wire and guage to 100a subpanel


  #1  
Old 02-26-04, 10:31 PM
P
Member
Thread Starter
Join Date: Apr 2001
Location: USA
Posts: 19
Upvotes: 0
Received 0 Upvotes on 0 Posts
wire and guage to 100a subpanel

I am wiring a new attached addition with a garage below it. My house is already over its max load, its a 100a service. So I am having an electrican upgrade it to 200a service. But I am wiring the addition to a 100a sub panel in the garage. My question is what wire and guage do I need to run from the main panel to the subpanel ? If length is a factor its 84 feet long to sub panel.I hope to have all wires run and in that cut cost of the elictrican.
Thanks
Phil
 
  #2  
Old 02-27-04, 02:49 AM
noxx
Visiting Guest
Posts: n/a
You'll be wanting to use #1 AWG copper. At this distance the voltage drop is well within limits. Your equipment grouning conductor size is #8 AWG.

Conduit size for this run is 1-1/4 inch for either EMT or schedule 40 PVC. (shcedule 80 runs you up to 1.5").

Hope this helps.
 
  #3  
Old 02-27-04, 07:09 AM
J
Member
Join Date: Sep 2000
Location: United States
Posts: 17,733
Upvotes: 0
Received 1 Upvote on 1 Post
A 100-amp feeder needs #2 copper (#3 if a certain set of conditions are met). noxx, where did you come up with #1? I'd like to figure out why you and I disagree.
 
  #4  
Old 02-27-04, 09:30 AM
noxx
Visiting Guest
Posts: n/a
Hmm... answering this before coffee is dangerous but I'll be short.

I used #1 because we're on 310.16 (this installation does not meet the conditions for a residential feeder to use 310.15). We're limited to a 60C rating because this wire will be fed from a residential style "plug-in" breaker, and if they manufacture those higher than their standard 40/60C rating, it's news to me.

Also should the actual load (not given) start to approach 100A, at the distance given on #2 wire we start to get close to our 3% Vdrop limit.

Also, I don't know where the person asking the question lives, who is the AHJ, and whether or not they are surly before they have their coffee.

Given these factors, there is no condition in which the use of #1 wire can be wrong, it is the "safe" answer to the question.
 
  #5  
Old 02-27-04, 01:29 PM
B
Member
Join Date: Dec 2002
Location: Connecticut USA
Posts: 130
Upvotes: 0
Received 0 Upvotes on 0 Posts
noxx

#2 Cu @ 60ºC is good to 95 Amps. 240.4(B) allows you to round up to the next standard size breaker (found in 240.6) at less than 800 Amps.

Most modern breakers are 60/75ºC rated and most lugs are 75ºC with some as high as 90ºC.

250.122(B) requires that when the ungrounded conductors are upsized for any reason that the Equipment Grounding Conductor must be upsized proportionally. The EGC with the #1 AWG's would have to be a #4 AWG.

Since the largest NM cable is #4 AWG, more than likely the #3 AWG will suffice for this installation, but there would be no harm with #2, which may be more readily available.
 
  #6  
Old 02-27-04, 07:18 PM
ampz's Avatar
Member
Join Date: Feb 2004
Location: Philadelphia
Posts: 513
Upvotes: 0
Received 0 Upvotes on 0 Posts
Just a thought,would either of you reccomend he use #2 aluminum,with a #4? If not why?
 
  #7  
Old 02-27-04, 07:28 PM
B
Member
Join Date: Dec 2002
Location: Connecticut USA
Posts: 130
Upvotes: 0
Received 0 Upvotes on 0 Posts
No

#2 Al is only good to 90A at 75ºC.
 
  #8  
Old 02-28-04, 03:56 AM
W
Member
Join Date: Jan 2004
Location: Oregon
Posts: 1,104
Upvotes: 0
Received 0 Upvotes on 0 Posts
Could someone double check me on my understanding of 240.4(B) and how it interacts with the other rules?

I thought that this exception would permit you to protect 95A conductors with a 100A breaker. However the _load_ on the conductors would still be required to be less than 95A.

For example, for the purpose of the present calculation, if the load on this panel were to be 97A 'non-continuous', then I would presume that the conductors would need to be rated at 97A or greater, no matter what 240.4(B) says.

On the other side of the coin, how does 240.4(B) interact with the various '125%' rules (eg. 215.3)? For example, if the load on this panel were 80A 'continuous', then 215.3 would require that the circuit be protected with a 100A breaker, and that the circuit conductors be properly protected by that 100A breaker. Since the load is <95A, and 95A conductors are considered protected by a 100A breaker, does the combination of 240.4(B) and 215.3 mean that for a continuous load of X you need a breaker of 1.25*X but may use conductors that are rated at slightly less than 1.25*X?

Just trying to understand these rules better; any clarifications would be appreciated.

-Jon
 
  #9  
Old 02-28-04, 08:03 AM
J
Member
Join Date: Sep 2000
Location: United States
Posts: 17,733
Upvotes: 0
Received 1 Upvote on 1 Post
People are always trying to apply the 125% rule (also often called by its inverse, the 80% rule) to situations where it doesn't apply. It applies only in the precise conditions listed in various code articles.

