A/C wire size.

[wwhitney]

trip curve

Two questions about trip curves:

1) Is the width of the region shown purely attributable to variation or calibration differences between different units? I.e. if we could repeatedly test a single given circuit breaker at the specified cold start ambient 40C temperature, would we give a single curve on the graph within that region?

2) Can anyone point me towards an annotated trip curve where an algebraic formula is given for the shape of the curve in each region? For a thermal magnetic breaker, my first thought is that there are only two regions, the thermal region where trip time t is proportional to 1/I², and a magnetic region, where trip time is constant. I imagine that's a bit too simplistic, though.

Cheers, Wayne

 

[Eddie702]

It also shows just how complicated answering a seemingly simple question can become.

It shows that the code is not written properly if all the smart minds on this forum can't come to a consensus, then how is the average contractor or inspector supposed to come to the correct answer?

The way I look at it (which is probably wrong) is that wires are selected for HVAC equipment in 440 and standard motors in 430 allow for 125% of the load on a standard motor or in the case of HVAC equipment based on 125% of the largest motor (usually the compressor) in multi motor equipment. The equipment ground is sized based on the overcurrent protection which in most cases exceeds the "normal" current rating of the wire so motors and HVAC equipment are the only equipment that I am aware of that in a normal installation the over current protection does not match the conductor size. Although I am sure there are other cases

In any case the equipment ground must be sized to a minimum as in table 250:122 That table has not changed in decades and #10 copper has been allowed for a 60 amp breaker forever. If a 60 amp breaker has been determined to be able to carry any fault that a 60 amp breaker will let through then oversizing the circuit conductors both hot and equipment ground should be allowed as ling as 250:122 is complied with;

How about Tap conductors? In the case if tap conductors the equipment ground may be undersized in relation to the over current protection ahead of the tap and in that case the equipment ground does not have to be sized larger that the ungrounded conductors of the tap.

 

[Elect117]

Reading and Configuring Time-Current Curves « Knowledge Base – Design Master Software

1) Is the width of the region shown purely attributable to variation or calibration differences between different units? I.e. if we could repeatedly test a single given circuit breaker at the specified cold start ambient 40C temperature, would we give a single curve on the graph within that region?

That is how I understand it.

2) Can anyone point me towards an annotated trip curve where an algebraic formula is given for the shape of the curve in each region? For a thermal magnetic breaker, my first thought is that there are only two regions, the thermal region where trip time t is proportional to 1/I², and a magnetic region, where trip time is constant. I imagine that's a bit too simplistic, though.

I can only find where they write it like piecewise functions or define each section of protection with their limit.

https://library.e.abb.com/public/114371fcc8e0456096db42d614bead67/2CDC400002D0201_view.pdf

 

[wwhitney]

Or am I mistaken, and we can construct a reasonable example where a normal sized EGC depends on a minimum impedance of the ungrounded conductors in order to be adequate during a fault?

If I had the time I would.

Thinking about this some more, it seems to me that if are concerned about conductors/EGC surviving a bolted fault, what we end up with is a criterion of a minimum conductor size based on the available fault current.

That is, if for sufficiently large fault currents, the OCPD opens (say magnetically) in a fixed time period (say 1 cycle), then the worst case I²t would occur when I is the full available fault current. E.g. for a fault immediately after the OCPD where the impedance contribution of the conductor and EGC between the OCPD and the fault is small enough to be negligible.

Then the temperature rise of any conductor carrying the fault current (including the EGC, which is likely the smallest conductor in the fault current path) will depend on I²t/A², where A is the cross sectional area. The square factor arises as one factor due to decreasing resistance with area (and thus decreasing power dissipation per unit length I²R and thus decreasing total energy I²Rt), and another factor due to increasing conductor mass per unit length, and thus increasing energy required to raise the conductor temperature a given amount.

The result is that if there is a maximum allowable conductor temperature rise (say, that the EGC does not get hot enough to damage the insulation of other conductors it may be in contact with), we get a maximum allowable I²t/A², which coupled with a worst case maximum fault current I and a fixed t in the instantaneous range gives us a minimum allowable A.

So then why doesn't the NEC have a rule of the form "given an AFC of X, the minimum conductor size is Y"? I mean, I haven't worked out an example of the above (partially because I'm unclear why the formula in post 100 has logs in it and is missing a square exponent on A; seems like for short time spans the heat generated has no time to be conducted away and the right hand side should just be proportional to the temperature rise), but surely it will say that with a 50,000A fault a #14 conductor would experience excessive temperature rise?

Regardless, getting back to the point of 250.122(B), it seems to me that the result is that if for a given circuit with a given size EGC, the EGC will avoid excessive temperature rise during a fault located anywhere on the circuit for any fault current in the instantaneous OCPD range up to the AFC, that same statement will be true even when the ungrounded conductor is upsized. As in either case, the worst case is that the fault is just after the OCPD and the fault current is the full AFC, and so it's the same worst case.

Cheers, Wayne

 

[mtnelect]

This might help to explain.