A comprehensive educational community devoted to trim carpentry, finish carpentry and architectural millwork. Hosted by nationally recognized author and finish carpentry specialist Gary M. Katz.
  Why Miters open  
  The following is a conversation (blog thread) on the topic of miters opening and the possible causes.  
  GaryM2: I recently had occasion to return to a job I did about a year ago as the customer wanted to have me do some additional work. While there I looked over my previous work. I noticed that almost all of my miter joints on the door casings had pulled apart. All were stock casings from the local lumber yard. They were apart about a 1/32 to 1/16 of an inch on most, but troubling none the less.I usually try to make the joints quite tight so was surprised to see this on each of the joints.Any thoughts on why this is happening would be appreciated. Thank you.

Joe Wood: Maybe you installed them during a real humid period and the stock had picked up a lot of moisture sitting at the lumber yard???

Gary Katz: Ditto. Must have been.Gary

WmP: Since its early spring late winter time period the wood has shrunk. Go back this summer and fall and youll see your joints probabably tight again. Another possibility is that the wood probably had a high moisture content at application. Not due to the normal conditions but thats the level it came from the mill. Might have been a bit "green" but who knows...You can slow this down some by putting the same finish on the back side, doesnt have to be perfect but it just has to be there. Also seal your end grain prior to setting up your miters (light coat of finish or sizing) This would help equalize your wood from expanding and contracting, also any possible cupping or warping. While this wouldnt be noticed by many, youd still see the trim pull away from the wall some, thinking the trim was installed loose, when in fact it has slightly cupped some.As for Marvins Q about spray stain, i dont think ive ever seen a solvent expand or move wood. However an alcohol based dye or stain could in theory do this since Alcohol absorbs water and it could have been old. But just because I've never seen it doesn't make it the Word of God...

GaryM2: I have considered caulking or applying glue prior to installation, but not sure of the impact there, so I have avoided it.

Gary W.: The wood needs a chance to acclimate to its new environment before it will stop moving around. I remember when the builds here in southeast Michigan would leave the trim in the house a week or more before installation so the wood can adapt, but they never do that any more and because or their hast I see a lot more miters open up.I see more miters open up when the trim was installed during the winter months as apposed to the summer months. This has always confused me, because winter air is drier then the summer air.Glue should be applied to all miters regardless of the weather conditions. However, gluing doesn?t mean it will not move after it?s installed. I have had a few miters open even when I glued them, but I leave the caulking for the painter.Gary W.

Rick: What type of heating system is there?Forced hot air systems take out all the humidity when used in the winter.This may be the cause.A humidifier placed on the system will help control this problem.

GaryM2: The area was heated with forced hot air.

Derrell Day: I live in NW Florida and this ugly situation rears it's head quite frequently. Trim is installed at about 18-22% moisture and immediately begins to dry when the AC goes on. I was locking out a house just this week that had the AC running. They hadn't finished running the condensation lines outside, so they had a 5 gal. bucket under the drip. It would fill up the bucket every 6 hours.We can't win under those conditions. It's not a reflection of your work.

GACC Dallas: GaryM2,What kind of glue were you using on the mitre joint?Ed. Williams

Gary,Most of the casings in this county are 4 1/2 x 5/4. The vast majority are biscuited, glued and clamped just prior to installation. Most everyone uses red topped titebond or Wilsonart ( made by Lockweld).Wood moisture content runs from below 6% to around 8% at delivery.The houses are huge and nearly all are trimmed by crews that do nothing but interior trim. Enough crews check moisture content with their own meters that suppliers have been trained. Most stack lumber away from open entries, and not in garages, and the trim starts going up quickly, before it picks up moisture, if the trim is going up in the spring.The trim is also painted quickly, and usually comes pre primed.As a result of these precautions, the miters stay together on wide casings. I presume you are using narrow casing like 2 1/2" if you get it from the lumber yard. If the heel of your joints are open, then the material was to wet when installed (probably the case). If the toe (apex) of the joints are open, then the material picked up moisture after installation (unusual).It is not the stain or water based latex paint causing swelling (which would cause the toe of the joint to open). If anything, alcohol stain would remove moisture from the wood.Good luck,Clampman

