Cupping of a Solid Wood Countertop
Woodworkers troubleshoot a case of moisture-related movement in a kitchen countertop. April 14, 2010
Question
About 3 1/2 months ago I installed a wood island countertop. A week ago I ran into the architect and he said it was cupping, so I called the contractor and he confirmed this. He said 2 - 3 weeks ago when he was there, he hadn't noticed anything. It's a new weekend house - slab floor with radiant heat which is left on all the time. The owner hasn't said anything but I don't think they've been there for about 6 weeks. The contractor checks it frequently, though, as they are now building a guest house next door.
The countertop is milled from 12/4 Eastern white maple to 2 1/4 thick and is 49" x 96". It was glued up from 5 - 7" planks (I know, I know, but that is what they wanted). The growth rings were alternated the best I could and still have the best appearance on top. The finish was The Good Stuff and the bottom was sealed, although not as much build there. There was an undermount prep sink routed about 1 1/4" up from the bottom and then sealed. There is also a refrigerator drawer unit at the other end and a single drawer dishwasher next to the sink cabinet. A lot of things going on to make it cup.
My thought is to run some saw kerfs in the bottom to hopefully release some of the pressure and then route some angle iron or channel crossways in the bottom to hold it down. Why do you think it cupped now?
Forum Responses
(Cabinetmaking Forum)
From contributor E:
The two things that come to mind immediately are the dishwasher drawer and the radiant floor. If the floor was poured recently, the concrete will give off a lot of moisture as it dries out. The dishwasher drawer is also suspect, as there always seems to be a lot of steam that comes out around the edges. Some of these drawers also require extra space above the top of the drawer to let steam escape, and it could be getting under the countertop.
From the original questioner:
The floor is probably dry - it was poured last winter and had all summer to cure. The heat was probably turned on in about June, although not used much in the summer. The dishwasher could be an issue and I wondered about putting a membrane under the countertop there.
From contributor S:
Did you finish both faces of the countertop? This may be an issue also. Does the house have a stable humidity, or now that it is winter is it drying out? I have a clause in my contracts that releases me from humidity issues. I have seen a few tops cup/warp/bow because of this. The dishwasher sounds like a culprit also.
From contributor M:
The dishwasher could be a big part of the problem but I also think the change in humidity from your shop to the house could be part of it. Around here if you buy a bunch of hardwood flooring, they will tell you to let it sit in the house with the boxes open to let the wood do what it will in your humidity. Same way with that top. I also could see the sink cutout relieving stress on that end and not on the other. I did one several years ago for my wife and we now use it as a bench in the shop; it is cupped also. As far as remedy goes, I can't see the saw kerf idea working too well, but the angle iron might. Fact is, that's a whole lot of thick wood and it will always want to move to some extent.
From the original questioner:
The humidity from my shop to the house was probably real close - the house is on the beach and I'm not too far from it either. Also, both had doors open around that time because of summer. I had thought about the humidity with winter. November was real wet and December has been pretty cold and dry, so that could be why now. I have radiant heat and because it's so even, it seems the humidity is more consistent, but I could be wrong. The bottom was sealed.
The dishwasher may not be the culprit. They haven't even been there lately and that's when the cupping has occurred. Also, being a weekend place and their secondary dishwasher, it probably doesn't get used that much.
Considering half my work is typically with this contractor and all my work for the next 6 months is with him, I'll do whatever it takes to make it right.
From contributor M:
Don't blame you for taking care of your contractor - we need the bread and butter work. Sometimes on this stuff we have to live and learn. A drum or wide belt sander will definitely take the cup out, resulting in a thinner top, but who is to say it won't cup again? And the same piece of wood in the same house probably would. If you make another top, it may cup also.
From contributor A:
How is the countertop fastened to the island?
Kerfing will show on the ends.
The Good Stuff is not a great product by any means. It's mostly known by people who have had to touch up Bally butcher block countertops. It's a very thin film finish. I bet it barely slows moisture diffusion.
6" wide maple planks are not the most stable, as you are aware.
I think most of us will throw one coat of shellac on the backside of a table and tell ourselves it won't warp because we've sealed both sides. The problem is we then throw 1/16" of hard bombproof plastic on the other side of the board. Surprisingly it warps.
From contributor G:
It sounds like the humidity in the home is too low. Set a humidistat on the table for a couple days and see what it says.
From the original questioner:
The top is attached through the cabinet tops with 1 1/4" screws - mainly just so it can't be bumped out of position. At about 350lbs, it would take a big bump to move it!
