Router - feed on wrong side?

[Steve Leverich]

I'm trying to figure out the best way to use Freud molding bits, specifically these

http://www.freudtools.com/p-185-casing-bits.aspx

I have two of these, and have yet to figure out how to follow the last comment on the page ("Note: For best results make two or more passes.")

It seems to me that as soon as you make a pass, there's nothing to fully support the work piece when you make the next pass.

When doing this with a straight bit, you can just add a shim to the outfeed table ("jointer mode") - but with a complex molding profile, how would you go about this?

Hence the title - it seems that if you were to feed the material between fence and bit (Obviously pushing the material the opposite direction and taking gentle bites) you would at least have the advantage of ALWAYS having the full width of the material against the fence, thereby having a better chance of getting the profile the bit is intended for.

I know this isn't considered a safe practice, and I understand why - but I've yet to wrap my head around a method that will be both safe AND practical.

Anybody doing this, and if so HOW??!?

Thanks for any enlightenment... Steve

[Jon Crowley]

Looks like you have to move the whole fence for each pass...

[Steve Leverich]

Yeah, but as soon as you make the first pass you no longer have a flat surface, so the second pass wouldn't have any support against the fence for the side of the work that's been cut the deepest... :confused:

[David Schnegg]

Sounds iffy... but after you make the first light pass... can you use the flat portion of the profile towards the end of the bit to rest against your fence? (looks like it's about 1/2" wide).

Other than that, I don't see how else you can run it through multiple times.

[Bill Hylton]

Why would you think "jointer mode" wouldn't work with a profile bit? When you set up the router fence for jointing, you shim out the outfeed half of the fence. And you position the fence so the face of the outfeed half is tangent to the bit's cutting edge.

With a profile bit, you set the fence so the outfeed face is tangent to some feature of the profile. On the final pass, you want the fence tangent to the smallest diameter of the cutter. On the first pass (or two), you can set the fence pretty much by eye.

What I do is use something like a 6" rule to align the fence. Unplug the router first. Lock down one end of the fence, swing the fence into what appears to be the correct alignment, and lock down the free end. Hold the rule against the fence so it extends across the smallest diameter of the bit. Turn the bit by hand. You want the cutting edge to just graze the rule without bumping it away from the fence or dragging it.

An alternative is to set the fence by eye, then rout about 6" of a 12" to 16" piece of stock. Hold the stock against the infeed side of the fence as you cut the power and let the router wind down. If there's a gap between the profile and the outfeed side of the fence, the fence is exposing too much of the bit. You've got to shift the fence to close the gap.

The trick with the sort of bit you linked to is keeping the stock from tipping on the final pass--that is, having the bottom edge kick in toward the fence, ruining the piece. If you are really making casing, the edge riding on the router-table-top is going to be only 3/8" to 1/2" wide on the outfeed side, with just a narrow band of the profile bearing against the fence 2" above the table. And the piece might well be 7 feet long. You can use featherboards, but they'll have to be blocked up 1 1/2" or so above the table. Attaching a low shim to the face of the outfeed side of the fence would also help.

You might also be prepared to sacrifice a couple of inches at the end of the stock to snipe.

Just don't do anything foolish, like trapping the stock between the fence and the cutter.

Bill

[Lee Schierer]

You can make this profile in two passes. On the first pass you will take away the bulk of the material on the top, leaving the boittom half of hte piece of wood still flat. Move the fence in a bit so it is tangent tothe large diameter and make hte second cut. The base of the moulding should still be in contact with the fence on the second cut. A low feather board on the infeed side should hold the piece tight to the fence and get you the full profile.

[Steve Leverich]

Lee, you've got things upside down - the bit is widest at the base, near the table top so it'll remove the most there, leaving the piece "top heavy" - so the feather board would need to be at the top of the workpiece. But I get your meaning, thanks.

Bill, your last full paragraph sums up my concerns about stability - I guess I'll just need to get more creative with outfeed shims. I'd hoped I was missing something but guess not.

Thanks for your safety concerns, they're mine too. That's why I posted the question in the first place. I play several musical instruments, and don't need to add to the challenge by subtracting digits :eek:.

