LumberJocks

Wood Shrinking and Swelling

  • Advertise with us

« back to Wood & Lumber forum

Forum topic by clin posted 07-29-2016 05:22 PM 524 views 0 times favorited 16 replies Add to Favorites Watch
View clin's profile

clin

513 posts in 460 days


07-29-2016 05:22 PM

I’m familiar with the basics of how wood shrinks and swells with changes in moisture. How different wood reacts differently. That there’s differences with flat vs quarter sawn. But I’ve got a question specific to an upcoming project.

I’m making a yard gate out of western red cedar. I’m gong to build a heavy frame from nominal 6” wide material with mortise and tenon joints. There will be a center stile dividing the gate into two panels about 12” x 44” each.

The panels will be made using floating tongue and groove boards. My question is how much should I allow for expansion?

I’ve actually studied the seasonal temperature and humidity data for my area (Albuquerque, New Mexico) and based on this and a calculation of wood moisture content to humidity and temperature, the expected range of moisture content is from 7% to 12%. Based on this, the expected wood change is about 1.2% across flat sawn boards.

My question has to do with not humidity, but rain and snow. The wood will have an oil finish on it (Penofin). So it will be sealed to a degree, but not as well as a coat of paint.

So do I have to account for rain? If so should I just use the worst case swelling with wood that is saturated (30% MC)?

This is mostly an issue because I’m putting the panel boards at a 45 deg angle. While the 45 deg does reduce and change to about 70% of max (cosine of 45 deg), it does mean that it will expand in both directions (horizontal and vertical). If I just have a 1.2% change of the wood, factor in the 45 deg angle, and the max 44” panel, I get:

44” X 0.012 X 0.7 = 0.37” (1/8”) which is very manageable.

However, if I have a maximum swelling, that is 3.8% and gives 1.2”.

Since the boards will settle in the bottom groove due to gravity, the 1.2” room needs to be in the top rail. Of course I need some additional depth to hold the board. So lets say I go with 1/5”. That seems extreme deep to me.

If I really need to go that deep, I would likely consider adding a horizontal rail int he middle to divide that vertical in half. Though that’s not the look I want and would be more work.

EDIT: After thinking about the geometry of this, because the boards are angled and expand across their width much more than their length, the total panel does not remain square when it expands. Think of it like stretching a rectangle by pulling on two corners.

The extremes of the expansion also depends on whether the boards are glued to each other or allowed to slide along each other. There is no simple it expands by cos(45 deg) of the width expansion as I had thought. It becomes a bit difficult to even define total expansion since the area doesn’t stay square.

-- Clin


16 replies so far

View MadMark's profile

MadMark

978 posts in 917 days


#1 posted 07-29-2016 05:33 PM

Your assumptions are incorrect. Wood does not expand lengthwise anywhere near as it does widthwise. Figure expansion without angular correction and THAT will be your width dimensional change. Take 5% of the width expansion as the elongation.

Refer to the USDA ‘Wood Handbook’ for more details on the maths and expansion curves.

M

-- Madmark - Madmark2150@yahoo.com Wiretreefarm.com

View Lazyman's profile

Lazyman

694 posts in 851 days


#2 posted 07-29-2016 06:12 PM

(Note: this is somewhat speculation on my part)
It seems to me that living in such a dry place, the max you computed would happen only when you consider that the wood will continue to dry and shrink some after you complete it, unless of course you dry it down to the minimum before you build it. Also, as long as you reapply the oil finish regularly, every 1 to 2 years so that it stays “fresh”, it will repel most water so should never get saturated enough to swell to the max. The only place that might not be true is at the ends sitting in a groove at the bottom of the gate frame so make sure you have a way for water to drain if you do have a groove in the frame.

