Go Down

### Topic: Voltage Reading With Arduino (Read 4070 times)previous topic - next topic

#### Dexterslab

##### Jul 25, 2013, 12:16 pmLast Edit: Jul 25, 2013, 12:19 pm by Dexterslab Reason: 1
Hi everybody;

I am working on a simple project. But I have problem.

I am trying to read a voltage with arduino uno. I have two pins as a + and - and when I read voltage by voltmeter from that two pins, I am reading a voltage around 3v. That is right value.

But when I read the voltage between ground and + pin that is not 3v. I am reading nonsense value like 0.60-0.70V. And when I read the voltage between grnd and - pin that is not 0V that is nonsense value again.

I read something like floating voltage.

So it means I cannot connect + pin to arduino's analog input. How can i connect it? What should I put between arduino and voltage source?

Thank you so much...

#### Grumpy_Mike

#1
##### Jul 25, 2013, 12:21 pm
Quote
I read something like floating voltage.

Yes it sounds like you do not have a floating voltage, you need one to measure it correctly.

What do you get when you measure with your voltmeter the voltage between the arduino ground and each of the two wires you have in turn.

#### Dexterslab

#2
##### Jul 25, 2013, 12:43 pm

Quote
I read something like floating voltage.

Yes it sounds like you do not have a floating voltage, you need one to measure it correctly.

What do you get when you measure with your voltmeter the voltage between the arduino ground and each of the two wires you have in turn.

I am trying to make my own THC for Plasma Cutter. That is coming from the divided arc volts of plasma cutter(powermax 45). So when the height of torch is too much the voltages goes up. And when the height of torch(during cut) is too low, arc volts goes down.

Actually the arc voltage can be between 50-250 volts. But I have an internal voltage divider inside of powermax. And that voltage divider's ouput ratio is 50:1. So when the arc voltage 150 I am reading 3V from divided arc volts pin.

And my ground was arduino's 0V grnd during test. 3V between + and - pins. But nonsense value like 0.60-0.70V between + pin and arduino's 0V grnd. So I did my old test with arduino's ground.

#### Grumpy_Mike

#3
##### Jul 25, 2013, 01:05 pm
Quote
But nonsense value like 0.60-0.70V between + pin and arduino's 0V grnd.

Those are not nonsense values those are the values you get between your voltage and what ever ground the arduino is on.

Quote
And my ground was arduino's 0V grnd during test. 3V between + and - pins.

I don't understand that.
You can measure between the + and - without getting the arduino involved at all.
Then you can measure between arduino ground an - and another between arduino ground and +

You can run the arduino off batteries and connect the - to the arduino ground. But then you need an isolated USB connection or serial connection to look at your results on a terminal, this can make debugging hard.

#### Dexterslab

#4
##### Jul 25, 2013, 01:29 pmLast Edit: Jul 25, 2013, 01:34 pm by Dexterslab Reason: 1
I don't understand that.

I think that is about my english, really so sorry.

I trying to say that: I read values between earth grnd(0V) and + pin and - pin. So earth ground and arduino's ground same thing, isnt it? I wanna be sure about the I wrote test results rightly.  I read that values;

(where Voltmeter pin connected to) --- (where Voltmeter pin connected to) =(result)

(+ pin of plasma output) --- (- pin of plasma output) =3V
(+ pin of plasma output) --- (grnd) =around 0.60-0.70
(- pin of plasma output) --- (grnd) =around 0.60-0.70

You can run the arduino off batteries and connect the - to the arduino ground. But then you need an isolated USB connection or serial connection to look at your results on a terminal, this can make debugging hard. .

So should I connect the - pin from plasma to Power input's grnd? And just use the rx and tx pins from usb. without any other power connection from usb.

Notice:The value of between - pin and ground is usually changing. And probably there are so much noise and interference at the - pin of plasma's output.

#### Grumpy_Mike

#5
##### Jul 25, 2013, 01:54 pm
Quote
So earth ground and arduino's ground same thing, isnt it

No not necessarily, it depends on how the arduino is powered. If it is through USB then it depends on how the host is being powered. A lot of systems do not have a mains ground point on them.

Quote
And just use the rx and tx pins from usb. without any other power connection from usb.

That is not going to work because the data pins also need a common ground reference point.

One explanation of the differences is if the signal is noisy or is floating. The meter will tend to average where as the arduino will read a sample anywhere in the cycle.
In addition to the measurements you have taken, what do you get with the - and arduino ground connected together on one lead and the + wire on the other. It looks like the voltage from the plasma is not isolated and the arduino is not isolated and therefore you can't get a common ground.

What you need to do is to look at your signal on an oscilloscope, that will tell you what is going on.

Another way round this is to have a voltage to frequency converter on the plasma end feeding it's output to an opto isolator going into an arduino pin where you measure the frequency and hence the voltage.

#### Dexterslab

#6
##### Jul 25, 2013, 03:00 pm
Ok Mike will do all of that tests. But I think firstly we need to look that;

Another way round this is to have a voltage to frequency converter on the plasma end feeding it's output to an opto isolator going into an arduino pin where you measure the frequency and hence the voltage.

That was my first idea. I think that is best one. Because we must isolate arduino. But I am not sure which optocoupler should I try? Have u got an any idea about it? I look so much datasheets but that is hard to select.

Input of optocoupler will be 0-7V analog and you know arduino's analog input's miliamps. And we dont want any delay. What is your idea about it?

#### Grumpy_Mike

#7
##### Jul 25, 2013, 04:14 pm
Quote
Input of optocoupler will be 0-7V analog and you know arduino's analog input's miliamps.

No you can't transfer a signal in an analogue manor trough an opto as the transfer function is very non linear.
You need to change the voltage into a frequency then you send that frequency through the opto. So the frequency you run at is going to determine the opto you need.
Something like the LM2907 / LM2917 to convert your voltage into a frequency. This needs powering separately from the arduino, possibly with a battery.

#### Dexterslab

#8
##### Jul 25, 2013, 05:12 pm

Quote
Input of optocoupler will be 0-7V analog and you know arduino's analog input's miliamps.

No you can't transfer a signal in an analogue manor trough an opto as the transfer function is very non linear.
You need to change the voltage into a frequency then you send that frequency through the opto. So the frequency you run at is going to determine the opto you need.
Something like the LM2907 / LM2917 to convert your voltage into a frequency. This needs powering separately from the arduino, possibly with a battery.

Ok. I will do some researches about it.

#### dc42

#9
##### Jul 25, 2013, 05:26 pm
I would use a digital isolator chip such as the ADUM6401, and an SPI ADC such as MCP3201 (or a more expensive ADC with a built-in voltage reference). The digital isolator chip both isolates the ADC from the Arduino and feeds power to the ADC. If you use voltage to frequency conversion and an opto isolator instead, you still need to power the voltage to frequency converter chip somehow.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

#### Dexterslab

#10
##### Jul 26, 2013, 10:33 am

I would use a digital isolator chip such as the ADUM6401, and an SPI ADC such as MCP3201 (or a more expensive ADC with a built-in voltage reference). The digital isolator chip both isolates the ADC from the Arduino and feeds power to the ADC. If you use voltage to frequency conversion and an opto isolator instead, you still need to power the voltage to frequency converter chip somehow.

Sounds like good idea. I ll have a look the datasheets.

Thank you so much.

Go Up