Program that can log from many multimeters.

[KungFuJosh]

I don't know if this matters, but the math box stops working when the grid box is closed (I usually pop them out).

You found a bug.

Each Math requires some local data for each window it is used in, I do create it for GridPanel, but not when you convert to SinglePanel.

It appears there's slightly more conditions for the bug to be avoided:

If the Grid panel, or the Single panel for the Vdc are closed, then the Math Single panel will not work.
or
If the Grid panel is open, with the Vdc Single panel inside the Grid panel (not popped out), then the Math Single panel will not work if it's popped out.

IOW, if all panels are in the grid panel it works. If the grid is open, and all relevant Single panels are outside the Grid panel, it works.

Thanks,
Josh

[HKJ]

It appears there's slightly more conditions for the bug to be avoided:

If the Grid panel, or the Single panel for the Vdc are closed, then the Math Single panel will not work.
or
If the Grid panel is open, with the Vdc Single panel inside the Grid panel (not popped out), then the Math Single panel will not work if it's popped out.

IOW, if all panels are in the grid panel it works. If the grid is open, and all relevant Single panels are outside the Grid panel, it works.

I believe I found the bug: All Grid panel and Single panel used the same data area for Math. When a panel is closed the area is removed and Math stop working, opening a panel again will reinstate the area and the Math will probably work again (I have not checked).

The fix is to use different names for each window (i.e. different areas) and due to a bug I did not do that. It was a easy fix and will be included in the next release.

All this is internal to TC and nobody (but me) is supposed to worry about it (Any test releases after you posted the bug will also include the fix).

[KungFuJosh]

I believe I found the bug: All Grid panel and Single panel used the same data area for Math. When a panel is closed the area is removed and Math stop working, opening a panel again will reinstate the area and the Math will probably work again (I have not checked).

The fix is to use different names for each window (i.e. different areas) and due to a bug I did not do that. It was a easy fix and will be included in the next release.

All this is internal to TC and nobody (but me) is supposed to worry about it (Any test releases after you posted the bug will also include the fix).

Yes, reopening the panel, or placing it correctly in or out of the grid panel does reinstate the math.

I'm happy to test for you if needed.

Thanks,
Josh

[NoDisintegrations]

Hi guys / HKJ,

I'm trying to get serial communication working for a Chinese PSU (KUAIQU-SPPS3010D-232) by creating a custom device file.

After many days of frustrated attempts, I've decided to throw in the towel and create a post asking for assistance.

So far I've managed to get it to successfully send commands to turn PSU's output on and off (tx), and to set the voltage or current (txrx), but I've hit a brickwall with receiving data (txrx?). I keep getting "Rx as numbers <No data (timeout?)>" errors in the Commands log after opening the Current Values tab. Running Test Controller via tcrundebug eventually leaves 'java.lang.NullPointerException: Cannot read the array length because "b" is null' in the log.txt, but this doesn't appear instantaneously after opening the Current Values tab, only after a while of leaving it open.

Unfortunately I haven't been able to find a solution within the documentation, or an example file that's close enough to reference.

If anyone's willing to point me in the right direction / highlight where I went wrong, that would be great.

I've attached the protocol document, along with my device file and the error log.

Please see the following threads, where two kind souls have provided protocol documents for the same series of PSU:
https://www.eevblog.com/forum/testgear/spps3010d-power-supply-softwareprotocol/
https://www.eevblog.com/forum/testgear/kuaiqu-programmable-switchmode-power-supply-(32v-10-2a)/

PS: HKJ, thank you for writing and releasing the software! Aside from the struggle of writing a custom device file, it's been spectacular!

Kind Regards,

[HKJ]

Hi guys / HKJ,

I do not have time today for checking your definition, but can give a few pointer:

Use the right driver, with a block format use the block driver: https://lygte-info.dk/project/TestControllerConfigDevice2%20UK.html#Binary_with_fixed_communication_blocks_(Block)  You can get away with other drivers, but generally it is easier if you use the correct one.

When you get a error in the log there is a reason for it, here it probably is because you do not receive any data, this means your definition of the data format do not match the received format.

