Program that can log/control many multimeters and other devices.

[rthorntn]

Thanks, does the bluetooth adapter not create a suitable serial port then?

https://www.eevblog.com/forum/testgear/howto-u1272a-firmware-update-with-u1177a-bluetooth-adapter/

The above seems to suggest you get a serial port?

[HKJ]

Thanks, does the bluetooth adapter not create a suitable serial port then?

https://www.eevblog.com/forum/testgear/howto-u1272a-firmware-update-with-u1177a-bluetooth-adapter/

The above seems to suggest you get a serial port?

Most Bluetooth adapters do not, if it does you can use it.

[Gertjan]

Please tell me how I can include 6 characters.

This is controlled from the #value lines, but they are already configured to show all digits. This means the meter do not send more digits on the interface.

You can temporary change the number of digit by right clicking on a table column and adjust the format.

Hi HKJ,

Indeed there seems to be an issue with using “SI” in the #value statement…
I checked, the issue is only on the 1V range (so not on 100mV, 10V etc ranges) It seems the problem only arises when there are more than 5 digits after the decimal point.

The #value  line in the 8846A definition is: #value Volt_DC VDC SI Volt_DC
The meter is sending all available digits: (sometimes even an extra digit)

Code: [Select]

;; FL8846A: Tx <READ?>
;; 192.168.1.34: Tx: <READ?.> 52 45 41 44 3F 0A
;; 192.168.1.34: Rx: <+1.23458200E+00> 2B 31 2E 32 33 34 35 38 32 30 30 45 2B 30 30 0D
;; FL8846A: Rx <+1.23458200E+00>
;; FL8846A: Rx as numbers <1.234582>
;; FL8846A: Tx <READ?>
;; 192.168.1.34: Tx: <READ?.> 52 45 41 44 3F 0A
;; 192.168.1.34: Rx: <+1.23457600E+00> 2B 31 2E 32 33 34 35 37 36 30 30 45 2B 30 30 0D
;; FL8846A: Rx <+1.23457600E+00>
;; FL8846A: Rx as numbers <1.234576>
;; FL8846A: Tx <READ?>
;; 192.168.1.34: Tx: <READ?.> 52 45 41 44 3F 0A
;; 192.168.1.34: Rx: <+1.23457600E+00> 2B 31 2E 32 33 34 35 37 36 30 30 45 2B 30 30 0D
;; FL8846A: Rx <+1.23457600E+00>
;; FL8846A: Rx as numbers <1.234576>
;; FL8846A: Tx <READ?>
;; 192.168.1.34: Tx: <READ?.> 52 45 41 44 3F 0A
;; 192.168.1.34: Rx: <+1.23457600E+00> 2B 31 2E 32 33 34 35 37 36 30 30 45 2B 30 30 0D
;; FL8846A: Rx <+1.23457600E+00>
;; FL8846A: Rx as numbers <1.234576>

The Table defaults to “SI format”:



Indeed the last digit is missing!
When setting the table to “SI7 format” the last digit is displayed:



When I change the #value  line in the definition to: #value Volt_DC VDC SI7 Volt_DC
The table defaults to “SI7 format”, and all digits are displayed by default.


I had the impression that when using “SI”, the number of digits in the table etc. is automatically adapted to the number of digits that is read out of the meter.
From your documentation:

SI will use SI prefix, i.e. Mkmu etc. after the number and will show from 0 to 7 decimals.
SI3..SI15 will use SI prefix and limit the number of digits (not decimals) to 3..9.

Using “SI” seems to be an elegant solution, as always the correct number of digits are displayed.
(I see this also used in other definitions)
What is going wrong? Is this a bug? Is there a cap on on the number of digits after the decimal point?
Is it better to use “SI7” (or even "SI8") in #value? (But then every mode has to be carefully checked for the possible number of digits…
Please advise.

Of coarse I know that always all available digits are saved in the table, and are included when saving the table .
I tested on Windows, using the latest versions of TC and Java

Regards, Gertjan.

[HKJ]

Indeed there seems to be an issue with using “SI” in the #value statement…
I checked, the issue is only on the 1V range (so not on 100mV, 10V etc ranges) It seems the problem only arises when there are more than 5 digits after the decimal point.


