Grounding question

The DC power input for PLCs generally is +V, 0V and ground. Usually, I connect both 0V and ground pin to the chassis.
Is this common practise?

I don't think I've ever seen one labeled '0V'. I typically go to + and - of my power supply, and then take the ground wire to a ground block or lug somewhere in my panel. Now, the way you do it should work just fine as long as the negative of the power supply and the chassis are tied to ground too. Thats not always done on the panels I've worked on.


-jeff

backtoplc said:

The DC power input for PLCs generally is +V, 0V and ground. Usually, I connect both 0V and ground pin to the chassis.
Is this common practise?

Click to expand...

+V and 0V means there should be the potential difference as required power between +V and 0V. Hence, you should NOT connect 0V to ground. 0V means the negative terminal of your power supply which should be connected to 0V and the positive terminal of the power supply is to be connected to +V.

I never connect the "0" or "-" terminal of switchmode power supplies to ground if they are powering PLCs and/or instrumentation and never had a problem. I have had problems when I have connected them to ground though as this disabled the isolation in the switchmaode due to the way it was designed - got noise.

O Volts and ground connection

Rather than tie the zero volt terminal directly to ground, I suggest you tie the zero volt terminal to ground thru a 10 meg ohm resistor. This way, they are at the same voltage potential with no current between them to cause noise effects

Thats all good and well until you hook your power supply to a device that is internally grounded. I head off the problem at the panel by grounding the 0V terminal.

Alaric said:

Thats all good and well until you hook your power supply to a device that is internally grounded. I head off the problem at the panel by grounding the 0V terminal.

Click to expand...

Finally a voice of reason...people don't realize how many devices do this or have small r or rc circuits tied to ground. Ground it and you know 100% what you got.

Nick

I don't think that there is a definitive answer to this. You need to look at the whole system design. Telecomm's systems used to be traditionally positive earth so connecting the -ve to ground on these would be interesting. One thing I do know, however, is that if you use analogue inputs on AB FlexIO and do not ground the -ve you will definitely see noise.

Andybr

Why is there a need to tie the AC ground to DC gnd? This might induce un-welcome AC ground noise into sensitive analog i/o if there is any in the system. I would vote to leave both the AC ground and Dc ground isolated where ever possible.

PLC newbie said:

Why is there a need to tie the AC ground to DC gnd? This might induce un-welcome AC ground noise into sensitive analog i/o if there is any in the system. I would vote to leave both the AC ground and Dc ground isolated where ever possible.

Click to expand...

Why? Is there some kind of a problem with the integrity of your ground?

Yeah, you cannot really isolate them. Ground is ground is ground is ground (an electrician taught me that). If any two "ground" points in your facility aren't tied together (get an ohmeter and measure continuity between them) then you have ground issues. If there's no continuity between the AC ground and the DC ground, you have probably set up a floating ground on one or the other, and while I don't actually understand all the reasons thats bad (someone smarter than me please fill us in), I do know it'll come back to bite you if you last long enough.


-jeff

I have seen this done both ways and have yet to see a serois performance difference between them. The machines I work on don't have 500 foot long analog runs, though.

The warning about devices that tie their DC common to ground internally is a very valid warning. If you are specifically looking for a floating DC system you will need to be very aware of this.

The one time I saw issues with a floating DC supply dealt with an RS-485 based communication link. One would think that this uses a differential input and is somewhat immune to common mode offset. This is only partially the case. We connected two separately powered devices together via the comm link and had all sorts of trouble getting solid communication. Once we powered both off the same supply the issues went away. We came to find out that the receiver we were using could only handle a limited common mode offset. Had both supplies been grounded we wouldn't have had this issue. Similar issue can occur with certain devices that have internal components explicitly grounded.

Keith

i ground where ever possible - and expect everyone working with
me to do the same. it is true that one must watch for devices
with internal connections to ground etc. (like with old G9S
safety relays which had PE terminal tied to one of power terminals).
the point is one should take this seriously, read the spec and
then check it out - before goes out to customer.

it is not professional to pass ticking timebombs... just closing
an eye on issues like this is the same thing...
do you stand behind your designs or hope that things will eventually
fall in places...somehow?

also there are engineers with mortal fear of not just grounding
but ground loops, analogs, testing circuits of their own creations etc.
these things are part of job, one can try avoiding them or learn
how to do them properly. grounding is part of electrical code in
every country that has electrical code. check if you work in such
country, get the book and read it...

as for industrial power supplies, i read the specs and use the
product to the letter, including grounding either of DC terminals,
shorting output etc.

what's the point in buying isolated power supply if you are afraid
of grounding any point (such as output's positive DC terminal for
example)?

what's the point in buying power supply with current limiting or
short circuit protection if you are not going to short it as soon
it's out of box and powered up? when do you expect this should
kick in and save the day? when you get accidental short three
years later and hopefully nobody remembers your name?

That makes no sense at all !

Alaric said:

Thats all good and well until you hook your power supply to a device that is internally grounded. I head off the problem at the panel by grounding the 0V terminal.

Click to expand...

If you want to create ground loop problems then go ahead.
Do you have any concept of what 10 megohms is. It's an open circuit. But this also forces both points to be at the same potential since the ir drop across 10M resistor would be zero.
Why would you want to create noise with your ground looping circuit. Have you ever heard of "single point grounding" and do you know what it means?