Building Roof Pass-Through

[PCHome]

Working on a metal building with heavy spray-foam insulation and no underlaying plywood between the insulation and the steel, I am looking for a roof cap or box of some sort that can handle the wires through to the inside in a neat and tidy way but nearly all I see are simply the two-wire type. If Im seeing it correctly, this setup will need 12 unless the negatives can be fed in as one in which case it would need 7 plus a couple grounds.

What is typically used to get multiple wires through or is it done with each pair through its own cap? I'm not asking how to get the wires inside through the spray foam but I need to know what to use on the outside that can handle multiple PV feeds. As much as I've looked and searched, I see no reference to this detail that is helpful.

• No conduits or junction boxes on the wall
• No roof peak-cap feed-through possible
• No eave feed-through possible
• No combiner box is being used *
  * the MPPT has multiple inputs for six or so separate arrays

 

[MTM98290]

If possible you might consider combining them in a junction box on the roof then continuing into building with just + - and earth in the conduit if they are all going to same charge controller.

Are you sure that your 435w panels won't exceed max PV amps? That is right on the edge.

 

[Daddy Tanuki]

run it down and to the side in conduit.. why would you create an ingress for water on the topside?

 

[Norwasian]

I'm probably not doing things so professionally, but I intend to run my PV wire directly under the ridge cap and down through the 4" vent at the peak. I have installed the ridge cap so that it overlaps the metal panels by about a foot, which should be adequate to keep windblown rain out while at the same time designed to vent hot air from below. The vent is there whether I add wires to it or not--and it's a convenient place to put the wires.

 

[PCHome]

Thank you for the replies. I do not want to install anything down the wall and the building has heavy spray-foam insulation so using the ridge cap won't work either. I know how to jury-rig something but I was asking primarily what is typically done on professional installations as I see nothing premade to do it. I did find some come larger boxes that were waterproof but I don't think those were for roof-mounting.

I already have a small 24v 200w solar panel on a different part of the roof which was put in as a stopgap to maintain the security system and remote cameras and, as it works well, I'll likely just leave it but I do need to get actual power to the receptacles as soon as possible. This 24v one uses one of the two-wire caps as often found on RVs and it is well-sealed under the edge of the panel so not directly in harm's way. I suspect I can use one of these for each group but I don't particularly want to put in so many possibilities for leakage.

Are you sure that your 435w panels won't exceed max PV amps? That is right on the edge.

I'm not sure what you mean on the edge as you don't know the specifications of the controller. If anything, this is at the lower end which is why possibly 12 more panels might be added later and I do have 24 panels to work with. In fact, the 3X4 configuration was suggested by the controller manufacturer based on the panel specifications.

 

[MTM98290]

I'm not sure what you mean on the edge as you don't know the specifications of the controller. If anything, this is at the lower end which is why possibly 12 more panels might be added later and I do have 24 panels to work with. In fact, the 3X4 configuration was suggested by the controller manufacturer based on the panel specifications.

I may have misunderstood, if your panels are 3S4P you would likely exceed the current spec of MC4 connectors etc. You must be using more than one MPPT or are you combining the series strings and using heavy gauge wire from the combiner to the SCC?

 

[Q-Dog]

It shouldn't be too difficult to push a conduit through the foam insulation in your roof, then use a can of spray foam to seal it back up when you are done.

Metal roof panel profiles and ridge caps are different, so its difficult to recommend a method to get the wires through the roof without actually seeing what you have.

Maybe you can use rigid conduit with a weatherhead and whatever flashing that works with your roof.

 

[PCHome]

I may have misunderstood, if your panels are 3S4P you would likely exceed the current spec of MC4 connectors etc. You must be using more than one MPPT or are you combining the series strings and using heavy gauge wire from the combiner to the SCC?

I thought the term 3S4P was for battery connections, not for solar panels but I could be mistaken as I'm quite rusty in these things. Althoug7h I have one from the original design, a combiner won't be used as the MPPT has connections for multiple arrays so the pair from each series group of four panels will have its own connection to it with more connections for the remainder of the panels at some later date. The voltage will be relatively high but not the amperage. Of course, the installers will know that, I'm sure, as they are supplying the wire and connectors. That is kind of the whole point of my question which is how to feed inside all the cables to the MPPT.

It shouldn't be too difficult to push a conduit through the foam insulation in your roof, then use a can of spray foam to seal it back up when you are done.

Metal roof panel profiles and ridge caps are different, so its difficult to recommend a method to get the wires through the roof without actually seeing what you have.

It's not a matter of getting it through the foam but rather what goes on the outside of roof above. It won't be pushed through it any event as doing so may dislodge its seal to the metal but rather an appropriately-sized hole will be bored as needed for a tight fit. Again, that's not the question or issue. I wasn't really clear but that was my main point in mentioning the small panel currently there.

The roof's peak caps are flush so out of the question and I wouldn't want to run the wires up-slope anyway.

