Results 1 to 8 of 8

Thread: Arc Flash Study

  1. #1
    Join Date
    Jun 2003
    Posts
    6,921

    Arc Flash Study

    Working on a proposal for an arc flash study.

    I've noticed that a lot of the older GE transformers do not have the impedance on the label. What to do?

    Also, what is the typical way to handle having multiple ways to configure a system? If there is a utility and a generator, I understand the energy should be calculated using each source, and the worse case used on the label.

    But what happens when the options get crazy? For example, a system with 2 utility feeds, a generator, 3 UPS's 3 transfer switches, and tie breakers??

    If I need to calculate it every which way, I'm sure the fault currents will be very low, and the breakers will trip very slowly, and everything is just going to be "dangerous".

  2. #2
    Join Date
    Jul 2018
    Location
    Dallas
    Posts
    29
    I use SKM for arc flash, not sure how it is for other software.

    If no transformer impedance is given then i use the default values for the transformer style and kVA rating. Include a disclaimer in the final report that typical values were used.

    I model the base scenario as normal utility power, then once the kinks are worked out i have to create scenarios for every possible significant system change. Ex - generator vs utility power, UPS in bypass. The software will generate an arc flash report using the worst case value comparing all scenarios.

  3. #3
    Join Date
    Mar 2013
    Location
    Rutland, VT, USA
    Posts
    349
    The software should be able to handle multiple scenarios. I will analyze all the reasonable scenarios and include at table of IE for each scenario. Labeling will be for the worst case scenario unless there are discussions with the owner about the other scenarios. For example: It may be possible to have work rules where if running on generator power, no work is done. Or prejob briefs where the report tables are used based on operating conditions to determine the PPE.

    Transformer impedance should be known from the data collection unless it is a small (<25kVA?) transformer or nameplate missing or illegible. Oh just reread, use a standard impedance. are these small transformers?

  4. #4
    Join Date
    Nov 2018
    Location
    Knoxville
    Posts
    1
    Are you using ETAP or SKM? ETAP uses two awesome tools called the scenario wizard and the study wizard, which make it easy to determine the worst case among multiple configurations.

    To provide the most accurate report, you'll need to analyze all of the options, but depending on the client, they may not want you to report on it. For example, generators have low fault current so you'll have high fault clearing times (perhaps beyond 120 cycles). In a situation like that, the customer will likely not ask for a label considering that no maintenance should be happening during generator or alternative power situations.

    There are situations though where a high fault clearing time is inevitable, like on the line side of emergency generator breakers.

  5. #5
    Join Date
    Feb 2003
    Location
    New York, 40.7514,-73.9925
    Posts
    4,588
    Quote Originally Posted by steve66 View Post
    I've noticed that a lot of the older GE transformers do not have the impedance on the label. What to do?

    Also, what is the typical way to handle having multiple ways to configure a system? If there is a utility and a generator, I understand the energy should be calculated using each source, and the worse case used on the label.

    But what happens when the options get crazy? For example, a system with 2 utility feeds, a generator, 3 UPS's 3 transfer switches, and tie breakers??

    If I need to calculate it every which way, I'm sure the fault currents will be very low, and the breakers will trip very slowly, and everything is just going to be "dangerous".
    older GE transformers do not have the impedance on the label - We use typical value for XFMR of that size and state as such in report.

    options get crazy - Yep, worst case on the label. Often it is generator scenario. Sometimes it is radial fed utility if at the end of long line and N+1 standby plant

    It is counter intuitive, as many think higher fault current scenario would be worst case, but most of the time, it is not.
    Ron

  6. #6
    Join Date
    Jun 2003
    Posts
    6,921
    I've used SKM in the past, but now we have EasyPower.

    The typical value for the xformer impedance with a disclaimer sounds like the way to go. They are mostly small transformers, but one is 112 KVA.

    I thought about trying different values, 4.5, 5, 5.5, 6, but will all the options for connecting the system, it would quickly become unmanageable. And I would have to manually keep track of the worst case for each label.

    Another question: What about the UPS'es? If there is a fault, I assume they would go into bypass mode, so are they normally just ignored? Or do I have to consider every scenario with and without the UPS driving the load?

  7. #7
    Join Date
    Feb 2003
    Location
    New York, 40.7514,-73.9925
    Posts
    4,588
    Quote Originally Posted by steve66 View Post
    Another question: What about the UPS'es? If there is a fault, I assume they would go into bypass mode, so are they normally just ignored? Or do I have to consider every scenario with and without the UPS driving the load?
    They transfer to static bypass in less than 20 ms, so we model them in bypass. Static bypass and maintenance bypass in parallel for short circuit analysis because most will transfer from one to the other in close transition and one or the other for arc flash since the resulting path impedance is higher when a single path.
    Ron

  8. #8
    Join Date
    Jul 2018
    Location
    Dallas
    Posts
    29
    Quote Originally Posted by steve66 View Post
    I've used SKM in the past, but now we have EasyPower.

    The typical value for the xformer impedance with a disclaimer sounds like the way to go. They are mostly small transformers, but one is 112 KVA.

    I thought about trying different values, 4.5, 5, 5.5, 6, but will all the options for connecting the system, it would quickly become unmanageable. And I would have to manually keep track of the worst case for each label.

    Another question: What about the UPS'es? If there is a fault, I assume they would go into bypass mode, so are they normally just ignored? Or do I have to consider every scenario with and without the UPS driving the load?
    My company models the UPS as a single bus, we do not use the typical UPS symbol/item

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •