Effect of Buck Transformer on SCCR?

[Shackled Designer]

Hello, esteemed forum members.

Whilst designing a panel for a 480/277 (Wye) feed, I have approximately 30 amps load of devices which require 230 VAC single phase to operate, so a phase-neutral buck transformer seems like the proper solution. With this drawn out on paper, I have proceeded to the short circuit current rating (SCCR) evaluation of the panel. Can anyone here tell me what effect on the SCCR of secondary-side devices the buck transformer has, if any? In its buck configuration, the transformer has an 8.6 kVA rating.

If I were using an isolation transformer, then I suppose I would not have to worry about devices on the secondary side since the secondary side available fault current is easily calculated from the rated kVA and the secondary voltage. (Please correct me if I err in this last statement.)

With an autotransformer, however, it seems to me that my secondary is not quite so isolated, particularly not so in a short circuit fault.

Does any part of the code address this scenario? If not, should I, for SCCR evaluation purposes, treat the devices on the secondary side of the buck transformer as if they were simply items on a normal (not transformed) branch?

Best regards,
The Shackled Designer

 

[templdl]

Hello, esteemed forum members.

Whilst designing a panel for a 480/277 (Wye) feed, I have approximately 30 amps load of devices which require 230 VAC single phase to operate, so a phase-neutral buck transformer seems like the proper solution. With this drawn out on paper, I have proceeded to the short circuit current rating (SCCR) evaluation of the panel. Can anyone here tell me what effect on the SCCR of secondary-side devices the buck transformer has, if any? In its buck configuration, the transformer has an 8.6 kVA rating.

If I were using an isolation transformer, then I suppose I would not have to worry about devices on the secondary side since the secondary side available fault current is easily calculated from the rated kVA and the secondary voltage. (Please correct me if I err in this last statement.)

With an autotransformer, however, it seems to me that my secondary is not quite so isolated, particularly not so in a short circuit fault.

Does any part of the code address this scenario? If not, should I, for SCCR evaluation purposes, treat the devices on the secondary side of the buck transformer as if they were simply items on a normal (not transformed) branch?

Best regards,
The Shackled Designer

After giving this some thought all the load is im series with the LV winding of the BB transformer. With that in mind you essentially end up with some what of an iron core reactor which should limit fault current to some extent. How much affect would it have would be any ones guess as I have no idea how it could be calcuated usless there is some way to determine what the impedance is. But never the less there should be some reduction of short circuit current.

 

[Shackled Designer]

Thanks for the response, templdl.

I figured that the buck transformer would yield a bit of current limiting, but some of the secondary side items will not present a constant load, so impedance is bound to vary. For example, one of the devices is a GFCI receptacle which should only be used when the panel's PLC programmer is sitting by the machine.

For SCCR, the receptacle is not UL listed, and so it is assumed to have a 2kA short-circuit current rating according to UL508A, supplement SB, table 4.1. With the buck transformer not being a complete isolator, I imagine that I'll have to either find a different receptacle or rate the panel SCCR at 2kA. (Yuck!)

This last option is very disagreeable to me.

Ah, well, happy Friday.
The Shackled Designer

 

[templdl]

Thanks for the response, templdl.

I figured that the buck transformer would yield a bit of current limiting, but some of the secondary side items will not present a constant load, so impedance is bound to vary. For example, one of the devices is a GFCI receptacle which should only be used when the panel's PLC programmer is sitting by the machine.

For SCCR, the receptacle is not UL listed, and so it is assumed to have a 2kA short-circuit current rating according to UL508A, supplement SB, table 4.1. With the buck transformer not being a complete isolator, I imagine that I'll have to either find a different receptacle or rate the panel SCCR at 2kA. (Yuck!)

This last option is very disagreeable to me.

Ah, well, happy Friday.
The Shackled Designer

There has to be something more interesting to do on a Friday than this. But if you take a look a wiring diagrams for either bunch or boost you us the same transformer but its how you configure the HV windings that would buck or boost the voltage. As such I would like to think that the affect either way would not make a very significant difference.

 

[kwired]

Hello, esteemed forum members.

