Straightening a 7" x 2" joist. (UK)

[Chris McBrien]

Guys, I've bee putting a new floor in my house. The actual wood used was metric size of... 170mm by 47mm and C16 standard.

Three of these 4 metre lengths bowed by as much as one inch in the centre. I came up with the following procedure to straighten them.

1) Starting at one end and about one inch down, draw a string line from one end of the joist/beam to the other end, again finish about one inch down. This should give you a shape that is one inch thick at each end and two inches thick in the middle. Assuming that you started with a one inch bow in the joist.

1. Using a table saw cut right down this straight line from one end to the other.

2. This thin section of wood can now be lifted out and placed on the under or opposite side of the joist/beam.

1. Now apply some good PVA glue to both faces and screw the two pieces together with suitable length screws, about 3 inches.

If you don't like the sound of this then just draw it on a piece of A4 paper first. It is so simple.

Regards Chris McBrien.

[John Manders]

Chris, One problem is that as soon as you machine the section of a strength tested piece of timber such as C16, you invalidate its certification. Assuming that you are using this in a strength critical application, this could be a problem with building inspectors if not in practice. You didn't say whether the bow was in the 47mm or the 170mm axis. If the former, then physical force should straighten it. If it's in the latter, you have a problem. It may be possible to correct the bow by drying etc.

John

[JackD]

Why not get more wood and cut the bent ones up for blocking/sills etc.? Seems like a waste of time and who knows if you still have the strength after you have cut it up like this.

-Jack

[Fred the Red Shirt]

By definition bow is about the 170 mm axis. If it were about the other it would be crook. About the log axis it would be cup. Of course not everyone uses those definitions.

It seems to me from the description that the boards are crooked. One inch of crook in 4 meters is not, IMHO enough to worry about. Most joists have some crook in them and the carpenter simply uses them crowned edge up so that the weight of the floor and subfloor strightens them out.

If the joists are bowed, then again I wouldn't worry. One inch of bow in a 4 meter board will easily be corrected by the cribbing.

If you are really unhappy with using the joists as is then unless you have to travel by slow boat all the way to Bumfuck, Egypt to replace them, buying new ones should be faster than butchering and reassembling the ones you have now.

[Chris McBrien]

the bow was in the 170mm axis of the joist. Cutting the C16 does of course loose you the thickness of the saw blade from the 170mm.

I'm also amazed that this C16 even has certification and is the bowing included in this specification. How is that I can cut the timber to length but not to width? This C16 was so poor that some of the 400mm pieces that I had left over could be snapped, in the 170mm axis, with my bare hands.

I will never use this type of wood again in anything requiring the slightest bit of accuracy. I think it just typifies the low standards that have now become the norm in this country.

Regards Chris.

[Lawrence Wasserman]

Put the crowned side up so the highest point is in the middle of the floor and in about 100 years it should just about be level. Seriously.

[Chris McBrien]

thaks for the jovial replies.......

1. I'll be dead in 25 years time.
2. There is 400mm. between centres of these joists. If the rest are OK then one, one inch higher, stands out like Dog's balls.
3. It took less than one minute to cut 4 metres on a straight line. Two minutes to spread PVA glue and say 10 minutes to screw the two pieces together again.
4. Buying more wood is no guarantee of a straight piece. I bought 14 lengths and 3 went bowed so that's quite a percentage.

Chris.

[Morgans]

Sound like you are using the table saw to cut a line, without the rip fence or the miter gauge as a guide. GIANT no no. Sooner or later, you will mess up and have a piece of wood hit you in the gut..

If you want to do this, tack a straight piece of wood on the crooked piece, and let the straight piece guide against the rip fence.

[Michael Daly]

You've got no guarantee that you've got enough shear strength in the glue to make the wood behave as one piece*. If it doesn't hold, you'll get three soft joists in the floor (squeek!). You might want to keep them separated by nice joists. You've also severed the outer chord fibers on both sides. As someone else pointed out, it's no longer to code. If you have to replace them after inspection fails, then you'd have been better off buying new now.

Can't you find a place that allows you to pick your own? It's usually not a problem a the local Borgs here in the colonies up to about 12' long.