GaryM2: Ed W.I have not used any glue, at least not yet.Thought I would throw out this question before I tried doing that. Toe and heel seemed to be open the same amount. I will try acclimating future casings, but is glue advisable?GaryM2

Gary Katz: All miters have to be glued. And glue is a good way of sealing the endgrain, too. Clampman is absolutely right. And in the big stuff, either biscuits or splines work great secure the joint--I prefer splines because they register everything perfectly flush, but whatever.But I don't understand that heel/toe stuff. Can you explain that in slow detail. I always suspected that either the long point of the miter would open first or the short point, as the wood dried out, but what's taking on ambient moisture from the house after installation have to do with it? I wouldn't mind pictures, too. I don't understand toe and heel. Can we use long-point/short-point?Gary

Jeff B: Toe = Long PointHeel = Short PointForced hot air can wreak havoc with wood. In my experience this extreme dry heat usually will shrink an unglued miter evenly at both points.

GACC Dallas: No glue? You're kidding right? No wonder the joints are opening up on you. You have to glue all wood joints. ALL. Except door stops or casings to jambs or base cap, stuff like that.Thing to be glued:* All mitred head casings or panel mould corners - We glue every mitre, scarf, jack mitre, stile and rail and every joint that you don't want to move. Lots of glue; wipe off the excess with a damp rag.* All returns - If you're returning anything back on itself, you have to glue it and a nail won't hurt if it's big enough.* All joints in the field - If you've got a long run of crown, you have to miter the joint (either 22 or 45 or whatever you choose; no butt joints please)and butter it up good. There is no such thing as too much glue. Squeeze out is a good thing.We use Elmers Yellow and white as well as Titebond 1 & 2, - gallons of the stuff. Seal up that end grain. Most of the time a nail is just there to hold things in place until the glue drys.Now you know why your joints are failing. You didn't glue them.Ed.

Gary W.: The short points open on my miters if their going to open at all. I have never seen where the long point opened unless it wasn't fitted properly in the first place. It probably has something to do with the climate here in Michigan. It?s not done much any more, but letting the trim sit for a period of time does make a big difference.In the 24 years I?ve been doing this work I have never used a moisture meter on any wood, I have never heard of one until I started posting here. I do high end homes sometimes and it?s not a practice used their either. However, that doesn?t mean it shouldn?t be done more. In fact, it would probably solve a lot of problems.Gary W.

SJP: Some good points been made here. I've never had a problem with trim that has been allowed to acclimate prior to installation, but sometimes a tight schedule makes that difficult.I'm wondering if poly glue might help solve this problem with the high moisture content boards. Anybody tried this for reg interior trim? (Maybe another use for the Hipurformer gun! not exactly practical for a whole house)....I've always used yellow glue and pins, and haven't had any problems with joints opening up other than on a few occasions. For stain grade I try to back and edge seal when possible. I'm a big fan of this with most trim, especially exterior.

Mike Nathan: Glue,biscuit,clamps won't stop moist wood from drying >shrinking, yes they help a little. The wood just has to be the right M.C. for the locality it's installed in. Once I had a large raised panel over a fireplace pull right off the wall after they turned the furnace on. :( Find out from your local floor guys what is the M.C. that they use and go by that. They have the most liability and knowledge about this issue.When I workrd in the East we would have wholehouse open up at the long point. The moisturecontent is so extreme between installation in the summer then the forced air furnace in the winter.

Dick Seibert: A couple questions about gluing:1) I use the Collins clamps, are they the best, or does anybody know of any better?2) I had to trim a window over a granite backsplash the other day, and I had to assemble little cap pieces on top of the granite (?" by 4" for 3?" casing to land on). I picked up some Gorilla Glue (for the first time) because I only had a couple of hours on the job and there was no time to clamp the assembled cap mold on top of the granite and install the casing. My carpenter didn't like the Gorilla Glue and just used Tight Bond Molding Glue. It set in a couple of hours and worked OK, but my question is: What is wrong with Gorilla Glue? Do you guys use it? BTW, we have a Hipurformer gun somewhere buried in the warehouse, but I couldn't see digging it out for a couple of little spots of glue, since we never seem to be able to reuse the stuff.I anticipate a lot more problems along this line, because we are going to more and more thin coat plaster over blue board, and there is quite a bit of moisture in the house when we trim out. One of the advantages of plaster is they put the color in the plaster and we are back trimming the day after the plasterers are out of there, but the humidity is something else!

