Asphalt shingle replacement, stop those roof leaks!

If you have a leaky roof, and your roof has standard asphalt shingles, you can repair it yourself. There are MANY instructions on the Internet, and an excellent Instructable How to replace a damaged shingle by sbrown. You should view these valuable instructions, they provide information that I will not cover here.

This Instructables provides a few tips learned on a recent roof repair. They worked for me, may or may not work for you. But I'm pretty sure there will be some things here you will not see elsewhere.

I started with the venerable blue tarp on my garage roof. Where I live in Washington State, it started raining last week. May stop next July.

Step 1: Pinpoint the leak source

I found the apparent location of the leak on the underside of the roof. No, this is not always where the water is entering above, the water can travel. But it should be close.

Next, I taped a stack of rare-earth (neodymium) magnets on the underside of the roof at the high point of the leak.

Then I went back on top the roof with a cheap compass. Knowing approximately the area of the leak, the compass quickly zoomed in on the magnets under the roof. To confirm, I placed another magnet on the roof to mark the spot.

Comments

[abenjam5]

(removed by author or community request)

[Bill WW]

If you mean asphalt shingles, they are available at all home centers like Lowes and Home Depot. Not expensive at all.

[Bill WW]

Getting a wee bit cold there now, be careful.

If you need any help as you get into the work, please let me know. Well, I meant advice, but anything (almost), please ask.

Bill

[spylock]

I read an article about a man who couldnt afford a roof for his cabin,so he used 300 thread count sheets got from the Goodwill,and 8 or ten gallons of latex paint.He had his help spread on paint and he laid the sheets on top,then when it was tacky,he started adding coats on top the sheets.He said in the article that his roof had been on 20 years.

[devynhogan]

instead of tar paper, ice and water shield. its far more water resistant. also, covered in asphalt so you wouldnt have needed the gallon to fix holes.

[Bill WW]

Thanks for the comment.
Did not know about ice and water shield, will try it next time.
However, I had another motivation of the gallon of roofing cement. I had a couple of caulking tubes in the cart; my wife says: why don't you just buy this gallon can of the stuff? So I say "yes dear, that's a good idea". That decision was more marriage 101 than roofing technology 101.

[devynhogan]

haha, done the same at times myself

[nicofrog]

wow scarty about the zink! hey sometimes you cannot get TO the area to hammer a nail so YAY magnets!!

[nicofrog]

I noticed a few "Glove free moments " DO NOT DO THAT I have fiberglass permanently imbedded in the backs of my hands from that(Makes you itch and break out whenever you wear any kind of rubber gloves) they are Fiberglass reenforced asphalt shingles good posting just a note LOVE MAGNET TRICK!! awesome!

[collegecultivator]

This is great! We live in Oregon and have a leak that has been driving us batty. Thank you

[Bill WW]

Thanks for the comment.
We do know our rain here in Oregon and Washington.
Single shingles can be bought at Home Depot, I did not have to but a whole bundle. And for a small job, the tube size roofing cement, for a caulking gun, makes more sense.
Go Ducks!

[collegecultivator]

Thanks again Bill,

yep....go Ducks!

[Burgje]

The moss could be the problem in the first place. Water will be sucked up by capillary pressure.
Magnetic field on the equator 3,1 • 10-5 T, on the poles 6,0 • 10-5 T. The magnets can be as strong 6• 10-2 T. So only when there is shielding by some metal there is a problem in detecting. Nice job.

[Bill WW]

Thanks for your comment.
My scheme would not work on a metal roof, that's for sure. What I did was kind of weird idea, I thought: "gee, I wonder if this would work...?".
Can you magine a magnetic field of 10^8 T, such as on a neutron star? Wow.

[chuckyd]

You never mentioned examining the roof sheathing in the suspected area of the leak. Was there any evidence that that part of the sheathing had ever been wet on the top side?

[Bill WW]

Thanks for your comment.
Actually, this was my second roofing project, the first was a few weeks earlier on a larger leak, where I had removed way too many shingles. On the earlier job, yes I noticed wet sheathing. The second job, no.
Our garage is located where falling tree branches nad damaging the roof, causing leaks. It appeared to me that branches had hit shingles, damaging them and tearing underlayment underneath. Strangely, I could see no evidence of damage when examining the roof.

[rncbme]

Bill, I love your magnet idea. Such marking will give you vertical line of/for your leak as the water will not travel very far horizontally in it's path downhill. Thanks!

