Derating (Large Project)


  #121  
Old 07-11-10, 12:06 PM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
I installed a grounding bar in my jbox to land all the grounds on but I lost the instructions for it.

Are the torque values for copper conductors pretty universal or should I try to locate the correct values for the bar I have?

Technical Library - Schneider Electric United States

The instructions they have available there just say "See equipment grounding bar label for torque requirements."
 
  #122  
Old 07-11-10, 08:46 PM
F
Member
Join Date: Mar 2006
Location: NE Wis / Paris France{ In France for now }
Posts: 4,364
Upvotes: 0
Received 2 Upvotes on 2 Posts
The toqure value is the same as standard load centre bussbar just don't go too crazy with toqure otherwise you will either strip the screw out or break the bussbar { I have see both ways }

Merci.
Marc
 
  #123  
Old 07-11-10, 09:01 PM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
If I've got a ground in each conduit and all the grounds from the NM cables and conduits are tied together and there's a fault... which one is going to carry the current?
 
  #124  
Old 07-11-10, 09:08 PM
F
Member
Join Date: Mar 2006
Location: NE Wis / Paris France{ In France for now }
Posts: 4,364
Upvotes: 0
Received 2 Upvotes on 2 Posts
All of the grounding conductor { bare or green and metal conduit of course } due they are bonded or tied together as one.

If you have PVC then yes you have to add a bare or green grounding conductor.

And during fault circuit mode it will go to the load centre and kick the breaker open.

Merci.
Marc
 
  #125  
Old 07-11-10, 09:16 PM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
I guess I need to pick up a book on electrical theory. you say all of the ground wires will carry the current? How would the #14 not get overloaded if there was a ground fault on a 50 amp circuit? even if it's distributed across the grounds, its not going to limit itself based on conductor size, is it?

I don't understand this stuff
 
  #126  
Old 07-11-10, 09:25 PM
F
Member
Join Date: Mar 2006
Location: NE Wis / Paris France{ In France for now }
Posts: 4,364
Upvotes: 0
Received 2 Upvotes on 2 Posts
Originally Posted by romexican
I guess I need to pick up a book on electrical theory. you say all of the ground wires will carry the current? How would the #14 not get overloaded if there was a ground fault on a 50 amp circuit? even if it's distributed across the grounds, its not going to limit itself based on conductor size, is it?

I don't understand this stuff
The grounding conductor is very tricky to deal with it escpally with multi circuits.

If you have to bring the grounding conductor what I do is sized it by the largest breaker of entire set if you have 50 amp breaker then it will be wise to use 6.0mm² { # 10 AWG } However I rather go larger due you have few on them.

I know you mention fexibale conduit so I really recommend that you bring out 16mm² { #6 AWG } for this purpose to hook up at the ground bussbar at the large junction box location.

That will take care the issue with grounding.

Merci.
Marc
 
  #127  
Old 07-11-10, 09:29 PM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
Originally Posted by french277V
The grounding conductor is very tricky to deal with it escpally with multi circuits.

If you have to bring the grounding conductor what I do is sized it by the largest breaker of entire set if you have 50 amp breaker then it will be wise to use 6.0mm² { # 10 AWG } However I rather go larger due you have few on them.

I know you mention fexibale conduit so I really recommend that you bring out 16mm² { #6 AWG } for this purpose to hook up at the ground bussbar at the large junction box location.

That will take care the issue with grounding.

Merci.
Marc
You're suggesting a #6 to be pulled inside the conduit? Not sure I understand that one.

Right now I've got a ground in each conduit that's sized to the largest circuit so I have no grounds larger than #10. Where the 2" flex conduit attaches to the main panel I am planning on using a bonding bushing with a short piece of #6 attached to the ground bar
 
  #128  
Old 07-12-10, 08:11 PM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
Does 300.14 apply to a junction box like the one I'm dealing with? I have about 1.5ft of free conductor from the front of the box. I thought it was a lot at first but then had to consider how far the conductors could have to reach across the box in the future if anything changes.

Trying to keep things as neat as possible. Wouldn't wanna violate 110.12 ! I can see this thing is going to look like a box of spaghetti pretty soon and I don't even have the NM cables attached to it yet.

