Rail and Stile help

  • Advertise with us

« back to Woodworking Skill Share forum

Forum topic by Overflowin posted 12-29-2016 06:12 AM 908 views 0 times favorited 9 replies Add to Favorites Watch
View Overflowin's profile


9 posts in 1237 days

12-29-2016 06:12 AM

I am starting my first raised panel door project, and i have a couple questions.

first, as you can see in the picture below, the joint appears to be pretty sloppy. i’m hoping someone can tell me if this is technique or possible some adjustment needed in the router bits.

i am using a benchtop router table with a porter cable 1.5 hp router, MLCS rail and stile bits. i know the MLCS bits have a less than perfect reputation in some corners of the web, but i didn’t want to drop serious coin on a set for my first go. i am running the stiles thru fairly slow, cutting in one pass with the fence set in line with the bearing. i have a featherboard holding the piece in the fence, but my current setup doesnt allow for a featherboard to hold the piece down to the table.

i am using an MLCS coping sled for the rails, roughly the same feed rate as the stiles, again cutting in one pass. i am not getting any burn on the long stiles, but the end grain copes on the rails i am getting some burn.

the copes on the rails fit pretty much perfectly with the setup block that came with the router bit set, but the long sides are sloppy when held to the setup block.

my theory is that the stiles are lifting off the table a little since i dont have a featherboard to hold them down tight, but this is my first rodeo so i thought i would ask here.

my other question is regarding the burn on the end grain cuts. do i need to slow down or speed up the feed rate, or do i need to make the cut in a few passes? i tried going super slow, and it seemed to burn more. i didn’t want to try to force it thru faster because i feel i should let the machine do the work. My current fence is a bear to adjust, and the sled doesnt seem like it would react well if i tried to set the wood farther back to make more passes, but i could be wrong on that.

thanks in advance,

9 replies so far

View mahdee's profile


3993 posts in 1915 days

#1 posted 12-29-2016 12:17 PM

Hi Donnie, I think your theory on the stiles are lifting off the table or they are bowed a little or the setup is not exactly at 90 degree is correct. With the end grain, I wouldn’t go any deeper than 3/16 at a time. Use a stick against the router bit as guide to achieve your final height.


View Rich's profile (online now)


3666 posts in 737 days

#2 posted 12-29-2016 01:11 PM

That’s not bad looking. You do need somewhere for the glue to go. I always use feather boards to hold down the stiles and also when I do the panel raising. If you can rig something up, it’s one less thing to worry about as you make your cut.

-- Half of what we read or hear about finishing is right. We just don’t know which half! — Bob Flexner

View rwe2156's profile


3093 posts in 1629 days

#3 posted 12-29-2016 02:27 PM

Yes it is too loose. You will have trouble keeping the faces flush during glue up.

This is not my favorite task and I feel your frustration.

Here’s the way I do it:

1) make absolutely sure all your stock is milled to EXACTLY the same thickness and there is no bow. It is critical to have all your stock is straight and not warped or bowed. I use winding sticks for this. If you don’t you will have trouble in the milling process and end up with twisted doors.

2) Have some extra pieces milled to the EXACT thickness as your stock for test pieces.

3) You want about 1/16” reveal on the cope so I would lower the bit.

4) You HAVE to have a feather board mounted to the fence to keep down pressure.


Once you’ve decided on the ogee reveal, run all the stiles.

Dial in the rail xcut using your extra stock.

Sometimes the bit sets just aren’t very good. I struggled with a cheaper bit set until I realized they just didn’t match up. Some can be shimmed. This my very well be what you discover.

Now I pretty much go with Whiteside bits.

-- Everything is a prototype thats why its one of a kind!!

View Rich's profile (online now)


3666 posts in 737 days

#4 posted 12-29-2016 03:01 PM

Maybe I misread your post, but you want to cope the rails first so that any tear out on the ends is removed by the stick cut. It’s pretty much standard practice for any router operation to start with the crossgrain cut for that reason.

I agree about getting the best bits. I used Amana for years, but Freud’s new system that allows cutting long tenons on the rails is pretty nice. It’s mandatory for interior doors, but makes sense for cabinet doors as well, due to the extra strength. You do have the extra step of cutting mortises in the stiles though.

Here are the kick rail, lock rail and top rail for an entry door I’m building for the entrance from the garage to the house.

-- Half of what we read or hear about finishing is right. We just don’t know which half! — Bob Flexner

View Madmark2's profile


392 posts in 736 days

#5 posted 12-29-2016 03:24 PM

Stock thickness is critical, any variance shows up 2x.
Press the stock down firmly when you route to eliminate bowing.
Run only one pass, cutting twice makes the joints sloppy.
Route the stiles first & set the top of the slot cutter to 5/8” leaving a 1/8+- thickness on the back.
Trim your rails to the opening plus 3/4 for the copes.
Flip the rails over and reset to cope bit.
Set a stile cut next to your bit and set the bottom cutter to the top edge of the 1/8” back. Set it exact as this is the most critical dimensions for a tight fit.
Use a backing board for all cope cuts to prevent blowout.
The resulting fit should be tight enough to hold together without glue.