Whilst designing a panel for a 480/277 (Wye) feed, I have approximately 30 amps load of devices which require 230 VAC single phase to operate, so a phase-neutral buck transformer seems like the proper solution. With this drawn out on paper, I have proceeded to the short circuit current rating (SCCR) evaluation of the panel. Can anyone here tell me what effect on the SCCR of secondary-side devices the buck transformer has, if any? In its buck configuration, the transformer has an 8.6 kVA rating.

If I were using an isolation transformer, then I suppose I would not have to worry about devices on the secondary side since the secondary side available fault current is easily calculated from the rated kVA and the secondary voltage. (Please correct me if I err in this last statement.)

With an autotransformer, however, it seems to me that my secondary is not quite so isolated, particularly not so in a short circuit fault.

Does any part of the code address this scenario? If not, should I, for SCCR evaluation purposes, treat the devices on the secondary side of the buck transformer as if they were simply items on a normal (not transformed) branch?

Best regards,
The Shackled Designer

As has been already mentioned, you need to know what the impedance of that transformer will be during a fault to know what kind of let through current you will have, plus you would also need to know what your available fault current is before inserting this impedance.

As far as an isolation transformer, available fault current on the secondary depends on VA rating of the transformer, impedance, and the current available on the supply side. Most short circuit current calculations assume infinite availability from the supply, but that basically gives you worst case possible, there will always be some limitation from the supply, whether it be impedance of a transformer or voltage drop on conductors.

 

[hurk27]

Thanks for the response, templdl.

I figured that the buck transformer would yield a bit of current limiting, but some of the secondary side items will not present a constant load, so impedance is bound to vary. For example, one of the devices is a GFCI receptacle which should only be used when the panel's PLC programmer is sitting by the machine.

For SCCR, the receptacle is not UL listed, and so it is assumed to have a 2kA short-circuit current rating according to UL508A, supplement SB, table 4.1. With the buck transformer not being a complete isolator, I imagine that I'll have to either find a different receptacle or rate the panel SCCR at 2kA. (Yuck!)

This last option is very disagreeable to me.

Ah, well, happy Friday.
The Shackled Designer

How do you intend to get 120 volts for a GFCI receptacle from 277 volts? I don't think your going to find a buck transformer that will reach that far, I would go with a 480-120/240 isolation transformer or better known as a SDS and fuse it's primary as well as the secondary or at least protect the receptacle at its rated current? done many times in control panels, buck/boost transformers will not give you both the 120v and 240 volts.

 

[Shackled Designer]

Hi, kwired.

Thanks for your comments. If I were planning to use an isolation transformer (which has surely become more of a possibility since this discussion began), the SCCR evaluation might be a bit more obvious, at least according to the UL508A supplement. Going from memory at the moment, I seem to recall that the supplement covers 10 kVA transformers or less for any (low) voltage and 5kVA transformers or less for 120V secondary voltages.

From what I have read, however, I cannot tell whether the supplement differentiates between an isolation transformer and an autotransformer. Even if it does not, knowing that one of the legs in the secondary is the same conductor as one in the primary, I think that a short circuit fault touching that leg has practically no current limitation at all from the transformer winding. If this is so, then there are a number of low SCCR devices, such as my GFCI socket, that should not be on a buck transformer branch, even if the code does not expressly forbid it.

As far as knowing available fault current in advance, I have the same quandary experienced by many OEMs. The initial panel design that occurs before there is a customer for it cannot be specifically tailored for the available fault current of a specific installation. So in this phase of the design, I'd like to achieve a panel SCCR that would meet or exceed maybe 75% of the possible customers' available fault current, but that would also allow for upgrading the SCCR without too much pain for the other 25% of prospects. This may prove more difficult than I imagine, but the cost of perpetual reengineering for multiple customers incentivizes me to invest in a better starting point on the design.

According to one vendor's literature, as many as 80% of U.S. industries have available fault currents of 25kA or less in their facilities. I suspect this may be somewhat dated information, but it may offer a reasonable starting point for pre-customer panel configuration.

Finest regards,
The Shackled Designer

 

[Shackled Designer]

Hello, Wayne.

You are exactly right. There is no standard buck transformer configuration that would take me from 277 to 120 VAC. There are many receptacles though that are rated to carry 230VAC and offer GFCI as well.

My intention for the receptacle is that it be labeled for 230V (+/-) and for use only as a programming port. Most laptop PC power adapters are rated to convert both 120 and 240 VAC. Additionally, I have the expectation that if the machine for which this panel is being designed is successful in the U.S., it would see use in other markets where utility sockets are commonly rated at 220 - 240 VAC in normal usage.

