Does 24 VDC have to be grounded?

[chrisplusian]

I have read through the NEC and I can't find a specific reference that says I have to have and equipment grounding conductor supplied to a device that uses 24VDC. Two things I am currently up against:
A control panel which has a 480 to 24VDC power source in it supplies 5 sub panels. The source is limited to 1000VA and the conductors used to feed the remote panels are Two #6's for DC+ and Two #6's for DC-. My question here is do I have to supply a grounding conductor to the sub panels? And second question: IF the sub panels feed individual devices used for controls (level probes, actuators, etc) do I have to provide and equipment grounding conductor to the devices themselves? The level probes are in metal housings. If you could please provide code references with any explanation I would appreciate it. My dilemma stems from the fact that currently the DC source has the DC- side connected to the ground bus in the cabinet it is supplied from. The sub panels have #14 Grounding wired instelled and connected to the eclosure. I am being told the #14 has to be replaced with two #6's per 250.122(B). I am reading article 250 part VIII in the 2014 NEC, and I am having a hard time deciding if any of the equipment (SUB panels, or devices fed from sub panels) are actually required to be grounded? Thanks in advance

 

[drktmplr12]

I have read through the NEC and I can't find a specific reference that says I have to have and equipment grounding conductor supplied to a device that uses 24VDC. Two things I am currently up against:
A control panel which has a 480 to 24VDC power source in it supplies 5 sub panels. The source is limited to 1000VA and the conductors used to feed the remote panels are Two #6's for DC+ and Two #6's for DC-. My question here is do I have to supply a grounding conductor to the sub panels? And second question: IF the sub panels feed individual devices used for controls (level probes, actuators, etc) do I have to provide and equipment grounding conductor to the devices themselves? The level probes are in metal housings. If you could please provide code references with any explanation I would appreciate it. My dilemma stems from the fact that currently the DC source has the DC- side connected to the ground bus in the cabinet it is supplied from. The sub panels have #14 Grounding wired instelled and connected to the eclosure. I am being told the #14 has to be replaced with two #6's per 250.122(B). I am reading article 250 part VIII in the 2014 NEC, and I am having a hard time deciding if any of the equipment (SUB panels, or devices fed from sub panels) are actually required to be grounded? Thanks in advance

There's nothing that requires either side of the 24 VDC power supply to be grounded. Any ungrounded conductors require overcurrent protection.

The enclosure of the sub-panels should be grounded according to 250.112 (2014)

Except as permitted in 250.112(F) and (I), exposed, normally non–
current-carrying metal parts of equipment described in
250.112(A) through (K), and normally non–currentcarrying
metal parts of equipment and enclosures described
in 250.112(L) and (M), shall be connected to an equipment
grounding conductor, regardless of voltage.

 

[Jraef]

By your mention of a 1000W power supply, before you go too far down this path, even if someone else designed it, there is something else you need to check.

A LOT of small 24VDC powered electronic devices get around some UL listing requirements by requiring that the power source be a listed “Class 2” power supply. That then means that the power supply is limited to 4A, which at 24V means 100W. You should check the data sheets on all of your devices to make sure none of them are needing Class 2 sources.

 

[Sahib]

There's nothing that requires either side of the 24 VDC power supply to be grounded.

Is it the 480V to 24V power conversion device of OP a separately derived source and so DC side to be grounded?

 

[drktmplr12]

Is it not the 480V to 24V power conversion device of OP a separately derived source and so DC side to be grounded?

If we read what exactly a separately derived source (SDS) is defined as, then a DC supply can be considered one. I think its a stretch and haven't yet seen an inspector make this claim. Although we had one say a UPS was a SDS and we explained there's no way to connect to ground without opening it. He backed off.

Slippery slope here.

In any case, the chassis of the power supply should be grounded.

 

[chrisplusian]

By your mention of a 1000W power supply, before you go too far down this path, even if someone else designed it, there is something else you need to check.

A LOT of small 24VDC powered electronic devices get around some UL listing requirements by requiring that the power source be a listed “Class 2” power supply. That then means that the power supply is limited to 4A, which at 24V means 100W. You should check the data sheets on all of your devices to make sure none of them are needing Class 2 sources.

Thanks for the Tip. I will take a look at that as well!

 

[chrisplusian]

Is it the 480V to 24V power conversion device of OP a separately derived source and so DC side to be grounded?

I am not sure what you were referring to when you said "OP". I did see (on another website) where someone said there was no requirement for the DC devices to be grounded, but if you chose to ground the DC side of the source then the entire DC side had to follow the rules for grounding provided in article 250, however, no code reference was provided.

 

[chrisplusian]

There's nothing that requires either side of the 24 VDC power supply to be grounded. Any ungrounded conductors require overcurrent protection.

The enclosure of the sub-panels should be grounded according to 250.112 (2014)

Except as permitted in 250.112(F) and (I), exposed, normally non–
current-carrying metal parts of equipment described in
250.112(A) through (K), and normally non–currentcarrying
metal parts of equipment and enclosures described
in 250.112(L) and (M), shall be connected to an equipment
grounding conductor, regardless of voltage.