Using “SI” seems to be an elegant solution, as always the correct number of digits are displayed.
(I see this also used in other definitions)
What is going wrong? Is this a bug? Is there a cap on on the number of digits after the decimal point?
Is it better to use “SI7” (or even "SI8") in #value? (But then every mode has to be carefully checked for the possible number of digits…
Please advise.

Of coarse I know that always all available digits are saved in the table, and are included when saving the table .
I tested on Windows, using the latest versions of TC and Java

I have not checked the code, but I believe I have added a default limit to SI at some time, to avoid calculated result showing up to 15 digits.
There is no principal difference between using SI7 or SI to format a number, they will both restrict output to actual significant digits, the number just put a limit on the maximum digits to show. You could say I believe the SI was the same as SI15, but it was actually SI6.

[Gertjan]

There is no principal difference between using SI7 or SI to format a number, they will both restrict output to actual significant digits, the number just put a limit on the maximum digits to show. You could say I believe the SI was the same as SI15, but it was actually SI6.

Hi HKJ,

Thanks.
I will adapt the definition to using SIx in the #values statement.
Later this week I will have a Tek DMM4050 for testing. Then I can test 2 things at the same time.

regards, Gertjan.

[EEVblog]

Thanks, does the bluetooth adapter not create a suitable serial port then?

https://www.eevblog.com/forum/testgear/howto-u1272a-firmware-update-with-u1177a-bluetooth-adapter/

The above seems to suggest you get a serial port?

Most Bluetooth adapters do not, if it does you can use it.

Is there a list of PC bluetooth adapters that provide a serial port under Windows?

EDIT: I asked Grok and got this list that supports SPP:
SENA Parani-SD1000 Bluetooth Serial Adapter
Plugable USB Bluetooth 4.0 Low Energy Adapter (USB-BT4LE)
ASUS USB-BT400 USB Adapter
Generic Bluetooth 2.0/4.0 USB Dongles with CSR Chipset

[HKJ]

Is there a list of PC bluetooth adapters that provide a serial port under Windows?

EDIT: I asked Grok and got this list that supports SPP:
SENA Parani-SD1000 Bluetooth Serial Adapter
Plugable USB Bluetooth 4.0 Low Energy Adapter (USB-BT4LE)
ASUS USB-BT400 USB Adapter
Generic Bluetooth 2.0/4.0 USB Dongles with CSR Chipset

I believe one of the Bluetooth profiles is a serial port, but DMM's do not use that profile, hence the problem getting a serial port.

When using the BM2257 DMM try this command on TC's command line:

#Readout BM2257.VoltageAC -28 2 7.6 red (153,255,0) 1

Rightclick allows for adjusting a lot of parameters.

[KungFuJosh]

I have an SML03 sig gen with an OCXO, and I wanted to get a more detailed stability chart of its 10MHz internal ref. It's currently connected to the counter in my SDG2122X.

In the Current values tab, it's set to SI. Also SI in the Scales for chart tab.
I've tried auto range with the min range values in a variety Hz, mHz, and uHz and nothing changes.
Even with the manual range as set in the screenshot, the chart will always only show 10M for the scale. I've tried a broad range of decimal places in the manual range. If I go too far, it just rounds up.
It doesn't matter how many decimal places I specify in the Scales for chart, it still shows 10M for the Y axis.

Am I missing something, or is this range not going to work in TC? It seems I can only go to 8 decimal places before it rounds it off in the manual scale range, and I need more.

Thanks,
Josh

PS. Yes, turning off SI gives less useful notation with more decimals on the Y axis.

[HKJ]

I have an SML03 sig gen with an OCXO, and I wanted to get a more detailed stability chart of its 10MHz internal ref. It's currently connected to the counter in my SDG2122X.

In the Current values tab, it's set to SI. Also SI in the Scales for chart tab.
I've tried auto range with the min range values in a variety Hz, mHz, and uHz and nothing changes.
Even with the manual range as set in the screenshot, the chart will always only show 10M for the scale. I've tried a broad range of decimal places in the manual range. If I go too far, it just rounds up.
It doesn't matter how many decimal places I specify in the Scales for chart, it still shows 10M for the Y axis.

