Help with Lock Miter Bit Issue

[fred everett]

This is my first here so hello all. I'm building a Sapele coffee table - 8/4 top with 6" square mitered legs. I'm looking for some suggestions on how to resolve an issue I'm having with a lock miter bit. I have Freud's large bit and the infinity lock miter master jig. The jig did an excellent job helping set the bit height and fence. When I do test cuts with boards lying flat on the router table the joints are seamless. The issue I have is when I make cuts with the board on the fence. Specifically, the tail edge of the cut has a divot 80% of the time - 1st pic below. In a effort to resolve this I built a jig that rides on the fence - 2nd, 3rd & 4th pic below. The idea is to keep the wood flat against the table, but the results are the same.....a divot on most every test cut. I would appreciate any suggestions on how to resolve this. At this point I'm ready to bail on the lock miter idea as I don't want to waste any Sapele which I have all machined ready for joinery - pic 5. Thanks in advance.

302607302608302609302610302611

[Art Mann]

As you make the profile cut, what used to be a wide blunt edge becomes a thin wedge shape on the bottom edge. As a result, there is very little support to keep the back end of the work piece from dropping down and making the bad cut you see. If you could either clamp or double sided tape a guide board to your work piece that will support the piece even with a sharply angle edge,then you can make a consistent cut. You will have to get rid of the flat guide piece evident in photos 2 and 3.

It is sometimes possible to cut the work piece a little long and start the profile an inch or so away from the end. Don't route all the way to the back edge either. This provides solid support to the work piece so that it can't drop down due to a flimsy wedge shaped edge. Afterwards, use a cross cut to get rid of the unrouted ends. Unfortunately, it looks like you don't have the spare wood to do that. Also, your wood is thick and it may not be safe to drop the work piece straight down onto the spinning bit. It isn't so bad with thinner wood used in box making.

[Dave Cav]

I wish I had a good answer for you, but I'm afraid I had the same problem. I spend the better part of a couple of days trying to make a set of table legs using a lock miter cutter on my shaper. I milled and test cut an enormous amount of poplar and every time I thought I had it right and used my white oak stock I screwed them up. I tried guide blocks, multiple featherboards and my stock feeder and I couldn't get anything to work. I finally gave up and ended up doing a conventional glue-up and laminating the two edge grain faces.

I also wasted most of a day trying to get the Infinity jig to work and eventually called them. They were very apologetic and told me that it wouldn't work on shaper cutters but their marketing people conveniently left that part out of the catalog. I was able to return it for a refund.

[Mel Fulks]

Art's explanation is good. I find it much better to not cut to sharp point ,sanding and slightly rounding the corners will
remove the slight hollow at corner. Couple of lengthy earlier threads on this subject might help you,too.

[ian maybury]

For sure inconsistent results can be a problem with these cutters. Might the fact that you are seeing intermittent issues Fred suggest that something is moving or springing in the set up as the part is being fed through? Possibly as below, or as Art describes above?

My experience (relatively recent) is on a shaper cutting a selection of mitre lock and double rebate joints. They worked out fine, but didn't require micron accuracies as on say fine box work.

One category of issue to watch out for is straightness of your pieces, or for that matter any slight dip or out of flatness in your router/shaper table or fence. Or spring/flexibility/deflection of the table/fence under pressure. Everything depends on simultaneously referencing cut heights accurately off the table and the fence at the cutter location for the full length/duration of the cut, and if a bow and/or dip intervenes (especially if the force you apply to hold the part down as is pretty much inevitable varies) then problems must follow.

An initial inconsistency caused by slight bowing in the ply strips i was using to build a tool cabinet was resolved using fairly stiff featherboards and/or a guide/hold down strip clamped to the fence for the long strips/cuts.

