LFMC instead of expansion fitting on long Rooftop EMT conduit Run

[electrofelon]

My company is investigating using a short section of LFMC (flex) in place of an expansion fitting to compensate for expansion and contraction of long runs of EMT conduit on Rooftops. The idea is that this will save cost. I'm not entirely supportive, but we have a few folks pushing the idea.

There are a few concerns:

- the LFMC might sag onto the roof. possible solution is to bend the EMT up a bit at the ends so the curve of the LFMC goes away from the roof, rather than sagging onto the roof
- the LFMC is difficult to bend and so it will build up pressure in the conduits and not releive the strain on the conduit and support system that the expansion fittings are meant to mitigate.

We had one of our subs build a little mock up with a 1" and 3" section. we are thinking it might be more feasible with the 1" because it's easier to bend...

Does anyone out there have experience with this? Thoughts?

Thanks,

View attachment 14109 View attachment 14108

I am in the camp of not worrying about expansion. If this is like most flat roof installations, the whole assemble is floating up on the roof on durablocks anyway so its got plenty of freedom to move. Unless you have straight runs that are ridgedly fixed at both ends, I think there is no issue at all. I think you'd be better off having the guys double check that all their fittings are tight vs having them spend time on this. I dont think Ive ever seen gas piping on flat roofs have flex or expansion fittings.....

 

[jaggedben]

That's a good point too. Perhaps why it separated, but its twin conduit didn't separate.

There's too many unknowns to be confident one way or the other. For all we know, the sister conduit is about separate but just hanging on by 1/32". (The vertical striations on both conduits might indicate as much. Or they might not.) Or the one in front continues in the same direction for 150ft while the one in back makes a 90 degree bend in another direction after only 10ft.

In general that sounds like it was written by someone who has no real world experience with installing conduit and wire or someone who has only inspected installations that were made by someone who had no real world experience with those types of installations. The only one that I would agree with is the last one...aluminum terminations need a bit more care in the installation that do copper ones.

FWIW, the author is quite widely respected in the PV world, and does not come off as inexperienced when you meet him in person.

 

[petersonra]

Yes, but you probably weren't dealing with 1,000 Vdc circuits that are energized whenever light falls on the roof. :thumbsup:

why would the voltage affect expansion of conduits due to thermal effects?

 

[petersonra]

personally, I think it is not a horrid idea if you actually needed it but my experiences with LFMC has not been real good. The fittings seem to just fall apart after a few years. Maybe poor installation technique or poor quality fittings. I don't know. Just seen enough of it to make we wary. It is the same reason I am wary of EMT outside. Just seen enough of it that has come apart not to trust it.

 

[shortcircuit2]

Here is a link to a chart for expansion for tubing...

http://steeltubeinstitute.org/steel...ds/sites/4/2013/04/Co-Efficient-Expansion.pdf

 

[iwire]

The rate of expansion and contraction between steel conduits and the steel or aluminum structures they are secured to are so similar it makes it unnecessary unless the goal is to overbuild the job for ones own reasons.

You PV guys do understand that we have been running long lengths of steel conduit on roofs long before PV was a thing right?

Yes, but you probably weren't dealing with 1,000 Vdc circuits that are energized whenever light falls on the roof. :thumbsup:

Whut? :?

 

[iwire]

FWIW, the author is quite widely respected in the PV world, and does not come off as inexperienced when you meet him in person.

So what do you think of that list?

 

[iwire]

The fittings seem to just fall apart after a few years. Maybe poor installation technique or poor quality fittings. I don't know.

The cheapest fittings are crap, barely grab the LFMC under ideal conditions.

On the other hand if you ask for say OZ-Gedney fittings there is no excuse for them to come apart unless the installer is a moron.

 

[SolarPro]

Whut? :?

My point is that the consequences associated with conductor damage are higher, and conductor insulation damage is more likely to result in a fault ar higher voltage levels. There are definitely elements of PV systems that are over-built as compared to a conventional ac electrical system serving customer-sited loads. If you have a fault in an ac circuit, a circuit breaker will trip and de-energize those circuit conductors. PV systems don't operate the same way. If you have a fault, the inverter might shut down and annunciate the fault. But the circuit conductors remain energized.

 

[jaggedben]

why would the voltage affect expansion of conduits due to thermal effects?

The point is that if thermal effects do result in a fault, the dangers are different and much higher. A typical AC circuit will be shut down by a breaker. A PV source circuit will not be.