Joe Carola: Dick,The GC that I frame for is also the trimmer and told nme that Gorilla glue isn't good to use if it gets on the wood that's going to be stained.

Steve: Gorilla Glue is great for lots of applications, but joining together molded trim is not one of them; it foams out of the joint and clean-up is a pain. Try a sample joint with some scrap and check it out. Also, one trick with Gorilla is to use way less glue than you normally do.

Dick Seibert: Steve:I bought it to bond the wood trim to the granite, how is it for an application like that? What is it really good for?

Steve: Dick;I suppose it would work if you use it sparingly and have a way to scrape or sand off the foamed out residue after it's cured. I don't recommend trying to remove any liquid glue with solvent as it would probably create a discolored smear on the granite. Perhaps a glue gun or construction adhesive might work better?Gorilla is best used for laminating up stock, and for stile & rail or mitered joinery with flat stock where you can thoroughly sand off the cured foam-out. Where it foams out of inside corners, it can be sliced out with a razor-sharp chisel.I use it a lot for cold-temp work; it's the only stuff besides epoxy that cures well down to about 40 deg F. It works best if you dampen one side with water and apply/spread glue to the other. The stuff's pretty slippery; often layers need to be bradded together so they don't swim around when you clamp them.

Gary Katz: You're probably lucky that you didn't use the Gorilla glue, as it would have foamed out over the granite and made a bad mess.There's a new hotglue availabe from FastCap. Derrel and I have some being shipped to us for testing/reviewing. I think it might be a great alternative to everything else. We'll let you know.Yes, Dick, that thin-wall application is going to cause you wicked problems, just like wet plaster used to cause problems, just like getting inside a house too soon after gypcrete has been poured.Gary

WmP: Also for reference sake, If your two mitered ends meet and have far off MC(one is 8 the other is 14%) your corners will look like crap because one piece shrank more then the other piece. And have pulled apart. Not all wood from the kiln is the same MC%Woodworkers have always been moisture meter fanatics. Finewoodworking did a review of moisture meters last year. Some are better then others and all have to be adjusted as per your species of wood.Its rather picky, but if youre the anal type or perfectionist type then you could save many a headache if you had a meter handy just to be sure.

Ross Welsh: Gary:Please keep us posted on the FastCap. A high strength fast set system has a lot of potential. I reviewed the Hypurformer (4-03) and like most everyone else thought it was great when it worked. The problem was that it wasn't quite ready for prime time. I have since started to pay attention to the package dating because I found a particular batch that didn't work right out of the package but other batch worked fine even when restarted. I'm sure we will all be interested to see how a new contender performs.Ross

Mike Nathan: Dick1.I like the Hartford mitre clamps.2 a..Poly glue is also inert to solvents after it dries, so it's impossible to clean up. This past winter we glued every piece of siding up with it.It cleans up well with alcohol or actone when wet.It get all over, sort of like epoxy. There is quite a learning curve for the amount to use. We found gorilla glue brand held better then the tite bond in lower temperatures. We did experiment to determime which brand to use. 45-50 degrees F is probably the lowest application temp. . it turns to a "sugar" when too cold. Regular yellow glue will hold wood better then Poly in our experience, with less mess. Poly does hold to more things, but seem to be less then perfect for most of the wood working projects.3.For fixing loose furniture its ideal. And if cleaned up well it won't seal V.G. Cedar from the exterior stain.I would never recommend it to be used on any moist wood.

GACC Dallas: Good Lord, can we complicate this WAY past where it needs to be? Sometimes the answer is right in front of us. But the lust for new technology just clouds the obvious. Now, I'm not locked in the past, but I'm not stupid enough to ignore what was learned and passed on to us either. Most of what we do has been done successfuly more times before in the past than we may ever do.If you were unfortunate enough not to have benefited from the generations of carpenters before us (1930 - 1960) then that's too bad. We are the next generation (1960 - 2000+) and had damn well better remember those before us or repeat their past mistakes.My advice is to simplify your techniques. Take the best of what has been and apply the best of today as to compliment the past, not ignore it.Maybe I'm just a dinosaur. Sorry to vent.Ed.