Time to go play "mad scientist" again... Steve

[Chris Padilla]

Here is what I think I might try:

Use your TS and rip a much off the board as you can so that you can cut the profile in one pass. Something like in the pic.

Leave it wider than you need so you have a reference to the fence and set up some good feather boards. Now rerip it to remove your excess.

Hmmm, seems like it would have been better to put the profile the other direction.

Are you having bad cuts doing this in one pass? Call Freud and see what they tell you.

[Peter Quinn]

Taking out that much material on a router in one pass is dangerous and foolish. In fact making moldings that big or bigger on a router is dubious to begin with. I'm seeing three options:

Use a straight fence, finish pass should be set parallel to the smallest cutting circle. Figure out the difference in thickness between the thickest flat (at the top) and the thinnest flat (at the bottom) and prepare a narrow shim for the out feed fence which supports the bottom of the stock so it doesn't toe in to the cutter. You will need to make a few test pieces, one for each pass you intent to take as you will need an out feed shim sized for each pass. I'd push 4" to 6" of the test piece in, shut of the router and use this partial pass to figure your shim thickness. Take light passes, use feather boards and make your stock at least 6"-8" longer than your finished molding.

Second way is to make a tapered out feed fence whose bevel matches the molding and supports it at at least two points, one high and one low. You really need a split fence to do this effectively. Not sure any standard router fence gives you that kind of set up flexibility?

Third way is to cut your stock 1/8"-3/16" wider than your finished width, run your molding in several passes, then rip off the ear you have created to support the stock on a table saw, clean up with one very light pass on the jointer if neccessary.

The real question in my mind is not how to do this but why? I've seen those Freud bits in my local suppliers display case and they put me somewhere between a chuckle and a shiver. Moldings of that size were meant to be made on a molder with the back down on a table, powerfed with the cutter head above. I might consider making this on the shaper in a limited run in a special species or for a color match that's hard to source otherwise, but hand fed on a router table? Just because Freud made these bits doesn't mean they are not foolish.

Taking out that much material is going to cause the wood to move a bit and feather boards are going to have a heck of a time maintaining proper allignment over the several passes neccessary. I predict you will be sanding till the cows come home and any deviation in width will cause havek come mitering time. And when its all over you will have a pile of saw dust to clean up and 2 1/4" X 11/16" colonial casing that is available at nearly every millwork supplier in the USA for less than it will cost you to make it.

I love wood working and I enjoy making moldings but I have yet to see the sense of those Freud (and other manufacturers) big molding bits for the router. I've made plenty of cool moldings with a router using small bits and creativity, but I doubt I could stand to make as much as one door's worth of casing using those monsters.

[Steve Leverich]

Peter, thanks for your candidness and ideas - I'd pretty much came to the same conclusions on difficulty/sanity of this, but I REALLY like your third method - no shims needed (other than for intermediate passes, which I've developed a neat method for) and pretty safe.

I'm getting ready to do a complete new kitchen and entire house remodel and have fallen into a deal on 200 beech doors and lots of face frame material, so in the ever-spiraling spirit of upgrading the "cracker box palace" into a livable home, wifey asked if we could continue the motif thru the rest of the house with beech moldings and window trim.

I have about 1700 lf of beech, most of which is 3/4 x 5-1/2 and about a fourth of which is 3/4 x 3. Don't need anywhere near that much just for cabinet face frames (even tho there will be a LOT of cabinets) soooo....

Anyway, once again the Creek flows knowledge like water - thanks everyone... Steve

Chris, I'm not having bad cuts yet - in fact, NO cuts yet. I tend to approach most of my more insane ideas with caution...

[Robin Cruz]

one of the scarier accidents I had was cutting a donut shape with a router. I made a jig to rotate a hex shape I glued up through a table router bit to make a perfect circle. the outside went great. Then I moved the jig to do the inside and I didnt realize now I was rotating with the bit vs as on the outside it was the opposite direction. As soon at the bit engaged, the donut jerked from my hands and the donut started to spin rapidly in the jig.