I've never had a problem using The Good Stuff before, but then I've never built a top with 6" planks before. I will put more finish on the bottom before I put it back in, though.
The problem with sanding the cup out of it is that with a 37" wide belt, I'd have to rip it first, and trying to rip that on a Unisaw doesn't really excite me!
I will check the humidity, but it is what it is.
From contributor A:
Ripping that monster on the Unisaw sounds kind of exciting. Ripping it then flatting each two-footer will be more effective than running the whole table through a 53" sander. How much cup are we discussing (i.e. 1/4" deflection 49")?
I would seriously consider grinding the countertop down so that no finish exists on either side. Then spray it both sides (2 coats backside/3 coats front side with 2k urethane). The other finish you might want to consider is pourable epoxy.
From contributor S:
I really think it is the difference in humidity. I had a raised door panel split on me last week prior to finishing. No big deal, just a new door, but it reminds me of many things I have seen over the years like this. It would be great to be able to control the humidity in the shop, but as you say, it is what it is.
Have you thought of just removing the top, ripping it again, re-gluing with cross strips underneath and then refinishing? This might be the only real way to relieve the stress. Lots of work, but if the client is a steady, I would do what it takes and move on. Where are you located?
From contributor J:
Did you allow the countertop to move? When you screwed it down to the cabs, did you use elongated holes so it can expand and contract freely?
Aside of that, I also think your problem may be in the finish. When you finish a top like that, you really should try to keep the film thickness the same on all surfaces.
For what it's worth, I'm not a believer in ripping pieces of wood down into narrow strips to make tops. With thinner wood it's tricky, but with thicker wood I've used 7" and larger on tops and haven't had problems. Wood is going to move regardless of how small you rip those pieces. If the wood has been dried correctly, and the top built, finished, and installed correctly, you should be fine using wider pieces.
May be worth finding a local shop that has a widebelt with a greater capacity as opposed to ripping the top down.
From the original questioner:
I haven't seen the top to know exactly how much cup there is, but it sounds like maybe 3/8". Running it through a wide belt in one piece would take too much off. Having it in two pieces is basically what I did when I made it, so I would only have one joint to sand out with portable sanders.
There were only 4-6 1 1/4" DW screws, so although the holes were not elongated, I think a countertop that big would move the screws in the plywood stretchers.
I think the plan is, when I get it here, to first try kerfing it in several places and stop just before exiting the ends. If I can then clamp it down flat at this point, I'll route either some U channel or angle iron into the bottom. If I can't flatten it this way, I'll rip it in half and run it through the wide belt. I'll also post on the Finishing Forum and see what they suggest.
I'm on Whidbey Island in Washington State.
From contributor R:
If you do rip and sand, just remember to re-joint those cut edges. Chances are they won't be at 90 degrees after ripping and sanding.
From contributor C:
Ripping and re-jointing will also change the width of the top. That may be a consideration if this is a critical dimension, like if there is a cutout for a sink or a specified overhang.
From contributor B:
The countertop is cupping because there is a difference in humidity between the top and bottom of the counter. Balance that and you will fix your problem. I lift my wood tops so there is a 3/8" gap between the bottom of the counter and the top of the cabinet. Yes, you can see it, but it is there for a reason. Don't like the 5"-7" planks, would have said so up-front.
From contributor J:
I would suggest elongating the holes through which the counter is fastened when you re-install. Do you really want the top to rip the stretchers apart when it moves? I would also suggest using screws with large washer heads as opposed to drywall (assuming that's what DW stands for?) screws.
Do yourself a favor and find a wood movement calculator. See how much a 49" wide slab of maple could move between winter and summer humidity. If you're not planning for your wood to move, you're going to be getting a lot of phone calls from unhappy clients down the road.
From contributor N:
The thick wood probably wasn't dry enough for that application. Most likely the top of the counter dried out more than the bottom, which is protected from drying somewhat by the cabinets. If the top was strips or planks, it would still warp if the moisture content of the two sides isn't balanced. A more water vapor resistant finish on both sides might slow the transfer of moisture somewhat and keep it more stable. More screws in slotted holes will also help.
From Professor Gene Wengert, forum technical advisor:
Moisture difference, top to bottom, is the only reason wood like this would cup. So, is there any way to increase the ventilation on the bottom side so that top and bottom see the same humidity? That will help in the future for this top and for identical tops.
As mentioned, avoiding moisture changes between the shop and installation is important, but you have addressed this. A $25 humidity sensor from Radio Shack will help determine what conditions you have in your shop and in the house. It is possible that while they are gone, the humidity is quite different.
