Seite 1 von 1

Doubts about conversion

Verfasst: Di 18. Sep 2018, 09:50
von imuguruza
Hi there,

I have a novice doubt about my setup:
I have a 50 N·M torque sensor (TS110a) and a GSV-3LS for amplifying and getting digital values from a serial communication.

I have doubts about the translation I need to make. In the GSV-3LS protocol datasheet explains that there are two bytes to take into account, the "highbyte" and the "lowbyte". And that, depending if you are using monopolar or bipolar mode, the "0" value changes.

As while the torque sensor is in my desktop, I'm reading a hexadecimal value of ~ 0x80 0xBA, I assume that I'm reading a bipolar mode.

In this case, if I read the next maximum values, the should match with the torque sensor max, like this:
0xFFFF => 50 N·M
and
0x0000 => -50 N·M
and
0x8000 => 0 N·M

So in code, I should use this relation to translate the hexadecimal to physical value. Is this right or do I need to use the "Input Signal" equation highlighted in yellow (attached)?

Thanks for the clarification in advanced,

imuguruza

Re: Doubts about conversion

Verfasst: Di 18. Sep 2018, 12:02
von QME
Dear imuguruza,
there exist different ways to get the measurement values:

#your way

Example:
You have a sensor with 1,2mV/V @ 50Nm and a GSV-3 with 2mV/V.

1) All replies of the bridge amplifier with A5 at the beginning are measuring values.
2) You take the two following bytes. The first one is the [highbyte] and the second one is the [lowbyte].
3) You convert [highbyte] to a decimal and the [lowbyte] to a decimal
4) You insert these values in the first formula you've found.

Code: Alles auswählen

output value (dez) = [highbyte] * 256 + [lowbyte]


5) you put the result of the first formula into the second formula

Code: Alles auswählen

input signal(dez) = [output value] - 32768 / 32768 * 2,10


6) Now you have your measurement value in mV/V.

7) You need the [scaling factor] of the GSV-3 to convert the mV/V in a display value in Nm.

You can set the right factor with the scaling dialog in GSVMulti for example, maybe we have set the settings already, mostly they are saved as user1 and user2.
https://www.me-systeme.de/en/software/gsvmulti/scaling


8)

Code: Alles auswählen

[final result in Nm] = [output value] / 2 * [scaling factor]


#using the DLL
If its possible, for your project you can also use our Windows-DLL. Some of the following projects/documents might be helpful for that.

https://www.me-systeme.de/en/software/programming/megsv
https://www.me-systeme.de/en/software/p ... ic-express

#measuring in text mode
up to 100 measurement values per second are possible

If you set the bridge amplifier to text mode (as Ascii) you've got the real measurement value without any extra mathematics.
The output character string corresponds to the indicator on display and may be shown e.g. with a terminal program (like docklight).
The data format is in the delivery state:
+50.0000 Nm CRLF

Re: Doubts about conversion

Verfasst: Di 18. Sep 2018, 16:45
von imuguruza
Thanks for your answer,

I will follow the first option.
Now I understand better, what I don't see clearly is the final step.

The TS110 has as "Rated Output" 1mV/V.
I have a test report, that among other values, says "rated load = 1.2130 mv/V".

Should I use this data to calculate the Scaling?

I understand that should be the next:

Code: Alles auswählen

Scaling Factor = (Physical Scale / Electrical full scale output) * Input Range = (50 N·m/1.213mV/V)*2mV/V = 84,4402


Is this math right?

Thanks,

imuguruza

Re: Doubts about conversion

Verfasst: Mi 19. Sep 2018, 09:41
von QME
Hello imuguruza,

Yes, perfectly. The best way is using the value from the test report (highest accuracy). Like this: (1.213 mV/V).

Code: Alles auswählen

Scaling Factor = (Physical Scale / Electrical full scale output) * Input Range = (50 N·m/1.213mV/V)*2mV/V = 84,4402

Re: Doubts about conversion

Verfasst: Mi 19. Sep 2018, 09:45
von imuguruza
Thanks for your help

I will proceed to code this math.

Regards,

imuguruza