If I understand your design and that you are using multiple tongue and groove boards sitting loosely and sort of stacked diagonally in a groove in the frame, one approach to mitigating some of this is by not having the boards floating completely loose in the frame of the gate. Instead, lock (2 screws at each end?) them in place with enough of a gap that each board could expand without having to move all of the boards above it. This approach allows each board to independently slide into the one next to them as they expand and contract. You are basically spreading the gap out over each board and so even if you decide to allow the max you calculated, it should easily fit within the depth of the tongue and groove of each board so you should never see a gap. So instead of worrying about the total expansion across the width, figure out what the expansion is for each board and space them accordingly.

-- Nathan, TX -- Hire the lazy man. He may not do as much work but that's because he will find a better way.

View bondogaposis's profile

bondogaposis

4028 posts in 1815 days


#3 posted 07-29-2016 06:14 PM

Lengthwise expansion is negligible and you can ignore it.

-- Bondo Gaposis

View Lazyman's profile

Lazyman

694 posts in 851 days


#4 posted 07-29-2016 06:35 PM


Your assumptions are incorrect. Wood does not expand lengthwise anywhere near as it does widthwise. Figure expansion without angular correction and THAT will be your width dimensional change. Take 5% of the width expansion as the elongation.

Refer to the USDA Wood Handbook for more details on the maths and expansion curves.

M

- MadMark

I think he was trying to compute how much of the expansion of diagonal boards would be reflected in the horizontal and vertical dimensions of the frame. My concern is that he will still get 100% of the expansion in one diagonal direction (across the width of the boards) but as you said, only 5% in the other diagonal across the length (shorter boards will obviously expand a little less than the longer ones) . I think that locking them into place so that the width expansion is only relative to the board on either side is the only way to deal with that.

-- Nathan, TX -- Hire the lazy man. He may not do as much work but that's because he will find a better way.

View Aj2's profile

Aj2

690 posts in 1262 days


#5 posted 07-29-2016 07:15 PM

I would allow for about 3/16 th.One the sides so your groove need to be at least 3/8 1/2 inch would be better.You don’t have to be so scientific.
If you can get quarter sawn for your panels then you will really be siting good.Wrc is very stable and a good outdoor wood.Because it’s soft it will also compress a lot.
If you can set your panels on a tongue,So the panel bottoms will have the groove.Sheds water better.
I also use gorilla glue.

I’ve made lots of gates.

Aj

View clin's profile

clin

513 posts in 460 days


#6 posted 07-29-2016 08:07 PM



Your assumptions are incorrect. Wood does not expand lengthwise anywhere near as it does widthwise. Figure expansion without angular correction and THAT will be your width dimensional change. Take 5% of the width expansion as the elongation.

Refer to the USDA Wood Handbook for more details on the maths and expansion curves.

M

- MadMark


Lengthwise expansion is negligible and you can ignore it.

- bondogaposis

You’ve both misunderstood me. I’m not suggesting there is significant change in length. The boards will be set at a 45 deg angle. So the expansion of the width of the board affects BOTH horizontal and vertical dimensions. Due to the 45 deg angle, the change in the witdh of the board, affects each direction by about 70%.


(Note: this is somewhat speculation on my part)
It seems to me that living in such a dry place, the max you computed would happen only when you consider that the wood will continue to dry and shrink some after you complete it, unless of course you dry it down to the minimum before you build it. Also, as long as you reapply the oil finish regularly, every 1 to 2 years so that it stays “fresh”, it will repel most water so should never get saturated enough to swell to the max. The only place that might not be true is at the ends sitting in a groove at the bottom of the gate frame so make sure you have a way for water to drain if you do have a groove in the frame.

If I understand your design and that you are using multiple tongue and groove boards sitting loosely and sort of stacked diagonally in a groove in the frame, one approach to mitigating some of this is by not having the boards floating completely loose in the frame of the gate. Instead, lock (2 screws at each end?) them in place with enough of a gap that each board could expand without having to move all of the boards above it. This approach allows each board to independently slide into the one next to them as they expand and contract. You are basically spreading the gap out over each board and so even if you decide to allow the max you calculated, it should easily fit within the depth of the tongue and groove of each board so you should never see a gap. So instead of worrying about the total expansion across the width, figure out what the expansion is for each board and space them accordingly.

