In which I finally cut acceptable crown molding

A Ph.D. in physics is necessary but not sufficient to cut crown molding.  But I think I’ve got it.

 

The problem is not in computing the angles although that is important – the problem is in visualizing which way the cut has to go so that it will be right when you get it up on the wall.  To that end, I made a model inside corner and a model of an outside corner.  Now, I hold these pieces up to the molding on the saw to make sure I have the saw blade and the molding itself oriented the correct way.

 

Then there’s the matter of the angles.  I got tired of never understanding why the power miter saws of today have detents where they do so I set out to re-derive the math for how those angles determined. I failed.  But I did find the equations that somebody else had derived which was good enough for me.  I made them up into a spreadsheet so that I could enter any wall angle/spring angles combination and have it spit out the proper saw angles.  The most important thing I concluded from this is that a slight change in the wall angle can translate to a large change in the saw angle.  But how large is large?

 

I have always known in an academic sort of way that no two walls are ever square but I decided to actually measure mine.  I found that as I traverse the walls in my hallway, the angles vary from about 88 degrees to 90.5.  This doesn’t sound like much but when you hold out the bevel gauge and look at it, you can clearly see that it isn’t square.  The human eye is apparently very good at spotting tiny changes from the ideal.  We may not be much good at measuring absolute values of measurements but we are very good at spotting changes.  So…

 

These small changes in wall angle translate to some rather small changes at the saw; mostly because it’s possible to vary the spring angle of the molding pretty widely as you hold it up there and still have it look OK.  “Rolling” the molding in this way allows you to make the miters meet up.  This would be easy to spot on a wall that was perfectly flat all the way across but mine are not and there are noticeable hills and valleys along the length of the molding so I have the freedom to roll it one way or another and know that any gap introduced will pale in comparison to the gap due to wavy wallboard.  Caulk is inevitable in my house.

 

It’s difficult on my saw (which actually belongs to my brother-in-law since I sold mind) to make such small changes accurately so I just left the saw set at the detents.  (The bevel angle is especially difficult to set.)  When I get the molding up on the wall, there is usually a pretty severe gap at the top or bottom but I find that if I have the length accurately measured (as in within 1/16”) then the miters will go together if I “roll” (or twist) the pieces. 

 

So, the important lessons here is that I first of all, made models of each type of corner: all together, four pieces labeled “inside left”, etc to help me visualize what I need.  Second,  I have to measure the length very accurately.  Plus, I sometimes need to take a chisel and shave off a bump or overly round inside corner on the wallboard texturing to make things work out.  Finally, even though the exact angle is very important, after taking into account the waviness of my walls and ceilings, I can get away with just accepting the default saw detents. 

 

All I have to do is measure accurately and tack the molding up with a single nail in the center.  Then roll it at the ends until the miters match and nail it in.  After all that analysis, I come back to the age-old practice of measuring twice, cutting once, and using a little trick at the end.  Plus caulk.