Has anyone tried this technique on raised panel doors?

[scott vroom]

No matter how careful I was in my door glue ups, I was unable to get the stile ends to flush up perfectly with the rails, at the tops and bottoms of the doors. I ended up having to sand even, which with hard maple is a PITA, particularly if the stile end grain is proud of the rail.

On my latest job I took a different approach that solves the problem: I cut the stiles 3/8" longer and cut the rails 3/16 wider. After glueup and sanding, and just prior to applying the outside edge profile, I simply run the doors through the table saw cutting 3/16" from each end, leaving a perfectly flush intersection of stile and rail.

Does anyone else use this method?

[Charlie Plesums]

I've been doing that for years, only I use 1/8" rather than 3/16. I publish a spreadsheet to calculate door part sizes, and by default it is set up for 1/8 inch trim all around.

[Gene Howe]

Not me, Scott. Every door I make is perfect.;)
Actually, I use an eighth, also.

[matt tennessen]

The second time I had to make inset doors I did this and now I do it regardless of inset or overlay. I, too, use about an eighth inch but mainly because 3/16th is an odd fraction for me. And now I make perfect cabinet doors everytime...:)

[Clint Olver]

I started doing this after I accidentally edge sanded a door into oblivion with 80g on my Progress edge sander. Plus, with a sled or sliding table, you can make sure your doors are all square regardless of any glue-up mishaps.

C

[Sean Rainaldi]

Nope but I will try that sometime, I finished up our kitchen and built 19 cabinets out of hard maple shaker style panels.

[Frank Drew]

I see no point in making the rails wider than the finished width, but I think it's absolutely best practice to make the stiles long, mostly to protect the mortises during the sturm und drang of glue up. I'd leave the "horns" on the doors until close to final fitting, in this case in order to protect the corners of the doors from getting banged up in the shop or on their way to the job.

[Larry Edgerton]

I take that idea a bit farther when it works out. If I have doors that are the same width I make my rails wide enough to be a top rail and a bottom rail. I usually make my bottom rails wider than my tops so I take that into consideration. I then glue up doors in as long as 12' sections, several doors at a time. When out of the clamps and sanded as one unit I cut the doors apart with either the slider or my OMGA. The benifit is that I can machine wider rails which are easier to manage and to get glued up. There are usually three doors to a glueup so you have to be fast, but you only have to check for square once.

Works for me......

Larry

[Dick Holt]

If my doors are going to be inset, I make them the same size as the opening. This gives me a little extra to cut off to make the door flush at the ends or to square them if necessary.


Dick

[Paul Symchych]

I did just that this morning. Everything a hair over sized with overhanging 'ears' left on. A run through the TS and all edges are clean.

[scott vroom]

I take that idea a bit farther when it works out. If I have doors that are the same width I make my rails wide enough to be a top rail and a bottom rail. I usually make my bottom rails wider than my tops so I take that into consideration. I then glue up doors in as long as 12' sections, several doors at a time. When out of the clamps and sanded as one unit I cut the doors apart with either the slider or my OMGA. The benifit is that I can machine wider rails which are easier to manage and to get glued up. There are usually three doors to a glueup so you have to be fast, but you only have to check for square once.

Works for me......

Larry


Larry, interesting idea and one that lends itself to production quantities. I'm presently building 27 raised panel doors and 9 drawer fronts for a refacing job; my goal is to build and finish all in 1 week. That's a challange for my shop: I've got a 10" cabinet saw, 15" spiral planer, 8" spiral jointer, 36" router table, 10" SCMS, 17" DP. I sort of "backed in" to a cabinet business and have backlogged several jobs that have me debating whether/when to add a slider, large capacity RAS, shaper, drum sander to increase efficiency and reduce cycle time.

[scott vroom]

I see no point in making the rails wider than the finished width, but I think it's absolutely best practice to make the stiles long, mostly to protect the mortises during the sturm und drang of glue up. I'd leave the "horns" on the doors until close to final fitting, in this case in order to protect the corners of the doors from getting banged up in the shop or on their way to the job.

Frank, I make the rails wider because I plan to shave 3/16" off each door end. To me, this is easier than trying to trim long stiles flush with the rails....without cutting into the rails.

[Chip Lindley]

Scott, do the math. If the exact tongue length/groove depth of your cutter set is known, no reason you cannot cut pieces to exact length and have the door fit perfectly if the panel is sized to fit exactly inside the rail/stile recess. The height of panels will remain exactly to fit inside the groove. But, I rip 1/8" off each side of the panel reveal (after it is finished) to allow for expansion. This way, the reveal remains constant all the way around, regardless, with the use of spaceballs etc.

Let's do some hypothetical math using these standards:

Rail and Stile width: 2-1/2"
Panel groove (and tongue): 3/8"
Door overlay: 1/2"

To make a make a door for an opening 30" tall x 16" wide:

Stiles (30" + 2 X 1/2" overlay = 31")
Rails (16" + 2 X 1/2" overlay = 17" total. 17" - 2 X 2-1/2" (stile widths) + 2 X 3/8" (tongue depths) = 12-3/4" )
Panel height (31" - 2 X 2-1/2" rail widths = 26" + 2 X 3/8" panel groove depth = 26-3/4" )
Panel width (17" - 2 X 2-1/2" stile widths = 12" + 2 X 3/8" panel groove depth = 12-3/4" )

See? Doors of like height, such as cabinet uppers and lowers will have standard vertical dimensions. Only the width of rails and panels will vary. I constructed a calculator in Microsoft Excel to do the calculations for me. I plug in the numbers and read the finished results of each piece needed to construct a 5-piece raised panel door. All pieces are cut to exactly those measurements and they all fit together like they should. No trimming; only final sanding!