 

[SolarPro]

These data from a voltage drop analysis for a 1.58MW PV system also illustrate the challenge:

Total distance of source circuit conductors [10 to 14 AWG]: 51,800 to 69,300 feet (depending on whether system is designed at 1,000 Vdc or 600-Vdc)
Total distance of PV output circuit conductors [1,000 to 300 kcmil] : 13,000 feet

Sure, 1.58 MW is a pretty big rooftop system. But MW-scale rooftop systems are a dime a dozen these days.

In other words, the opportunity for a failure is greatly increased or concentrated on these rooftops as compared to what you might otherwise find.

 

[electrofelon]

The point is that if thermal effects do result in a fault, the dangers are different and much higher. A typical AC circuit will be shut down by a breaker. A PV source circuit will not be.

Do you have data to back that up? Im not saying you are wrong and maybe there are studies that show that, but it is not necessarily intuitive to me and I could see it either way. Yes with PV, a source or output circuit wont shut off, but you also have the fact that it is current limited where a typical feeder or branch circuit is not. A fault on a non current limited raceway is not necessarily going to trip instantly especially if there is poor workmanship. I recently had a fault on a 150 amp feeder in 2" EMT and before the breaker tripped it blew a 2 inch hole in the pipe, blew melted metal everywhere - the surrounding area could have easily caught fire in the right conditions. Have any "experts" or studies declared that a conduit containing PV conductors is more of a fire risk?

 

[ggunn]

Have any "experts" or studies declared that a conduit containing PV conductors is more of a fire risk?

Well, yes. It's the driving force behind the new rapid shutdown requirements.

 

[iwire]

Have any "experts" or studies declared that a conduit containing PV conductors is more of a fire risk?

Well, yes. It's the driving force behind the new rapid shutdown requirements.

I do not believe that for a moment.

The reason for the rapid shutdown is not because PV conductors are more of a fire risk it is just that it had been impossible to shut them down at all for any reason short blacking out the sun.

 

[iwire]

The point is that if thermal effects do result in a fault, the dangers are different and much higher.

You think a PV circuit presents a much higher danger faulting than say a 100 amp 480 volt feeder to an RTU?

I would like to see some numbers to back that up.

A typical AC circuit will be shut down by a breaker. A PV source circuit will not be.

The fault current levels are much lower in PV, the current is self limited by the panels, beyond that by the OCPD in the combiners. I am not seeing this higher level of danger.

 

[electrofelon]

The reason for the rapid shutdown is not because PV conductors are more of a fire risk it is just that it had been impossible to shut them down at all for any reason short blacking out the sun.

I dont spend much time studying the ROP's, comments, etc, but this was my understanding too: that rapid shut down is for firefighters who dont want energized circuits, or at least want to know where energized circuits are and are not. I was not aware that there was an assumption that there are more likely to respond to a building BECAUSE OF the PV system.

 

[ggunn]

The reason for the rapid shutdown is not because PV conductors are more of a fire risk it is just that it had been impossible to shut them down at all for any reason short blacking out the sun.

And you don't think that makes them more of a fire hazard? Rapid shutdown would likely have prevented the Bakersfield fire.

 

[Smart $]

Yeah, I guess I realize that it is conduits subject to the same temperature situation. The one thing you can't see in that photo, is that the intact raceway could be a much shorter length, with therefore less of an issue of thermal expansion.
...

Tightness of compression nut, among other possible variables, will determine what joint, or even what side of what joint will slip the most. Say insertion depth is 5/8" for both conduits. That gives up to 1-1/4" of play if both sides slip equally. However, slipping out could occur in as little as 5/8" (or less if not completely inserted :angel if one side is held rigid while the other side slips.

If all other variables were equal in the two runs shown, the contraction may have been shared by both sides of the upper run, while the right side of the lower run could have been rigid, resulting in the left side slipping out.

 

[Smart $]

Here is a link to a chart for expansion for tubing...

http://steeltubeinstitute.org/steel...ds/sites/4/2013/04/Co-Efficient-Expansion.pdf

I don't believe anyone has said steel conduit does not obey the laws of thermal expansion and contraction. However, the substrate to which the conduit is attached also expands and contracts. It is the difference that we need be concerned with.

Look up the coefficient of expansion for concrete. You'll find that it varies widely rather than a set coefficient for steel (though steel will vary some with composition just like everything else)... and that steel's coefficient is typically within the range exhibited by concrete.

Given the choice, I'll choose an expansion joint over LFMC any day of the week. I've been involved with too many bad wire pull experiences where LFMC was used as a substitute as the less expensive option. FWIW, they were all 3" or 4" conduits with large conductors.

 

[iwire]

And you don't think that makes them more of a fire hazard?

No I do not think that the fact conductors are supplied by a PV system makes them more dangerous and requires running the conduits differently.

If anything was more dangerous it was just the simple fact de-energizing them was not easy.