Derrell Day: Jeff B,Don't feel like a lone ranger.... I've always refered to the long and short point of angles as heel and toe.They come freshly cut with a long point and a short point. When an adjustment is necessary, you take a little off the toe or heel.I learned that from my father who was a fine carpenter, as was his dad.

Marvin: I second Gary's request for more info regarding the why's of what part of the mitres open up. I don't understand from what I have read so far, why then if the heal (I've always used inside/outside and back and front to describe my mitres ...but each to his own) would open up in the bathroom on a house where all the other joints did not open up.I've been using Weldbond on all the prefinished stuff. It really dries clear, which is nice.Up here, 3 hrs north of the US border, it is becoming fashionable to have floor heat. There have been a number of instances where it seems that the entire wall is drying out and shrinking from standing on the heated concrete. Thus the bottom of the door opening are spreading enought to almost allow the bottom of the door to swing through past the stop. Any expereince with that?

Jeff: Well, I might be a lone ranger as I always prefer to work by myself but... I actually use the terms long and short point. I was just trying to clarify for Gary because lots of guys I know use the toe/heel system. Of course it goes all throughout carpentry from the heel of a rafter cut to the toe and heel on a bench plane. I might even have a few toes and heels myself.

clampman: A long post follows. Marvin- mechanics as I see them. A 3 inch wide casing measures 3 inches wide at the short point of a miter and narrows in width to zero inches wide at the long point. So as you progress from the long point to the short points of miters, there is an increasing width of wood to either expand or contract. Thus, when a three inch casing loses moisture and shrinks, there are 6 inches of wood to shrink at the short point (leg and head combined) decreasing to nothing at the long point. This results in a tremendous load in tension on the glue joint at the short points decreasing to no tensile load at the long points, and the glue joint fails at the short point, creating a gap at the short point. BUT........When that same casing takes on moisture and swells, I believe there are two VERY important differences.Six inches worth of casing at the short point swells. This puts a tremendous COMPRESSIVE load on the short point glue joint, diminishing in strength as you progress toward the long point, at some point changing to a load in tension .BUT, casing porfiles are generally thinner at the short point, and get progressively THICKER toward the long point. This I think results in some fiber compression at the short point (resulting in less of a load in tension on the long point). Further, the long point (with a greater cross sectional surface area than the short point) has a stronger glue joint to begin with than the short point, and can thus handle a greater load in any direction than the short point.Further, since most carpenters have not yet switched to clamping their casings, they are not getting true glue joints any way. Only the long points are accessable for nailing miter to miter and getting the pressure needed for a glue joint. Thus, the long points are bonded much more tightly than the short points.My point to all this "way complicated" flogging a dead lizzard, Dallas?- TOO DRY IS BETTER THAN TOO WET.Why your heel opened Marvin is anyone's guess. Maybe that casing was real wet when installed. Maybe someone slammed the door so hard to move the door stop deeper that they shifted the head jamb towards the hallway and opened the miters inside the bath. Without a moisture meter, it may remain a mystery forever.Dick, aliphiatic resin (titebond,etc.) for interior casings, gorillas for other stuff. Mike likes Hartfords, I like Clam Clamps - course I'm biased on that score. Hartfords were patented in 1888, so clamping miters is not that new a concept.Regards,Clampmanhttp://miterclamp.com

