a isolated UPS energizing Sensors, PC, floating signal problem?

[askay]

Hello All,

I have my  4 AC motors connected to main power supply of 380V. Torque sensors are attached to it with NI DAQ and PC controller test bench.

Now what I have is a 220V AC/AC UPS with built in isolator transformer. This UPS has its clear AC power and an isolated gound new ground.  I use this to supply  industrial PC (NI DAQs are installed) and small AC/DC converter (220AC/24DC V) to supply torque amplifier. Basically everywhere within control cabinet (signal level)

The question is:  Can I safely use the isolated ground in my measurements without touching the field ground?

The same ground is attached to PC power supply, used as Signal ground, used signal shield ground. Basically I use the ground everywhere as a Signal ground within control unit, i.e. PC, sensor Amps. Etc.)

I read in the NI helps about floating signal (I am supposing that I have floating signal now by having a UPS, I know that it has ground, makes me confusing )

Do you think that I have a floating signal ? I have a new ground reference. I think it is not a floating signal, i.e. floating ground. I measure the signal differentially.

Last question, Do you thing isolated  signal grounds and Field ground points should be connected somewhere at the mains supply, so that all of them flowing to the earth? or not at all touch to isolated signal ground?

Is there any risk/recommendation that I should follow?

I have attached a schematics of what explained above.

Best regards,

Desmond,

[Efrain_G]

Hi Desmond,

Thank you posting a screenshot of your system configuration. In answer to you rquestions: "Can I safely use the isolated ground in my measurements without touching the field ground?"

You must keep in mind that in order to build a safe electric system you must ensure that the voltage between channels (when using a differential measurement) and the common voltage from your channels to ground must be within the specified safety ranges of the manufacturer of your DAQ board. In your case, you must ensure that the voltage between signal (+) and signal (-) is within the "channel to channel" safety specifications. Also, the voltage between any of the previous channels and your UPS ground is within the "common to earth" safety specifications.

"I read in the NI helps about floating signal (I am supposing that I have floating signal now by having a UPS, I know that it has ground, makes me confusing )

Do you think that I have a floating signal ? I have a new ground reference. I think it is not a floating signal, i.e. floating ground. I measure the signal differentially."

The short answer for this is that it depends. It is all relative. A voltage measurment is by definition a measurement relative to a reference. For must purposes, the Earth ground is used as an absolute reference such that it is considered to be at 0V and all other voltage levels are referenced to that. However, it is also possible that another reference could be chosen to be our 0V reference.

This means that from the Earth ground reference perspective, your entire system is "floating" because it is not directly connected to Earth ground. However, from the perspective of your PC which has the DAQ board, the ground reference is your UPS ground. Your sensor power supply is also using your UPS ground as the "ground reference". Therefore, if you reference signal (+) and signal (-) to your UPS ground, you will have created a "ground referenced" measurement, and if you do not reference them to your UPS ground they will be considered "floating signals".

I recommend you to take a look at the Field Wiring and Noise Considerations for Analog Signals tutorial.

"Last question, Do you thing isolated  signal grounds and Field ground points should be connected somewhere at the mains supply, so that all of them flowing to the earth? or not at all touch to isolated signal ground?"

Here we go back to the concept of relative voltages. Since everything that you have connected to your UPS is technically floating relative to Earth ground, that means that there could be a voltage difference buildup between the two references. So anything connected to your UPS might actually be at a higher voltage than what you expect relative to Earth ground. Therefore, you must be careful when connecting anything from your UPS reference to your main power supply. As a matter of fact I would try to avoid it.

I hope this helps,

[askay]

Hi Efrain,

Would you suggest powering the control &sensor devices via isolator transformer  that is not providing ground isolation but power isolation. I still use the same ground as in mains grounding.(My aim is just to supply a clean AC to the instrumentation(sensors, industrial PC) side since I have power swtc. devices like motor drives connected to the same mains in a test bench).

Also,  Do you think it is good idea to have a seperate grouding bed (ground termination bar) for signal and power grounding, then finally getting them together somewhere at a star point. Should I keep these grounding bed as close as possible to eachother in a test bench with a 1x1x2 m size.

One more question, Considering an industrial PC controlling a motor,  If I drive enable an actuator  externally, and then power up the industrial PC, I see an unexpected behavior of the motor (like PC DAQs sending 3.3V or something  as a reference signal) . This is happening before everything is under control of a test bench GUI. I heard that it is due to the fact that before DAQs chips are energized and control its I/Os, DAQ's I/O s are not controllable. One of my friend suggested me to to consider an external power supply for DAQs live 5V. So that DAQs are energized via ups, battery, before the control PC.

Thanks for your helps,

Desmond.

thanks for your help.

[Efrain_G]

Desmond,

In answer to your questions:

"Would you suggest powering the control &sensor devices via isolator transformer  that is not providing ground isolation but power isolation. I still use the same ground as in mains grounding.(My aim is just to supply a clean AC to the instrumentation(sensors, industrial PC) side since I have power swtc. devices like motor drives connected to the same mains in a test bench)."

Yes, I would suggest to have your sensor devices and DAQ device referenced to the same ground, which in this case is the isolator transformer.

"Also,  Do you think it is good idea to have a seperate grouding bed (ground termination bar) for signal and power grounding, then finally getting them together somewhere at a star point. Should I keep these grounding bed as close as possible to eachother in a test bench with a 1x1x2 m size."

Just make sure that you are not creating different ground levels, you should have the same ground level for everything.

"One more question, Considering an industrial PC controlling a motor,  If I drive enable an actuator  externally, and then power up the industrial PC, I see an unexpected behavior of the motor (like PC DAQs sending 3.3V or something  as a reference signal) . This is happening before everything is under control of a test bench GUI. I heard that it is due to the fact that before DAQs chips are energized and control its I/Os, DAQ's I/O s are not controllable. One of my friend suggested me to to consider an external power supply for DAQs live 5V. So that DAQs are energized via ups, battery, before the control PC."

What is the DAQ device that you are using? To be honest, I think this behavior is independent from your power source and grounding level.