Maybe you can use rigid conduit with a weatherhead and whatever flashing that works with your roof.

that was my original plan a decade ago and I do have one along with a flexible boot but am trying to see what more rthe modern options are. A device like that isn't particularly unobtrusive either so I was hoping to have something smaller than can be tucked under the panels out of direct weather. Also not sure what to use in the weatherhead's helmet to seal the cables individually as most rubber blocks for these have only a limited number of holes.

 

[MTM98290]

I thought the term 3S4P was for battery connections, not for solar panels but I could be mistaken as I'm quite rusty in these things. Althoug7h I have one from the original design, a combiner won't be used as the MPPT has connections for multiple arrays so the pair from each series group of four panels will have its own connection to it with more connections for the remainder of the panels at some later date. The voltage will be relatively high but not the amperage. Of course, the installers will know that, I'm sure, as they are supplying the wire and connectors. That is kind of the whole point of my question which is how to feed inside all the cables to the MPPT.

That makes sense, disregard my comments then. If your MPPT has multiple inputs for each string then you will be fine. I was mistakenly thinking that you were combining the strings into one input.

 

[Q-Dog]

There are a couple of boxes that I know of for getting wires through a roof, and probably a few I don't know of. I used EZ Solar for my shingled roof. Maybe they have something for metal? Again, we might have to see the specific roof to make a suggestion.

 

[PCHome]

Thank you. I had looked at the EZ Solar boxes at one point but now I see better, clearer photos and one might just work. I hadn't thought of putting multiple wires through a single connector. The cap I have for my smaller 24V system allows only one per for good sealing so presumed that part would be the same.

That makes sense, disregard my comments then. If your MPPT has multiple inputs for each string then you will be fine. I was mistakenly thinking that you were combining the strings into one input.

No worries but I also misspoke. While we're still using multiple feeds from the panels, it's not as many individual ones as I had thought as I misread the instructions sent to me by the manufacturer for my particular arrangement. It is apparently grouping together in parallel two of the 4-panel groups for one of the MPPT inputs and leaving the remainder 4-panel group to another input. That will allow room for the rest of the panels whenever they are installed to be connected similarly.

Anyway, moving in the right direction with everything, so far under control now that I have found an (expensive) installer willing to take on the job when I bought the panels elsewhere. Most will work only with their own products.

 

[Robar]

For what it's worth, I'm planning to do something like this with my metal roof installation:

 

[Norwasian]

For what it's worth, I'm planning to do something like this with my metal roof installation:

For what it's worth, an experienced handyman here in Thailand told me that silicon lasts about 3 years or so, then loses its stick to the roof due to heat damage. I think this would likely vary based on latitude, but I have found it true here that the stuff has a limited usable life. My recommendation, based on this, would be to do all you can to ensure the roof penetration location would be a cooler location, and shade it if possible. If it's near the peak, it should have less impact if it does actually leak--but that may not be the coolest spot. Otherwise, of course, just be prepared to redo the silicon every few years.

If there were a way of using rubber with a clamp-type fitting so that it remained tight, I would think it might last better. You may find better advice from local experts. I've been unable to find tar here--perhaps that would last well, I don't know.

 

[Q-Dog]

For what it's worth, an experienced handyman here in Thailand told me that silicon lasts about 3 years or so, then loses its stick to the roof due to heat damage. I think this would likely vary based on latitude, but I have found it true here that the stuff has a limited usable life. My recommendation, based on this, would be to do all you can to ensure the roof penetration location would be a cooler location, and shade it if possible. If it's near the peak, it should have less impact if it does actually leak--but that may not be the coolest spot. Otherwise, of course, just be prepared to redo the silicon every few years.

If there were a way of using rubber with a clamp-type fitting so that it remained tight, I would think it might last better. You may find better advice from local experts. I've been unable to find tar here--perhaps that would last well, I don't know.

Silicon is the wrong thing to use on a roof. There are roof adhesives and sealants with the correct properties, designed specifically for roof use. Look in the roofing section of your local home center or hardware store.

 

[Norwasian]

Silicon is the wrong thing to use on a roof. There are roof adhesives and sealants with the correct properties, designed specifically for roof use. Look in the roofing section of your local home center or hardware store.

You are in the land of the blessed. Virtually any product that exists is available in America. Here, the options are often much more limited. As I said, I have yet to see tar for sale anywhere here. I bought some roof paint that claimed 100% waterproofing seal, but when the rain came, the paint leached through the cracks with the rainwater, staining the floor below the roof. Obviously, I'm less inclined to try roof paint again. I really don't know what would be better here than a 100% silicon sealant. I could try liquid nails, a lacquer spray-on adhesive/sealant, or one of several types of caulk, but I sure would have no guarantee any of it would work better than the silicon. That's why, for me, I prefer to put in a rubber seal, or make no hole through the metal at all.

Actually, perhaps an epoxy would work better here? Even superglue is not 100% insoluble in water, and will wash away over time.