You can see at the bottom of the box where I've mounted the grounding bar. Since I have the bar, I don't need to attach a grounding lug to the threaded hole provided in the box do I? I tried to screw a 10-32 ground screw into the hole and was struggling to thread it in. When I unscrewed it, the threads were crushed. Not sure what happened there.

 

Last edited by romexican; 07-12-10 at 10:03 PM.
  #129  
Old 07-12-10, 10:35 PM
F
Member
Join Date: Mar 2006
Location: NE Wis / Paris France{ In France for now }
Posts: 4,364
Upvotes: 0
Received 2 Upvotes on 2 Posts
Did you use the self tapping machine screws ?? they are diffrent than standard self tappers and yes it will crush a little but once you get it on good and tight you will be fine.

Per the NEC code you need at least 2 threads in the sheet metal.


Merci,
Marc
 
  #130  
Old 07-12-10, 10:36 PM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
No they werent self-tapping. The hole was already threaded.
 
  #131  
Old 07-12-10, 10:43 PM
F
Member
Join Date: Mar 2006
Location: NE Wis / Paris France{ In France for now }
Posts: 4,364
Upvotes: 0
Received 2 Upvotes on 2 Posts
You should be using the 8-32 or 10-32 depending on which size hole you bore out. that are two most common size.

Merci.
Marc
 
  #132  
Old 07-12-10, 10:46 PM
mukansamonkey's Avatar
Member
Join Date: Nov 2007
Location: Indianapolis
Posts: 114
Upvotes: 0
Received 0 Upvotes on 0 Posts
300.14 does apply. However, note that it says "6 inches from where it emerges from the raceway". This is just so down the road you don't end up with a tiny stub of wire that can't be worked with... the wires do not have to reach across a 12" box. Although like you said, some extra length makes it easier to work with in a box this big.

Okay! Time for fun explaining grounding.
Firstly, the ground on an individual circuit can be smaller than the hot/neutral wires because it does not normally carry any current. A short to ground should result in the breaker throwing in a fraction of a second, so the small ground wire doesn't have time to heat up and melt. If it were expected to hold the ground fault current for a significant amount of time it'd have to be bigger, but this shouldn't happen in a residential system.

Second, every circuit should have its own ground wire run with it, and be sized off the largest circuit that ground is protecting. Looks like you've got that one under control, just want to point out that it's a bad idea to have a hot and its ground separated. Causes resistance problems, can lead to overheating or bad breaker operation.

Third, don't rely on flex or EMT to function as a ground. Flex has a laundry list of limitations, starting with fitting requirements. EMT is a crappy ground as well, because at each fitting you're relying on the screw to maintain contact. One loose screw you'll never notice, but it'll raise the resistance at that junction enough that a ground fault may not throw the breaker anymore. And then there's the fittings at each box, can't be sure just how much tight metal-to-metal contact there is. Grounding only works if it's a very low-resistance network, it doesn't take much resistance to mess it up.

In fact, even with rigid conduit the conduit isn't really a reliable ground. The fittings aren't always wrench-tight. Some folks set up a ground test rig, using a typical industrial power source (200A transformer at something around 2kV I think) and made a bolted short at the far end of the pipe using a variety of ground paths. The only one that worked right was a ground wire inside the conduit. Using the rigid as a ground resulted in an impressive display of fittings arc-welding themselves, and a voltage drop of over 100V between one end of the conduit and the other. Not good.

The upshot of all that is, if you have a hot conductor, run a ground with it. In the case of your two foot of flex between box and panel, run a wire from the ground bar in the box to the one in the panel. Then strip off a bit of insulation near the bushing at each end of the flex, and slip the conductor into the clamp on the side of the bushing. That's some good reliable grounding.
 
  #133  
Old 07-13-10, 08:42 AM
I
Forum Topic Moderator
Join Date: Feb 2005
Location: Near Lansing, Michigan
Posts: 9,785
Upvotes: 0
Received 45 Upvotes on 43 Posts
Originally Posted by romexican
I guess I need to pick up a book on electrical theory. you say all of the ground wires will carry the current?
Yes, current takes all available paths. That stupid "path of least resistance" quote that everyone learns is wrong.