- Lazyman

The wood appears to be as dry as it will ever get. As measured with 2 moisture meters it is maybe 6%. Supplier said they’ve had it for several months, it was stored covered, but outside and we’ve had bone dry humidity for months. And cedar dries quickly. So I think it pretty much will only swell from it’s current size.

There will be a groove in the bottom rail, but I’m planning to put weep holes in that. So that should help somewhat.

When you say “2 screws at each end?”, do you mean 2 screws total, one at each end? That would make sense, simple put a screw in the middle of each board and allow it expand sideways. But if you mean 2 screws at each end (4 total), that would seem like a mistake.

In any case I understand the idea behind it and I think that makes a lot of sense. Probably work better anyway. Just because there are no fasteners, doesn’t mean the wood couldn’t bind if it has to move all the boards above it.


I would allow for about 3/16 th.One the sides so your groove need to be at least 3/8 1/2 inch would be better.You don t have to be so scientific.
If you can get quarter sawn for your panels then you will really be siting good.Wrc is very stable and a good outdoor wood.Because it s soft it will also compress a lot.
If you can set your panels on a tongue,So the panel bottoms will have the groove.Sheds water better.
I also use gorilla glue.

I ve made lots of gates.

Aj

- Aj2

I agree. 3/16” on each board adds up to plenty across 8 boards.

If I attach the each board, with space when I built it, what is the best way to to that? Screws were mentioned earlier. However with the board resting in a groove, channel, dado (whatever you want to call it), I can’t see a good way to to this without the fastener showing.

I suppose I could put a fastener in the edge at the top corner of each board. It would be inside the groove and covered by the board that goes above it. Sort of like laying wood flooring.

This may sound weird, but what about putting a screw or cutoff nail into the groove in the frame, that each of the panel boards rests on. Similar to the way a shelf rests on shelf pins. So I’d slide in a board, then put this screw or nail in above that board, then slide the next board in and so on. Each board rests on the fastener, not the board below it.

-- Clin

View Lazyman's profile

Lazyman

694 posts in 851 days


#7 posted 07-29-2016 09:10 PM

You could be right about 1 versus 2 screws at each end but personally, I think that 2 is better and won’t be a problem. Here is why. If your boards are 6” wide for example and you put 2 screws so that they divide the board into 3rds, that is really only 2” between the screws. Even at 3.8% expansion you’ll get at most about .076” movement between the 2 screws, less if your 70% assumption for the 45 degree angle is correct. 2/3rds of the total expansion on each board will be outside the screws. Because WRC is a soft wood, .076 should be easily within the compression tolerances of WRC, especially since each hole will absorb half of that. If you are still worried about that you could pre-drill oversized or oblong holes much like you would on a breadboard table top end but I doubt that is necessary, especially since the finish should prevent some if not most of the expansion, especially on a vertical surface.

I understand that the screws don’t match your design aesthetic. I suppose you could try dowels but you’ll have similar expansion concerns and they could split or get loose over time. At a minimum use screws that won’t react with the wood such as stainless steel. Home Depot carries some 2 1/4” trim head stainless screws that will leave a minimal entry hole. You could try to put a dab of paint on the heads to hide them a little if they really bug you. I would just screw them straight through the side of the groove on which ever side is the least visible rather than trying to toenail or some other strategy. By having each board firmly screwed to the frame, it may actually make it stronger and help prevent sagging that could cause you to add a turnbuckle later especially if the boards angle upward toward the hinge side of the gate. If the longest board goes corner to corner that would be even better.

-- Nathan, TX -- Hire the lazy man. He may not do as much work but that's because he will find a better way.

