OK to Tie Powersupply Common to Chasis Ground?

skyfox

Lifetime Supporting Member
Join Date
Nov 2006
Location
CA
Posts
276
Wondering what are the pros/cons of tying Power +24VDC supply Common to chassis (earth) ground. Is it OK to do so? I have seen some control panels where this is done and others without. I have also seen Dual polarity supplies +/-15VDC commons tied to chassis ground as well. What determines if a power supply common should or shouldn't be tied to a chassis ground?

Thanks
 
Why? I never tie 24VDC or 24VAC to ground. That being said my 24VAC is always through a double wound transformer and 24VDC through a switchmode power supply - grounding the negative removes isolation. No pros - only a huge con - remove isolation. For example if the 24VDC negative is grounded there is a path to ground in the field - if not no path - no shorts - no blown up input cards.
 
I almost always tie 24 vdc supplies to ground. Yes it removes isolation, but that can be safer if somehow a different voltage gets on it. Example would be 480 somehow vets tied to it. If not grounded then theoretically no circuit protection would open and an unsuspecting person could come into contact with 480AC. I've also seen it clear up noisy analog signals by tieing to ground.
 
I almost always tie 24 vdc supplies to ground. Yes it removes isolation, but that can be safer if somehow a different voltage gets on it. Example would be 480 somehow vets tied to it. If not grounded then theoretically no circuit protection would open and an unsuspecting person could come into contact with 480AC. I've also seen it clear up noisy analog signals by tieing to ground.


I have even had the analog out on some devices not work unless the 24VDC negative is grounded and yes also to noisy analog signals if not grounded.
 
Last edited:
Do a search here, you will find the subject has been covered many many many times, there are entrenched views on both sides. In my understanding one side is right and one side is wrong, but I am not telling which is which.
 
I've seen it done both ways, and depending on which set of regulations you're working to it might be interpreted as mandatory. I've never personally solved a problem by grounding an ungrounded system, but I have solved problems by removing the ground.
 
I've seen it done both ways, and depending on which set of regulations you're working to it might be interpreted as mandatory. I've never personally solved a problem by grounding an ungrounded system, but I have solved problems by removing the ground.

AutoMax, I have had the opposite of what you stated and have solve several issues by applying the proper grounding to systems that had intermittent problems.
 
I have done it both ways but prefer to leave it ungrounded. I've never seen grounding solve a problem with noise or anything else. Not that I doubt that other people have.

Also it would seem to be hazardous to our delicate 24VDC electronics in the event of a major ground fault of nearby high voltage equipment.
 
Talking about grounds.
Not a 24 VDC ground, but I have a 120 volt AC isolated ground system powering several computer systems.

Jackleg electrician tied one of the isolated grounds to the ground in another panel.
In two days I had two computers corrupt their files.

I had to get a REAL electrician to find it and remove the 'extra ground'.
Once my isolated ground panel was, once again, isolated (with only the ONE
ground) my computers were happy and corruption free again.

MadPoet.
 
WOW! Seems 50/50. I was hoping for 90/10 sort of response. Thanks to all of you that replied. I guess more research is needed on the subject.

Cheers!!!!
 
OK, can't believe I am being pulled in to this again. The danger of not earthing a system is that one electrical fault can go undetected but can kill someone, simple as that. If you choose not to earth the system then you must use a design system that separates the dangerous voltage and the 'safe' voltage so that no fault can ever cause them to meet. Generally it means separating the mains and controls in separate parts of a panel. If you have a mixed panel as most industrial boxes are with contactors switching mains next to relays working at 'safe' voltages then you must earth the secondary.

but I have solved problems by removing the ground.
You didn't fix the problem you just painted over it with dangerous paint. The problem is still there, you have a lousy earth system that needs to be fixed, people have died because of bad earths.

Sometimes you have to have a floating system, but you must be very aware of the consequences.
 
Respectfully, Bryan-

How do I keep hazardous voltage from being introduced on the load side of an ungrounded signal wire from (for example) auxiliary contacts on a motor starter overload? I totally understand what you are saying, but are you suggesting as a matter of course electricians should touch bare conductors without verifying the absence of voltage?
 
I will only tie the 24VDC common to ground if required by some oddball piece of hardware in the system that has it done that way internally. I can think of one recent device like this that would lose its marbles if I didn't ground my DC negative side.

The only advantage I can think of aside from the above is that if you are troubleshooting a field device where the DC common is not present and you want to see if the right 24+ voltage is present, you can touch one meter lead to a good ground.

The disadvantages have already been described: 1) higher susceptibility to ground fault noise. 2) potential for false (sinking) inputs if a conductor shorts to ground.

There would not be a shock hazard with 24vdc. I'm with BobB on this but will make exceptions.
 
Don't worry you don't need to be respectful, I have had this argument many many times.

but are you suggesting as a matter of course electricians should touch bare conductors without verifying the absence of voltage?
Don't think I said that anywhere.

Let me bounce a question back to you. Are you saying that you would be happy to stand up in court and say 'It would have been easy to design it to be safer, but he should have used a meter before going near that low voltage circuit'.

Another issue I forgot to mention before is that removing the earth connection may or may not create a floating system. Each component that has electrical noise filtering will have a connection between the supply and earth. I have found various devices that connect the negative to earth within the device. So unless you test each component you don't create a floating system you create an unknown system.

I am always open to hearing arguments and am happy to be persuaded of the error of my thinking. Bobs argument is that you loose isolation, I would say that if you are designing an isolated system and you understand the consequences of that design then that is great. But if you are leaving it unearthed because that is the way 'Fred' always designed them then you are heading for trouble.

I know that it is easy for me to tell people to fix their earth problems, I live in a wet country with a fairly solid supply system, others don't have that luxury.
 
The only advantage I can think of aside from the above is that if you are troubleshooting a field device where the DC common is not present and you want to see if the right 24+ voltage is present, you can touch one meter lead to a good ground.
It is safer, you are less likely to kill someone, to me that is quite a big advantage.

The disadvantages have already been described: 1) higher susceptibility to ground fault noise.
So you are saying that you do it because it is easier, that is a really bad reason. Fix the earth problems.

2) potential for false (sinking) inputs if a conductor shorts to ground.
Are you sure that your are not just trying to wind me up now. You shouldn't be using inputs that can be fooled by a short to ground fault with a negatively earthed supply, they are for positively earthed supply system. That just means you have bought the wrong input card for your PLC, or you wired it wrongly. Generally in the West we use a PNP system, generally in the East they use an NPN system, obviously that is very generalised.

I knew I shouldn't have got back in to this argument.

Has anybody got a reasoned argument about why it is safer not to earth the control supply output.
 

Similar Topics

So Mil-Ram gas detection closed down, and now we need to find another gas detection maker to hook up to our SCADA. Just wondering if anyone here...
Replies
11
Views
2,660
on my Allen Bradley HMI i want to put a login and logout screen for when my operators want to reject a part in production i want to eliminate this...
Replies
3
Views
1,468
I am new to programming rockwell and could use some help. I am using compact logix controller with an Ethernet IP Servo Drive. The servo has 400+...
Replies
13
Views
3,093
We have belimo valve that is getting powered by 24VDC power supply. That same power supply is feeding the DC+/- of a OF8C analog output module...
Replies
2
Views
1,732
Anyone have any recommendation for small wire ties guns... other than dykes and needle nose I use a lot of small wire ties (4 inch) and had a...
Replies
13
Views
3,453
Back
Top Bottom