Questions about SCCR and transformers for air conditioning units

Status
Not open for further replies.
Shackled Designer

Shackled Designer

Member
Location
Decatur, AL USA
Suppose I want to install a relatively small air conditioner on my industrial panel enclosure, and the AC unit is to be powered as a branch circuit from the mains feed into the panel. The air conditioner has a fairly low short circuit current rating of 5 kA, and the AC vendor has not bothered to test any fuse or circuit breaker combinations with the AC unit for a higher combo rating.

From a UL perspective, in order to increase the total panel SCCR, is it easier to simply put a transformer between the branch supply and the air conditioner (with no voltage change), or am I likely to be as successful improving the SCCR by simply putting a MCB on the branch circuit?

Other than testing the panel themselves, does UL provide any means for me to quantify panel SCCR with either method I have mentioned? I ask because whatever I do, at the end of the panel design, I have to document and label accurately the panel SCCR.

If the use of the transformer provides significantly higher SCCR as I expect, and/or a better solution for other reasons, then I have another question about the transformer sizing: must I size the transformer for the startup current of the air conditioner, or can I instead select a transformer based on the AC unit's running current?

The unit I have in mind has a startup current of about 12-13 amps, but its running current is not more than 2.5 amps. I imagine that the startup does not last for more than 2 to 4 seconds. Maybe another way of asking the transformer sizing question is whether a transformer's rating should reflect the expected inrush or some nominal margin above the expected constant load, say 125% of continuous current.

The NEC seems to provide direction on overcurrent protection for the transformer but, so far as I can discern, less instruction on transformer sizing with respect to the load itself.

Comments, questions, and criticisms are welcome.

Kind regards,
Shak
 
Shackled Designer said:
Suppose I want to install a relatively small air conditioner on my industrial panel enclosure, and the AC unit is to be powered as a branch circuit from the mains feed into the panel. The air conditioner has a fairly low short circuit current rating of 5 kA, and the AC vendor has not bothered to test any fuse or circuit breaker combinations with the AC unit for a higher combo rating.

From a UL perspective, in order to increase the total panel SCCR, is it easier to simply put a transformer between the branch supply and the air conditioner (with no voltage change), or am I likely to be as successful improving the SCCR by simply putting a MCB on the branch circuit?

Other than testing the panel themselves, does UL provide any means for me to quantify panel SCCR with either method I have mentioned? I ask because whatever I do, at the end of the panel design, I have to document and label accurately the panel SCCR.

If the use of the transformer provides significantly higher SCCR as I expect, and/or a better solution for other reasons, then I have another question about the transformer sizing: must I size the transformer for the startup current of the air conditioner, or can I instead select a transformer based on the AC unit's running current?

The unit I have in mind has a startup current of about 12-13 amps, but its running current is not more than 2.5 amps. I imagine that the startup does not last for more than 2 to 4 seconds. Maybe another way of asking the transformer sizing question is whether a transformer's rating should reflect the expected inrush or some nominal margin above the expected constant load, say 125% of continuous current.

The NEC seems to provide direction on overcurrent protection for the transformer but, so far as I can discern, less instruction on transformer sizing with respect to the load itself.

Comments, questions, and criticisms are welcome.

Kind regards,
Shak
Click to expand...
So you are wanting to power the air conditioner at the incoming line power voltage level then? If so, would that be a device inside of the cabinet or a load fed from it? By that I mean if you build a control panel, and it feeds an air conditioner compressor motor that is 1 foot away, you would not include that in the SCCR rating, correct? So how is that different from one that is 1/8 inch away? Because technically that AC unit is not INSIDE of the control panel is it?
 
Bob, Jraef,

Thank you both for your responses and for helping to push back the frontiers of my own ignorance.

Yes, Jraef, I need to power the AC unit at the incoming line voltage (480V, 3-ph).

The air conditioner is a device outside of the enclosure, as you say, although it is attached to the enclosure. By the NEC article 100 definition (at least as of 2014) for an Industrial Control Panel, the components of the panel include those "mounted on, or contained within, an enclosure . . . " So with that definition in mind, I am inclined to treat the AC unit as part of the panel.

