before you try Interrupts.

Yes, I have more work to do on the Interrupt Query Async......but for now it is good enough.

Ian.

[yuhar]

Now the PC and meter connected with simple usb-cable (printer).

Two relative short (10 min long) tests have been completed 120 x 5sec measurement (without interrupt as You suggested) then make a calculation in excel, and I get same result; there is not a single second difference.

I tested again the interrupt; now without Query Async (duration: 20sec, period: 1min and sample period: 5sec);
- the interrupt-Autocal start at 35 sec (after 6 samples) - why so early? 
( tests done with 2, 3 and 4 mins interrupt period too; the interrupt start 25 sec earlier than set every time; - the formula:  The interrupt start time = (intrrupt period) - (interrupt duration+5sec)  )
(In fact, it would be necessary to autocal after hours or if the temperature has changed permanently.)

- then Autocal succeeded, then 4 pending reading happened in quick succession.   - I think; the pending readings can be deleted during the interruption. - in fact, it is necessary to pause
the logging is continue  (-410 Query INTERRUPTED errors accumulate in meter for known reasons)



Now it is better than Query Async (there is no Autocal and the log also stop)

Thank You

Istvan

[IanJ]

New version in Post #1 available for download. - PLEASE READ THE INFO IN POST #766 BEFORE UPGRADING TO THIS VERSION IF YOU ARE COMING FROM V3.177 OR EARLIER.
WinGPIB_V3_184.zip WinGPIB_V3_185.zip
[Virus scanned using BitDefender AntiVirus]
Need help installing, upgrading or moving to a new PC, then see instructions in Post #1

Bug fix & further development of the Interrupt system.
- Fixed the issue with period timing not working properly. The duration is now appended onto the period setting, i.e. Period then duration then restarted period again.
- The Query Async option no longer requires any duration to be set and is greyed out accordingly.
- The Interrupt Enable and Query Async interrupt checkboxes are now saved to the profiles.

I have only lightly tested this update (for now) on Device 1 with my 3458A:
Interrupt Send Async = NPLC 10
Interrupt Query Async = TEMP?
I have not tested a Send Async such as ACAL DCV, nor a longer duration Query Async, nor have I tested Device 2.

EDIT:
Small update to V3.185 - When duration text box is greyed out the label text below it is greyed out also.

Ian.

[yuhar]

without Query Async:

timing is ok now...
- looks to me the interrupt duration does not mater; if I set 3 sec or 35 sec; the Autocal succeeded every time, and then every time 4 pending readings happened... (5sec sample period)

Query Async:
still not working; at the interrupt time; - the logging is not continue at all, and the Autocal not start.

- WinGPIB not send the command!
The command not visible in "Keysight IO Monitor"

(if I send, same command with Query Async button before the interrupt; the command goes out.)


Thank You
Istvan

[IanJ]

without Query Async:

timing is ok now...
- looks to me the interrupt duration does not mater; if I set 3 sec or 35 sec; the Autocal succeeded every time, and then every time 4 pending readings happened... (5sec sample period)

Query Async:
still not working; at the interrupt time; - the logging is not continue at all, and the Autocal not start.

- WinGPIB not send the command!
The command not visible in "Keysight IO Monitor"

(if I send, same command with Query Async button before the interrupt; the command goes out.)


Thank You
Istvan

That’s quite weird, because it works for me on my 3458A. Both Interrupt Send Async and Query Async are working as planned.

The fact you are getting 4 pending commands make me think there is still some USB Visa issue on your PC……so to test I will set up my 34461A via USB tomorrow to see what happens. A shame the 34461A doesn’t support Autocal though, but I am sure there are some other ‘time consuming’ commands I can find.

Ian

[chilternview]

Latest version (185) is marked as a trojan by Windows Defender. Please can this be fixed...

[IanJ]

Latest version (185) is marked as a trojan by Windows Defender. Please can this be fixed...

False positive.....can you give me the trojan name etc so I can submit the EXE to Microsoft.
Thanks,

Ian.

[chilternview]

It's Trojan:Win32/Wacatac.B!ml

Unfortunately Defender really, really doesn't want to let you run the exe. And probably best not to if you're not certain.