Vertical/fence supported cuts in the end of pieces were handled with a right angle sled clamping fixture running on the table and against the fence, with the horizontals clamped on to and cut off the shaper slider. Both worked out very well, but need care as they bring added potential for issues - the slider needs to be set coplanar with and only a shade above the level of the table (or the work is raised above the table and left unsupported in the gap between the slider and the cutter - a tenoning plate would in some cases be an option, but i had long fence guided and table supported cuts i wanted to make from the same set up as well), and the sled also needs to be dead square and flat/straight on both faces so that it also runs flat against the table and the fence and doesn't cause bridging.

Another category of issue if you are relying on blocks/solid strips as opposed to sprung guides (featherboards etc) to guide the work over the cutter is the possibility that the stock is not of absolutely consistent thickness - the thinner sections may be permitted to lift off the table/fence. This was initially a factor in my case with poor quality Eastern made so called marine ply.

As ever the devil is in the detail…

[Steve Jenkins]

Rather than messing with the lock miter I just use a regular miter and glue them up. It is a long grain joint just like edge gluing so the glue joint will be stronger than the wood. For clamping I use clear packing tape. I just wrap it around the assembly giving it a good stretch. I don't spiral the tape just wrap it about every 6"-8" or so. I like the clear packing tape because it will stretch and provides a lot of pressure. With accurate meters the joint is self aligning as you tape it.

[Bill Huber]

I am with Steve, just use a miter joint.

I made a table a while back and use the lock miter, I did all my test on some poplar and had things set to a great joint. I then started to work on the real legs of ash and things were not going to good.
It did not take me long to just give up and use regular miters, as Steve said it is long gran to long grain and that is as strong as it gets.

I used blue 1" tape and laid the parts out on the tape like I was making a small miter cornered box and then rolled them all together and had a really nice fit.

I have done some other lock miter joints but I have never had good luck with the Frued large bit, the one I like and use the most is the Infinity and I use the lock miter master jig.

[fred everett]

Many thanks to everyone for the responses - much appreciated. At this point I'm ready to consider the bit/setup jig a $130 mistake. I decided on the miter lock bit to avoid the safety issue cutting bevels on the table saw. It's back drawing board trying to figure out how to safely cut 32 perfectly accurate miters in 3/4" x 14" stock.

[Larry Edgerton]

Art's explanation is good. I find it much better to not cut to sharp point ,sanding and slightly rounding the corners will
remove the slight hollow at corner. Couple of lengthy earlier threads on this subject might help you,too.

Ditto......

Before you give up try this method. It works well for me and I use lock miter cutters often. A sixteenth flat left is plenty and an 1/8 or 3/16 roundover will take care of it.

[Dave Cav]

Once I started using my Wixey box to set the blade angle I quit worrying about the accuracy part.

For the actual cutting, cut them face down with the blade angled away from the fence. On a right tilt saw, put the fence on the left side of the blade. Put a featherboard on the rip fence to hold the stock flat to the table and it should work fine.

[Ralph Okonieski]

One approach that has worked for me was to clamp a support piece to the vertical piece so that at the end of the cut the weight of the vertical piece is held by the support piece. The support rides on the top of the fence. I also attach the horizontal piece to a trailing support piece with double stick tape; this prevents tearout as well as make a nice long support (and keeps hands safely away).

[George Bokros]

Why not a 45* bevel bit in the router table running the inside face down on the table.

[fred everett]

I am with Steve, just use a miter joint.

I made a table a while back and use the lock miter, I did all my test on some poplar and had things set to a great joint. I then started to work on the real legs of ash and things were not going to good.
It did not take me long to just give up and use regular miters, as Steve said it is long gran to long grain and that is as strong as it gets.

I used blue 1" tape and laid the parts out on the tape like I was making a small miter cornered box and then rolled them all together and had a really nice fit.

I have done some other lock miter joints but I have never had good luck with the Frued large bit, the one I like and use the most is the Infinity and I use the lock miter master jig.

