ryan_618
Senior Member
- Location
- Salt Lake City, Utah
Re: Parallel Conductorology 101
From the 2005, 310.4:
From the 2005, 310.4:
Parallel Conductorology 101
- Thread starter lile001
- Start date
- Status
- Location
- Mission Viejo, CA
- Occupation
- Professional Electrical Engineer
Re: Parallel Conductorology 101
Larry, "other" Bob, Ryan, Don, et al.,
I believe I see the position everyone on either side of this issue is taking and they are all reasonable from the perspective of our various functions.
One thing I have genuinely appreciated about this forum is the sincere desire for safety I have seen exhibited from our regular contributors.
As an engineer, it was "obvious" to me that this is a current divider issue and I don't see a simple, practical, "common" method to safely protect the conductors [edit add: in some odd parallel configurations] that I can readily express in "Code text." (Other than "engineering supervision," that is
)That makes "enforcement" tough. It can definitely be done, but it would take some fairly sophisticated ampacity studies and/or relatively complex relaying schemes.
I hope by now, you know my philosophies well enough to know that "enforcement" is not by itself my top priority but I'm still sympathetic with those who have the AHJ function. I do not believe the 5-point laundry list is exhaustive. To me, "physical characteristics" is not only structural/mechanical but anything (including significant inductive impedance variations) that affects the even distribution of current between the conductors.
In truth, assuming all phases are represented in each conduit, running one set in 3" IMC and another in 4" IMC would make little difference. Assuming all phases are represented in each 3/c cable, running one 3/c cable in one conduit and two 3/c cables in another probably wouldn't make much difference either. But running three 1/c in one conduit and six 1/c in another (2 per phase) could be significant. The problem is - you just don't know readily.
[ November 04, 2004, 11:01 PM: Message edited by: rbalex ]
Larry, "other" Bob, Ryan, Don, et al.,
I believe I see the position everyone on either side of this issue is taking and they are all reasonable from the perspective of our various functions.
One thing I have genuinely appreciated about this forum is the sincere desire for safety I have seen exhibited from our regular contributors.
As an engineer, it was "obvious" to me that this is a current divider issue and I don't see a simple, practical, "common" method to safely protect the conductors [edit add: in some odd parallel configurations] that I can readily express in "Code text." (Other than "engineering supervision," that is
)That makes "enforcement" tough. It can definitely be done, but it would take some fairly sophisticated ampacity studies and/or relatively complex relaying schemes. I hope by now, you know my philosophies well enough to know that "enforcement" is not by itself my top priority but I'm still sympathetic with those who have the AHJ function. I do not believe the 5-point laundry list is exhaustive. To me, "physical characteristics" is not only structural/mechanical but anything (including significant inductive impedance variations) that affects the even distribution of current between the conductors.
In truth, assuming all phases are represented in each conduit, running one set in 3" IMC and another in 4" IMC would make little difference. Assuming all phases are represented in each 3/c cable, running one 3/c cable in one conduit and two 3/c cables in another probably wouldn't make much difference either. But running three 1/c in one conduit and six 1/c in another (2 per phase) could be significant. The problem is - you just don't know readily.
[ November 04, 2004, 11:01 PM: Message edited by: rbalex ]
Re: Parallel Conductorology 101
Ryan's Post
Ryan's Post
The code says "Where run in seperate Raceways or Cables the raceways or cables must have the same characteristics. The circuit proposed is not a cable. It is sets of conductors and they meet the list 1 thru 5 as stated in 310.4. We are not using 2005 code yet. I do not see how you can turn this down using 2002 wording. If possible I think an extra conduit should be installed and do the job right if the construction site will allow.
sandsnow
Senior Member
- Location
- Southern California
Re: Parallel Conductorology 101
Thanks Don and Ryan
I'll look it up on Monday
Bob
I see your position on things here, respect your experience and knowledge and your right, I'm interested in safety primarily. One thing that my experience and this forum has made me realize is that invocaton of 90-4 is not to be taken lightly and you have to have facts or evidence to back it up.
I am also in the position of not being an engineer. I am told (by state law) I cannot check engineering calculations for that reason. We don't have an electrical engineer on staff. So when an issue comes down to engineering supervision we have only the text of the code to rely on and/or have an engineer stamp and sign the calculations and accept it. Using the text of the 2005 code takes the support (in writing) of the chief building official. Which is not a problem, just something I don't want to use at the drop of a hat.
That being said, I think the best solution for the original problem is the pin connectors. Which is one of the posters original questions. Any reason these would not work for him?
Thanks Don and Ryan
I'll look it up on Monday
Bob
I see your position on things here, respect your experience and knowledge and your right, I'm interested in safety primarily. One thing that my experience and this forum has made me realize is that invocaton of 90-4 is not to be taken lightly and you have to have facts or evidence to back it up.
I am also in the position of not being an engineer. I am told (by state law) I cannot check engineering calculations for that reason. We don't have an electrical engineer on staff. So when an issue comes down to engineering supervision we have only the text of the code to rely on and/or have an engineer stamp and sign the calculations and accept it. Using the text of the 2005 code takes the support (in writing) of the chief building official. Which is not a problem, just something I don't want to use at the drop of a hat.
That being said, I think the best solution for the original problem is the pin connectors. Which is one of the posters original questions. Any reason these would not work for him?
- Location
- Mission Viejo, CA
- Occupation
- Professional Electrical Engineer
Re: Parallel Conductorology 101
Ach "Forest for the trees" - it's a fine solution and I had planned to say so earlier.