Bonus tip: use ‘space balls’ in the grooves to float the panel.


View Rich's profile (online now)


3666 posts in 737 days

#6 posted 12-29-2016 03:36 PM

+1 on the space balls. You can get a glimpse of them on the top rail in the photo above. I use gel CA glue to hold them in place. One less thing to deal with during glue-up.

I still say to cope first. Search around and you’ll see I’m not alone on that. Each to his own, though. It’s the result that counts.

-- Half of what we read or hear about finishing is right. We just don’t know which half! — Bob Flexner

View EricTwice's profile


237 posts in 681 days

#7 posted 12-29-2016 04:02 PM

The space may be caused by vibration, but as long as you can make the face that shows look good, I wouldn’t worry too much about it. The burn on the end grain may be caused by your feeding speed being to slow, try a little faster and see what happens. Listen to your router if it starts slowing down you are going too fast.

I wouldn’t try to push a warp out with any kind of hold downs. I run everything face down, with the belly of the board down on the table. this is much easier to control. It also means that any warping will cause the corners of the door to rest nicely against the cabinet. This is much nicer than having the center touch and the corners lifted off. This also has the added effect of putting any thickness variation on the back side where it is less noticeable, and easier to fix.

If you are afraid that the joint will not be strong enough because it is too loose, add a dowel to it. (This is easier to do if you put them in before you run the profile, but it can be done.)

-- nice recovery, They should pay extra for that mistake, Eric E.

View JBrow's profile


1366 posts in 1068 days

#8 posted 12-31-2016 02:21 AM


I noted that what I assume is a stile (top piece of wood in the photo) has a shoulder that contacts the shoulder on the rail (bottom piece of wood in the photo). If a little of the shoulder (near your thumb) were removed from either piece, the joint would be a little tighter. This suggests either a pair of router bits that are poorly manufactured or a slight bevel on edge of the stile and/or the end of the rail exists.

Therefore the first potential issue is that the edge and ends of the work pieces are not exactly 90 degrees to the face that registers against the router table. This could be caused by not jointing the edges or the jointer fence not set square to the jointer bed. Likewise the saw making the crosscuts on the rails may not be exactly 90 degrees to the table. I may also be worthwhile to check that the coping sled is holding the downward facing face of the rail parallel to the table.

The second potential issue is the slow feed rate. A slow feed rate gives the router bit more time in each spot of the wood. As suggested by EricTwice, any vibration (the router not firmly anchored to the table, run-out in the router bit and/or router) that could result in a sloppy fit could be exaggerated. The vibration could account for the sloppy fit of the tongue and groove. Also a slow feed rate causes burning since the bit is in longer contact with the wood, heating the wood to the charring point. I would not be surprised if the extra heat could also change the dimensions of the router bit and contribute to a sloppy fit.

Feeding the stock through the router bit as quickly as the router will allow might eliminate some problems. A larger router (3hp or more) could also help with increasing the feed rate. Also making two passes could help, but the two pass problem that must be solved is how to set up for the second pass so that that stock follows the exact same path relative to the router bit as it did during the first pass. If it does not follow the same path, the resulting joint could be a sloppy fit.

Although possible, I doubt the work piece(s) lifting off the table produced the joint you show. But it can be checked by closely inspecting the rail and stile cuts. Both should have straight lines all along the milled shoulders and all along the ogee (no waviness). Nonetheless, it may be worthwhile to install an auxiliary fence tall enough so that a feather board can be clamped to act as a hold down. The hold down and hold against feather boards would allow greater focus on feed rate and keeping fingers out of harms way.

View Overflowin's profile


9 posts in 1237 days

#9 posted 12-31-2016 02:23 AM

the stock i have prepared for the doors is, as near as i can tell, the exact same thickness, however after looking at it, the test piece in the photo is actually a little thicker, maybe 1/16th.

the table is a cheap benchtop deal, so i suppose the thin metal plate that the router is mounted to could be vibrating. any ideas on proving or overcoming that? i have plans to build a new router table, but the wife wants the doors installed before i take that on.

i did do the copes first, as my research showed that i can take off the tear out this way. the coping sled has built in backer blocks that are supposed to eliminate tear out, but it wasnt entirely effective, probably because they are plastic blocks. maybe some wooden backer blocks instead?

while i would like to get whiteside, amana or frued sets, they are almost triple what i paid for this ‘starter’ set. i am usually a ‘buy the best and buy it once’ kinda guy, especially when it comes to tools, but with a baby i have to watch my spending. i’m hoping i can make adjustments to my technique, or fiddle with the washers and shim this set to at least get results i can live with for a painted aquarium stand.

I will try to figure out a way to get a featherboard on the vertical and give it another run over the weekend with some of the stock milled to the correct thickness and post about my progress. i appreciate all the advice and input, and i am already a believer in in the space balls. i read about them and, well, they just make sense. i like the idea of putting a little dab of CA glue on them, i will definitely be using that idea!


Have your say...

You must be signed in to reply.

DISCLAIMER: Any posts on LJ are posted by individuals acting in their own right and do not necessarily reflect the views of LJ. LJ will not be held liable for the actions of any user.

Latest Projects | Latest Blog Entries | Latest Forum Topics