On the other hand, you, Bob, direct me to the details of the supplement (UL508A SB, I presume), which, in the interest of full disclosure, I must admit that I do not presently possess. So I make a semi-educated guess that maybe you both are talking about the idea that equipment being controlled is not actually considered part of the panel, and therefore not part of the SCCR evaluation. Is that what you meant?

This concept of excluding controlled devices from the panel SCCR used to be more explicitly stated in NEC 409.2, but I note that in NEC 2014, 409.2 has been removed. I expect though, that supplement SB is still an approved method of determining SCCR, and so even with the deletion of 409.2 from the NEC, if supplement SB allows me to ignore the SCCR of an air conditioner, then that makes my life easier.

Please correct me if I misapprehend your meanings.

Backing up to the 10,000 foot view, let's suppose I install on my air conditioner branch an MCB, as the AC unit vendor recommends, that is rated for 65 kA. Then, assuming that the 5 kA-rated air conditioner may be rightly considered as controlled equipment and therefore not part of the panel SCCR, it is attached to the outside of the enclosure and its power line connected to the MCB. So I apparently have an arrangement that meets code. But what if, say, a forklift impales the AC unit and causes a short circuit that draws more than 5 kA? Another way of asking this question is: what does confining the scope of the code to panel devices intend to accomplish?

I know my example is a bit contrived, but my company has seen a touchscreen controller station get stabbed by a forklift, so I guess what I've imagined is not outside the possible . . .

Thank you for your thoughts.
Shak
 
Last edited:
Shackled Designer said:
Suppose I want to install a relatively small air conditioner on my industrial panel enclosure, and the AC unit is to be powered as a branch circuit from the mains feed into the panel. The air conditioner has a fairly low short circuit current rating of 5 kA, and the AC vendor has not bothered to test any fuse or circuit breaker combinations with the AC unit for a higher combo rating. ...
Click to expand...

Shak -
Just so I am clear. This is my translation:

You are adding componets to a listed panel. The stuff you are adding does not have the the withstand that the panel can deliver.

Your question: What can you do to bring the modified panel up the the required withstand?

Is this correct?

ice
 
The marked SCCR on the industrial control panel is based on the "as shipped" configuration.

In a previous post you suggested adding a MCB (not sure what this is). My suggestion is that you will gain nothing by doing so as it would not change anything.

In any case, the supplement says this:

Exception No. 3: Enclosure air conditioners or multimotor and combination load equipment that is
cord-and-attachment-plug connected or supplied from a branch circuit protected at 60 A or less is not
required to have a short circuit current rating.
Click to expand...

So as long as the A/C is powered from a 60A or smaller branch circuit you should be OK if you are worried about not violating UL508a rules on SCCR. I am not sure it matters much since you are probably not authorized by UL to apply a label anyway.
 
Last edited:
Bob,

Please forbear with my previous lack of clarity. The motor circuit breaker (MCB) is the recommended over-current protection by the air conditioner unit's manufacturer.

Also, for both Bob and Iceworm, I am not attempting to modify an existing panel that already has a UL listing. Instead, I am trying to design a panel that eventually, when a future customer requests it, will be reviewed by UL.

Furthermore, because I anticipate multiple customers for the panel design even where a UL listing is not requested, I am attempting to make the design suitable for many/most expected installation sites, and their anticipated available fault currents, by designing for a panel SCCR of 22 to 50 kA.

In either case, I want to preclude as much future redesign as possible.

My original question presupposed that, because the air conditioner was a device attached to the panel enclosure, it would affect the rating of the panel. Moreover, since the air conditioner has an SCCR of only 5 kA, and because the manufacturer has documented no combinations of over-current protection with the unit to give it an increased combination rating, I was concerned that there might be no way apart from UL testing to rate the panel above 5 kA.

Interestingly, Bob, the exception you have quoted looks similar to NEC 440.4 B, Exception 3, though that covers some stuff outside of panel design. More to the point, though, the exception does not explicitly indicated that the air conditioner has no effect on the panel's SCCR, but rather that the air conditioning unit need not have its own SCCR. Maybe, since the normal method of determining SCCR required either a marked or a documented SCCR, this exception means that the air conditioner cannot by virtue of electrical code affect the panel's rating. If the AHJ does not see it that way, though, then I have problems.