To send actual voltage/current values you will have to do some formatting in (), i.e. using (formatDouble(value,2,2,1,1)) when formatting the message to the supply.

[thephil]

Code: [Select]

#interfaceType PS:1 PS:2 PS:3
#interface setOn OUTPUT:STATE CH(channel),(value)                                     
#interface getOn OUTPUT:STATE? CH(channel)                                           
:string:

That seems to work. However, I have two more "channels" for serial and parallel mode, respectively. Unfortunately, they are not referred to as "CH4" and "CH5" but "SER" and "PAR". So e.g. #setVoltage would have to be APPLY SER,(value)V. At the moment I have no idea how to accomplish that...

Any hints welcome.

You need to use a bit scripting and then channels must be called 4 and 5:
https://lygte-info.dk/project/TestControllerFunctions%20UK.html#getElement

Instead of CH(channel)
use CH(getElement(channel,"1 1 2 3 SER PAR"))
The reason for the two 1 is because the list is zero based, this way both 0 and 1 will return 1.

Sorry for getting back so late (busy with other things in the meantime). I read the function documentation that you had linked to but still cannot get this to work – I must be missing something.

First of, it seems to me that it should be

Code: [Select]

getElement("CH1 CH2 Ch3 SER PARA", channel) (i.e. reverse order of aguments) – correct? I played with many variations but to no avail...

This works fine for channels 1-3:

Code: [Select]

#interfaceType PS PS:2 PS:3 PS:4 PS:5
#interface setOn OUTPUT:STATE CH(channel), (value)
#interface getOn OUTPUT:STATE? CH(channel)
:string:

But this does not (even for the regular channels 1-3):

Code: [Select]

#interfaceType PS PS:2 PS:3 PS:4 PS:5
#interface setOn OUTPUT:STATE (getElement("CH1 CH1 CH2 CH3 SER PARA",channel)), (value)
#interface getOn OUTPUT:STATE? (getElement("CH1 CH1 CH2 CH3 SER PARA",channel))
:string:

Trying to set the value I see no error message but the channel is not turned on and reading the value back gives no result.
I read through a few examples of using getElement in other config files but could not find an example in the context of defining the interface. So maybe there is something special about it that I am not aware of?

EDIT:

I even tried the process in the command window and this works as expected:

Code: [Select]

OUTPUT:STATE? (getElement("ch1 ch1 ch2 ch3 SER PARA",5))

OUTPUT:STATE? PARA
;; OFF

So it seems I am using channel wrong or it is not accessible that way or ...

[thephil]

BTW: is there a way to get some kind of error message if something does not work? At the moment, all I know is that it does not do what I expected. Or could I somehow get the string (SCPI command) that my non-functional interface definition produces. Maybe that would help me understand why my definition does not work.

[HKJ]

First of, it seems to me that it should be

Code: [Select]

getElement("CH1 CH2 Ch3 SER PARA", channel) (i.e. reverse order of aguments) – correct? I played with many variations but to no avail...

I did make a mistake with the parameters and I can see you also fixed the detail with first element being 0 (not 1).

Code: [Select]

#interfaceType PS PS:2 PS:3 PS:4 PS:5
#interface setOn OUTPUT:STATE (getElement("CH1 CH1 CH2 CH3 SER PARA",channel)), (value)
#interface getOn OUTPUT:STATE? (getElement("CH1 CH1 CH2 CH3 SER PARA",channel))
:string:


But I cannot see why this fails (But see my next answer)

[HKJ]

BTW: is there a way to get some kind of error message if something does not work? At the moment, all I know is that it does not do what I expected. Or could I somehow get the string (SCPI command) that my non-functional interface definition produces. Maybe that would help me understand why my definition does not work.

Yes, you start TC with the testControllerDebug.bat file, it will list a lot of the communication and error message will be listed in the DOS window.

[thephil]

Yes, you start TC with the testControllerDebug.bat file, it will list a lot of the communication and error message will be listed in the DOS window.

Excellent. Thanks.

I'm on LINUX so I ran it with

Code: [Select]

java -jar TestController.jar debug
(BTW: the equivalent tcrundebug file lacks the debug argument.)