Me three......a straight up miter is where I'm going. I'm going to build a smaller 45* sled and use hold downs to keep the work stable and me safe. The path that led me to the LM bit goes back to my last project......a billiard rack below. I'm happy with the way that project came out BUT I cut miters in 7/8 stock on the table saw with the stock against the fence. I didn't have a "scare" or anything like that, but as you can see it just wasn't smart from a safety perspective.

302652302653

[Art Mann]

Steve Jenkins and Bill Huber are exactly right in my opinion. Lock miter joints are hard to do and I only use them if there is no other choice. If you are gluing long grain to long grain, then complicated joints are unnecessary. I use the lock miter bit on mitered corners of jewelry and keepsake boxes where the alternative is end grain to end grain adhesion.

[Mel Fulks]

For small projects the lock mitre might be more trouble than it is worth, and even too modern for some projects. But it is the best way to do things like column wraps or square newel posts. Carving is hard to do ...along with a lot of other skills;
there is nothing difficult about lock mitres. The detailed instructions I posted in two earlier threads will work and over the years I have shown a number of guys how to use it. We have also had good reports from those who have bought Alan
Shafler's set up gauge . Since the set up is referenced off the back of the material instead of the face all parts need to be
exactly the same thickness, and the process works best with a power feeder.

[fred everett]

Thanks Mel. Using Alan's jig made the setup a non-issue. Post setup I made cuts in pieces machined to the same thickness lying flat on the table - the fit was so perfect the joint nearly disappeared. My fence is less than .002 sq to the table. The table itself is dead flat....though off by .004 in some places across the router plate.

I'd consider myself an advanced beginner with occasional forays into the "intermediate" realm so there's no doubt my skills may be at fault....lol. Anyway, I think you are correct regarding power feeder....considering everyone's feedback and advice my issue seems to be all about the feed.

[Mel Fulks]

Carbide being less sharp than steel, and the angular zig zag pattern of the cutter, can make tearout a problem with woods
such as birch, maple, or cherry. If you use a 3/8 round over on the inside corners prior to using the lock mitre head you will
not lose any material to tear out. That round over must be made feeding backwards (climb cutting), but can be done with a pretty fast feed. With that square corner removed the "sweater will not unravel", you will not lose even one piece of
material no matter how difficult the grain.

[Jim Jakosh]

I used to have that problem when I just pushed the board through because it cuts way most of that bottom surface and the thin edge won't hold the board up. I don't have that problem any more. I made what I would call a vertical holding sled to attach that vertical board to with 2 DeStaco toggle clamps running in a T track on the sled. It rides on the top of the fence and keeps the board in that same position all the way through the cut. I have to use a board the same thickness as the sled up against the fence when running the horizontal board past the cutter so I am typically setting up to the same surface.
Send me an e mail at jcj7815@att.net and I'll find a photo of it for you. I'll be traveling for 5 days so I won't be on E mail until after that..

Jim

[fred everett]

Carbide being less sharp than steel, and the angular zig zag pattern of the cutter, can make tearout a problem with woods
such as birch, maple, or cherry. If you use a 3/8 round over on the inside corners prior to using the lock mitre head you will
not lose any material to tear out. That round over must be made feeding backwards (climb cutting), but can be done with a pretty fast feed. With that square corner removed the "sweater will not unravel", you will not lose even one piece of
material no matter how difficult the grain.

Okay you have me thinking now. So a 3/8 round over with the vertical piece in the same position it would be in for the lock miter cut? Do you know why this eliminates tearout and why it must be a climb cut?

[Mel Fulks]

Fred, when you have sharp square edges the cutter can find a "sweater thread" in unfavorable grain and just quickly rip off a large piece. First time I tried lock mitre on a shaper with birch I tore out not small pieces but huge long pieces .
Rounding over just gets rid of those starter points. Climb cutting might not be necessary ,but it is certainly possible to tear
off a big piece even using a round over bit. I've seen sanders used ,too. It's good to be skeptical about new info;but there
is nothing original or new here,just a very old tehcnique well known to those who make curved stair rail parts and such.
I hope it is obvious that the lock mitre cutter must be used only in normal forward cutting, be safe

[Pat Barry]

My suggestion is a bit different. Go with the flow. Since you get a divot that seems to be pretty repeatable work with it. Cut your pieces a bit longer than you need, then do the lock miter as you have been and then cut the pieces to finished length to removse the divot.