[IanJ]

It's Trojan:Win32/Wacatac.B!ml

Unfortunately Defender really, really doesn't want to let you run the exe. And probably best not to if you're not certain.

Are you using Win11 or Win10......the MS online form needs the info about which of their security products detected it, one of the following:
System Center Endpoint Protection
Win10 Windows Defender AV
Win11 Windows DefenderAV
Microsoft Defender SmartScreen
Smart App Control

Ian.

[chilternview]

I believe it's Win10 Microsoft Defender SmartScreen, it's the one that has the icon on the taskbar that has 'Windows Security' if you hover the cursor over it.

[IanJ]

I believe it's Win10 Microsoft Defender SmartScreen, it's the one that has the icon on the taskbar that has 'Windows Security' if you hover the cursor over it.

Ok, I have reported the EXE to Microsoft......they usually take less than a day to turn it around, and then advise users to update AV Definitions to clear the false positive.

It's ALWAYS Windows Defender, never the other AV solutions. Oh well!

Ian.

[IanJ]

New version in Post #1 available for download. - PLEASE READ THE INFO IN POST #766 BEFORE UPGRADING TO THIS VERSION IF YOU ARE COMING FROM V3.177 OR EARLIER.
WinGPIB_V3_186.zip
[Virus scanned using BitDefender AntiVirus]
Need help installing, upgrading or moving to a new PC, then see instructions in Post #1

Bug fix to Interrupt Cmd Query Async.
- Testing with a Keysight 34461A I found that Query Async wasn't working properly, this is now fixed.

I used TEST:ALL? command which takes about 1.7secs so seems a long enough command to test and confirm that logging now halts until a message is received back and before allowing logging to continue.

What I haven't done yet is to test simultaneous Device 1 & 2 Interrupts running, mixing Interrupt Send Async & Query Async etc.
I also want to revisit the method I am using to 'wait' for the Query Async to finish, because I think it currently freezes the GUI until a reply is detected.

NOTE: I do notice that connecting to my 34461A is produces an error on the DMM Lcd (top right corner), and when opening it I get "203 Command Protected; Feature not available on this model"........but everything runs fine!

Here's a screenshot captured just when the Interrupt Query Async is run.



Ian.

[yuhar]

Yes, the interrupt with Query Async working now.

I wanted to suggest logging the interrupt; but now that's included:
"6,34465A,2024-06-05_14:04:30,+0,0.0,0.0"

without Query Async:
There are still pending reads, and duration has no effect; but that doesn't really affect me.

Thank You:
Istvan

[chilternview]

The Y axis scale is still somewhat lacking...

[IanJ]

New version in Post #1 available for download. - PLEASE READ THE INFO IN POST #766 BEFORE UPGRADING TO THIS VERSION IF YOU ARE COMING FROM V3.177 OR EARLIER.
WinGPIB_V3_187.zip
[Virus scanned using BitDefender AntiVirus]
Need help installing, upgrading or moving to a new PC, then see instructions in Post #1

Bug fix - Device 2 Interrupt duration problem now fixed.
Bug fix - Attempt at stopping the GUI freezing temporarily when an interrupt Query Sync is waiting for a reply.

I have also disabled the interrupt system from working when Dual logging Device 1 & 2 as it will need more work.

Ian.

[IanJ]

The Y axis scale is still somewhat lacking...

Yeah, I'll get to that at some point.......Microsofts MSChart graphing stuff absolutely sucks!

The way around it is to auto-scale, then make a note of the values, then uncheck auto-scaling and paste in the values as a starting point, then tweak till you get what you like.

Ian.

[IanJ]

New version in Post #1 available for download. - PLEASE READ THE INFO IN POST #766 BEFORE UPGRADING TO THIS VERSION IF YOU ARE COMING FROM V3.177 OR EARLIER.
WinGPIB_V3_188.zip
[Virus scanned using BitDefender AntiVirus]
Need help installing, upgrading or moving to a new PC, then see instructions in Post #1

Bug fix:
- Y-axis scaling on the Live Chart didn't produce a fixed amount of intervals/labels when manual scaling - Fixed.

This is a quick fix but appears to be ok.
Did I say I hate Microsofts MSChart!

Ian.