Ups neutral

I often use a 120vac ups to back up my plc's some people have told in the past to never connect the neutral and the ups neutral to the same row of terminal blocks and i have also been told by others that it should not matter.


Any thoughts?

I am assuming you are actually referring to grounding. The 'identified conductor' is supposed to be at ground potential. I think what you want to do is tie the UPS neutral to gound, so that the hot doesn't float to a much higher potential. Tying the two neutrals might give you some strange loop currents.

No i agree the neutral on the control transformer needs to be bonded to ground and then i normally leave the neutral of the UPS floating.

But then tying both neutrals together might cause issues you really think?

interesting any reason how?

Technically if both neutrals are tied to ground, they are tied together. The problem that I experienced was only tenuously related. Someone wired some circuits so that the load was connected to the hot of one circuit, and the neutral of another circuit. The result was a ground current between the two. This was before ground fault breakers became popular. The rationale I learned for tying the neutral to ground was to prevent the secondary hot from floating. The (bizarre) instance that I was describing would be like if a load was wired to the UPS hot and the supply neutral - hot of secondary to neutral of primary - very unlikely.

How can you have the two neutrals NOT tied together? If the normal line and neutral are connected to a load, say a PLC power supply, and the UPS line and netutral are connected to that load to provide power if normal supply fails, then the line and the neutral of the UPS are tied to the line and neutral of the normal supply at that point.

If it's a double-conversion UPS, I would be especially careful to not tie its output "neutral" to the "input" neutral. Without knowing how the UPS is designed internally, there's little to tell you what might happen, no matter the topology. Earth ground will always be earth ground, but the neutral should go through the UPS.

That's fundamentally no different than plugging a computer into a UPS, and then plugging the UPS into the wall outlet. How is connecting a PLC and associated components any different? The fact that the UPS's output is brought to terminal blocks shouldn't make any difference.


-rpoet

First let us define terms
GROUNDED conductor is in the circuit carries current aka neutral colored white in USA
GROUNDING conductor is NOT in the circuit and does NOT carry current normally colored green in USA.

I think a good comparison of this is a transformer 480/277 4 wire fed to primary of transformer. Secondary is 208 / 120 wye again 4 wire.

In both cases teh neutral GROUNDED of the 480 distribution and the neutral GROUNDED are bonded to ground at the grounding bus. It can be argued and I agree that techically they are tied together via the bonding. However in all other aspects they are two completely separate circuits.

IF the UPS has an selection switch relay where the load supplied from AC line OR from UPS output and there is a switch mechanically ie "air break" separating the two sources then all should be OK. If they are isolating the two sources from each other with just solid state I would not personally buy that equipment.

I really wonder about this grounding stuff and if we are not requiring too much to be grounded therefore tieing too many supposedly separate systems together.

It is a completely separate system with other design consideration but US Navy ships electrical was not grounded distribution when I served and I believe still is ungrounded. We as best as I recall never had a neutral and in very few cases had a common or center tapped transformer in our generation distribution and motor control.

Dan Bentler

Dan,
Depending on the UPS topology, tying the neutrals together might be fine, or not.

http://en.wikipedia.org/wiki/Uninterruptible_power_supply

I think it would probably OK to tie neutrals together with an offline style, as it usually has the "airgap" style relay that you described.

Double conversion UPS topology is actually pretty similar to switch-smode power supplies and VFDs. The UPS rectifies the AC to DC which charges the battery, and then the DC is inverted back to AC to power the load. I'm sure simplifying things greatly, I'm sure but the concepts hold.

I have no idea what tying the neutral of a double-conversion UPS to the input side neutral would do. I'd hazard a guess it wouldn't do anything good, but if the OP is has his heart set on doing so, at least call the manufacturer to find out first. It might keep the magic smoke in. I imagine it might do the same thing as tying an output leg of a VFD to neutral - hopefully the UPS would just fault and shut down, but I'm not sure.

