Nicolas:
Before you give up entirely, please entertain the proposals below. Both options require you to do the following.
That aside, purchase applicable hall effect sensors from Polou or however you spell it. Add the required capacitor and a a 10K 10 Turn trimmer potentiometer that you place on the output, so you can convert the full scale value to engineering units. 30 A full scale would translate to 300 mV. If your really lucky, you might be able to reduce the scale to read percent, i.e. 0.500 is 50% of full-scale.
We would have to look at sensor scales and the full scale current desired and possibly use the 1-2-5 sequence, or say 6A full scale. So, you get readings like 25% of 6 for real currents or say 0-100% or 0-1 V. It's still a number you can use.
Your sensor now outputs 0-1.0 volts or 0-100% of the full scale you trimmed it too.
Now supply every hall effect sensor with a single 5V supply, but it would be upgraded to the hard way later.
Hard: DC-DC converter. 5V to isolated 5V. Easy: 5V power supplies. Drawback: bigger parts. All the sensors need is about 0.2 A at 5V per sensor. The sensor output would be isolated, so in this case, they could share the same 5V supply.
Let's say your looking at $15.00 per device being measured an 1 power supply that needs to supply about 25 mA * n devices.
Now, you have two options:
Option 1.
Purchase 2 voltmeters. With Digital Panel Meters (DPM's), you usually get the ability to set the decimal point and the calibrated range. Assume for now, that the meters can read 1 Volt and would indicate in percent based in the info earlier.
Each DPM also gets it's own isolated 5V power supply.
Purchasing two separate meters. Biggest drawback? To much panel space.
Option #2
Purchase the same voltmeter/current meter you were thinking about. The one with the 75 mV input. There was stuff I read on the net, that you can basically short the proper pins, so that the meter reads zero, i.e. short the terminals that would read the 75 mV from the shunt. So, the ammeter always reads zero, Later you can upgrade.
Now with option #2, you can add a DPDT switch at each meter with a Volts-none-% max torque. Current is roughtly proportional to torque, so lets label the switch something meaningful.
That DPDT switch, would connect the measuring leads of the voltmeter to the voltage being measured or the output of the current sensor.
Drawback: The meaningless zero. The extra switch, but it doesn't take up a lot of room. The mods are not proven.
Advantages: Later when the "ADD-ON" is proven, you can plug the switch hole with a plastic rivet or chrome plug or even a meaningless screw head.
The point is, everything purchased would be used later. My scenereo would replace the single sensor supply with a DC-DC converter for each meter.
Later, that simple divider from say 2.5 V to 1V which was say a 10 A full scale meter would have to be changed to < 75 mV full scale. The potentiometer could proabably stay.
---
I do find the project intriguing. Do, you see a benefit to fellow CNCers to turn a cheap meter into a high side sense? I think your looking at monitoring the X, Y and Z axis. I see nothing wrong in using whatever the unit is.
When the voltmeter isn't designed to read say 199.9 mV full scale and have an adjustable decimal point, getting engineering units can be tough. I did just that once with a slight glitch and it was the multiple ground issue, but it worked stand-alone and not computer controlled. I had a calibrate switch, which would present 5V to the meter input and you would adjust the potentiometer for the correct engineering unit and then use a DIP switch to set the decimal point. It worked out suprisingly well.
We had this habit of using any mass flow controller and calibrating it on the system in engineering units for whatever gas or mixture we were using. If it as 15.3 slm FS or 16.3 slm FS, the calibration was correct and we didn't have to fuss with the individual units. They were initially calibrated for Nitrogen or Argon.
In this design 2.5 V at the sensor is full scale. So, we start getting complicated again. If this were a Printed Circuit Board (PCB), it could be a calibration jumper.
Whatever you do, Don't follow the video.
I could see some use for a 30 V, 2 to 5 A meter.
If you haven't noticed, I like to sometimes put all of the likes on the table and all of the options even if they are silly. Then reduce them to something manageable or something that "could" be upgraded. Anticipating and planning pay dividends.
I don't mind going full circle either. By doing so, you have said this isn't important.
I hope you learned one thing. Connecting ground to different potentials creates trouble. The best way to eliminate ground loops is to design them out. You have to realize that it's not a perfect system. There is Earth or protective ground and it's supposed to handle only fault currents. Ground is also a reference and it might be called analog ground. It's also quiet. Digital ground is considered a noisy ground where logic signals are returned to.
During construction, these grounds are connected within themselves. Then, usually at on point on the chassis, they are connected together.