~~Chip~~

[scott vroom]

Scott, do the math. If the exact tongue length/groove depth of your cutter set is known, no reason you cannot cut pieces to exact length and have the door fit perfectly if the panel is sized to fit exactly inside the rail/stile recess. The height of panels will remain exactly to fit inside the groove. But, I rip 1/8" off each side of the panel reveal (after it is finished) to allow for expansion. This way, the reveal remains constant all the way around, regardless, with the use of spaceballs etc.

Let's do some hypothetical math using these standards:

Rail and Stile width: 2-1/2"
Panel groove (and tongue): 3/8"
Door overlay: 1/2"

To make a make a door for an opening 30" tall x 16" wide:

Stiles (30" + 2 X 1/2" overlay = 31")
Rails (16" + 2 X 1/2" overlay = 17" total. 17" - 2 X 2-1/2" (stile widths) + 2 X 3/8" (tongue depths) = 12-3/4" )
Panel height (31" - 2 X 2-1/2" rail widths = 26" + 2 X 3/8" panel groove depth = 26-3/4" )
Panel width (17" - 2 X 2-1/2" stile widths = 12" + 2 X 3/8" panel groove depth = 12-3/4" )

See? Doors of like height, such as cabinet uppers and lowers will have standard vertical dimensions. Only the width of rails and panels will vary. I constructed a calculator in Microsoft Excel to do the calculations for me. I plug in the numbers and read the finished results of each piece needed to construct a 5-piece raised panel door. All pieces are cut to exactly those measurements and they all fit together like they should. No trimming; only final sanding!

~~Chip~~

Chip, I'm already using a spreadsheet to determine stile and rail lengths based on the givens...that's not what I'm addressing here. My problem is that no matter how careful I am at glue up I typically end up with stile ends proud of the rail or vice-versa. My solution is to add 3/16" in my spreadsheet to the calculated rail width, and 3/8" to the calculated stile length. Doing so allows me to trim exactly 3/16" off each door end, leaving each door precisely the correct height and with perfectly smooth intersections of stile and rail.

[Chip Lindley]

Chip, I'm already using a spreadsheet to determine stile and rail lengths based on the givens...that's not what I'm addressing here. My problem is that no matter how careful I am at glue up I typically end up with stile ends proud of the rail or vice-versa. My solution is to add 3/16" in my spreadsheet to the calculated rail width, and 3/8" to the calculated stile length. Doing so allows me to trim exactly 3/16" off each door end, leaving each door precisely the correct height and with perfectly smooth intersections of stile and rail.

Scott, if some stiles protrude beyond the rails, and some rails are proud of the stile ends, the variables I see are these:

- rails not ripped to consistent width.
- stiles not cut to constant length
- panels not cut to same height
- panels not trimmed perfectly square
- groove depth in rails not consistent

Any variation as above will cause ill-fitting door parts. One trick I learned is to mount the top and bottom rail on a raised panel and measure the exact height when all are fitted together. If that measurement varies from the calculated height of the door, I cut all stiles to the "real" length. I use a cutoff stop on the RAS to crosscut all parts to identical length. A stop can likewise be employed if a tablesaw crosscut sled or chop saw is used. Do not rely on pencil or even scribe marks; there will be quite a variation in length of "identical" parts.

With all parts cut exactly as I outlined above, there is absolutely no reason for everything not to fit together as it should. Top and bottom panel tongues should fill the rail grooves completely since wood has very negligible expansion with the grain. A perfectly square panel automatically helps draw the door square.

Check again for perfect squareness of crosscuts. Use the same measuring tape for the whole project. Cut all same-length parts using a cutoff stop for consistency. Rip all parts to width at the same time and fence setting if possible. Respectfully, to my mind, the extra operation of having to trim the perimeter of every door is a waste of time and good hardwood. But, to each their own...

~~Chip~~

[Neal Clayton]

i agree with chip. you don't have to build too many doors and windows to find out that pencil marks and tape measures aren't good enough. it doesn't matter as much that the dimensions are precise as it matters that they're all the same.

only way to make them the same is to cut with stops rather than measurements and marks.

for windows (which are always bridle joints or through tenons) i do leave the overall size proud by an 8th and trim, but that's mainly just for clean looking joints, it's easier to cut than to sand glue residue off.

[John Piwaron]

I do now. I just did that on my most recent project. Call it learning everything the hard way. In my case, I chose to make both doors a little big on both ends (as you did) plus a bit more on the edge the hinges were mortised into. Why do that? Well, I realized the carcase was slightly out of square. Not much, but enough that hanging doors made "perfect" and then mating them to the not square cabinet would get less than pleasing results. To me if no one else. I'd see it. Every time I looked at it. So I decided on my too large door plan in combination with a lot of critical measurements. Then, with the help of a CAD program, I worked out a plan to shape the doors to get an even reveal around each and locate the hinges so they'd open smoothly and stay that way until closed.

It all worked out wonderfully. I'll remember this self taught trick going forward.

What you did makes a lot of sense.

[Jerome Hanby]

I just did something similar on drawers to go in a lathe stand. Trimmed a bit off left and right sides until they worked perfectly with the drawer slides.

[Neil Brooks]

That's how I do breadboard ends, on cutting boards. Works great, and saves a lot of needless headaches :)