Regarding the power conversion, I may have to use an isolation transformer instead of a buck transformer. There seems to be no other way to achieve a reasonable SCCR for a receptacle, as I have not yet found any vendor's documentation that offers either standalone SCCRs or series-rated combinations for their sockets.

Chained to the panel,
The Shackled Designer

 

[templdl]

Hello, Wayne.

You are exactly right. There is no standard buck transformer configuration that would take me from 277 to 120 VAC. There are many receptacles though that are rated to carry 230VAC and offer GFCI as well.

My intention for the receptacle is that it be labeled for 230V (+/-) and for use only as a programming port. Most laptop PC power adapters are rated to convert both 120 and 240 VAC. Additionally, I have the expectation that if the machine for which this panel is being designed is successful in the U.S., it would see use in other markets where utility sockets are commonly rated at 220 - 240 VAC in normal usage.

Regarding the power conversion, I may have to use an isolation transformer instead of a buck transformer. There seems to be no other way to achieve a reasonable SCCR for a receptacle, as I have not yet found any vendor's documentation that offers either standalone SCCRs or series-rated combinations for their sockets.

Chained to the panel,
The Shackled Designer

The purpose of BB transformer is to make a relatively small change in voltage without incurring the cost in an isolation transformer. When you need more than 32v change there ends up to be very little or no cost advantage to using A BB transformer over an isolation transformer.

 

[Shackled Designer]

Hi, templdl.

The purpose of BB transformer is to make a relatively small change in voltage without incurring the cost in an isolation transformer.

That is how understand BB transformers as well.

When you need more than 32v change there ends up to be very little or no cost advantage to using A BB transformer over an isolation transformer.

Going from 277 to 230, I'm looking at not quite 48 volts of change. However the cost advantage for me is still very clear because in an autotransformer configuration, a relatively low kVA transformer can be treated as a higher effective kVA device. In my case, a 1.5 kVA transformer (single phase) wired in a buck configuration yields 8.6 kVA.

Without considering SCCR, I can pay in the ballpark of $250 for such a transformer, whereas if I want a 10 kVA isolation transformer (the next lowest standard size is 7.5 kVA, less that what I need), I'm looking at nearly $650.

This is only the monetary cost difference though. The area footprint of the isolation transformer is triple that of the buck transformer, and saving panel space in this particular instance is rather important.

The SCCR question may make it a moot point, however.

Happy Monday,
The Shackled Designer

(Would it be pretentious of me to adopt a nickname of Shak?)

 

[templdl]

Hi, templdl.



That is how understand BB transformers as well.



Going from 277 to 230, I'm looking at not quite 48 volts of change. However the cost advantage for me is still very clear because in an autotransformer configuration, a relatively low kVA transformer can be treated as a higher effective kVA device. In my case, a 1.5 kVA transformer (single phase) wired in a buck configuration yields 8.6 kVA.

Without considering SCCR, I can pay in the ballpark of $250 for such a transformer, whereas if I want a 10 kVA isolation transformer (the next lowest standard size is 7.5 kVA, less that what I need), I'm looking at nearly $650.

This is only the monetary cost difference though. The area footprint of the isolation transformer is triple that of the buck transformer, and saving panel space in this particular instance is rather important.

The SCCR question may make it a moot point, however.

Happy Monday,
The Shackled Designer

(Would it be pretentious of me to adopt a nickname of Shak?)

Then you would be getting a custom auto transformer because I don't believe that it would be a commonly available design.
If you went that rough you could ask the manufacturer to advise what the CL ability it has to reduce the fault current for you.

 

[Shackled Designer]

Then you would be getting a custom auto transformer because I don't believe that it would be a commonly available design.

The auto transformer lists in the manufacturer's catalog, so I thought that it must not be that custom . . .

As I am quite new to this forum, I'll ask if there is any problem with me citing specific product part numbers and naming their vendors.

Respectfully offered,
The Shackled Designer

 

[templdl]

The auto transformer lists in the manufacturer's catalog, so I thought that it must not be that custom . . .

As I am quite new to this forum, I'll ask if there is any problem with me citing specific product part numbers and naming their vendors.

Respectfully offered,
The Shackled Designer

I don't think so. To ask opinions about or to review products is a normal practice as that is how we gather and share opinions and ideas.We may also suggest or recommend a product to address specific application issues. But to sell, promote and advertise products which promotes their sale is not acceptable.
I would assume that the manufacturers catalog you are referring to is from a manufacturer that we are familiar with.

 

[Shackled Designer]

OK. Thank you, templdl, for clarifying on my question of acceptable vendor citation. I can assure you that I am a transformer agnostic, and so I do not intend to be advertising.

The buck transformer I had in mind is from Hammond Power Solutions, part number Q1C5DTCF. This is actually a single-phase isolation transformer with a 240 x 480 primary and a 24 x 48 secondary. However, it is easily wired as a buck transformer, and Hammond provides a HV:LV selection chart for a wide variety of voltage choices, including 276:230.

In the interest of impartiality, I observe that at least two other transformer vendors, Acme and Jefferson, provide very similar selection charts in their buck-boost catalog sections that list the 276:230 voltage configuration along with many other conversion ratios.

Amongst the various providers and transformers used, the wiring diagrams for BB configurations are not all identical. Given that there is conductivity between the primary and secondary in buck/boost transformers, they plainly cannot be treated as isolating devices. However, while some schemas make at least one connection directly between primary and secondary, others make both connections (single-phase) through the windings of the transformer. In those cases, an AC short circuit might receive a bit more current limiting.

Ultimately, however, whether such a transformer actually does limit current in a short circuit event may be merely academic if indeed UL508A precludes its use as a remedy for low SCCR. I have now found one document that refers to the UL508A and which explicitly rules out the autotransformer as a current limiting device. Siemens released this publication covering UL Guidelines in November, 2010, entitled, Industrial Control Panels for the North American Market (Order # A5E02118900). Page 117 is very clear in its wording on the subject. Other documents I've read have been rather vague on that point.

Best regards,
The Shackled Designer

 

[templdl]

Auto transformers having CL capabilities should point out the it is insignificant enough to be considered as being irrelevant. That's probably the reason that you don't see any references to it.
I worked with some excelled transformer design engineers over the who I'm sure had the capability to calculate what the current limiting capabilities were. But, in my 21 years in the industry that has never been a concern.
The only thing left for you to do here on out is to talk to Hammond and/or the other manufacturers regarding the CL capabilities.
I would anticipate that an autotransformer would be simulate to an iron core reactor to some extent. It may end up to be insignificant enough to even matter.

 

[kwired]

The line to ground available fault current on the "common" will not change as it is connected directly to the source voltage. The other line will have reduced available fault current - amount will depend on impedance in the windings, same for line to line on the load side of the transformer. Now if the auto transformer happened to have the "common" connected to the grounded circuit conductor, then the "load" side of the transformer would have transformer impedance limiting fault current on the "load" side of the transformer.

 

[Shackled Designer]

Hi, kwired.

The line to ground available fault current on the "common" will not change as it is connected directly to the source voltage. The other line will have reduced available fault current - amount will depend on impedance in the windings, same for line to line on the load side of the transformer. Now if the auto transformer happened to have the "common" connected to the grounded circuit conductor, then the "load" side of the transformer would have transformer impedance limiting fault current on the "load" side of the transformer.

I agree with what you are saying, but as templdl points out, current limiting capabilities of an autotransformer are likely . . .

insignificant enough to be considered as being irrelevant.

I'm guessing that there would maybe be 5 ohms or so of resistance across the coil with the rest of the impedance of the winding coming from inductive reactance. As long as the short circuit fault current comes from an AC source, there may be a measurable amount of impedance, but I'm not certain that code written for short circuit fault handling makes the assumption of an AC source. For (an extremely unlikely) example, suppose the asteroid hits my panel and causes an unusually high DC current short . . . OK, I know, we've got bigger issues than NEC in that scenario, but SCCR code is surely written for what ought to be exceptional rather than normal circumstances.

My original question concerning SCCR was raised because I had thought about adding a GFCI receptacle (PLC programming socket) to the buck-generated 240V line which supports several other devices. The UL508A assumed rating of the unmarked GFCI receptacle is 2kA, which is really low if I am attempting to design for a sensible panel rating. In view of the discussion so far here as well as commentary I have since discovered regarding SCCR and autotransformers, I'm now thinking that I should simply specify a separate, dedicated isolating control transformer for the receptacle. If I properly understand what I have read so far, putting the socket behind a control transformer should completely remove it from SCCR considerations.