This might be an easier question to my mind if NEC 409.2 had not been deleted.

I appreciate and welcome further comments, questions, and advice.

Kind regards,
Shak
 
Shackled Designer said:
.... The motor circuit breaker (MCB) is the recommended over-current protection by the air conditioner unit's manufacturer.

... This might be an easier question to my mind if NEC 409.2 had not been deleted. ...
Click to expand...

Shak -
I'm not going to be much help, I'm definitely not a specialist in panel design. I've done a few per NEC409, but I'm not a whiz. But I am interested.

Couple of questions:
What is a Motor Circuit Breaker (MCB)?

As for 2014 NEC delecting 409.2 - Have you read 2014 NEC 100, Definitions, Industrial Control Panel? I'm not seeing much difference.

ice
 
I will suggest this.

It is generally not cost effective to design and build a panel and then have UL come in and do a one time inspection. There are a fair number of things that can trip you up if you are not real familiar with UL508a. It can and has been done but is rarely if ever cost effective to do so.

If you want to use UL508a and the SCCR supplement found in it as a way of determining the panel SCCR, you have to abide by ALL of the requirements found in UL508a. You cannot pick and choose which ones you want to follow. The standard tells you what components are acceptable, including what circuit breakers can be used and what A/C units are acceptable. I am not familiar with any A/C units acceptable to UL508a that require or allow anything other than normal branch circuit protection. I am not saying it is impossible that such a thing exists, but I have not run across one.

I would also ask what you mean by motor circuit breaker. It is a term that might mean several different things, none of which are properly referred to by the term.
 
Last edited:
Bob, Iceworm,

Thank you both again for your responses.

A motor circuit breaker is an overcurrent protective device that operates similarly to a normal circuit breaker in that it responds to thermal and/or magnetic tripping. Typically, it protects smaller loads than a molded case circuit breaker. In applications with which I am familiar, it is used to protect, among other things, the conductors of VFDs, servo power inverters/converters, and some air conditioner units.

I grant that there are probably technical reasons that the term, motor circuit breaker, is wrong, but in the automation industry it is an accepted term by many vendors and customers. The following link is offered not as a particular vendor endorsement but because the webpage shown gives a nice visual sampling of several different styles of MCBs.

http://www.schneider-electric.com/p...rs/3055-motor-circuit-breakers/684-tesys-gv2/

Iceworm, thanks for reminding me of the actual definition in NEC 100 of an industrial control panel. I see from my copy of the NEC that this definition was edited in the 2014 version, so that may explain why 409.2 was deleted.

Bob, my company has already had a customer that requested, prior to purchase, a UL review of the panel design on a similar product to the one presently being developed. For this reason, we expect it will happen again.

As far as using UL508A and its supplement for determining panel SCCR, I have no desire to cherry pick acceptability criteria, especially since I expect that the AHJ would tend toward a fairly strict interpretation of things. Again, my goal is to eliminate as much future redesign as I possibly can by proper design up front.

You comments underscore the need for us to go ahead and obtain some accepted standard (and procedure) for determining SCCR, be it UL508A or some other accepted criteria. In my experience, UL508A and its supplement SB is the only acceptable standard that is named for this purpose. However, if you know of some other standard, I'd be happy to look into it.

From what you say about UL508A in relation to air conditioners, it looks like I need to proceed with a great deal of caution.

Further ideas are welcome.

Cheers,
Shak
 
Thanks for the link, Jraef. That is a pretty interesting interpretation, especially in light of the conversation so far.

From the link on Hoffman's website:
You may use a 5 or 10 kVA isolation transformer between the customer?s panel and our air conditioner and not have an effect on the customer's 65 kA rating.

You may use a fuse or circuit breaker with a 5 kA short circuit rating on the line side of the ACU and its branch circuit protective device and not have an effect on the customer's 65 kA rating.
Click to expand...
My own emphasis is added to the above quote.

From what I have read elsewhere, using a low SCCR rated fuse is not normally considered a solution for increasing overall panel rating.