Bottom line is that a 100-amp breaker on #2 copper is permitted in almost all cases. The exceptions might be if that feeder supplies a known 100-amp load (virtually non-existent in a residential setting), or at least 80 amps of continuous load (virutally impossible in a residential setting).

The conditions in which you could not put a 100-amp breaker on #2 are so rare that they are hardly worth discussing.

If you have marked 75-degree terminations and are using individual wires rather than a cable assembly, you can even use #3.
 
  #10  
Old 02-28-04, 11:01 AM
noxx
Visiting Guest
Posts: n/a
Originally posted by John Nelson
The conditions in which you could not put a 100-amp breaker on #2 are so rare that they are hardly worth discussing. [/B]
This is true, however the conditions in which you cannot put #1 wire on a 100A breaker do not exist. As I said it's a sure thing.

However, I think I've demonstrated that I should probably not be calculating wire size while uncaffeinated.
 
  #11  
Old 02-28-04, 11:53 AM
ampz's Avatar
Member
Join Date: Feb 2004
Location: Philadelphia
Posts: 513
Upvotes: 0
Received 0 Upvotes on 0 Posts
Thanks Bolted,I would think that the instances that a #2 AL come close to 90 amps is extremly rare. For instance a 100 amp main panel in a residence will never have all 20 circuits energized simultaneously.
 
  #12  
Old 02-28-04, 11:54 AM
J
Member
Join Date: Sep 2000
Location: United States
Posts: 17,733
Upvotes: 0
Received 1 Upvote on 1 Post
Um... that's not a sure thing either. How about when the wire runs through an area with 160 degrees F ambient temperature?

The point is that there are exceptions to every rule. I don't want to start splitting hairs in this forum or we'll never answer anybody's question.
 
  #13  
Old 02-28-04, 11:57 AM
J
Member
Join Date: Sep 2000
Location: United States
Posts: 17,733
Upvotes: 0
Received 1 Upvote on 1 Post
ampz, Nothing at all wrong with using #2 aluminum. Just don't protect it with a 100-amp breaker. Then it's even less likely that the current will exceed its rating.
 
  #14  
Old 02-28-04, 12:23 PM
ampz's Avatar
Member
Join Date: Feb 2004
Location: Philadelphia
Posts: 513
Upvotes: 0
Received 0 Upvotes on 0 Posts
Like you said "not to split hairs" but what about 100a SEU {type XHHW} is rated as Bloted said for 90a at 75 degC & is protected with a 100a breaker. Granted,not every circuit will be on at the same time.
 
  #15  
Old 02-28-04, 03:19 PM
J
Member
Join Date: Sep 2000
Location: United States
Posts: 17,733
Upvotes: 0
Received 1 Upvote on 1 Post
The solution is simple. If you have 90-amp wire, then use no more than a 90-amp breaker.

I'm confused by your reference to 100a SEU rated for 90a. It's either 100 amps or 90 amps, not both.
 
  #16  
Old 02-29-04, 07:42 AM
W
Member
Join Date: Jan 2004
Location: Oregon
Posts: 1,104
Upvotes: 0
Received 0 Upvotes on 0 Posts
John,

I think that you said it best with ' I don't want to start splitting hairs in this forum or we'll never answer anybody's question.'

To summarize the answers to the original question, 'wire and guage to 100a subpanel?', the possibilities are:

1)Overkill that is _almost always_ okay: Use 1 ga wire for the two hots and the neutral, and using a 7 ga wire for the equipment ground. ( This is combining noxx's answer with bolted fault's reference to 250.122(B)

2)Acceptable in almost all expected cases: Use 2ga wire for the current carrying conductors, use 8ga for the equipment ground, presume 60C limits (John's answer above.)

3)Acceptable if you use 75C limits, meaning 75C terminations, and 75C cable assemblies or 75C conductors in raceway, with suitable ambient temperature limits: Use 3ga wire for the current carrying conductors and 8ga for the equipment ground. (John's alternate answer above.)

I double checked the voltage drop (84 feet, 240V, presume worst case 100A load) and it is acceptable in all cases.

Now, with the question answered, I personally would like to continue 'splitting hairs' to improve my understanding of some of the issues discussed above. I will not be offended if everyone else wants to go off and actually answer DIY questions, which after all is the main purpose of the forum. But if you want to indulge my curiosity, continuing along these lines would be appreciated.

I believe that what ampz was calling '100A SEU' is #2 AL service entrance cable, which _when used as the main power feed_ in a _dwelling_ can be used for 100A service. When used in all other applications, this cable has a normal 75C ampacity of 90A. My understanding is that this would not be applicable to the present question, unless the main service coming in was similarly sized at 100A with conductors sized using 310.15(B)(6). If the main service were 100A using #2AL, then feeders to subpanels could be 100A using #2AL. This same rule permits the use of #4 CU conductors in similar cases. Did I get this correct?