[ November 05, 2004, 10:23 AM: Message edited by: rbalex ]
Ach "Forest for the trees" - it's a fine solution and I had planned to say so earlier.[ November 05, 2004, 10:23 AM: Message edited by: rbalex ]
Re: Parallel Conductorology 101
Sandsnow
I never said this was a proper installation. 310.4
in 2002 code does not go far enough. My only argument was that since the code was vague I could not disagree with the origional posters proposal with the exception of trying to install 8 500 kcm conductors in a 3 1/2/" C. As Ryan and Don have pointed out the change in the 2005 code makes it clear that this istallation is not permitted.
Sandsnow
I never said this was a proper installation. 310.4
in 2002 code does not go far enough. My only argument was that since the code was vague I could not disagree with the origional posters proposal with the exception of trying to install 8 500 kcm conductors in a 3 1/2/" C. As Ryan and Don have pointed out the change in the 2005 code makes it clear that this istallation is not permitted.
sandsnow
Senior Member
- Location
- Southern California
Re: Parallel Conductorology 101
bob #20
I think I confused my Bob's in my reply. No problem.
bob #20
I think I confused my Bob's in my reply. No problem.
- Location
- Mission Viejo, CA
- Occupation
- Professional Electrical Engineer
Re: Parallel Conductorology 101
First, I think Bob (#20)?s solution (a fourth conduit) would very likely be a good one too. We would only need to assure ourselves that the fourth conduit resulted in the fourth conductor?s impedance being substantially equal to the other three for proper current division.
Both his and Larry?s latest response though, bring up one of my concerns with the current ?way of things.? I don?t believe a competent PE would certify the design as we originally discussed it (4 sets in three conduits) without enough investigation to ultimately conclude that there would be a better, more economical solution (either of the two proposed by Larry or Bob (#20), for example). The original proposal was not a ?poor? design, it was an ?unsafe? one; yet, because there is some apparent vagueness in the NEC wording, at least one competent inspector felt it must be accepted. Conversely, another competent inspector would reject an installation that, in all probability, would be ?safe?; i.e., one set in a 3? conduit and one set in a 4? one. (Although thankfully, he didn?t say he would ignore an ?engineered solution? .)
This bothers me no end. Personally, I don?t believe the wording is vague from an engineering standpoint. I know exactly why the requirements are there and I know how to design for them. The NEC is becoming more and more a design specification geared primarily for the benefit of those not technically responsible for design. Section 90.1 (all of it) just isn?t true. The NEC is often impractical, has MANY requirements well beyond the minimum necessary for safety, and it IS becoming a design specification. Yet, with all that, we still have an ?unsafe? design that some inspectors believe they must accept.
[ November 08, 2004, 11:52 AM: Message edited by: rbalex ]
First, I think Bob (#20)?s solution (a fourth conduit) would very likely be a good one too. We would only need to assure ourselves that the fourth conduit resulted in the fourth conductor?s impedance being substantially equal to the other three for proper current division.
Both his and Larry?s latest response though, bring up one of my concerns with the current ?way of things.? I don?t believe a competent PE would certify the design as we originally discussed it (4 sets in three conduits) without enough investigation to ultimately conclude that there would be a better, more economical solution (either of the two proposed by Larry or Bob (#20), for example). The original proposal was not a ?poor? design, it was an ?unsafe? one; yet, because there is some apparent vagueness in the NEC wording, at least one competent inspector felt it must be accepted. Conversely, another competent inspector would reject an installation that, in all probability, would be ?safe?; i.e., one set in a 3? conduit and one set in a 4? one. (Although thankfully, he didn?t say he would ignore an ?engineered solution?
.)This bothers me no end. Personally, I don?t believe the wording is vague from an engineering standpoint. I know exactly why the requirements are there and I know how to design for them. The NEC is becoming more and more a design specification geared primarily for the benefit of those not technically responsible for design. Section 90.1 (all of it) just isn?t true. The NEC is often impractical, has MANY requirements well beyond the minimum necessary for safety, and it IS becoming a design specification. Yet, with all that, we still have an ?unsafe? design that some inspectors believe they must accept.
[ November 08, 2004, 11:52 AM: Message edited by: rbalex ]
Re: Parallel Conductorology 101
So, forgive me for raising an elementary question, but what is the calculated load to be served by this set of feeders?
The conductors should be sized for the load, not the panelboard ampacity. Rather, the panelboard is to be sized not less than the calculated feeder load.
If the non-continuous load + 125% of the continuous load is 1140A or less, the panel can be fed from 3 parallel sets of 500MCM w/ 3/0 EGC's. (Assuming 75C terminals, CU conductors and no ambient or bundling derating - all of which seem reasonable from the posting.)
That 60A is causing a whole lot of difficulty...
I assumed in reading this that 1200A is the panelboard rating.
So, forgive me for raising an elementary question, but what is the calculated load to be served by this set of feeders?
The conductors should be sized for the load, not the panelboard ampacity. Rather, the panelboard is to be sized not less than the calculated feeder load.
If the non-continuous load + 125% of the continuous load is 1140A or less, the panel can be fed from 3 parallel sets of 500MCM w/ 3/0 EGC's. (Assuming 75C terminals, CU conductors and no ambient or bundling derating - all of which seem reasonable from the posting.)
That 60A is causing a whole lot of difficulty...
ryan_618
Senior Member
- Location
- Salt Lake City, Utah
Re: Parallel Conductorology 101
Kenneth: What kind of overcurrent protection do you have in mind and how will 240.4(C) be satisfied?
Kenneth: What kind of overcurrent protection do you have in mind and how will 240.4(C) be satisfied?
ryan_618
Senior Member
- Location
- Salt Lake City, Utah
Re: Parallel Conductorology 101
Thats alright Kenneth!
Thats alright Kenneth!
- Status