View rwe2156's profile

rwe2156

2193 posts in 944 days


#8 posted 07-29-2016 09:23 PM

Clin, I think you ARE correct. There will vertical AND horizontal expansion NOT due to boards expanding lenthwise, but in taking the whole panel in to account. Due the 45ºangle of the boards, the whole panel will push at 45º to the frame. The main force vector will be 45º across the frame but will also be pushing vertically expecially as you get the corners. Hope this makes sense.

I think if you build it considering what the wood will do in the future, that will dictate the spacing.

Summer and humidity 10% and you expect it to go up (you should be able to find date for seasonal avg humidity) then build it loose. If building it in winter, build it tight.

I would probably glue the entire panel together and put panel in a groove with substantial depth. I would not screw it I would let the panel float.

Good idea about the 1/4 sawn wood.

-- Everything is a prototype thats why its one of a kind!!

View Aj2's profile

Aj2

690 posts in 1262 days


#9 posted 07-29-2016 09:29 PM

Whats the purpose of the screws?To keep the boards centered or held in place and allow expansion and contraction?
If so then it’s up to your design when you build do what feels right.
Keep in mind places where water gets trapped causes the wood to swell and crack.
I have found holes drilled for drainage don’t work very well they clog.

Good luck Sounds like a fun project.

Aj

View pintodeluxe's profile

pintodeluxe

4855 posts in 2277 days


#10 posted 07-29-2016 10:27 PM

Go to the shrinkulator and enter your wood species and dimensions etc. Then you can compare expansion and contraction at maximum / minimum expected moisture content in your area. It’s a very helpful tool.
http://www.woodbin.com/calcs/shrinkulator/

-- Willie, Washington "If You Choose Not To Decide, You Still Have Made a Choice" - Rush

View AandCstyle's profile

AandCstyle

2571 posts in 1721 days


#11 posted 07-29-2016 11:09 PM

Clin, you asked about rain & snow. Both are very fleeting in ABQ, as you know, so they would only have a very temporary and minimal impact on the moisture content (MC) of your panels. That is, the rain would dry up before the wood’s MC had increased significantly, then the MC would start to decrease again. I doubt there would be any measurable change in the wood’s dimensions even during the monsoon season. Personally, I would be more concerned about the sun’s effect than the ambient humidity. FWIW

-- Art

View JBrow's profile

JBrow

818 posts in 384 days


#12 posted 07-30-2016 02:55 AM

clin,

An alternative to screws to hold each field plank in place within the frame is a dab of water resistant glue at the center of each plank. If confined to the center of each plank, wood movement would be accommodated.

I would think ensuring the end grain of the planks are well sealed would go a long way to reducing moisture exchange within the planks. As moisture strikes the panels it can run into the groves where it will remain damp for a good while. I doubt this is much of a problem for the top and side grooves, but water can collect in the bottom groove. If the end of the planks setting in the bottom groove are held up from contacting the bottom of groove by perhaps 1/8” or so, some additional protection can be afforded, but only if some means of draining water out of the bottom groove can be found. Weep holes on the bottom or in the sides of the bottom rail would help drain water. A chamfer formed on the bottom rail, if added, can direct some of the water away from the bottom groove.

It is impossible to predict whether the gate might require a repair in the future. If it does and a plank is damaged and the planks are captured by the frame, replacing a plank would be very difficult. Forming the grooves in a way that the panel planks can be accessed would address this potential problem.

View clin's profile

clin

513 posts in 460 days


#13 posted 07-30-2016 04:01 AM



clin,

An alternative to screws to hold each field plank in place within the frame is a dab of water resistant glue at the center of each plank. If confined to the center of each plank, wood movement would be accommodated.

I would think ensuring the end grain of the planks are well sealed would go a long way to reducing moisture exchange within the planks. As moisture strikes the panels it can run into the groves where it will remain damp for a good while. I doubt this is much of a problem for the top and side grooves, but water can collect in the bottom groove. If the end of the planks setting in the bottom groove are held up from contacting the bottom of groove by perhaps 1/8” or so, some additional protection can be afforded, but only if some means of draining water out of the bottom groove can be found. Weep holes on the bottom or in the sides of the bottom rail would help drain water. A chamfer formed on the bottom rail, if added, can direct some of the water away from the bottom groove.