[Steve Clardy]

Hmmm.
Just thinking here. I agree with Peter. This needs to be run with a molder.
Thats how I would do it.


But if I had to do it on a router table, I'm thinking cut the excess off like Bro Chris suggested, then set the fence on the table back away from the bit the distance that the molding thickness needs to be.

Now you would have to feed from left to right, instead of the normal right to left.
Reason being, is that the material is between the cutter and fence.

You also would need a straight fence, not a 2-piece fence.

[Steve Leverich]

Yeah, climb cuts are rarely fun :(

Well, not only did Peter's third suggestion work like a charm (my stock is 3" exactly, so plenty of room to leave "rails") but it wasn't scary at all.

I have 3 Jessem fences, two of which are on Jessem table tops with lifts and Dewalt 618's, the third is on a HF cast iron table with a Freud 3000VCE (what can I say - I'm a toolaholic...) -

I set the bit up with the fence I'd added a built out insert on (taller and deeper, since the Jessem fence has a bridge across the top that won't take this tall a bit) - I used a straight edge (probably got that tidbit from one of Bill's books, have 3) to set max depth of cut as Bill also mentioned earlier, then set my two secondary fence stops and locked them down so that's as deep as the fence would go - then loosened the primary fence stops and backed off about a half inch for first pass, then made 4 passes total to the final cut (against the secondary stops for repeatability) and it came out perfect.

Used a featherboard, reset for each pass, and push sticks - one against the top pushing against the fence to augment the featherboard, and one at the rear pushing down/forward.

I can see from this that if I'm careful to set bit height, I could use the same TS setting to rip both sides off after routing. With as much of this stuff as it appears I'll be making, mass production techniques are important to me - that's also why the 3 tables/6 routers, 2 TS's, RAS, 2 jointers, SCMS, DP, etc, and soon to be a second planer - if I work it right, each phase of the project will require only one or two setups of most tools. I'm 63, so this ambitious a project needs minimum wasted time on setups IMHO.

Again, thanks for all the suggestions - I spend more time "down at the creek" these days than when I was a kid, and lovin' it just as much... Steve

Just realized that my comment about using the same TS setting for both sides would mean having the waste on the wrong side of the blade - scratch that... Steve

[Steve Clardy]

Sounds like you got it worked out A-ok. :D

My suggestion doesn't involve climb cutting. ;):D

[Peter Quinn]

Peter, thanks for your candidness and ideas - I'd pretty much came to the same conclusions on difficulty/sanity of this, but I REALLY like your third method - no shims needed (other than for intermediate passes, which I've developed a neat method for) and pretty safe.

I'm getting ready to do a complete new kitchen and entire house remodel and have fallen into a deal on 200 beech doors and lots of face frame material, so in the ever-spiraling spirit of upgrading the "cracker box palace" into a livable home, wifey asked if we could continue the motif thru the rest of the house with beech moldings and window trim.

I have about 1700 lf of beech, most of which is 3/4 x 5-1/2 and about a fourth of which is 3/4 x 3. Don't need anywhere near that much just for cabinet face frames (even tho there will be a LOT of cabinets) soooo....

Anyway, once again the Creek flows knowledge like water - thanks everyone... Steve

Chris, I'm not having bad cuts yet - in fact, NO cuts yet. I tend to approach most of my more insane ideas with caution...

I guess you just met one of the few conditions under which I would stretch to get it done...special wood that has to match other millwork. As a possible fourth option are there any guys in your area with a small molder operation or a Williams and Hussey? Might make sense to prepare the stock your self and find some one if possible to run the footage on a molder. That's a pretty common profile so you probably wouldn't incur a knife charge or have much trouble sourcing knives. Even a guy with a shaper can speed things up for you a bit and keep things clean. It can be challenging at least to run enough 90" pieces (80"+2 1/4"+snipe) for a whole house, but its a comparitive breeze on a small molder (W&H, Shop Fox, Logosol, Woodmaster, etc.).

Good luck and work safe. Post pics when your done!