Going back to the 125% (or 80%) rules, which John mentioned are often mis-applied. If I do a feeder calculation for a dwelling unit, adding up the various loads, including the general lighting loads and applying the various demand factors, I end up with a load in VA. I divide this number by the nominal supply voltage and I get a nominal current requirement. My gut feeling is that this current requirement represents an approximation to a continuous load, since that is the point of applying demand factors as a way to combine several diverse non-continuous loads into a single overall load calculation. 230.42(A) or 215.2(A)(1) would seem to apply. But John's answer suggests that this 125% rating of the conductors and breakers would not apply. My question: do the calculations for dwelling feeder loads already work in the '100% noncontinuous load plus 125% continuous load' factors?

Thanks
Jon
 
  #17  
Old 02-29-04, 08:23 AM
ampz's Avatar
Member
Join Date: Feb 2004
Location: Philadelphia
Posts: 513
Upvotes: 0
Received 0 Upvotes on 0 Posts
Thanks Win,John,thats exactly what what I was refering to.I truly hope you dont mind these pesky questions but I {like Winnie}just want another perspective on the code interpretations. A guy I know {also an electrician} asked me why I go online & "talk shop",he said when he goes home the last thing he want to do is even talk about work or the trade.I enjoy the trade & just want to absorb all the knowledge I can. Thanks again for taking the time to field my questions.
 
  #18  
Old 02-29-04, 09:47 AM
J
Member
Join Date: Sep 2000
Location: United States
Posts: 17,733
Upvotes: 0
Received 1 Upvote on 1 Post
Excellent posts guys. Just some random thoughts.
  • I don't think we should refer to cable or wire as xxx-amp cable or wire. It isn't really rated in this manner, since ampacity depends on application. So we should just stick to defining a cable or wire by specifying the type of construction, they type of metal and the gauge of the wire.
  • While I agree with almost all of winnie's excellent post, I think you went too far when you said, "My gut feeling is that this current requirement represents an approximation to a continuous load." A demand load calculation does not determine continuous load. A demand load calculation is really meant to approximate the expected maximum load (my opinion).
  • Table 310.15(B)(6) is much debated as to exactly what it applies to. Winnie, how exactly did you conclude that, "If the main service were 100A using #2AL, then feeders to subpanels could be 100A using #2AL"? I do know that several posters here have said that #2AL is commonly used in their area for 100-amp feeders, but I'm can't reconcile that with the NEC.
  • I also agree that once we've dealt with the poster's question and he understands our answer, that it's fine to delve into some of the finer points to increase our understanding.
 
  #19  
Old 02-29-04, 05:33 PM
W
Member
Join Date: Jan 2004
Location: Oregon
Posts: 1,104
Upvotes: 0
Received 0 Upvotes on 0 Posts
John,

Thanks for the 'random thoughts'.

When I said "My gut feeling is that this current requirement represents an approximation to a continuous load.", I could have perhaps worded that "My initial understanding was that this current requirement..." I guess that I got this understanding from other situations where the calculation of large numbers of diverse loads results in an average continuous load. But I can see how it would be better read as "This calculated current requirement represents that required rating for conductors and overcurrent protection, and implicitly includes the necessary capacity for peak loads, as well as any overrating required for continuous loads." I don't think that it is actually made clear anywhere in the code. I think that it would be fair to call this number a 'nominal' value, that somehow incorporates various peak and continuous load terms.

I read table 310.15(B)(6) as saying that if a particular feeder supplies the _main_ power to a dwelling, then the specified conductor types can be used with the specified protection. So the feed to an apartment with 100A service could be 4ga copper. I then read in paragraph 310.15(B)(6) "The feeder conductors to a dwelling unit shall not be required to be larger than their service-entrance conductors.", which I interpret as meaning that _any_ feeder in a dwelling unit can use the same reduced wire size. In other words, if the service to an apartment is 100A on 4ga copper, but then off the main panel there are several circuits and a 100A subpanel, I believe that the 100A subpanel could also be fed with 4ga copper.

This might not be a good idea in the long run, because if a later capacity increase changes the main service to 200A, then the 4ga would no longer meet the requirements of 310.15(B)(6), and would have to be rated using 310.16. For this, 4ga copper has an ampacity of 70A, 85A if you use 75C components.

-Jon
 
  #20  
Old 03-01-04, 07:03 AM
ampz's Avatar
Member
Join Date: Feb 2004
Location: Philadelphia
Posts: 513
Upvotes: 0
Received 0 Upvotes on 0 Posts
John you are right,application is everything in determining wire gauge.The terms 100a or 200a cable are used daily in my neck of the woods so its just habit.In fact going into any supplyhouse asking for one of them will automatically get you AL type cable.Some houses only stock CU at their main location.Winnie got me to think of another question when he mentioned apartment services:When I change a service on a duplex apartment{orignal feed was 60a SEU}I go into the 2 gang meter with 100a SEU & use the same for the sub-mains.This is because 60a main breaker panels are not readily available & cost more than 100a.This is an accepted practice in this area,I was wondering what you feeling is & if this type of service is done differently.
Thanks guys,this is 1 heck of a great thread
 
 

Thread Tools
Search this Thread
 
Ask a Question
Question Title:
Description:
Your question will be posted in: