LumberJocks Woodworking Forum banner

Bowed shelf

3K views 24 replies 6 participants last post by  JBrow  
#1 · (Edited by Moderator)
I am trying to construct a wall-hung decorative shelf unit I saw in a Woodworker's Journal. There are three shelves of differing lengths. I decided to use hickory and planned down my boards (that's another long story - learning opportunity on flipping the cutter knives to the other, apparently sharper, side). One of the longer shelves, 31 inches, has bowed a bit. I have read about laying the board out in the sun or using heat of some sort. I have not had the opportunity to try that (of course, a cloudy day is forecast!). Few questions - if I can "unbow" it using such a method, can I expect it to remain flat? I understand I need to make sure I don't lay it flat on my worktable but allow air flow around the board. As another option, would it work to rip the board and do a glue-up of sections? The article had a tip that stated "To minimize warping, glue up the shelves from at least three boards…" I don't have three different boards, so am not sure if reversing parts of this board could achieve that same outcome? The hickory is really pretty so am hoping to be able to figure this out and complete the project. Otherwise, may need to convince my husband that another trip to the hardwood store is in order. :)

Image
 

Attachments

#2 · (Edited by Moderator)
Having built a large kitchen with hickory, I found it can be a very unforgiving if warped or bowed. So I would not expect it to straighten.

I would try wetting the underside (concave side) and sticker -in clamps- for a week or so. You could also try heat on the convex side. But if this board is kiln dried I would not be very optimistic this will work.

It could also be stress in the board and have nothing to do with moisture content.

That being said, since it is going to be a shelf, install the crown up, and one could expect eventually gravity and the weight of what's on the shelf will work in your favor (or not) ;-)

I would try placing the board on a couple stickers and put some weight for a week or so in the middle see what happens.
 
#3 ·
A single flaw-sawn board like that is prone to cupping. That's the point in the article's tip about three boards. If it were me, I'd rip that board into four equal pieces, joint and plane them flat, and re-glue them with the growth rings alternating.

Going with four pieces will do two things - it will minimize the amount of wood you need to do to flatten them, and it will even out any future cupping so that the full board doesn't cup end to end.
 
#4 · (Edited by Moderator)
I like the idea of letting gravity do its work…but this is more decorative (plan to place some of my husband's woodturning pieces for display) so likely need to address the issue. Thanks for the suggesting of adding moisture and weight.

Having built a large kitchen with hickory, I found it can be a very unforgiving if warped or bowed. So I would not expect it to straighten.

I would try wetting the underside (concave side) and sticker -in clamps- for a week or so. You could also try heat on the convex side. But if this board is kiln dried I would not be very optimistic this will work.

It could also be stress in the board and have nothing to do with moisture content.

That being said, since it is going to be a shelf, install the crown up, and one could expect eventually gravity and the weight of what s on the shelf will work in your favor ;-)

I think I would try placing the board on a couple stickers and put some weight in the middle see what happens.

- rwe2156
 
#5 ·
I don't have a jointer, but expect I should be able to get a good edge with my table saw cut (?). Unfortunately, I'll lose width. Will have to check what "extra" I have left from the cuts I made so far and maybe I can work something into the design (lot of "craftsmanship" in my projects it seems!). Thank you for the suggestion.

A single flaw-sawn board like that is prone to cupping. That s the point in the article s tip about three boards. If it were me, I d rip that board into four equal pieces, joint and plane them flat, and re-glue them with the growth rings alternating.

Going with four pieces will do two things - it will minimize the amount of wood you need to do to flatten them, and it will even out any future cupping so that the full board doesn t cup end to end.

- RichTaylor
 
#6 ·
BB1,

Although it would change the look of your project, the front edge of the shelf could be straightened by adding a reinforcing rail along the front edge, typically a strip of wood that Âľ" thick X 1-1/2" wide, glued to the edge to the underside of the front edge of the shelf. To determine whether this might be an option, a scrap piece of wood more or less the length of the shelf and 1-1/2" wide could be clamped to the front edge of the shelf. If it fails to work, the shelf which looks to be about 1" thick, could be planed to a thickness of Âľ". This would make the shelf a little more flexible and easier to straighten with a reinforcing rail. If this method works, it may require either a dado in the back of the shelf unit or a similarly installed reinforcing rail along the back edge of the shelf. This would depend on whether the front reinforcing rail is effective in bringing the shelf to flat across its entire width.