[Steve Leverich]

Pics will happen as things progress :)

Steve, the climb cut comment was for Robin, I musta been asleep while typing :rolleyes: Also, I'd considered the "backward on the wrong side of the fence, but I quit intentionally putting the old bod in harm's way when I buried the last of our horses about 8 years ago, hence the first question ;)

My next "tweak" for this setup is coming up with a featherboard mount that fastens to the FENCE so I don't have to reset the featherboard for each deeper pass - anybody used "board buddies", and didja like 'em??!?

(Quietly gloating over the super-nice looking 15" piece of molding in my hand...:D) Steve

[David DeCristoforo]

Maybe too late but if I absolutely had to run this on a router table here's how I would set it up:

88557

Not sure if this is what Peter was talking about. Need more coffee.....

[Steve Leverich]

David, that's pretty much what I did except per Peter's suggestion I left a bit of full-depth wood ABOVE the bit as well - that way I always had original material face to bear on the fence. If my material were slightly narrower, your diagram would also work as well.

As I mentioned, this went smooth as silk - not scary at all, and the finished test piece could almost be stained without sanding, it's that smooth. I'd venture a guess that I'll be able to mill all the molding for the entire house in a day or less using this method... Steve

[Cary Swoveland]

This has been a very instructive thread. Leaving material above and/or below the cut, then ripping the waste after, would seem to be a useful technique for a range of router applications.

Cary

[Steve Leverich]

I agree, Cary - this molding bit that's too tall for some stock fences now isn't anywhere near as scary as a panel raising bit - and the results, as I mentioned, are as close to perfect as I could have asked. I think this'll cause me to look at problems a bit differently from now on... Steve

[David DeCristoforo]

Just for future reference, it is also not a bad idea to make your stock wide enough to mill two pieces of moulding. You can run one edge, flip the piece "end for end", run the other edge and then separate the two. You can even flip the stock over and mill the second profile on the "reverse" side if you have issues with grain direction. Sometimes, this can eliminate some steps and lessen the amount of waste. Depends on your finished moulding size and the widths of your rough stock.....

[Peter Quinn]

Maybe too late but if I absolutely had to run this on a router table here's how I would set it up:

88557

Not sure if this is what Peter was talking about. Need more coffee.....

That picture is actually exactly what I was suggesting in my distracted english. I envisioned milling the stock to match the thickness of the thickest part of the finished molding, setting the fence for the final pass to the depth of that part of the bit, and setting the bit height to make that radius on the top (outside edge of finished work) and leaving only an ear at the bottom to reference the fence and be ripped later. This reduces the table saw cut to one.

It seems leaving stock at both edges wouldn't hurt, just takes an extra TS setup. Glad to hear you have achieved success. Sounds like a beautiful upgrade for your home.

[Steve Leverich]

Peter, to my mind leaving a strip on both sides of the pattern is the only way I can see to ensure a constant reference against the fence without the possibility of anything shifting as I deepen each successive pass - and since the material I'm using has enough width to do that, and since my first attempt using 4 passes got me near perfection, about the only improvement I'm after now is more consistent board control - sooo...

I'm now working on a feather board setup that will reference ENTIRELY off the fence, so that each successive pass will NOT require resetting the horizontal featherboard like it does when using the miter slot. If anyone's interested, I'll post pix of the setup when it's complete... Steve

[Lee Schierer]

Lee, you've got things upside down - the bit is widest at the base, near the table top so it'll remove the most there, leaving the piece "top heavy" - so the feather board would need to be at the top of the workpiece. But I get your meaning, thanks.

Yep, you are right. You can still do it in two passes. Make a sample cut on a piece of scrap only a short distance in from the end and then stop. Measure the height difference in the step created by the set up ( A pair of digital calipers will make this easy and accurate. Harbor Freight sells them for under $20). Cut a shim on your TS to exactly that thickness and stick it with double sided tape to the outfeed side of the fence so the moulding rides against it on the exit side of the cut. Then test your shim by finishing the cut on thepiece of scrap. Repeat this procedure for the second pass.