GACC Dallas: Hey Clamps, I get your point somewhat. Thin to thick doesn't really make that much difference in my opinion. The moisture content of the wood your working with does. As most of the lumber we buy today is kiln dried to under 11% it's really a non-issue. However, having said that, I've seen wood cup coming right out of the moulder if it's too wet. I own a moisture meter - I think everyone in our business should if they are worried about the stability of the wood products they're using.Too dry is better than too wet? Why? The end grain of any wood sucks up glue like a sponge. Your lucky if a thin application has any real bonding ability at all after it wicks down into the end grain.Are you clamping up these casings before you nail them to the jamb? If so, this just adds extra wasted time to the project in my opinion.Securing the legs and head casing to the jamb acts as to stabilize the short point (if the jamb is installed firmly). The joints for the head piece should be paralell to the legs, right? Unless the mitre box blade is drifting during the cut, there should be no gap to begin with. A little hand pressure and a good bead of glue added with a lace nail will act as the clamp.You don't need 50lbs of pressure to insure the glue joint will work. Just enough pressure to cause some squeeze out of the glue is enough.This whole thread started because the original poster didn't glue his mitres and was wondering why they failed. Duhh (to use my neices slang).I agree that Gorilla glue has no place here, and I'm really shocked that someone else here uses contact in a wood to wood application. In my opinion, good old Elmers. It's still the right choice.Ed.

clampman: Hi Ed, I think too dry is better than too wet for the reasons outlined in the first 5 or six paragraphs in my post above. And there is another reason which is too complicated for me to elaborate on.I also disagree that pre-clamping casings wastes time. As do hundreds of trim crews in the New England area. There are descriptions of the clamping techique I, and many others in the area, use in FHB issue 104; FWW issue 121; Tools of the Trade, Spring 1997. And even right here on the JLC site, the Dec 1998 issue. The JLC one used to be free but there is now a charge for it.I'm not sure what lace nails are, but unless they are 8" long, I fail to see how they are going to put any pressure on the inside corner of a 4 1/2" casing miter, or maintain pressure applied by hand. If temporary hand pressure is all that's necessary for a good glue joint, then a whole bunch of cabinet makers have been wasting a lot of time any money on pipe and bar clamps for a great many years.In a New England home, with dry winters, and no humidifier an 11% moisture content is at least 3% too wet to install 4 1/2" casing without biscuited or splined and clamped joints. And it would be really pushing the limits with the biscuits. Most of our millwork has been installed so dry that the needle on my meter barely moves, yet I've never seen a joint open on the long points even though we put our biscuits as close to the short points as we can get them.The only long points of miters I used to see regularly open where those on exterior turned wood columns made from glued up 12 sided hollow poles, and on exterior newels.So what can I say Ed, we just disagree. And the chances are that the whole reason for it is the difference in climate.Clampman

GACC Dallas: I guess so.Right now in Dallas it's about 10:30 pm, 85 degrees with 70% humidity. My moisture meter was still reading milled trim work at about 9% in the unairconditioned penthouse we are in the process of trimming. You'd think that this winter when they turn the heat on everything would shrink. But I've never seen it happen. I see movement in doors, but not running woodwork with glue sealed end-grain.A lace nail is just a finish nail at the long point holding the two mitred casing pieces together until the glue dries. I'm sure different parts of the country call them different things.After 29 years as a carpenter, I've never seen anyone pre-assemble door casing. Believe it or not, the joints down here aren't failing. Go figure. No biscuts, no splines, just good tight joints and lots of Elmers glue.As long as the joint doesn't fail, I guess it doesn't really matter how you get there.Ed.

clampman: Ed,I did them the way you do from 1970 until about 1992 when I subbed some work for a guy from the Hartford area who pre assembled all his. After seeing the quality, speed and ease of his work, I bought 3 dozen Hartford clamps before my next big job.After using them for a couple years, I decided I could make better ones, and now instead of having just 4 or 5 thousand pounds of woodworking equipment, I also have approximately 15 tons of metalworking machinery. So now I can pay for two shops instead of only one.Never get one of anything, always get two (except women).Clampman

Gary W.: I'm with Ed, glue and a nail throught the miter. The only place I see any casings pre-assembled is on TV.When you pre-assemble your casings how do you get them tight to the jam and wall and still keep a tight miter when the jams are not lined up with the wall perfectly?If you cut them at 45's and pre-assemble them, when you have to nail it tight to a jam thats not lined up with the wall you will break open the miter unless you cut the miters acordingly. If this is your method then it seems to me fitting one miter on the spot would be faster.Gary W.

clampman: Gary, the answers to your questions can be found at this link.http://miterclamp.com/page11.html
  Supported by corporations who care about education in the construction industry.  
  Mastering FInish Carpentry DVDs