Aside from that, this quote suggests that even with the "controlled device" not being strictly in the panel, it might have an effect on the panel SCCR without some other protective device. I would guess that the 65 kA rating is an upper limit for the panel supplying the air conditioner.

I also see that Hoffman states that this information applies to the installation of their particular air conditioners, so this may not necessarily apply to others.

For those who read this post and wish to use the quoted information as a guide for their own design, I would advise that you refer to the rest of the information on the linked site since it includes some important qualifiers on fuses or breakers that are used. More importantly, please bear in mind that an AHJ may not consider Hoffman to be authoritative when it comes to justifying design decisions.

Another point of interest from a downloadable document on Hoffman's linked page: referring to accessory items such as air conditioners, their "Standards Summary and Directory Spec Sheet" states,
If the accessory contains both a load and controls, the controls would need to be assigned a short circuit current rating. An example of this would be an air conditioner that has a contactor included in the design. In this case, if the contactor is not marked, the SCCR could be assigned from Table SB4.1 in UL508A, or be tested per UL508, or the standard that applies to the component being evaluated, in this case the contactor.
Click to expand...

This would seem to restrict the UL508A supplement SB exception regarding air conditioners that was previously mentioned. Even that speaks of "load equipment" without any reference to built-in controls.

Food for thought. . .
Shak
 
Allow me to correct a partially erroneous statement I made earlier. I said

From what I have read elsewhere, using a low SCCR rated fuse is not normally considered a solution for increasing overall panel rating.
Click to expand...

By virtue of the supplement's Article SB 4.3.3, if the peak let-through current of a feeder fuse is not greater than the lowest component SCCR in the branch, the combination SCCR of the feeder/branch is the available RMS current of the fuse.

My earlier statement was made without considering the feeder fuse. In a panel such as my own with multiple branches, I have not found any general provision that says installing current limiting fuses on one of the branches necessarily increases the panel's SCCR, unless possibly the fuse is replacing a protective device with the lowest component SCCR of the panel. This is what I had in mind with my earlier statement.


Bob,

My comment on the air conditioning exception has as its context the statement made by Hoffman about A/C units with built-in controls as I expect many or most enclosure A/C units do. If this mixture of load and control devices is what is meant by "combination load equipment" as stated in the SB supplement exception, then I guess the A/C unit on the branch is no longer a factor in the SCCR. So, yes, maybe the Hoffman text is a restatement of the exception. I hope that is correct.

Best regards,
Shak
 
I just spoke with a UL engineer about this subject.

He confirmed that the Hoffman approach is generally correct for air conditioners. Specifically, there must be branch circuit protection that is short-circuit rated for not less than the indicated SCCR of the panel. Of course, since I am in the design process, that means that the fuse or breaker that I select will help determine the panel's SCCR, even considering that the SCCR of the A/C unit itself does not figure into the determination.

Naturally, the branch OCP that I use must also, by the Hoffman statement, have a peak let-through value of less than 5kA according to the rating of the panel at which I am designing.

I do appreciate all of the contributions to this discussion. Thanks for the help.

Very finest regards,
Shak
 
Again, I have assaulted the reader by my lack of clarity. Apologies all around.

Yes, GoldDigger. Peak let-through will have to match the SCCR of the air conditioner. 5 kA happens to be the SCCR of the A/C unit.

When I wrote
the branch OCP . . . must also . . . have a peak let-through value of less than 5kA according to the rating of the panel at which I am designing
Click to expand...

I had fuses in mind where the peak let-through varies according to the available fault current supplied to the panel. So, for example, my branch OCP might be a Class T, 30-amp fuse if my panel has an SCCR of 50 kA or less, but if my panel instead was rated for 100 kA, the peak let through on the same fuse would be 7 kA and no longer acceptable as protection for the A/C unit.

It escapes me as to whether circuit breakers have similar variation according to the available fault current on the supply. Doing a quick check, I am so far unable to find documentation that says so one way or the other . . .

Shak
 
Circuit breakers are only directly rated according to the current they can safely interrupt, which must meet or exceed to available fault current.
If you want to use a high interrupting rating main or feeder breaker to protect a branch breaker with a lower rating they must be tested together by the manufacturer and listed as "series rated".
 
Status
Not open for further replies.
Top