[Scott Loven]

Here is a video (http://www.freudtools.com/videos/WCM-Video.wmv) on how they do it on another bit.
Scott

[Steve Leverich]

Lee, Not a bad way to do it if the material was only the exact width needed - but mine is 3" wide and the molding ends up at 2-1/4", so I can either leave solid wood on BOTH sides (as I did in the trial cut, which came out perfect) and make two TS trims, or maybe do as Peter and David suggested and leave ALL the scrap on the fat end of the bit, resting the opposite side of the molding on the widest part of the profile (narrowest part of the bit) - either way, no shims to measure and cut and only ONE fence setup that needs to be precise (the final pass)

I'll probably try Peter and David's suggestion before I'm done, but it's hard to argue with perfection :D so I'm now designing a different jig to hold the horizontal featherboard that attaches to the fence instead of the miter slot - that way there's no need to adjust featherboard for each successive pass because the distance between featherboard and fence stays constant no matter how deep a cut is made.

I've already made fence stop blocks that are part of the Jessem fence for each table, these allow precision repeatability for final pass depth so it's a piece of cake to do multi-pass work on large profiles like this - it's as much fun playing "mad scientist" with this stuff as it is seeing the looks on LOML's face at the results (can you say, "brownie points" ?)

Later... Steve

Scott, just saw your vid post - downloading it now, thanks.

[Chris Padilla]

Steve,

This seems to be A LOT of work to run the amount of molding you need! I'm happy you found something that works but how well will it work once 1000 lf of birch is run through the bit? More to the point: How will YOU feel!! ;)

Keep the bit nice and clean and good luck to you!

[Steve Leverich]

Chris, I'll feel tired - already know that :rolleyes: nothing new there.

I'm in "hold" mode right now, so this is just play time finding the most efficient way to get all the molding I need from the material I have - which is beech, not birch - otherwise, I'd be looking for a source. But after seeing the price per foot of lesser wood moldings, I'm willing to spend a couple days doing this.

Keep in mind that the "deal" I found was 200 various sized raised panel doors and approx. 17-1800 lf of 3/4 beech, about 3/4 of which is 5" wide and the rest is 3" wide by 3/4 thick - throw in ANOTHER 100 oak doors, 500 sq ft of 2x6 cedar decking, 7 18" white painted closet doors (still wrapped) and a couple of bathroom fixtures, all for $1400 and 3 trips with my 1-ton van from about 15 miles away, and I feel I can afford a bit of time to work on methods - even if I have to get a second molding bit to finish the job, it's cheap IMO.

Plus, unless you're doing it for a living, when is it NOT fun to make lotsa dust??!? :D Steve

[Chris Padilla]

Here is a video (http://www.freudtools.com/videos/WCM-Video.wmv) on how they do it on another bit.
Scott

Well there you go! They sure make it look easy, which if they designed the bit, hopefully they do make it look easy! :)

[Steve Leverich]

Just watched the vid - cool stuff, although I already have a better way of getting repeatability than a pencil mark - and if I'm not mistaken, doing crown that way is still a bit simpler than using a bit that's as unbalanced (in profile, that is) as the one I have - but that's one of the many cool things 'bout the creek - so many helpful/knowledgable people willing to help out. And all this with almost NO "'tude" - ya can't beat that with a stick... Steve

[Lee Schierer]

Lee, Not a bad way to do it if the material was only the exact width needed - but mine is 3" wide and the molding ends up at 2-1/4", so I can either leave solid wood on BOTH sides (as I did in the trial cut, which came out perfect) and make two TS trims,

I guess I missed the width part too, but I would rip it to the needed finish width first, joint the cut edge and then cut the profile in two cuts. That way you won't have saw marks on the exposed edges of the finished moulding. If you set p the first pass right you would still have 1-1/8 of flat against the fence on teh outfeed side at the top of the cut and the start of the second pass.