Before you give up entirely, please entertain the proposals below. Both options require you to do the following.
That aside, purchase applicable hall effect sensors from Polou or however you spell it. Add the required capacitor and a a 10K 10 Turn trimmer potentiometer that you place on the output, so you can convert the full scale value to engineering units. 30 A full scale would translate to 300 mV. If your really lucky, you might be able to reduce the scale to read percent, i.e. 0.500 is 50% of full-scale.
We would have to look at sensor scales and the full scale current desired and possibly use the 1-2-5 sequence, or say 6A full scale. So, you get readings like 25% of 6 for real currents or say 0-100% or 0-1 V. It's still a number you can use.
Your sensor now outputs 0-1.0 volts or 0-100% of the full scale you trimmed it too.
Now supply every hall effect sensor with a single 5V supply, but it would be upgraded to the hard way later.
Hard: DC-DC converter. 5V to isolated 5V. Easy: 5V power supplies. Drawback: bigger parts. All the sensors need is about 0.2 A at 5V per sensor. The sensor output would be isolated, so in this case, they could share the same 5V supply.
Let's say your looking at $15.00 per device being measured an 1 power supply that needs to supply about 25 mA * n devices.
Now, you have two options:
Option 1.
Purchase 2 voltmeters. With Digital Panel Meters (DPM's), you usually get the ability to set the decimal point and the calibrated range. Assume for now, that the meters can read 1 Volt and would indicate in percent based in the info earlier.
Each DPM also gets it's own isolated 5V power supply.
Purchasing two separate meters. Biggest drawback? To much panel space.
Option #2
Purchase the same voltmeter/current meter you were thinking about. The one with the 75 mV input. There was stuff I read on the net, that you can basically short the proper pins, so that the meter reads zero, i.e. short the terminals that would read the 75 mV from the shunt. So, the ammeter always reads zero, Later you can upgrade.
Now with option #2, you can add a DPDT switch at each meter with a Volts-none-% max torque. Current is roughtly proportional to torque, so lets label the switch something meaningful.
That DPDT switch, would connect the measuring leads of the voltmeter to the voltage being measured or the output of the current sensor.
Drawback: The meaningless zero. The extra switch, but it doesn't take up a lot of room. The mods are not proven.
Advantages: Later when the "ADD-ON" is proven, you can plug the switch hole with a plastic rivet or chrome plug or even a meaningless screw head.
The point is, everything purchased would be used later. My scenereo would replace the single sensor supply with a DC-DC converter for each meter.
Later, that simple divider from say 2.5 V to 1V which was say a 10 A full scale meter would have to be changed to < 75 mV full scale. The potentiometer could proabably stay.
---
I do find the project intriguing. Do, you see a benefit to fellow CNCers to turn a cheap meter into a high side sense? I think your looking at monitoring the X, Y and Z axis. I see nothing wrong in using whatever the unit is.
When the voltmeter isn't designed to read say 199.9 mV full scale and have an adjustable decimal point, getting engineering units can be tough. I did just that once with a slight glitch and it was the multiple ground issue, but it worked stand-alone and not computer controlled. I had a calibrate switch, which would present 5V to the meter input and you would adjust the potentiometer for the correct engineering unit and then use a DIP switch to set the decimal point. It worked out suprisingly well.
We had this habit of using any mass flow controller and calibrating it on the system in engineering units for whatever gas or mixture we were using. If it as 15.3 slm FS or 16.3 slm FS, the calibration was correct and we didn't have to fuss with the individual units. They were initially calibrated for Nitrogen or Argon.
In this design 2.5 V at the sensor is full scale. So, we start getting complicated again. If this were a Printed Circuit Board (PCB), it could be a calibration jumper.
Whatever you do, Don't follow the video.
I could see some use for a 30 V, 2 to 5 A meter.
If you haven't noticed, I like to sometimes put all of the likes on the table and all of the options even if they are silly. Then reduce them to something manageable or something that "could" be upgraded. Anticipating and planning pay dividends.
I don't mind going full circle either. By doing so, you have said this isn't important.
I hope you learned one thing. Connecting ground to different potentials creates trouble. The best way to eliminate ground loops is to design them out. You have to realize that it's not a perfect system. There is Earth or protective ground and it's supposed to handle only fault currents. Ground is also a reference and it might be called analog ground. It's also quiet. Digital ground is considered a noisy ground where logic signals are returned to.
During construction, these grounds are connected within themselves. Then, usually at on point on the chassis, they are connected together.