[Bill WW]

Thanks for the comment.
The magnet idea was kind of a wild idea; I had done another repair previously and had removed a LOT more shingles than I really needed to. So when I noticed another leak, I thought "how could I pinpoint the leak and simplify the job?". When I thought about the magnet, it was like, gee, I wonder if that would work?

[ElZorro]

As a carpenter of 35 years and a roofer for quite a few. All tradesmen, roofers carpenters, plumbers and electricians would just drive a nail through the roof at the proper point. Reasoning is that you are going to either cut or repair that spot anyway.
clever but absolutely unnecessary. Old timers would say you're "overthinking" it.

[Bill WW]

Thanks for the comment, I understand.
I qualify as an "old timer", but not regarding roofing. I am very much an overthinker, believe I am an expert at it. May be too late to change, but appreciate the suggestion (and common sense) of the nail idea.

[ccmore666]

sunshine also stops moss. only damps roofs will have moss growing on them.

[rimar2000]

What a clever method, that of the magnets and compass!

[Bill WW]

I am rather proud of the magnet method, for what it is worth (not much). But only los locos appreciate such things.

[rimar2000]

My roof is done in "tejas coloniales". I suppose I can't apply your method because the thickness of them (and timbering) would weaken too much the magnetic field.

Regarding this I have a question, I could not find in the web, maybe you know the answer: the strength of a magnetic field is inversely proportional to the CUBE of the distance? I think it is so, but I am not sure.

[Tim Temple]

You could still do it with tiles; you need a bigger magnet!

The magnetic force is inverse to the square of the distance. (I knew my old engineering manuals would come in handy.)

[Bill WW]

Now that there are so many of us (three, maybe) interested in how magnetic field varies with distance, I will do an experiment and post results in an Instructable in the near future.

[rimar2000]

Genial. I guess it's not too hard to do that with a sufficient approximation to differentiate 1/r2 of 1/r2.

[rimar2000]

Tim, the tickness of the roof is about 15 cm, I would need a REALLY BIG magnet. Anyway, maybe it is possible. A good compass can be very sensitive.

The magnetic force is inverse to the cube of the distance. You can "feel" this when playing with magnets. Read the comment of BillWW, please.

[Bill WW]

Interesting question; I would have thought magnetic field strength varies inversely with SQUARE of distance, as is true for light and gravity. But, Osvaldo, seems you are quite correct, magnetic field is inversely proportional to the CUBE of the distance, according to the PHD physicists I located on the web:
http://www.physicsforums.com/showthread.php?t=522223
The Instructables site is great, where else can we discuss roof leaks and the physics of magnetic field strength on the same page?

[clgonsal]

Inverse-cube is an approximation that works reasonably well when you're far enough away that you can pretend that the north and south poles are in the same place. When you're closer to the magnet than the poles are to each other then apparently inverse square is a better approximation.

From Wikipedia: "The magnetic field of permanent magnets can be quite complicated, especially near the magnet. ... The equations are non-trivial and also depend on the distance from the magnet and the orientation of the magnet."

In any case, your magnet trick is pretty nifty.

[Bill WW]

Thanks for adding to the magnet force at distance discussion. Good thought about the closeness of the poles. And as far as orientation of the magnet, that makes sense, especially if you look at the lines of force, such as with iron fillings on a paper sheet near a magnet. Very non-trivial. Much easier to visualize the inverse square law regarding light intensity.

[rimar2000]

Congratulations, Bill, your instructable was yesterday in the bulletin.

[Bill WW]

"Bulletin?" No se "bulletin". Que es (en Instructables)?

[rimar2000]

Bill, I get a weekly summary of the most important publications of Instructables. I name it "bulletin".

[clgonsal]

The official name is "newsletter". Go to the main instructables.com page, scroll to the bottom, and you'll see "Join Our Newsletter".

Sadly, there doesn't seem to be a consistently maintained archive of the newsletters (somebody please prove me wrong). Here's a screenshot of the one containing this Instructable:

[rimar2000]

OK, thanks.

[blkhawk]

Correct me if I'm wrong but I know that nailing zinc strips at the top of peaked roofs prevents the growth of moss on shingles.

[Bill WW]

Yes, I'm sure that is correct. I have a test roof or two to check such things out. Where we have zinc roof penetrations (vents, etc) there is no moss downhill from them. I sprayed moss killer a few days ago, so most of the moss may be on the way out.