It is impossible to predict whether the gate might require a repair in the future. If it does and a plank is damaged and the planks are captured by the frame, replacing a plank would be very difficult. Forming the grooves in a way that the panel planks can be accessed would address this potential problem.

- JBrow

I had also considered a dab of glue in the center. However, I was hoping to assemble three sides of the frame and slide the panel boards into place. So I can’t see how to get glue in the center of each board. I think trying to glue it all together at one time would just be too much.

Sealing the ends of the boards is a great idea, simple and would likely help. I was already planing to chamfer the bottom frame a bit. I can’t really see that helping much, but better to have the surface slope away from the bottom channel than towards it.

Also, tacking some stainless steel bards into the the boards where they touch the bottom would be a way to keep the board off the bottom. But then again, it creates an an area that can hold a lot more water. So perhaps that would do more harm than good.

I can’t see any reason to be too concerned about future repairs. If I build it right, it shouldn’t have any problems. And I can’t think of any reason the gate will actually get damaged from something hitting it. Always possible of course. There are always ways to get creative when doing repairs if I had to replace an entire board.


Clin, you asked about rain & snow. Both are very fleeting in ABQ, as you know, so they would only have a very temporary and minimal impact on the moisture content (MC) of your panels. That is, the rain would dry up before the wood s MC had increased significantly, then the MC would start to decrease again. I doubt there would be any measurable change in the wood s dimensions even during the monsoon season. Personally, I would be more concerned about the sun s effect than the ambient humidity. FWIW

- AandCstyle

Humidity is not the issue. I had already looked into that. What really got me thinking about this was a rain a few days ago, and I noticed one of our existing gates swelled quite a bit. That’s what got me thinking about rain/snow vs just humidity, which of course we don’t generally have any high humidity for any length of time. Now that gate is a POS primarily made of painted plywood. It is on the list to be replaced. But I have to get a new gate built for a new opening in a wall.

Obviously we don’t get a lot of rain and snow. But there is the occasional wet spell where we might have rain for several days. That’s pretty much the question, design for humidity, or design for rain?

Simplest thing might be to just toenail a stainless steel brad in the center, at the end of each board. The hole would be quite small, and of course could fill it if needed. Doing this I could spread the gap across all the boards and avoid having an unusually large groove in the top rail.

-- Clin

View splintergroup's profile

splintergroup

828 posts in 686 days


#14 posted 07-30-2016 01:19 PM

Simple test would be to toss a scrap of know width into some water for a day, then measure the width change.

Per your calculation:


44” 0.012 0.7 = 0.37” (1/8”) which is very manageable.

you need 3/8” (0.375 = 3/8”)

View DwightC's profile

DwightC

26 posts in 240 days


#15 posted 07-30-2016 02:04 PM

Chapters 6 and 7 of Bruce Hoadley’s book, Understanding Wood, will give you a great deal of organized information on how wood moves in response to moisture (tangentally, radially and longitudinally) and practical ways of coping with that movement—principally allowances for movement and finishes. We used that book in a Characteristics of Wood course and it’s well worth having in your library. As far as exposure to water, I’d remind you that you’re building a gate to be used in the desert, not a boat that’s going to float on a pond. As far as how you pin your panel, multiple point of contact tend to add more stress than strength (think breadboard construction). On that note, I’d be inclined to use dowels rather than screws or glue, and give the dowels space to slide (again, like a breadboard), but I’ll cheerfully admit that I may be overthinking it).

showing 1 through 15 of 16 replies

Have your say...

You must be signed in to reply.

DISCLAIMER: Any posts on LJ are posted by individuals acting in their own right and do not necessarily reflect the views of LJ. LJ will not be held liable for the actions of any user.

Latest Projects | Latest Blog Entries | Latest Forum Topics

HomeRefurbers.com