[Art Mann]

For small projects the lock mitre might be more trouble than it is worth, and even too modern for some projects. But it is the best way to do things like column wraps or square newel posts. Carving is hard to do ...along with a lot of other skills;
there is nothing difficult about lock mitres. The detailed instructions I posted in two earlier threads will work and over the years I have shown a number of guys how to use it. We have also had good reports from those who have bought Alan
Shafler's set up gauge . Since the set up is referenced off the back of the material instead of the face all parts need to be
exactly the same thickness, and the process works best with a power feeder.

Why are you saying that a lock miter joint is the best way to do square posts? The time I would spend setting up a lock miter bit on my router table could be spent actually gluing up the pieces. Why insert an extra useless step?

[Mel Fulks]

Art, I did qualify my answer in the first sentence. I possibly failed to mention I was referring to shapers with power feeders. Not router tables.

[ian maybury]

I think it'd be a pity to reject mitre locks out of hand when they can be very useful in some situations - but as ever the choice depends on a lot of factors including the equipment on which the job is being done.

To Fred's issue. Table/fence out of flatness is as before probably in the mix. It's hard to see what's going on in the pics, but there's a couple of possibilities to explain the intermittent 'divot' at the end of some cuts - which looks much like planer snipe, and may have similar origins.

One possibility might be that there is some irregularity/out of flatness/possibility of bridging or tipping in the router plate/router table/insert ring set up. That depending on where and how downward pressure is applied to the workpiece that the end maybe sometimes drops down into the cutter opening in the router plate to create the divot when the end of the work clears its lip/loses that support.

It could also be happening as a result of a similar gap or depression in the fence which allows the work to move a bit more into into the cutter, but that seems less likely as the cutter is running a more or less zero clearance and the effect of a dip would probably be more gradual.

I've found I did better with fairly stiff finger boards rather than rigid guide strips - presuming a slightly flexible workpiece which may include some minor thickness variations spring pressure is likely to improve consistency a bit by keeping the work tight to the fence and the table at the point where it passes over the cutter. As Mel (?) a properly angled power feeder or set of angled spring or roller stock guides like the Jessem items would likely deliver a similar benefit - presuming they can be fitted in.

Another potential issue is getting the fence adjustment precisely set. I fitted the fence on my shaper with threaded jacking screws at each end so that it can be incrementally adjusted/accurately returned to a previous setting if needed. I'd rather not try to set a loose fence on this sort of job by tapping it over and clamping it down.

My sense is that it's down to your choice of poison. A plain mitre has the advantage that it can be cut off a single set up (presuming the second piece can be flipped around) - but beyond that it surely requires similar accuracy of table flatness/precision of set up etc to deliver a good result? Cutting slots for biscuits introduces another set of possible issues, and then assemply may or may not be awkward with tape depending on the layout...

Fence asdjuster:

302736

[Mel Fulks]

I didn't say anything about divots, it was "huge long pieces". It can sound like a shot and travel almost as fast. For some problems there are multiple solutions ,this is not one of them. After a climb cut round over the cuts will be succesful.
I have even seen smooth cuts over obvious existing cracks in a direction opposing the cutter; though I will not use such
defective material. Even with metal I've heard of engineers requiring some things to be polished to get rid of "incipient cracks".... and I listen to 'em .....don't believe in "engineer bashing".

[fred everett]

You guys have convinced me to give the bit another try. I'll get back to it after the holiday visiting and such. My plan is to build a fence that immobilizes the work piece as suggested by Jim.