Although you could wet one side and get it to straighten, when the wetness dried it would pull out the screws, etc. Kerfing will work fairly well, but with humidity changes, face to face, you may see more cupping later or maybe even crowning unless the RH is identical on both faces.
To be safe, I suggest that you use slotted holes, as suggested previously.
From contributor I:
At around 4.00 a bf for new stock, with everything you are going to try which might or might not make a difference, I would just make a new top and move on. Do you thickness your stock all at once or bring it down in stages? Unless you seal the entire top with a good quality finish, not mineral oil or some such, you are asking for more problems.
From contributor G:
If a new top is made it is very likely the same thing will happen. As mentioned in earlier posts, humidity differences have to be addressed and allowed for to solve the problem.
From Professor Gene Wengert, forum technical advisor:
It would be so helpful if there was a finish that could seal wood, but there is not such a finish, even high quality, available for products such as this one. A topnotch finish may slow moisture changes, but not prevent them. And of course, if the finish top and bottom is different, it is likely that the moisture will also be different at times.
The problem is the humidity differences, face to face, which creates moisture differences which results in size differences (= warp). If the humidity is the same on both faces, you could even use a piece with no finish.
If there is a slightly wet core in the material, you can make a moisture difference by planing one side more than the other, but such an event would show up as rather prompt warping, not delayed, as in this case.
From the original questioner:
I finally got to see the countertop today. It has about 3/16 cup at one end and about 3/8 at the other. The worst part of the cupping is over the subzero refer drawers. I talked to the contractor and we decided to kerf the bottom and install angle iron or channel to help keep if flat. Then we'll hold it up above the cabinets with spacers. We'll probably go 3/16 - 1/4". Where there is a bookcase and a wine rack with full tops, I'll drill some holes to help with ventilation and air flow. Hopefully this will be enough to even out the relative humidity. After kerfing, I'll add enough finish on the bottom so the film thickness is the same from top to bottom.
My two concerns are the refer drawers and the DW drawer. I don't know if putting in a membrane of some sort or sheet metal would help. It will probably be a couple of weeks before I work on it, as the owners will be gone then. Thanks for all the replies.
I really doubt I'll have to build a new top. Reworking this one will take a lot less time than starting from scratch. Routing the undermount sink in probably took more time than the rework will. Material alone with waste was $980.00 120 bf total - 96 bf finished. 12/4 select and better was $6.21 BF.
From contributor S:
If you are going to take it out and do this work - kerfing, angle, refinishing - I would still be concerned about the initial glue joints. You could be adding stress on the joint, which will more than likely split in time as it dries out.
I checked the humidity levels where you are and they have a rather high fluctuation rate over the seasons.
The drawers are constantly producing heat, keeping the top drier around this area. As mentioned before, you may wish to rip at the glue joints with a narrow kerf blade and re-glue.
I agree there is no need to remake from scratch. Maybe add a piece or two. I still have my concerns about the initial glue joints being stressed by trying to force flatten the piece.
From contributor J:
I agree that if you have to force the top into submission, there's a pretty good chance of failure down the road. Having said that, if you can flatten it the way you described and build up the finish on the bottom, you may be okay with a moderate amount of restraining force.
Again, key thing to remember is that the top is going to move. If you try to restrain it completely, something's going to give.
Not sure you'll get too much trouble from the DW drawers, as most of the moisture should be directed out from the front of the unit. I have a regular DW and haven't had any countertop problems. The refrigerator could be an issue, with the constant warm air circulating against the bottom of the top. Might not be a bad idea to try and insulate it somehow.
From contributor F:
Curious about all the comments relating to 5" to 7" planks. This is done all the time in furniture making.
From contributor P:
Did I miss something? Which direction is the cupping? The inside of the curve is the dry side. Allowing the same air to reach both sides of the slab (which must be finished equally) is the cure - I know, this can be tough when you have screwed the slab to a countertop.
From contributor G:
That last post made me realize something (other than assuming the cup was up). I have never seen anything cup down. Has anyone seen a top or anything else cup down?
From contributor F:
I have. I usually speak of cups or even bows in terms of round side or hollow side. As contributor P said, the hollow side (concave side or inside curve) will be the dry side. The wetter side will be round (convex or outside curve). I am looking at the solid wood top of an antique dresser in my house and it is cupped down, if that is what you mean by convex side up.