Now I got debugging output that was quite helpful:

Code: [Select]

;; Found Uni-T UDP3305S on 192.168.0.66 sn: ADP**********
;; UDP3305S: Tx <OUTPUT:STATE (getElement("CH1 CH1 CH2 CH3 SER PARA",channel)), 1>
;; UDP3305S: Tx <OUTPUT:STATE? (getElement("CH1 CH1 CH2 CH3 SER PARA",channel))>
;; UDP3305S: Rx Timeout
So, while "(value)" is correctly translated into "1", the getElement expression is just handed over verbatim. For the heck of it, I removed the channel variable and just added a number, instead:

Code: [Select]

#interface setOn OUTPUT:STATE (getElement("CH1 CH1 CH2 CH3 SER PARA",2)), (value)
#interface getOn OUTPUT:STATE? (getElement("CH1 CH1 CH2 CH3 SER PARA",2))
:string:
and still, it is not parsed:

Code: [Select]

;; UDP3305S: Tx <OUTPUT:STATE (getElement("CH1 CH1 CH2 CH3 SER PARA",2)), 1>
;; UDP3305S: Tx <OUTPUT:STATE? (getElement("CH1 CH1 CH2 CH3 SER PARA",2))>
;; UDP3305S: Rx Timeout
If I copy&paste this exact expression into the command window, it works just fine:

Code: [Select]

OUTPUT:STATE? CH2
;; UDP3305S: Tx <OUTPUT:STATE? CH2>
;; UDP3305S: Rx <OFF>
;; OFF

At that point, I fired up a Windows 11 VM just in case this is a LINUX-only problem but there was no difference. In an act of despair, I opened my config file in a hex editor to ensure that the interface definition does not contain any weird characters that might confuse the expression parser but found everything to be well-behaved ASCII.

I tried removing the outer parenthesis, I tried adding an extra pair, I added parenthesis around the channel variable. The latter did improve things a bit: Now, at least the channel variable was translated to a number.

Interface definition:

Code: [Select]

#interface setOn OUTPUT:STATE (getElement("CH1 CH1 CH2 CH3 SER PARA",(channel))), (value)
#interface getOn OUTPUT:STATE? (getElement("CH1 CH1 CH2 CH3 SER PARA",(channel)))
:string:

now results in this:

Code: [Select]

;; UDP3305S: Tx <OUTPUT:STATE (getElement("CH1 CH1 CH2 CH3 SER PARA",1)), 1>
;; UDP3305S: Tx <OUTPUT:STATE? (getElement("CH1 CH1 CH2 CH3 SER PARA",1))>
;; UDP3305S: Rx Timeout

So our final challenge is getting the getElement expression to be recognized as a function and parsed/processed.
Trying to overcome this last hurdle, I changed double quotes to single quotes, added and/or removed spaces in random places, cast ancient spells and sacrificed my only child to the gods of Java but nothing helped. 

So at this point, I am out of ideas. Any ideas, anyone?

[HKJ]

So at this point, I am out of ideas. Any ideas, anyone?

It looks like a bug in TC.

Try this updated version: http://lygte-info.dk/pic/Projects/TestController/TestController.jar
Note: It is only the .jar file.

[thephil]

It looks like a bug in TC.

Try this updated version: http://lygte-info.dk/pic/Projects/TestController/TestController.jar
Note: It is only the .jar file.

Thanks for the quick response!

I tried the new version but the issue persists:

Code: [Select]

;; UDP3305S: Tx <OUTPUT:STATE (getElement("CH1 CH1 CH2 CH3 SER PARA", 2)), 1>
;; UDP3305S: Tx <OUTPUT:STATE? (getElement("CH1 CH1 CH2 CH3 SER PARA", 2))>
;; UDP3305S: Rx Timeout

[HKJ]

I tried the new version but the issue persists:

Code: [Select]

;; UDP3305S: Tx <OUTPUT:STATE (getElement("CH1 CH1 CH2 CH3 SER PARA", 2)), 1>
;; UDP3305S: Tx <OUTPUT:STATE? (getElement("CH1 CH1 CH2 CH3 SER PARA", 2))>
;; UDP3305S: Rx Timeout