I originally considered this specific code when originally contemplating this question. My dilemma is 250.112 (I) allows for class 1 circuits rated less than 50 volts to not be grounded. Looking at 725.41 I see no requirement for a source to be listed, which kind of surprises me, and makes me wonder if there is another place in the code that would require the source to be listed as class 1. IN this circumstance the source IS less than 50V AND is "not greater than 1000VA". Which makes me believe article 250.112 permits the sub panels to not be grounded. The biggest problem for me is not pulling the two #6's but the fact that I only have a 3/4" conduit that has existing class 1 wiring in it to pull them through. Following class 1 requirements for de-rating I would have to replace 3/4" conduit with minimum 1" conduit to be code compliant. I was considering pulling the ground if it is not required (and if I can't find any safety risks in so doing). For reference the sub-panels have a 10amp breaker in the panel they are sourced from, and in the sub-panels themselves.

 

[chrisplusian]

By your mention of a 1000W power supply, before you go too far down this path, even if someone else designed it, there is something else you need to check.

A LOT of small 24VDC powered electronic devices get around some UL listing requirements by requiring that the power source be a listed “Class 2” power supply. That then means that the power supply is limited to 4A, which at 24V means 100W. You should check the data sheets on all of your devices to make sure none of them are needing Class 2 sources.

You now have me wondering about this. The original intention was to add power sources to each of the sub-panels which would really have simplified things, but there was talk about maintenance personnel needing to open the panels to calibrate load cell controllers, and the arc flash rating when we added the source. For that reason the decision was made to use really large conductors (to accommodate for voltage drop) rather than the source in the sub panel. Could this be what you were referring to? Thanks in advance for your input and help.

 

[drktmplr12]

Check 725 for class 1 and class 2/3 circuit wiring requirements. Don't have a code book in front of me but you might find what you are looking for there.

I believe it says for a circuit to be considered class 2 or class 3 the power supply must be listed as such.

Sent from my Pixel 2 using Tapatalk

 

[chrisplusian]

If we read what exactly a separately derived source (SDS) is defined as, then a DC supply can be considered one. I think its a stretch and haven't yet seen an inspector make this claim. Although we had one say a UPS was a SDS and we explained there's no way to connect to ground without opening it. He backed off.

Slippery slope here.

In any case, the chassis of the power supply should be grounded.

I would tend to disagree with you on this. The definition of Separately derived System is :

An electrical source, other
than a service, having no direct connection(s) to circuit
conductors of any other electrical source other than those
established by grounding and bonding connections.

Typical DC power sources, in today's industry, are switched power supplies. Meaning they use a form of filtered pulse width modulation to derive the DC side of the source. In such case when a switch closes there is a direct electrical connection to the primary side of the source, unless opto-couplers or similar are used for isolation of the supply. In medical applications isolation is typically used, but in my experience that is not the case in most industrial applications. An isolation transformer would be an example of a Separately Derived System.

 

[chrisplusian]

Check 725 for class 1 and class 2/3 circuit wiring requirements. Don't have a code book in front of me but you might find what you are looking for there.

I believe it says for a circuit to be considered class 2 or class 3 the power supply must be listed as such.

Sent from my Pixel 2 using Tapatalk

In this case I was referencing Class 1

 

[chrisplusian]

I originally considered this specific code when originally contemplating this question. My dilemma is 250.112 (I) allows for class 1 circuits rated less than 50 volts to not be grounded. Looking at 725.41 I see no requirement for a source to be listed, which kind of surprises me, and makes me wonder if there is another place in the code that would require the source to be listed as class 1. IN this circumstance the source IS less than 50V AND is "not greater than 1000VA". Which makes me believe article 250.112 permits the sub panels to not be grounded. The biggest problem for me is not pulling the two #6's but the fact that I only have a 3/4" conduit that has existing class 1 wiring in it to pull them through. Following class 1 requirements for de-rating I would have to replace 3/4" conduit with minimum 1" conduit to be code compliant. I was considering pulling the ground if it is not required (and if I can't find any safety risks in so doing). For reference the sub-panels have a 10amp breaker in the panel they are sourced from, and in the sub-panels themselves.

Sorry I meant to say I see no requirement for a "Class 1" source to be listed.

 

[jaggedben]

A DC power supply can have a transformer. AFAIK just about any 120VAC to 24VDC power supply is going to have a transformer because of the voltage difference. If you're converting AC to DC without a big voltage difference then you're likely not to have a transformer.

 

[chrisplusian]

A DC power supply can have a transformer. AFAIK just about any 120VAC to 24VDC power supply is going to have a transformer because of the voltage difference. If you're converting AC to DC without a big voltage difference then you're likely not to have a transformer.

In the last few years, nearly all the 24VDC power supplies I have installed are switching power supplies. In years past they were almost all transformers. In any event, this installation uses a switching power supply, and doesn't meat the definition of SDS.

 

[Sahib]

In the last few years, nearly all the 24VDC power supplies I have installed are switching power supplies. In years past they were almost all transformers. In any event, this installation uses a switching power supply, and doesn't meat the definition of SDS.

Suppose a fault condition in which 480V appears on 24V side due to insulation break down. If a 24V conductor is grounded, an upstream 480V side OCPD would operate and remove the hazard. OP means opening poster ie you.