 

[Robar]

You are in the land of the blessed. Virtually any product that exists is available in America. Here, the options are often much more limited. As I said, I have yet to see tar for sale anywhere here. I bought some roof paint that claimed 100% waterproofing seal, but when the rain came, the paint leached through the cracks with the rainwater, staining the floor below the roof. Obviously, I'm less inclined to try roof paint again. I really don't know what would be better here than a 100% silicon sealant. I could try liquid nails, a lacquer spray-on adhesive/sealant, or one of several types of caulk, but I sure would have no guarantee any of it would work better than the silicon. That's why, for me, I prefer to put in a rubber seal, or make no hole through the metal at all.

Actually, perhaps an epoxy would work better here? Even superglue is not 100% insoluble in water, and will wash away over time.

I used this for the rail footings so far, and plan to use it for the conduit flashing gasket as well: https://www.amazon.com/dp/B0042U7PUM

It is silicone-based, but has other components as well and is specifically designed for metal roofs, though I will be checking on it periodically over the years to see how it holds up. The combiner box, conduit, and flashing will all be underneath one of the panels, so they will not be exposed to direct sunlight. Not sure how the heating from the panel will compare to the heating if it is exposed, but this seemed to be the best option all things considered.

 

[Q-Dog]

You are in the land of the blessed. Virtually any product that exists is available in America. Here, the options are often much more limited. As I said, I have yet to see tar for sale anywhere here. I bought some roof paint that claimed 100% waterproofing seal, but when the rain came, the paint leached through the cracks with the rainwater, staining the floor below the roof. Obviously, I'm less inclined to try roof paint again. I really don't know what would be better here than a 100% silicon sealant. I could try liquid nails, a lacquer spray-on adhesive/sealant, or one of several types of caulk, but I sure would have no guarantee any of it would work better than the silicon. That's why, for me, I prefer to put in a rubber seal, or make no hole through the metal at all.

Actually, perhaps an epoxy would work better here? Even superglue is not 100% insoluble in water, and will wash away over time.

You need to talk to a professional roofer. I don't believe for one second there are no professional roofing sealants available in your country. Roofs are everywhere.

 

[Ivelostmytailagain]

Pitch pocket?

 

[Norwasian]

You need to talk to a professional roofer. I don't believe for one second there are no professional roofing sealants available in your country. Roofs are everywhere.

In my area, the local "professional roofer" is just a regular handyman, a day laborer who gets minimum wage for any form of construction work that comes up. Without supervision, he won't do a satisfactory job. I had to call the guys back who put in the metal flashing on the siding because it was leaking when the wind blew the rain in. They just used silicon around the edges to solve that. I am far more confident with my own construction skills than with anyone here.

Same is true for the engineering: I engineered the Warren trusses for the roof myself. I did a decent enough job on it, such that when the building department stepped in and shut us down until we had an engineer's signature put to it (long story...welcome to Thailand), with the sketches, photographs of the as-built trusses which were already in place, and the metal thicknesses/sizes, we were able to secure an engineer's signature within five days (the engineer also had to prepare about three dozen pages of plans for it based on my design, so it wasn't going to be a one-day thing). The engineer made no change whatsoever to what I had designed...in fact, the engineer wanted to add more support columns, but then I said those were already in my plans--we had just been stopped before this had been finished.

I simply do not trust the Thai "professionals" to be anything close to the meaning of that word. Again, Americans are blessed to have a good educational system and professional workers. Ask any Westerner what they think of the "professionals" in Thailand and I think you will get a similar opinion to what I share here. This is basically a third-world country.

 

[Q-Dog]

In my area, the local "professional roofer" is just a regular handyman, a day laborer who gets minimum wage for any form of construction work that comes up. Without supervision, he won't do a satisfactory job. I had to call the guys back who put in the metal flashing on the siding because it was leaking when the wind blew the rain in. They just used silicon around the edges to solve that. I am far more confident with my own construction skills than with anyone here.

Same is true for the engineering: I engineered the Warren trusses for the roof myself. I did a decent enough job on it, such that when the building department stepped in and shut us down until we had an engineer's signature put to it (long story...welcome to Thailand), with the sketches, photographs of the as-built trusses which were already in place, and the metal thicknesses/sizes, we were able to secure an engineer's signature within five days (the engineer also had to prepare about three dozen pages of plans for it based on my design, so it wasn't going to be a one-day thing). The engineer made no change whatsoever to what I had designed...in fact, the engineer wanted to add more support columns, but then I said those were already in my plans--we had just been stopped before this had been finished.

I simply do not trust the Thai "professionals" to be anything close to the meaning of that word. Again, Americans are blessed to have a good educational system and professional workers. Ask any Westerner what they think of the "professionals" in Thailand and I think you will get a similar opinion to what I share here. This is basically a third-world country.

You problem sounds unsolvable. I guess its a good thing that the original poster is not in Thailand.