From the original questioner:
The outside edges of the counter are lifted up. Interestingly enough, there is a dining room table across the room that is a large 1 piece slab that has crowned, even though it is open on all sides. The reason there, though, is that it is one piece and it followed the direction of the growth rings.
My thought is, after I kerf it, I will let it sit for maybe a week to see if some of the cup will come out on its own. Then I will give it some gentle persuasion to flatten it more if need be and then put stiffeners in with elongated screw holes to keep it there.
From contributor U:
Interesting and all too familiar problem! You surely have different moisture levels on the two sides of the slab. So, if time is not an issue, I'd fix like this... Take the countertop off, sand the finish off both sides. Put it in a room with constant temperature and humidity, on some sawhorses, so air can circulate around the whole slab. Leave it there for a few days or weeks or months or years (choose one) until moisture reads same on both sides. Run it through a wide belt sander, then seal it immediately on both sides evenly. Reinstall. If time is an issue, get a new top from some sucker who will guarantee it to stay flat!
From contributor T:
In layup I prefer a top to cup down, not up. Look at the growth rings, both ends. Is this top moving in relation to the growth rings? Check each board with a small (7") straight edge. If you find a problem board in the center of the cup, try cutting it out and flip it over. For this top to cup up at the outboard ends, it can not be screwed down well. Also, the air that is trapped inside that cab, along with a heater (refrigerator) is never good. Maple moves a lot, in width and cup, as RH changes. At 49" 1% RH = 3/16"? Maybe 1'' total seasonal. Take it back to the shop, try wetting the face side with damp towels. Steam with an iron if you need to. Understand the cup! Work slowly, as you can split the top. Clamp it and let it dry. Try AL angle fasten to cab and top, in 4/5 places. Slotted holes, and use a number of real screws.
From contributor Q:
A pin meter would go a long way in answering your question as to why it cupped. Either the top dried or the bottom gathered moisture, resulting in the cup. If the home had a higher RH than your shop, the bottom could gather moisture because of the finish not being equal. If the home had a lower RH, the top could dry faster than the bottom due to lower circulation on the bottom after installation. Measure the RH in both places and the EMC of the top and bottom of your work to figure the best way to move forward.
From Professor Gene Wengert, forum technical advisor:
A moisture meter with pin-type or needle-type probes can indeed be useful, but you need to know the MC before a problem occurs. In this case, it is possible to have a moisture gradient in the piece prior to installation. Or, it is possible that there is a moisture gradient that develops after installation. In this case, if the moisture gradient is induced rapidly, then when the MC gradient flattens out due to slow drying, there will still be warp. Yet the MC meter will show no moisture gradient. So, moisture meters can be helpful indeed, but care must be used when trying to figure out what the readings, taken after a problem shows up, mean in a situation like this.
From contributor O:
I may have a way to keep it from cupping again after reinstatement. With the countertop removed, install 1/4" thick 3/4" wide spacers on top of the existing cabinets, leaving numerous spaces to allow the moisture inside the cabinets to escape. On top of the 1/4" spacers place a thin piece of sheet metal the same size of the outside dimensions of the cabinets. On top of this install more 1/4" thick 3/4" wide spacers in the same places as the spacers that are attached to the cabinets. Drill and elongate the countertop mounting holes through the sheet metal. When re-installing the counter top, place a dollop of silicone sealer on top of each of the mounting holes in the sheet metal. Hopefully this will stop any moisture traveling to the underside of the countertop from inside the cabinets after the fasteners are reinstalled through the silicone.
Now to cover up this 1/2" gap between the countertop and the cabinets. Every foot or so around the perimeter of the cabinets, attach small spacer blocks (you choose the length) 3/16" to 1/4" thick to the cabinets. Mount decorative trim to these spacers 1/4" below the underside of the countertop to allow for air circulation. Hopefully the spacers with the trim attached to them, along with the countertop overhang, will offer another shadow line and the 3/16" to 1/4" space will not be noticeable. Hopefully this will allow the moisture being generated by the dishwasher and refrigerator to be harmlessly reintroduced back into the room and not affect the natural movement of the countertop.
From contributor V:
I'll be willing to bet that a sheet of plastic was placed on top to protect your finish and was left there during and after several cycles of the dishwasher and/or the cycling of the fan motor of the mini-fridge. The resulting moisture increase at the dishwasher and the fan drying the bottom side at the fridge with the relative constant humidity under the plastic will make the wood cup. Your kerf and angle solution should work as the wood moisture has balanced. A humidistat controlled fan for compensating for the moisture introduced by the dishwasher should prevent further problems.