Please verify the version number: On configuration page press About, the number must be 2.47


I did test with this (On a unrelated device, i.e. it did not know the commands):
#interface setX OUTPUT:STATE (getElement("CH1 CH1 CH2 CH3 SER PARA",channel)), (value)
#interface getX OUTPUT:STATE? (getElement("CH1 CH1 CH2 CH3 SER PARA",channel))

From log:
=setX("SDG2122X:1",0)
;; 17522.57ms SDG2122X: Tx <OUTPUT:STATE CH1, 0>

=getX("SDG2122X:1")
;; 71857.00ms SDG2122X: Tx <OUTPUT:STATE? CH1>


I uses setX/getX instead of setOn/setOff to avoid problem with existing interface commands.

[thephil]

OK – retested and things got interesting:

Version is 2.47 – I'm running the correct jar file.

Code: [Select]

#interface setOn OUTPUT:STATE (getElement("CH1 CH1 CH2 CH3 SER PARA", (channel))), (value)
#interface getOn OUTPUT:STATE? (getElement("CH1 CH1 CH2 CH3 SER PARA", (channel)))
:string:
results in the old problem:

Code: [Select]

;; UDP3305S: Tx <OUTPUT:STATE (getElement("CH1 CH1 CH2 CH3 SER PARA", 2)), 1>
;; UDP3305S: Tx <OUTPUT:STATE? (getElement("CH1 CH1 CH2 CH3 SER PARA", 2))>

So I copied your setX definition into my config, too and low and behold it works:

Code: [Select]

;; UDP3305S: Tx <OUTPUT:STATE CH2, 1>
;; UDP3305S: Tx <OUTPUT:STATE? CH2>
;; UDP3305S: Rx <ON>

Upon closer Inspection, your definition does not put channel into its own parenthesis so I removed them form mine, too. And voila – it's working!

Thanks a lot for your help! I'll conduct some more testing of my config and then come back here to contribute it to the project.

[HKJ]

Upon closer Inspection, your definition does not put channel into its own parenthesis so I removed them form mine, too. And voila – it's working!

It is a optimization in TC that is the reason for this issue:
If (channel) and (value) is present in the string, it will skip the parser and directly replace the strings with the value (The parser has considerable higher overhead than a direct replacement).

[thephil]

OK – I spent another hour or two testing my config and I believe it does what it's supposed to.
 
So at this point I'd like to submit my config file for inclusion in a future release of TestController. It supports Uni-T UDP3305S and UDP3305S-E programmable power supplies. It should be feature complete with respect to direct control of all parameters but does not support things like list or delay mode, because I believe those don't make sense when controlling remotely, anyway.

 UNI_T UDP3305.txt (11.42 kB - downloaded 12 times.)

Is there anything else I am expected to supply – or is this thread even the right place to do this?

I am maintaining this in a github repository, so if anyone finds a bug or has ideas for improvements, I'll be happy to accept issues there.
https://github.com/philpagel/udp3305s

[HKJ]

OK – I spent another hour or two testing my config and I believe it does what it's supposed to.
 
So at this point I'd like to submit my config file for inclusion in a future release of TestController. It supports Uni-T UDP3305S and UDP3305S-E programmable power supplies. It should be feature complete with respect to direct control of all parameters but does not support things like list or delay mode, because I believe those don't make sense when controlling remotely, anyway.

It will be included in the next release, thanks.

Is there anything else I am expected to supply – or is this thread even the right place to do this?

No, posting a definition in this thread or mailing one to me is enough to get it included.

I am maintaining this in a github repository, so if anyone finds a bug or has ideas for improvements, I'll be happy to accept issues there.
https://github.com/philpagel/udp3305s

To get the definition included with TC updated, it must be posted here or emailed to me.

[jmurray]

Hi HKJ,

I'm struggling a bit with the rxalluntil? syntax for SCPIx.