How would the #14 not get overloaded if there was a ground fault on a 50 amp circuit? even if it's distributed across the grounds, its not going to limit itself based on conductor size, is it?
Yes the current is governed by the conductor size. When the current flows through two paths that are in parallel, the majority of the amps flow on the largest conductor.

The theory is very much the same if you have a pressurized water system with a garden hose and fire hose hooked up. If you open the nozzles on both hoses, you will get water flowing from each but the majority of the gallons come out of the fire hose by virtue of its larger diameter. Yet you will still get some flow through the smaller hose. No matter how many hoses of various sizes you add to the system all of them will have at least some flow even though the small ones will eventually become a trickle.

Parallel ground paths work in exactly the same way. The more paths you add, the more the current divides across all available paths, but the largest diameter paths get the majority of the amps.

This site explains the math behind exactly how much current flows on each path: Current in a Parallel Circuit
 
  #134  
Old 07-13-10, 04:00 PM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
Originally Posted by ibpooks
Yes the current is governed by the conductor size. When the current flows through two paths that are in parallel, the majority of the amps flow on the largest conductor.
How does the smaller conductor not get overloaded at first?

When I think of the hose analogy, I think of the idea that the water will enter both hoses, and keep filling them up until it backs up to where it started. Only a certain amount can be pumped into the hose because of the diameter. How does a conductor reach this limit?
 
  #135  
Old 07-13-10, 09:04 PM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
Wow.. okay so I got a little too eager and decided to start swapping out circuits today.

I disconnected every circuit in the panel and then worked on attaching the jbox with the 2" flex. Didn't go as planned. The existing 2" KO was so far over that I couldn't get my locknut on the flex connector. I had to punch a new one inside the existing one about a 1/2" over to get more room.

First mistake was starting at around 4pm. "How long could it possibly take?" I was starting to run out of daylight. Not fun.

Second mistake was thinking I could get away with using a box only 4" tall. A bundle of solid #14 doesn't pull well across a box at a sharp angle. Pulling the #12 should be a real treat.

Third mistake was starting the swap before I had all of the new conductors. I had planned on doing some of them later on but not this many. I had to bring a bunch of NM cables back into the panel through the top KOs so I could reconnect them. Made a huge mess of the panel inside but I got everything reconnected safely. I was moving fast, but I certainly didn't rush it and make it unsafe.

Debating weather I should get a taller jbox...

More to come...
 
  #136  
Old 07-17-10, 02:42 PM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
So now I'm toying with the idea of ditching the jbox to transition from EMT to 2" flex and just attaching 3/4" flex to each EMT. I don't wanna play with my bender in the garage all day trying to bend the EMT into what I could easily do with the flex. Any reason not to do it this way?

I'll come down about a foot short of the panel and then use EMT to flex coupler. How much of an offset can you get in 3/4" steel flex in only 1ft? Haven't played with it yet.
 
  #137  
Old 07-17-10, 03:13 PM
mukansamonkey's Avatar
Member
Join Date: Nov 2007
Location: Indianapolis
Posts: 114
Upvotes: 0
Received 0 Upvotes on 0 Posts
Originally Posted by romexican
How does the smaller conductor not get overloaded at first?

When I think of the hose analogy, I think of the idea that the water will enter both hoses, and keep filling them up until it backs up to where it started. Only a certain amount can be pumped into the hose because of the diameter. How does a conductor reach this limit?
The hose analogy only gets you so far. Look at it like this: you can increase the flow in each hose if you increase the water pressure. But eventually the hose will physically fail due to too much pressure. The electrical equivalent, forcing more amperage down a wire, creates more heat. Enough heat builds up and the insulation melts, which is the wire failing.

Where the analogy falls apart is that a sudden pressure spike in the water moves relatively slowly, and probably makes the hose fail the moment the pressure spike hits it. A spike in amperage moves very fast, but the heat builds up over time. So a small conductor has more resistance and so carries less current than a big conductor run parallel to it, but the small one can also take less heat before failing. In either case, if the short circuit protection disconnects fast enough (1/240th of a second is common), even the relatively huge current of a bolted short circuit won't make enough heat in that short time to cause insulation failure.

Incidentally, this is why overload protection is entirely distinct from short-circuit protection. It just happens that both devices are present inside a residential circuit breaker. The overload device is usually a biwire strip that heats up and disengages if there's a bit too much current for a significant period of time. The short-circuit device is a solenoid that yanks open a contact very fast if a lot of current is applied. The solenoid won't operate unless the overcurrent is very large, and it operates much faster than a mechanism that uses thermal buildup to function.
-------------
A foot of 3/4" flex can offset about 3-4" no problem. So if you come reasonably close you should be okay. If for some reason you can't get your EMT that close, it'd probably be best to buy a few feet of flex and experiment... nothing beats actually holding a piece of flex in front of where you're working and seeing what shape you can get it into.
 
  #138  
Old 07-21-10, 09:42 AM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
Ok one concern I have now is supporting the #8 wires in the conduit. My conduit system isn't compete yet, but just as a test I pushed in 3 #8, yeah I said pushed... this stuff is so slick it's pretty impressive. It's so slick that it doesn't always wanna stay put in the conduit if I get too much weight on one end of the wire. Anyways, what is done typically to keep conductors from sliding out of a conduit under their own weight?

I let a little bundle of the excess wire hang at one end of the conduit and it started to slide the entire run out! I can't imagine what it'd be like in a vertical run. Is something tied around the conductors to keep them from falling down the conduit?
 
  #139  
Old 07-21-10, 08:13 PM
F
Member
Join Date: Mar 2006
Location: NE Wis / Paris France{ In France for now }
Posts: 4,364
Upvotes: 0
Received 2 Upvotes on 2 Posts
Get couple hevey zipties that will hold in the place until you get the rest done and the new coating on the conductor you experince some called sim-pull I think that is from Southwest wire If I am correct on this part.

Merci.
Marc
 
  #140  
Old 07-21-10, 09:27 PM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
I'm looking at the T&B Shrink-Kon catalog for some white shrink tubing to put over the bare neutral on the SE cable. I'm not seeing what I think I would need though. They have it in the catalog, but not in the color I need. Should I be looking for 600v rated tubing to match the insulation rating of the other wires in the box?

http://www.tnb.com/pubint/docs/shrinkkon.pdf
 
  #141  
Old 07-28-10, 03:40 PM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
the EMT transitions to steel flex conduit for about the last foot or so. Am I understanding 250.118(5)(b) correctly? For any of the conduits that contain circuits which are more than 20 amps, the flex conduit cannot be relied on for grounding. So I'll need to use a bonding bushing?

Edit: I don't think I am understanding it right. I can't use the flex conduit as ground unless it meets the criteria of 250.118(5). Since the conduit will contain circuits of more than 20 amps, I have to pull a ground. Which I already am. So I don't think a bonding bushing is required. Just a copper ground.
 

Last edited by romexican; 07-28-10 at 04:24 PM.
  #142  
Old 07-31-10, 04:08 PM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
progress... only a few circuits left to bring into jbox.





Looks a lot better with the cover on. Not sure how to make that many wires any more organized. More slack?
 
  #143  
Old 08-01-10, 10:46 AM
S
Member
Join Date: Dec 2006
Location: orange county
Posts: 268
Upvotes: 0
Received 0 Upvotes on 0 Posts
Cool

It is my understanding that anytime one has more that 3 current carrying conductors in a" particular" raceway de-rating is required. In post #56 I believe something else to the contrary was suggested??
 
  #144  
Old 08-01-10, 11:31 AM
S
Member
Join Date: Dec 2006
Location: orange county
Posts: 268
Upvotes: 0
Received 0 Upvotes on 0 Posts
Ahhh, make that de-rating must be considered
 
  #145  
Old 08-01-10, 12:45 PM
R
Member
Thread Starter
Join Date: Aug 2009
Location: Northern California
Posts: 128
Upvotes: 0
Received 0 Upvotes on 0 Posts
Yep, derating all taken into account, which is why I ended up with 5 conduits.

One thing I didn't take into account and I'm not sure about is ambient temperature correction factors. Not sure how hot my garage gets but I'm guessing at least 110F on the hottest days.

Do I derate and then multiply by the correction factor?

Assuming an ambient temp range of 105–113F:

For my conduit with the 4 15 amp circuits I would have:
25 * 0.7 = 17.5
17.5 * 0.87 = 15.225

Does that look right?
 

Last edited by romexican; 08-01-10 at 01:04 PM.
 

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