Circuit breaker practical trip time under short circuit

[kallen88]

Hello,

I have a question regarding trip times of MCCBs under short circuit conditions in practical terms. This is to use in calculation for wire sizing as I need the amount of time the short circuit will take to clear. Even though the tripping curves for the breakers I have found indicate a trip will occur after 1 cycle or 20 ms for a 50Hz supply im not sure if I can take this as the time to clear the fault. Is it ok to use this as the fault clearing time for calculation purposes?

Thanks.

 

[kallen88]

Hey, sorry not sure if this question really belongs here I re-posted in circuit design.

 

[jraef]

No, this is a fine place to post this, you have now cross posted in Circuit Design, so that's the one you should probably red flag yourself for removal.

Generally, it's not necessary to factor in fault clearing time for conductor sizing on circuits where you would use an MCCB. Those issues tend to be incorporated into code related conductor sizing requirements. But if you must know, you look at the total clearing time, which is often not necessarily clearly stated in an MCCB (again, because it's generally deemed unnecessary). So what you will see is the instantaneous trip SENSING time on the TCC curve, then you must find and add the mechanical unlatching action and opening times and then the arc extinguishing time. If you know a specific circuit breaker THAT YOU INTEND TO USE, someone at the mfr. level will be able to eventually find all of that information for you, but what you find for one will not translate over to another one, they are all going to be slightly different.

"Will work for (the memory of) salami"

 

[healyx]

Is this to size for short-circuit temperature rise? I usually take the upper tolerance value from the trip curve for this. The breaker may also have a let-through energy profile where the I^2t at a given fault level is usually much lower than the calculated value using clearance time.

 

[kallen88]

Hello, thanks for the reply's.

I am using Australian Standards AS 3008.1.1 for cable selection up to 1kV, to specify cable sizes for a control panel build. I normally just focus on the control side of things but am now required to take some of the power and protection design work so I am a little out of my practical field of experience. Using the guidelines in the aforementioned standard there is a short circuit power calculation in order to size the conductors so they wont be damaged under the short circuit. This requires the prospected short circuit current, duration of short circuit and a constant K provided in a table. There are MCCBs being used to protect some of the branch circuits. In the case of the a fault on these branches after the breaker I believe I would need the breaker to also protect the conductors hence the need for this calculation.

Yes this is for the thermal protection of the cables. I will have a look at the data sheets again.

 

[healyx]

Hi Kallen88,

I'm in Australia too. What size circuit are you trying to design for (in Amps)? You should really have an idea of fault levels (both max and min) before you size cables. If you can give me some more detail I could run a calc for you.

 

[kallen88]

Thanks Healyx.

Ill start by providing some context. The control panel is for a small piece of process plant, for mineral separation, to be designed and operated in house. The full calculated load on the incoming is around 35 A for 6 motors (2 VFDs, 4 DOL), a couple of instruments and some of the plant lighting. I have asked the electrical contractor that installed the distribution panels on site and they had no idea of the exact amount of fault current available only that the protection devices in the panel are rated to 6kA, which I was taking as the max. There are 6 single phase branches with rated loadings of about 0.5A to 2.5A, the two VFD motor branches have 19A and 4.7A input loads and the DOL motors are from 0.32A to 3.4A with manual motor starters as the protection device. Following the AS 3008.1.1 guidelines the low amp branches require a small x section with respect to current carrying capacity so I figure I need to size them based on short circuit considerations.

Thanks for any assistance you can provide.

 

[healyx]

OK, so are all those loads fed from the one panel? If so, does that panel already exist? If the panel doesn't exist, are you needing to size the incoming cable to the panel and is this panel going to be fed from one of the other installed distribution panels that have the 6kA switchgear or from the the larger board that feeds these panels?

Firstly, if you are only feeding these small loads from an existing panel, it is likely based on what you have described that you are using MCBs (Miniture Circuit Breakers) not MCCBs (Mould Case Circuit Breakers) as stated previously. If a circuit breaker is rated at 6kA is is almost certainly an MCB. MCCBs are much larger and have higher ratings (usually at least 25kA).

If they are 60898 MCBs (as per BS EN 60898) a class 3, C Curve, 6kA (should be labelled with 6000 in a box with a 3 in the box under) device rated 20A to 32A should limit let-through energy to 52kA^2s at the high end.

A PVC 2.5mm2 cable can withstand about 77kA^2s (k=111, S=2.5). So should be OK.

Bear in mind the use of 6kA breakers does not mean the calculated maximum fault level is at or less than that. The MCBs may have been cascaded with a breaker upstream that is protecting the smaller breaker at a fault level above its native breaking level. When cascaded, the let through energy shown before, should still hold, but be sure to use the same make of breaker throughout.

If you are protecting the incomer to the board with a smallish breaker for the 35A load (maybe a 40A MCB?? although i'd go larger) it will be very difficult to achieve discrimination between that breaker and the ones feeding these small loads. A trip on one breaker may trip the upstream breaker too, taking out the whole board. This might be OK if nothing is critical and there are no safety circuits. There are rules of thumb in AS3000 for this, but they will unlikely produce a true discriminated solution.

Of course, none of this can really answer if the disconnect time requirement from AS3000 is satisfied. Again you could refer to AS3000 to add up impedances but that is a guesstimate at best but would cover your backside. To be strictly accurate you'd need to know all the cables back the the transformer.

If you are sizing the circuit breakers, don't forget to size the DOL motor breaker to 2.5x to 3x the running current to allow for starting current.

 

[kallen88]

Yes sorry, Miniature Circuit Breakers was what I was referring to. The control panel will be fed from a distribution board which is already in place. All the switchgear on this board is rated at 6kA and I do understand that this is not necessarily the SC capacity but its the only figure I have to work with. I should have mentioned the design is for a refurbishment of an older control panel which is currently in use. We need to bring it up to standard and essentially redesign the system for PLC control. I believe the breaker on the distribution board feeding the control panel is rated at 50A. I need to specify everything down stream of that such as the cables, mains isolator, contractors and branch circuit protection.

I am aware some of the concepts of power protection such as coordination and discrimination but I suppose I lack the practical experience and understanding to be confident in the decisions I make. My studies are in mechatronic engineering, not pure electrical.

From a safety standpoint there is no real concern if the the trip occurs at the distribution panel breaker (there would if it goes any further that that as the dist board is feeding other equipment in the area) however there are some critical aspects from a process standpoint. Even though the plant is operated for a 3 to 5 days every few weeks if one particular circuit goes down at any time it will mean a couple or more tedious hours in maintenance trying to unblock pipelines and pumps and few $$ in material.

Thanks for the comment about the motor protection but already on top of that one.

I will check with the breaker datasheets we use and confirm their let though energies and size the cables that way. The current capacity calculations for the main circuit of 35 A I came out to 16mm^2 and the 19 A VFD circuit at 4mm^2.

 

[kallen88]

I have access to the AS 60898.1 but I cannot find information anywhere on the A^2s classes. I have found a reference to the EN 60898 annex ZA with the class tables. The MCBs being used in our application are from Hager, which according to their catalogue are to 60898. From the tables it states that a class 3, c curve, 6kA device rated from 16A to 32A has a let-through energy of 55 kA^2s and for devices up to 16A 42 kA^2s. Im wondering where you sourced that information from as it is slightly different to what you have stated.

So in order to size the cables to account for protection against short circuit would it be enough to use the calculation you provided earlier so long as the branch protection device is enough to limit power let though during the fault?

As far as discrimination is concerned I suppose I will have to investigate what is up stream of the panel and dig further into the wiring rules to make a calculated decision.

 

[healyx]

I don't have either of those standards, but I had been advised that those figures came from BS EN 60898 annex ZA. I dont think they appear in AS 60898 but I think that is just a case of the AS standard not being amended. The AS 60898 devices have the class 3 label on them anyway. Manufacturer data takes precedence anyway - you don't need the actual figures from the standard. Do you data for the Hager devices? I've always had some trouble getting data for them.

Yes, as far as I know, provided the thermal limit of the cable is greater than the let-through energy you should be OK. Bear in mind that to be complete you should check let-through across the range from min fault to max fault. Those let-through energies are for the maximum fault. If the minimum current was in the overload region you theoretically could get a higher let-through up to 5s.

For discrimination with breakers the size you are proposing, provided you aren't servicing any safety services, AS3000 2.5.7.2.3 would deem discrimination to be achieved if the ratio of the rating between successive breakers is at least 2x. So if you have a 50A breaker upstream, provided you use 25A or less in the control panel, you are OK per the standard. In reality that will probably only hold for fault currents up to about 400A. It is very difficult to achieve total discrimination between two MCBs.

 

[waross]

The impedance of small conductors generally limits the short circuit current so that conductor damage is rarely an issue for control panels.
Conductor damage may occur in a control panel but it is invariably the result of a code violation or a failure of the protection devices.
I am not surprised that your contractor was unable to provide information on the damage curves of the conductors.
By "code" I mean, in North America, the National Electrical Code or in Canada the Canadian Electrical Code. These are the codes that contractors work to and AHJs inspect to. The code tables for the maximum ampacity of conductors and for the maximum size of breakers and or fuses provide protection against conductor damage.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[dpc]

I have asked the electrical contractor that installed the distribution panels on site and they had no idea of the exact amount of fault current available only that the protection devices in the panel are rated to 6kA, which I was taking as the max.

In general, this is a dangerous assumption. At the very least, you should try to find the upstream transformer that is serving this system and get some idea of its size.