[Steve Leverich]

I appreciate the lack of saw marks and don't intend to have any - but I'd thought that, since this bit tends to get a little "hoggy" when in full contact, that leaving the "rails" top and bottom will ensure a positive positioning without having to worry about WHERE to put the side pressure -

So I'm thinking I'll keep the center cut by

1. Cut the profile, centered on the piece (all 1000 feet of it :eek:)- This is gonna take 3 passes - third one will be thin cleanup pass.
2. Rout the back side (see the bit profile) on one of the other two tables so setup doesn't change - with a 3/4" bit I can set up so I flip each board end for end and get a wide enough shallow groove.
3. Trim the thicker side on the TS
4. Joint that cut, making sure the piece is held against the fence AND table
5. Trim the other side on the OTHER TS (I have a PM66 and a 1973 Craftsman 12" contractor saw) - that way no setup changes between trims.
6. Joint the thin side, holding against fence and table.

7. Have a Henry's and gloat a bit :D

8. Empty the DC (several times) Steve

Oh, 9. Learn to cuss better so I can miter all that molding :confused:

[Peter Quinn]

Steve, the thickest face of your molding, what would be the part of the face nearest the out side edge of your particular casing, can be your second reference edge if your final pass is made with the fence set flush with the cutting circle of that part of your bit. You would have to prepare the face of your stock as finished work (slow feed rate on planer or drum sander) which may add more work than its worth depending on your set up and might get marred anyway with successive passes.

Your adaptation of this method allows you to produce a full jointing cut with a parallel fence which will probably produce a nicer molding face on the router table without any tricky set ups. Its actually quite ingenious. Bravo!

I hear a baby power feed (1/8HP) is highly effective at this sort of work on the router table if that is in your budget. My main weapon is a shaper now but the Router table still gets plenty of use.

[Steve Leverich]

" I hear a baby power feed (1/8HP) is highly effective at this sort of work"

Why not, it's only money - not like I haven't already learned how to spend it :rolleyes: and actually I HAVE been sporadically drooling over power feeds - just to give you an idea of the level of madness in my "mad scientist" comments, I've briefly considered a way to back off the cut on my planer and use THAT to feed stock thru the router table - then I realized that it's cheaper to just buy a power feed than to wear out a planer that way :o

"Your adaptation of this method allows you to produce a full jointing cut with a parallel fence which will probably produce a nicer molding face on the router table without any tricky set ups. Its actually quite ingenious. Bravo!"

Thank you Peter, but most of the credit for that belongs to you and David - I merely made a slight modification and it worked REALLY well.

Another part of the success is due to my double lock fence method - instead of messing with different thickness of shims in FRONT of the fence for successively deeper cuts, I made some 1-1/2" square blocks out of 1/2" Micarta, drilled 9/32 holes in the centers, and used some 1/4-20 jig knobs and nuts to be able to lock them behind the fence (riding in the same track as the big Jessem hold-down knobs)

Now, when I set up I use the straight edge across the smallest diameter of the bit (as Bill and others have suggested) - I then lock the fence down, and then I push the micarta blocks up against the back side of the fence and lock THOSE down. Those blocks then NEVER MOVE til I'm done with that setup - they become my "zero reference" instead of a pencil mark or piece of scrap clamped somewhere.

I then loosen the main fence hold-downs and keep one end of the fence against the micarta, bringing the other side forward maybe 1/2" or so - make a pass, then set the gap between block and rear of fence to 1/4" (gives me 1/8" cut) and proceed til there's only a small gap between micarta block and fence - my next to last pass is done with 4-5 thicknesses of paper between the micarta blocks and the fence on both sides - then I pull the paper and push the fence right against the micarta blocks for the final pass. Ends up looking like 320 grit finish, or close.

Doing this keeps every piece finished at exactly the same depth of cut without having to think about it much. Mainly it keeps me from having to pay attention to yet another detail (shims in front)

I like this method enough that I made 3 sets of these blocks, one for each table. The Jessem method of moving scales with lock screws isn't bad, but mine works better for me.

Speaking of shapers, the guy in the cabinet shop where I got my PM66 sold his shaper but couldn't bring himself to sell the cutters - I've been tempted to try and talk him out of them but if I were to get started down that road I'd probably never get my shop to a usable state much less remodel my entire home :( so at this point I'm resisting. Just don't ask me tomorrow :confused:

Well, I'm up at 4am for work tomorrow so better think about some shuteye - thanks again for all the help, I really appreciate it... Steve