Am I missing something, or is this range not going to work in TC? It seems I can only go to 8 decimal places before it rounds it off in the manual scale range, and I need more.

SI works different than the other formats: you do not specify decimals, but digits.

But instead of fighting to get as many decimals as possible, what about using math and subtract 10M from the value. Then you do not need many digits to see variations.
Internal precision in TC is double, that is about 15 digits.

[Gertjan]

I have an SML03 sig gen with an OCXO, and I wanted to get a more detailed stability chart of its 10MHz internal ref. It's currently connected to the counter in my SDG2122X.

I can only go to 8 decimal places before it rounds it off in the manual scale range, and I need more.

I'm afraid the problem is with the counter in your SDG2122X......
These freebee counters in ARB generators are good at producing a lot of digits on the display, but do not have the real resolution.

Looking in the SDG2122X spec heet, counter resolution is not mentioned at all... (nor gate time).

For looking at detailed stability of a good 10MHz OCXO, one needs ppb (parts per billion) resolution.
1ppb @ 10MHz is 10mHz. So you need at least 10 digits resolution. Only the better dedicated counters will have this resolution. (with a reasonable gate time).

regards, Gertjan.

[KungFuJosh]

I have an SML03 sig gen with an OCXO, and I wanted to get a more detailed stability chart of its 10MHz internal ref. It's currently connected to the counter in my SDG2122X.

I can only go to 8 decimal places before it rounds it off in the manual scale range, and I need more.

I'm afraid the problem is with the counter in your SDG2122X......
These freebee counters in ARB generators are good at producing a lot of digits on the display, but do not have the real resolution.

Looking in the SDG2122X spec heet, counter resolution is not mentioned at all... (nor gate time).

For looking at detailed stability of a good 10MHz OCXO, one needs ppb (parts per billion) resolution.
1ppb @ 10MHz is 10mHz. So you need at least 10 digits resolution. Only the better dedicated counters will have this resolution. (with a reasonable gate time).

regards, Gertjan.

How accurate it is, I dunno, but it shows an average of -0.028ppm in the screenshot I shared. After running 24 hours, it's currently showing +15ppb. 🤷

I'll try HKJ's math suggestion.

Thanks,
Josh

[KungFuJosh]

But instead of fighting to get as many decimals as possible, what about using math and subtract 10M from the value. Then you do not need many digits to see variations.
Internal precision in TC is double, that is about 15 digits.

I tried guessing, but nothing worked. How do I subtract 10M from SDG2122X.Frequency2?

I tried subtracting 9999999 instead, and the result is just 1.00000. Does that mean it doesn't output the resolution needed to see anything?

Thanks,
Josh

[HKJ]

I tried guessing, but nothing worked. How do I subtract 10M from SDG2122X.Frequency2?

I tried subtracting 9999999 instead, and the result is just 1.00000. Does that mean it doesn't output the resolution needed to see anything?

It looks that way and I supposed the problem with subtracting 10M was that the result was 0.

[KungFuJosh]

I tried guessing, but nothing worked. How do I subtract 10M from SDG2122X.Frequency2?

I tried subtracting 9999999 instead, and the result is just 1.00000. Does that mean it doesn't output the resolution needed to see anything?

It looks that way and I supposed the problem with subtracting 10M was that the result was 0.

Exactly. Oh well, thanks for your help either way.

[Taimer81]

Can you tell me what the problem might be? .

[HKJ]

Can you tell me what the problem might be? .

A B and a non-B model.

In the file  "Chauvin Arnoux CA 5293.txt" try adding this section, it will add a non-B model.

#metadef
#idString Metrix,MTX3293,
#name Metrix MTX3293
#handle MTX3293
#replaceText Brand Metrix

Check it out and report back if everything works (Maybe the non-B model uses a slightly different protocol and some stuff do not work).

[Taimer81]

Can you tell me what the problem might be? .

A B and a non-B model.

In the file  "Chauvin Arnoux CA 5293.txt" try adding this section, it will add a non-B model.