 

[healyx]

Waross - there is a lot of debate as to whether you use the fault at the origin or at the end of a conductor to perform the short circuit temperature rise calculation. I always do the check across the whole range, that way a fault in the cable itself will not permanently cause the maximum temperature to be exceeded. Our standard (the standard the OP is using - AS3008)is vague on this, so it is generally accepted that the calculation include the origin fault level. IEC60898 devices have sufficient current limiting abilities to pass this check.

dpc - I agree that in a perfect world the whole system should be included in the analysis. If there are existing IEC60898 6kA breakers installed on the site, I would consider the use of same (make and model) to be a defensible decision. At some point we need to assume that the person who came before did the right thing.

 

[kallen88]

Thank you all for your input.

healyx, there is some information I can find on Hager devices although very general and have not seen any figures in their documentation relating to a specific devices let-though energy. I suppose the next step would be to contact our suppliers or Hager themselves. As far as the evaluating the fault current available at the transformer I would believe this would be a question for the utility supplier?

Id like to assume that someone has done the right thing but seeing as I could not get any information about this from the contractor, whether or not they were the designer, invokes a little doubt and we know what happens when we ass-u-me. As you mentioned I would like to cover my behind should anything go wrong.

I have not had any specific training on this matter only what little power design I did at uni. I did attend a training workshop however it was more geared toward the high voltage substation level and did not cover much on end user equipment design. I think I may get some help until I can rely on some personal, practical experience. Or I could just be thorough in my own research and use of the relevant standards although this may take a lot of time.

 

[healyx]

Kallen88,

All you need to know is the kVA rating of the transformer, from that we can make a good assumption of the % impedance and thus the maximum fault level at the LV terminals. If you are in a industrial plant, I'd be surprised if you didn't have a dedicated transformer (or more). I'd also be a bit surprised if you didn't have as-built documentation showing the plant's single line diagram.

If you can get the transformer kVA and the make/model of the upstream circuit breakers, you'll have enough info to guarantee that the 6kA breakers are sufficiently rated. In fact the breaker feeding the board that feeds the control panel may be the only one you need to check. If cascading is achieved you are good to go. Will need to be the same make of breakers to check the cascading tables. Checking cascading will mean you don't have to worry about cable sizes (excluding disconnect time analysis).

I'd be surprised if you find it to be done incorrectly but kudos for being thorough. Most people would just install more 6kA breakers without giving it a second thought.

Don't worry about not knowing this stuff from uni, no unis seem to teach LV systems protection. I certainly didn't learn it there. If you are interested, I'm pretty sure NHP or Schneider would run courses on it.

 

[kallen88]

Thanks again,

From the training workshop I did perform some example calculations and were given resources to estimate the transformer % impedance from its power and primary/secondary voltages to find the prospected fault current so if I had the information I would be comfortable. We operate in the metropolitan area but we are actually a registered mine site due to the amount of pilot scale mineral processing we do. We are in an industrial area so I don't think our facility we have a dedicated transformer. As a pilot testing facility we house a variety of smaller plants that are run on an as need basis for client requirements. We did have a larger plant installed that required more power than available and I believe that a new switchboard and larger cables were put in by the supply company, but I had no involvement with that.

If needed I will contact the supplier as I am sure they have all the information. As far as documentation that is kept on site, I am not sure.

This does raise another question, what about dedicated plant purchased from either in or outside Australia? As the manufacturers do not know for sure where there plant would end up how would they design protection systems? I suppose it would be up the purchaser to make sure they comply.

 

[healyx]

By dedicated plant, are you saying plant that has integrated switchboards and switchgear? Not too sure. I suppose you'd spec the equipment to us IEC devices. Using fault current limiters (HRC fuses) or cascading circuit breakers provides a guarantee for maximum fault rating such that actual location is not that important.

 

[kallen88]

By dedicated plant I mean plant that is for a dedicated purpose that may or may not be incorporated into a larger system, controlled on a local level. Not necessarily with its own dedicated switchboard but has it's own switchgear to protect it's circuit branches for equipment such as motors heating elements.

As it turns out there was a new substation transformer installed on our site to service our processing equipment and have managed to obtain the design drawings from the contractor. The transformer is a 630kVA/22kV to 415V but no listed % impedence rating. From the resources I have been given, an estimate can be made for %Z based on transformers of 0.25MVA, 0.5MVA, 1MVA up to 10MVA. The stated value of %Z for a 22kV primary voltage for a 0.5MVA is 4.5%. Does this seem accurate seeing as the installed transformer is 630kVA and not 500kVA? After that I would still need to assume the cable resistance at the control panel to calculate the actual fault current.

I did not get any information of the switchgear for the substation or main distribution board, only that the sub board feeding the plant I am designing will more than likely be a hager 6kA 50A breaker.

Thank you very much for all your input, it has been most helpful.

 

[healyx]

4.5% for a 630KVA transformer is a reasonably safe estimate. That will give you a fault level at the LV terminals of about 20-21kA.