If you elect to rip, alternately flip, and re-glue the shelf suggested by RichTaylor and the shelf has no twist, using a feather board to keep the shelf firmly against the fence would probably be the best bet for a straight glue-ready edge. A sharp and clean saw blade set perfectly 90 degrees to the table saw table and a steady feed rate can also help. If all does not go well, then some more material would have to be removed as the edges are prepared for glue-up, further narrowing the overall shelf width. If the shelf has a twist, the board may rock when ripped and getting a good glue line would be very difficult.

If the shelf is a little too narrow after ripping, some width could be added with the reinforcing rail. A rabbet cut in the rail that would engage the front and bottom surface of the shelf and then gluing in place could restore the width you need.

Like rwe2156, I am pessimistic that efforts short of planning one face flat and running the opposite face through the planer, will bring the hickory to flat and straight. I suspect the hickory has assumed its forever shape. Also if the shelf unit has sides to which the shelves attach, shelves that are bowed to differing extents could cause problems during assembly since the shelves measured from end to end could be different lengths. This could lead the non-parallel sides and/or if one shelf straightens out more than others, stress would be added to the joinery.

Another trip to the hardwood store would probably lead you back to where you are now; wood that may look good at the store but after a day or two in the shop it begins to move a little and does not look quite so good.

Although I am not sure you are interested, I found a fairly good article published by Popular Woodworking in 2009 about lumber movement. It explains the cause and results of different types of wood movement and methods to milling these boards…

http://www.popularwoodworking.com/article/why-wood-warps
 
#7 · (Edited by Moderator)
Thank you so much for all the helpful suggestions and that link on wood movement. Some issues with a typical shelving unit may not be an issue with this more "open" design. This is the item (from Wood magazine http://www.woodstore.net/plans ...not the Woodworker's Journal like I thought):

Image


Haven't had a chance to check what materials I still have to see if I have extra wood do a glue up.

BB1,

Although it would change the look of your project, the front edge of the shelf could be straightened by adding a reinforcing rail along the front edge, typically a strip of wood that Âľ" thick X 1-1/2" wide, glued to the edge to the underside of the front edge of the shelf. To determine whether this might be an option, a scrap piece of wood more or less the length of the shelf and 1-1/2" wide could be clamped to the front edge of the shelf. If it fails to work, the shelf which looks to be about 1" thick, could be planed to a thickness of Âľ". This would make the shelf a little more flexible and easier to straighten with a reinforcing rail. If this method works, it may require either a dado in the back of the shelf unit or a similarly installed reinforcing rail along the back edge of the shelf. This would depend on whether the front reinforcing rail is effective in bringing the shelf to flat across its entire width.

If you elect to rip, alternately flip, and re-glue the shelf suggested by RichTaylor and the shelf has no twist, using a feather board to keep the shelf firmly against the fence would probably be the best bet for a straight glue-ready edge. A sharp and clean saw blade set perfectly 90 degrees to the table saw table and a steady feed rate can also help. If all does not go well, then some more material would have to be removed as the edges are prepared for glue-up, further narrowing the overall shelf width. If the shelf has a twist, the board may rock when ripped and getting a good glue line would be very difficult.

If the shelf is a little too narrow after ripping, some width could be added with the reinforcing rail. A rabbet cut in the rail that would engage the front and bottom surface of the shelf and then gluing in place could restore the width you need.

Like rwe2156, I am pessimistic that efforts short of planning one face flat and running the opposite face through the planer, will bring the hickory to flat and straight. I suspect the hickory has assumed its forever shape. Also if the shelf unit has sides to which the shelves attach, shelves that are bowed to differing extents could cause problems during assembly since the shelves measured from end to end could be different lengths. This could lead the non-parallel sides and/or if one shelf straightens out more than others, stress would be added to the joinery.

Another trip to the hardwood store would probably lead you back to where you are now; wood that may look good at the store but after a day or two in the shop it begins to move a little and does not look quite so good.