#metadef
#idString Metrix,MTX3293,
#name Metrix MTX3293
#handle MTX3293
#replaceText Brand Metrix

Check it out and report back if everything works (Maybe the non-B model uses a slightly different protocol and some stuff do not work).

I did as you suggested, but the same error occurred. They write that they all have the same management MTX328x. Maybe there is an error in the file? Please help.

[HKJ]

I did as you suggested, but the same error occurred. They write that they all have the same management MTX328x. Maybe there is an error in the file? Please help.

Sorry, I did not look careful enough, your meter do not report as a Metrix, but as a Chauvin Arnoux

#metadef
#idString Chauvin Arnoux,MTX3293,
#name Metrix MTX3293
#handle MTX3293
#replaceText Brand Metrix

[Taimer81]

I did as you suggested, but the same error occurred. They write that they all have the same management MTX328x. Maybe there is an error in the file? Please help.

Sorry, I did not look careful enough, your meter do not report as a Metrix, but as a Chauvin Arnoux

#metadef
#idString Chauvin Arnoux,MTX3293,
#name Metrix MTX3293
#handle MTX3293
#replaceText Brand Metrix

I changed some settings. I deleted the device and re-added it. But the result was the same . Maybe I made a mistake?

[HKJ]

I changed some settings. I deleted the device and re-added it. But the result was the same . Maybe I made a mistake?

The text after #idString must match the answer from the meter, try copying the text shown on the screen and insert it (Including HV version and everything else).

Usually there is a comma after the name, but it must be missing here.

[Taimer81]

I changed some settings. I deleted the device and re-added it. But the result was the same . Maybe I made a mistake?

The text after #idString must match the answer from the meter, try copying the text shown on the screen and insert it (Including HV version and everything else).

Usually there is a comma after the name, but it must be missing here.

;; CP2102 USB to UART Bridge Controller (COM5) Device "Chauvin Arnoux,MTX 3293,, HV B  FV 1.08," do not match answer: "Chauvin Arnoux,MTX 3293,, HV B  FV 1.08"

#metadef
#idString Chauvin Arnoux,MTX 3293,, HV B  FV 1.08,
#name Metrix MTX3293
#handle MTX3293
#replaceText Brand Metrix

[HKJ]

;; CP2102 USB to UART Bridge Controller (COM5) Device "Chauvin Arnoux,MTX 3293,, HV B  FV 1.08," do not match answer: "Chauvin Arnoux,MTX 3293,, HV B  FV 1.08"

#metadef
#idString Chauvin Arnoux,MTX 3293,, HV B  FV 1.08,
#name Metrix MTX3293
#handle MTX3293
#replaceText Brand Metrix

If cannot really get closer to an answer, TC tries to match the two strings and fails.
There may be an extra space somewhere or a special ascii character.

[Gertjan]

I think the problem is the comma at the end of the #idString line..... (the last comma is not in the device answer)

From the documentation:

The idstring is the brand name and the device name and must exactly match the text returned from the device, the two last parameters must not be included.

A bit of reading in the documentation answers a lot of questions..... 

[Taimer81]

I think the problem is the comma at the end of the #idString line..... (the last comma is not in the device answer)

From the documentation:

The idstring is the brand name and the device name and must exactly match the text returned from the device, the two last parameters must not be included.

A bit of reading in the documentation answers a lot of questions..... 

You did well. Everything worked out.

[Gertjan]

Hi HKJ,

I added support for the Tektronix DMM4040 and DMM4050 to my Fluke 8845A / 8846A configuration file.

The differences between the Fluke and Tek meters are mostly cosmetic.


IMG_3041  Tek DMM4050-2000pix.jpg

I also updated the "SI" in the #value lines to SI5...SI8. Now all available digits will be shown in the Table and Current values tabs by default. (In some modes the 8846A will send a digit more than it is showing on it's display)

I found that "SI5" is functioning the same as "SI". So "SI" is capped to 5 digits. (the documentation suggests 7 digits...)

Please find the updated definition file here: Fluke8845A-8846A.zip

If you prefer a separate definition file for Tek DMM4040/DMM4050, I will create it for you.

Regards, Gertjan.