Mike

*there's a difference between satisfying a woodworker and satisfying building codes.

[John Manders]

Chris, I'm a bit out of date now but I used to be the Engineer in a sawmill which also carried out strength testing. The process was quite simple. Pass the timber through 3 rollers, 2 on one side about 2 ft apart and one pressing from the other side in the middle of these 2. This was to try to bend the wood between the 2 rollers. The deflection was measured all the way along the length and if it exceeded a certain value, that piece failed and was rejected. The 2 ft at either end were not tested of course. As the wood is continuously tested along it's length, it can be certified for the whole length so shortening is OK. Cutting into the section is a different matter though. The wood is strong enough in total but there is no way of knowing where that strength is concentrated across the section of the wood. The wood you are using would only have been tested in the shorter direction. Hope that answers your question. I have to admit that UK grown timber is generally of low quality. There are reasons for this, most centre on the climate and the wood before it gets to the saw mill. Basically, due to our mild climate, out trees grow quite quickly and are therefore very open grained and of poor quality. Good UK wood is obtainable but it requires a lot of selecting. I would tend to agree with others though that 1" of deflection in 4m is fairly straight for softwoods regardless of their origin.

John

[Chris McBrien]

Jim, you're correct there. Just eyeballing down the drawn straight line with me guiding/pushing and my mate holding the two pieces after the cut.

I'm not sure how you mean to attach the straight piece of wood to act as a guide against the fence. Can you enlarge.

Chris.

[Fred the Red Shirt]

There is very little sheer stress near the upper or lower surface of a beam. The Shear due to bending is maximun on the neutral plane.

I do not think there are any woodworking glues today that would be insufficient for the job.

I guess if Chris' dog were to walk backwards it would be obvious.

[Michael Daly]

That depends. If the glue line is on top of the joist and it is solidly attached to the floor, the neutral axis is shifted up to the floor and close to the glue line.

Like I said, satisfying woodworkers isn't satisfying code. If not inspected, (building or insurance - the latter being pickier in my experience) then it doesn't matter. YMMV

Mike

[Morgans]

You do need to stop that practice. Bad in everyone's book.

Let the attatched board overhang the crooked one. The fence will follow the straight board, and the cut will be parallel to the new board.

-- Jim in NC

[Chris McBrien]

Jim, OK I get the idea of attaching a striaght piece and of course it doesn't have to be the same thickness either as it is just a guide.

Guys, let's take this loss of strength a bit further. Let's say I want this 'banana' shaped piece of 7" x 2" to be a straight piece of

5" x 2". So this time I cut off my one inch strip but off of both edges and just throw them away. Now I have a nice straight piece of 5" x 2". Are you now telling me that this will be weaker than a stock piece of 5" x 2" from the yard just because the edges have been trimmed?

Chris.

[Chris McBrien]

John, I assume you are load testing with the beam/joist in its normal vertical usage position with the rollers on the 2" faces. I did some rudimentary testing when I started. I supported one beam at each end, that's 4 Metres apart. I marked a zero deflection point, in the middle, on a tape measure. I then made a rope loop over the beam and then suspended my body weight of 102kgs on the single beam at a single point. The measured deflection was 3/16" or say 5mm. Assuming the relationship is linear then 5 x 5mm = 25mm = 1" so I need to be 5 x

102kgs or 510kgs. in order to straighten the beam.

Also if the joists are layed at 400mm. centres then a gradient between two adjacent joists of 25mm. in 400mm is one in sixteen or

6.25%.

Chris.

[JackD]

Wood is typically stronger along the grain than across the grain, so the answer is yes. It will be weaker. Maybe not a lot, but it won't make it any stronger.

-Jack

[Fred the Red Shirt]

Yes, and the subfloor nails or screws will go though the glue line mooting the issue of sheer due to bending in that plane.

Yepper.

[Andy Dingley]

Yes. The grain is still banana shaped, and is now shorter. Talk to a longbow (or crossbow) maker.

-- Die Gotterspammerung - Junkmail of the Gods

[JackD]

Despite what Jim says, he is wrong.

-Jack "doesn't make a noticeable difference, but he is wrong"