Probably a silly question...

[Rick Densing]

How about this one:

or

 

[drewcrew6]

I have also had problems with thermocouples that didn't have a ground on the power supply common. Is an older post actually.

I agree with Tim Moulder that grounding it (I'm talking at point of source) will reduce noise in the electrical panel. If its left to float the noise can induce some odd voltages.

To me it makes sense that not grunding it would mean more noise not less. Because any noise now has a place to run to and get "drained" intead of causing a "safe" voltage to be able to float and possibly raisng that low potetial to of a potetial of any supply voltage on premises.

It was stated that not all areas are at the same ground potetial. I agree but the difference should be minimal with proper grounding bonding procedures. But how is that diffence being transmitted back into the plc?

All this brings up another question but I'll start a new thread!


Drewcrew6

 

[rogerhollingsworth]

Grounding

Grounding
In 3-phase, 480V applications, large inverters, drives, and motors can introduce significant distortion and interference on the power lines. Using a single grounding system can, and will, introduce unwanted problems. Depending on the application, system integrators choose between a single ground point that ties all AC and DC voltages together, or separate grounds for the AC and DC voltages. In creating a common ground, engineers typically connect the incoming AC voltage ground with the negative leg of the DC voltage.

Of course, this common ground compromises the isolation between the AC and regulated DC and ultimately negates the requirement of an isolating transformer. Because a power supply that uses transformer isolation achieves single (and even double) isolation, the common-ground approach comes further into question.

By using the right circuit layout and components that feature touch-proof connections, system designers can create insulating boundaries between different voltages. These boundaries maintain the integrity of the power supply and all of the devices connected to its DC voltage output. In short, it's possible to design a safe system with true isolation between AC and DC and two separate grounding systems.

All DC power supplies offer a plus (+) and minus (-) output connection that is isolated from the AC input. With the DC circuit, a common plus or minus connection is possible using touch-proof, DIN-rail terminal blocks that allow voltage isolation from the DIN-rail itself. By keeping different voltages physically and visually separated, the need to keep a common ground is no longer important. The physical distance between the various voltages negates the possibility of creating a dangerous situation, unless it's intentional.

Because 24VDC would normally power analog 4mA to 20mA loops, or 24VDC relays, contactors, and proximity sensors, it's important to make sure that all metal casings and shields are tied to ground (unless otherwise specified). However, the AC ground should in no way be connected to either the positive or negative connection of the 24VDC power supply.

 

[Peter Nachtwey]

I am in Roger's camp on this one.

In addition , analog input devices should have a separate isolated power supply. Sharing the the digital on/off power supply will result in noisy analog signal. This will prevent one from using the differential gain in the PID.

Oakiebob also brought up naval systems. Our boat's 3 phase power was routinely checked for shorts to the hull. ( ground ) I know this is AC and not DC but then we also had isolation transformeters on top of that so the DC was floating too. In an emergency we could run with one phase grounded, but obviously not with two. I just wanted to point out that it is not uncommon to have everything float as Roger suggests.


Noise? Our analog inputs use differential inputs have a input+, input- and common and use use the difference between V+ and V-.
See the attached file. If done properly out customers get very clean signals.

 

[fareast]

Hi,
Whilst I never use a chassis a part of a power supply return , I do like to see -Ve power grounded , as you said , it makes testing on a machine so much easier if you can test between chassis and say PLC input or other device . I have just completed a big Profibus job , and as you may know , each leg of a Profibus network (shield) should be grounded to the cabinet . I have found that I have never had much problem using a grounded power supply in fact , I now recall that Siemens S7 analogue inputs must have the M terminal grounded and tied to power supply -ve if there is any PD detected.