Although I am not sure you are interested, I found a fairly good article published by Popular Woodworking in 2009 about lumber movement. It explains the cause and results of different types of wood movement and methods to milling these boards…

http://www.popularwoodworking.com/article/why-wood-warps

- JBrow
 

Attachments

#10 ·
RichTaylor,

I think your rip and re-glue idea is a viable method for dealing with both cupped and bowed lumber. The tricky part is getting a good glue line on the table saw with misbehaving lumber when a jointer is unavailable.

When the board is bowed, it could be ripped and re-glued, alternating bow up and bow down. With some luck, the forces within the lumber causing the bow can be counteracted when the glue-up is as BB1 showed in the photo; clamps applied to keep the surfaces of the boards flush while the glue sets.

Either way, I thought yours was a good suggestion.
 
#11 ·
Not sure my cuts were as good as I would like. I will see how it looks when I pop off the clamps.

As a secondary question, what are some good blades to promote clean cuts? I have a Freud thin kerf glue line rip that I used on my Bosch but cannot use with my new PM saw as the riving knife is too thick. I used a combination blade for this cut. I found a sale on Forrest blades but even on sale the 20 tooth was $80 and the 40 tooth was around $111. Now that I have a better saw, I am looking at upgrading to all full kerf blades but am not sure the best way to go even after reading lots of reviews.
 
#12 ·
BB1,

It may be worth a call to Powermatic to check whether a thin kerf riving knife is available for your saw.

I am sure there are a number of good saw blades designed for various cuts on the market. I use Forrest blades because I have been satisfied with these blades (Woodworker II 40 tooth full kerf combo blade on table and radial saws and the Forrest Plywood full kerf blade). The Forrest blades use quality carbide that can be re-sharpened several times and the blades are flat. The arbor holes fits the arbor snugly.

Whether you go with Forrest or blades from another manufacturer, this Forrest 22 minute promo video carries a lot of useful information that may be helpful as you sort through the mountain of info.

 
#13 ·
Out of the box, the higher-end Freud blades cut beautifully. One of the reasons for the difference in price compared to Forrest is the thickness of the carbide tips. The Forrest can be sharpened many times more than the thinner tips on less expensive blades.

By the way BB1, I think you are going to have some problems with your glue-up. You have no where near enough clamps on there.
 
#14 ·
Thank you for the info and link. Will check that out.

I have a thin riving knife ordered (back ordered I guess until later this month).

BB1,

It may be worth a call to Powermatic to check whether a thin kerf riving knife is available for your saw.

I am sure there are a number of good saw blades designed for various cuts on the market. I use Forrest blades because I have been satisfied with these blades (Woodworker II 40 tooth full kerf combo blade on table and radial saws and the Forrest Plywood full kerf blade). The Forrest blades use quality carbide that can be re-sharpened several times and the blades are flat. The arbor holes fits the arbor snugly.

Whether you go with Forrest or blades from another manufacturer, this Forrest 22 minute promo video carries a lot of useful information that may be helpful as you sort through the mountain of info.


- JBrow
 
#15 · (Edited by Moderator)
Glue up turned out ok (yes…this may be used for rationale for purchasing more clamps!) - much better than the bowed board I was staring at yesterday! On to the next steps of the plan!

Image


Out of the box, the higher-end Freud blades cut beautifully. One of the reasons for the difference in price compared to Forrest is the thickness of the carbide tips. The Forrest can be sharpened many times more than the thinner tips on less expensive blades.

By the way BB1, I think you are going to have some problems with your glue-up. You have no where near enough clamps on there.

- RichTaylor
 

Attachments

#17 ·
More clamps would have actually been helpful (and I'm always looking for sales on clamps so any excuse, serious or not, is valid). :) I don't have many that are "deep" so had to pull out my two Kregs and then didn't realize they would stick to the boards. Thankfully was able to pry them off without real damage. Every project is a learning experience!

Thanks again for the suggestions. Made some progress and hope to deal with the assembly this afternoon .

Image


That looks great. I was kidding about the clamps. Not that any of us needs a rationale for buying more.