I'm with you on the grounding thing. Earth ground should always be earth ground, but tying everything together sometimes leads to more problems than it solves; I never tie the 0v of my 24v power supplies to earth. I leave them floating. This isolates my 24vdc power from all that noise running around from drives, etc...


-rpoet

As you said rpoet - the UPS inverts DC to AC - you can not bond both neutrals together or by earthing (grounding).
Notice the Wikipedia bypass dwg actually disconnects the UPS from the supply
Most UPS neutrals are not grounded in this situation

I think this requires a call to the factory.

I am also basing my comments on Natl Elect Code in USA. NEC requires all separate derived power to be bonded to ground. If my equipment I would be spending quite a few hours studying up on everything to determine what to do. Without the studying I am assuming the UPS is a separate derived power source and its output must be grounded to comply with NEC.

I read the wiki referance and in all the drawings there is a "switch" to select the source feeding the load.

Got a bit confusing regarding the online double convesion section. Sure sounds similar to VFD setup where normal AC feeds a rectifier which feeds DC bus. DC bus feeds a battery and the output IGBTs (or other solid state switch technology) which then feeds load. So in theory at least the load is completely separated from the normal AC line by a bunch of solid state stuff. I am leery on this because I am not real knowledgable on solid state ability to provide isolation and have seen enough failures to not totally trust it and prefer a complete isolation transformer or an air break type switch.

Dan Bentler

As you said rpoet - the UPS inverts DC to AC - you can not bond both neutrals together or by earthing (grounding).

I am also basing my comments on Natl Elect Code in USA. NEC requires all separate derived power to be bonded to ground.

Click to expand...

Looking at the NEC Handbook, it does state in 250.30 that "A separately derived system that is grounded" shall comply with 250.30(A)(1) through (6). (These are requirements for bonding jumper, grounding electrode conductor, and so on). The Handbook Fine-Print Note says that a converter is a separately derived system, "although these systems do not have the same wide use as ... transformers". Ha! so much for that!

Still there are some exceptions that would eliminate the requirement for a grounded neutral for some inverter-type power supplies. For example, 250.36(A)(2) says that if the "transformer" for a separately-derived system is 1000 VA or less, then no grounding electrode conductor is required.

The NEC is often a little behind the technology and needs a few years to catch up. When some of these grounding rules were made, there were no small UPS inverter-converter-type power supplies, or at least they were not common and scattered all over a site everywhere there is a PLC or a computer.

leitmotif said:

Sure sounds similar to VFD setup where normal AC feeds a rectifier which feeds DC bus. DC bus feeds a battery and the output IGBTs (or other solid state switch technology) which then feeds load. So in theory at least the load is completely separated from the normal AC line by a bunch of solid state stuff. I am leery on this because I am not real knowledgable on solid state ability to provide isolation and have seen enough failures to not totally trust it and prefer a complete isolation transformer or an air break type switch.

Click to expand...

I think some UPS's, especially small units like one might find powering a desktop computer, may have transformer(s) even if they have double conversion topology. Small units don't have enough battery cells in series to create a full-voltage DC link. I'm not too sure what those units do, but I'm guessing it's much like a switchmode power supply on the output.

As far as the OP's question, I'd love to know what model # UPS we're talking about here; is it something that sits in the bottom of the cabinet and is the size of a shoebox, or is it room-sized and needs a forklift to replace individual battery cells. I bet those have different requirements.

If it's a "desktop" style, just plug it in to an edison receptacle in the panel, and plug a male edison into it. That male edison plug then goes to a row of terminal blocks where the OP can power whatever he wants. The UPS sits on a little shelf in the bottom of the cabinet; ideally the UPS is ethernet enabled, so it can complain to the HMI or the SCADA if it has a problem.


-rpoet

Correct me if I'm wrong but I don't think the US Navy follows NEC.

KAJ said:

Correct me if I'm wrong but I don't think the US Navy follows NEC.

Click to expand...

Nope I believe they are in the exemption for maritime
Other exclusions include
Phone, power generation and rail

Dan Bentler

the laws are similar here - if a site i solely Gen powered then the authorithies do not care - but the electricians keep to the wiring standards.