The only way I see around this is if the autotransformer were to be treated as part of the feeder circuit of the panel. In that case, each of the several devices on the secondary of the buck transformer could be treated (and protected) as individual branches. I question whether this is legitimate, however, as a transformer seems more like its own branch with respect to the feeder circuit. Is there any NEC or UL language which clarifies this?

Kind regards,
The Shackled Designer

 

[kwired]

Hi, kwired.



I agree with what you are saying, but as templdl points out, current limiting capabilities of an autotransformer are likely . . .



I'm guessing that there would maybe be 5 ohms or so of resistance across the coil with the rest of the impedance of the winding coming from inductive reactance. As long as the short circuit fault current comes from an AC source, there may be a measurable amount of impedance, but I'm not certain that code written for short circuit fault handling makes the assumption of an AC source. For (an extremely unlikely) example, suppose the asteroid hits my panel and causes an unusually high DC current short . . . OK, I know, we've got bigger issues than NEC in that scenario, but SCCR code is surely written for what ought to be exceptional rather than normal circumstances.

My original question concerning SCCR was raised because I had thought about adding a GFCI receptacle (PLC programming socket) to the buck-generated 240V line which supports several other devices. The UL508A assumed rating of the unmarked GFCI receptacle is 2kA, which is really low if I am attempting to design for a sensible panel rating. In view of the discussion so far here as well as commentary I have since discovered regarding SCCR and autotransformers, I'm now thinking that I should simply specify a separate, dedicated isolating control transformer for the receptacle. If I properly understand what I have read so far, putting the socket behind a control transformer should completely remove it from SCCR considerations.

The only way I see around this is if the autotransformer were to be treated as part of the feeder circuit of the panel. In that case, each of the several devices on the secondary of the buck transformer could be treated (and protected) as individual branches. I question whether this is legitimate, however, as a transformer seems more like its own branch with respect to the feeder circuit. Is there any NEC or UL language which clarifies this?

Kind regards,
The Shackled Designer

Calculating the impedance of the transformer winding is a little bit outside of what I am used to doing so I am not real sharp with details there, but in comparison to how much the available fault current will drop with just a little additional length to the run of conductors in most cases, 5 ohms would be a pretty high resistance. Quite often if available fault current at a service is too high at a service you can fix the problem by simply moving the source a little farther away and all it takes is maybe 20-30 feet in many cases.

I also see this transformer acting like (maybe not entirely like) a line reactor, which are used for the purpose of limiting fault current levels.

 

[templdl]

Calculating the impedance of the transformer winding is a little bit outside of what I am used to doing so I am not real sharp with details there, but in comparison to how much the available fault current will drop with just a little additional length to the run of conductors in most cases, 5 ohms would be a pretty high resistance. Quite often if available fault current at a service is too high at a service you can fix the problem by simply moving the source a little farther away and all it takes is maybe 20-30 feet in many cases.

I also see this transformer acting like (maybe not entirely like) a line reactor, which are used for the purpose of limiting fault current levels.

To the point that I made initially in response to the OP regarding an iron core reactor. Also, to the point that I made previously if the BB manufacturer has support from good transformer design engineers it is those engineers that could address this issue. I know for a fact that the two transformer manufacturers that I worked as a sales and application engineer my engineers that I relied on for design issues had the ability of answering these types of question with a ball park figure to see if farther calculations would be practical or not. They would either reply that the CL capability is insignificant to be even worth doing a calculation on or that there may be a benefit of knowing what it may be.
But, rather than beating around the bush with theory and guesses simply run this issue by the BB transformer manufacture and get an answer directly from them the manufacturer themselves. I know that if I was still in my old positions I would have no problem addressing this issue with my engineers. It may be that talking directly with the manufacturer will result in an accurate answer but an answer that you don't want to hear. What frustrates me is that I once had access to a brain trust that could address these design issues which is simulate to not having the key to the lock of the door anymore.

 

[kwired]

What frustrates me is that I once had access to a brain trust that could address these design issues which is simulate to not having the key to the lock of the door anymore.

How long has it been since your wife left you?:lol:

Sorry, that may not be very funny to some people, but also may be hilarious to others.