The command I have as an example is:

Code: [Select]

#scpiCmd FUNC2A? rxalluntil? countMatch(value,">")>0 FUNC2?
It appears to just be transmitting the entire string "rxalluntil? countMatch(value,">")>0 FUNC2?" to the instrument.
I've tried a few things but I'm sure I'm missing something very obvious.

Code: [Select]

FUNC2A?
;; F45DEV: Tx <FUNC2A?>
;; F45DEV: Tx <rxalluntil? countMatch>0 FUNC2?>
;; COM2: Tx: <rxalluntil? countMatch>0 FUNC2?.> 72 78 61 6C 6C 75 6E 74 69 6C 3F 20 63 6F 75 6E 74 4D 61 74 63 68 3E 30 20 46 55 4E 43 32 3F 0A
;; COM2: Rx: <?>> 3F 3E
;; F45DEV: Rx <?>>
;; ?>

[HKJ]

I'm struggling a bit with the rxalluntil? syntax for SCPIx.

The command I have as an example is:

Code: [Select]

#scpiCmd FUNC2A? rxalluntil? countMatch(value,">")>0 FUNC2?
It appears to just be transmitting the entire string "rxalluntil? countMatch(value,">")>0 FUNC2?" to the instrument.
I've tried a few things but I'm sure I'm missing something very obvious.

It is not really surprising, the SCPIx do not have the command, it is only supported in the "Ascii" driver.
I have no idea why I have listed it as supported for SCPIx

[jwells777]

Has anyone been able to log multiple Brymen 869s meters at the same time? 

I think that they are both reporting identical HID info and thus unable to be read simultaneously.  If I have understood the documentation correctly, the solution is to use remapping somehow, but I haven't been able to figure out how to do so.

Also, is it possible to log the secondary display information from the 869s or will this program only log the primary display information?

[HKJ]

Has anyone been able to log multiple Brymen 869s meters at the same time? 

I think that they are both reporting identical HID info and thus unable to be read simultaneously.  If I have understood the documentation correctly, the solution is to use remapping somehow, but I haven't been able to figure out how to do so.

You need to load multiple BM869 meters, TC will use the cable ID as address (It is filled in automatically).
If you want different names for the meters in the tables and charts, you can use the remapping (TC will stick a number after the default handle, if you don't).
Note: Remapping has one issue, TC may not be able to setup the correct number format for reloaded data. When that happens you can fix it for the loaded data on the "Table" page.

Also, is it possible to log the secondary display information from the 869s or will this program only log the primary display information?

Only primary display.

[jwells777]

Thanks for the quick response.  I figured that it could only log the primary display, which is OK for my current need.

Unfortunately, I think that the issue is that the Cable ID is reporting the same for the two meters, which then gets used as the address and prevents me from using the remapping page successfully.  Both meters show up in the list box, but I can only add one to the remapping page.

In any case, I have attached a couple of images in case you see something obvious that I am doing wrong. 

Hoping that I can get it working, as this seems like a pretty nice utility.

[HKJ]

Unfortunately, I think that the issue is that the Cable ID is reporting the same for the two meters, which then gets used as the address and prevents me from using the remapping page successfully.  Both meters show up in the list box, but I can only add one to the remapping page.

Usually TC will automatic rename duplicate devices by sticking _xx (xx is a number) on the handle. It fails because the number that is added to their name is based on their address and that means they will get the same number.
I have made a new version of TC that will assign a random number to each in that case, but that means the remapper cannot be used, because you will not know the full name of the device, before it is loaded.
Jar only file: http://lygte-info.dk/pic/Projects/TestController/TestController.jar

[KungFuJosh]

This isn't too big of a deal, but it's probably worth mentioning: I can't add Math functions when the log is running. I had to disable the log, add the function, and then restart TC for everything to work right. I dunno if that's a bug, or just a limitation. 🤷

Thanks,
Josh

[HKJ]

This isn't too big of a deal, but it's probably worth mentioning: I can't add Math functions when the log is running. I had to disable the log, add the function, and then restart TC for everything to work right. I dunno if that's a bug, or just a limitation. 🤷

It is a designed interlock, some stuff is disabled while logging. This is basically anything that can affect the number of columns on the Table page.