- RichTaylor
 

Attachments

#18 ·
Freud also makes a full kerf glue line rip blade for about the same price as the thin kerf. As for flipping growth rings, I'm not a fan. You will loose the figure and grain pattern by doing this. I do not pay attention to growth rings and have not had any issues with any panels I've glued up and I have glued up quite a few over the years. The secret is properly seasoned wood and careful preparation and handling of the finished product. Also you need to treat both sides of the panel with the same amount of finish.
 
#20 ·
If you want a REALLY nice ripping blade, try a Freud LM74M010.

As far as I'm concerned, its the bomb for smooth ripping. Especially like it in conjunction with tenoning jig.

May be psychogenic, but it seems to be a better blade then the red version (R010).

Glad everything worked out on your project. Did you use biscuits for alignment?
 
#21 ·
BB1,

The glue-up looks good and I like the design!

Regarding clamping a panel glue-up and the effort to keep adjacent boards flush, I thought I would mention the clamping cauls I finally found time to make. The cauls I made were flat pieces of scrap of various lengths. I made the cauls in pairs. One caul is placed under the glue-up and the second is placed on top of the glue-up. Clamping pressure is applied with C clamps to the cauls to sandwich the glue-up tight between the cauls. Once the cauls are clamped down, the edges of the panel are then clamped.

I put a little effort into making the cauls. They are as flat as possible. The cauls were sanded smooth (I sanded at 150 grit only). The edges were slightly eased and then a couple of coats of polyurethane were applied. Once the finish was cured a coat of paste wax was applied. The cauls work fairly well and lift right off the surface, even if the glue has cured. I made several pairs in various lengths to accommodate panel glue-ups for panels that vary in width and length. I find that I use them for every panel glue-up, so for me it was worth the effort.
 
#22 · (Edited by Moderator)
I had thought about cauls. Another one of those items that would be good to have ready. As it was, I used clamps I had (and as my pic showed, a lot of "variety" and applications that were not really optimal). I expect cauls would have been easier to arrange . What type of material and how thick.of wood did you use?
BB1,

The glue-up looks good and I like the design!

Regarding clamping a panel glue-up and the effort to keep adjacent boards flush, I thought I would mention the clamping cauls I finally found time to make. The cauls I made were flat pieces of scrap of various lengths. I made the cauls in pairs. One caul is placed under the glue-up and the second is placed on top of the glue-up. Clamping pressure is applied with C clamps to the cauls to sandwich the glue-up tight between the cauls. Once the cauls are clamped down, the edges of the panel are then clamped.

I put a little effort into making the cauls. They are as flat as possible. The cauls were sanded smooth (I sanded at 150 grit only). The edges were slightly eased and then a couple of coats of polyurethane were applied. Once the finish was cured a coat of paste wax was applied. The cauls work fairly well and lift right off the surface, even if the glue has cured. I made several pairs in various lengths to accommodate panel glue-ups for panels that vary in width and length. I find that I use them for every panel glue-up, so for me it was worth the effort.

- JBrow
 
#23 ·
Thanks for the blade suggestions. I have a number of Freud (all thin kerf except one) and I have found them to be good. As a new woodworker, I do not have much experience with other brands or blades for comparison so appreciate comments from others.

Freud also makes a full kerf glue line rip blade for about the same price as the thin kerf. As for flipping growth rings, I m not a fan. You will loose the figure and grain pattern by doing this. I do not pay attention to growth rings and have not had any issues with any panels I ve glued up and I have glued up quite a few over the years. The secret is properly seasoned wood and careful preparation and handling of the finished product. Also you need to treat both sides of the panel with the same amount of finish.

- ric53
If you want a REALLY nice ripping blade, try a Freud LM74M010.

As far as I m concerned, its the bomb for smooth ripping. Especially like it in conjunction with tenoning jig.

May be psychogenic, but it seems to be a better blade then the red version (R010).

Glad everything worked out on your project. Did you use biscuits for alignment?

- rwe2156
 
#25 ·
BB1,

I made the cauls from red oak because that is what I had. But hard maple might be better. The shorter cauls are Âľ" thick and run a 1-1/2" to 2" wide. I also made some 48" long cauls that are also red oak and are 1-1/2" x 1-1/2". The cauls can be used on edge but work fine for me when laid on their faces (for the wider clamping surface).