[lowell holmes]

I do have a shaper, but I've never needed nor used a lock miter, so what I'm about to say probably has no relevance. I have experienced the last inch or two of a piece being out of shape. I have always made the pieces longer and then cut off the bad stuff.

If your wood is already cut to length, it's not an option.

[ken masoumi]

You guys have convinced me to give the bit another try.
I admire your persistence,I threw my set in the" Do not ever try again" box with some other router bits,it had been a source of aggravation until I banished them.

[Alan Schaffter]

Sorry, I've been offline for awhile.

If you are getting a divot at the end of the cut, it is most likely that the bit or fence is not set properly. Possible, but less likely, you have too large of table opening and pushing down too hard on the trailing edge of the stock causing it to dip into the opening. By the way, divots caused by improper settings can start at almost any time but most often once at least half of the board has passed the cutter. It happens because the bit is too high or the fence is too far back which allows the cutter to remove too much material. That leaves the knife edge edge unsupported on the outfeed side of the cutter. It allows the outfeed end of the board to pivot slightly down toward the table or towards the fence so the cutter removes even more material for the remainder of the cutting pass.

Even using the Lock Miter Master (LMM), there are still a number of things that can cause a bad lock miter joint. I have addressed most, if not all, of them in the downloadable manual on Infinity's website. I'll try to list some of the more common ones here:

1. First, I don't remember if the website mentions it, but I am almost certain I included it in the manual- the LMM can not be used with a few lock miter router and shaper bits whose carbide is a chord of the cutter and not a radius. It should work with any bits where the cutter is, or close to being, a radius (see drawings below) The large Infinity Lock Miter shaper cutter falls in this category.

Carbide is a radius:

http://www.ncwoodworker.net/pp/data/2580/medium/Shaper_New.jpg

Carbide is a chord:

http://www.ncwoodworker.net/pp/data/2580/medium/ShaperB-1.jpg

2. Be as precise as possible marking the centerline of your stock and aligning your marks the reference lines on the LMM. Get the stock close to the jig (but don't bump the jig) and get up close and personal to avoid parallax.

3. When setting the fence, rotate the bit so the LMM is perpendicular with the fence. In their video, Infinity uses a small square to check this, but I have found using your eyeball is typically good enough.

4. It is easier if they are, but the boards DO NOT need to be the same thickness for a good joint. But, you must follow the procedures I put in the manual for this situation- the process is different than cutting standard lock miters in boards that are the same thickness. Cutting lock miters in boards of unequal thickness is a two step process- you cut all thinner boards first, readjust the fence/bit (depends on what profile you plan to use with each board), then cut all thicker boards.

5. If you are using the LMM you should ensure it is seated so the diagonal ledge is always against the 45° carbide and the center point is against the underside of the diagonal of the profile. If you are not careful you can bump the jig and cause it to slip from the correct position. If the recess for the carbide in the bit's body is too shallow or improperly machined the LMM may not sit on the carbide properly.

6. Unless you are using the LMM, if you have a bad test joint, NEVER, NEVER attempt to adjust just the bit height or fence position alone. Always check both settings. Since the main cutting edge of a lock miter bit is at 45° to both the table and the fence, if you change the position of one it will affect the other! You will be forever chasing the proper settings. This is one procedure that all the magazines articles I have read about setting lock miter bits get wrong!! Once you have set the bit height and fence position, if you go back and raise or lower the bit to adjust the joint for a board cut in the horizontal position you have likely screwed up the fence position for any boards you will cut in the vertical position! Both settings must be done and verified at the same time!

7. Feed rate, and pressure on the stock, and position of the stock can affect the cut of the joint. If you press too hard, you can deflect the knife edge and remove too much stock. If you attempt to move the stock too quickly, you may end up removing too little stock, especially if you allow the stock to separate from the bit in either the horizontal or vertical direction.

8. Since the knife edge can be fragile, especially with soft wood, consider, as someone has already mentioned, leaving blunt tips on both boards- that is easy to do with the LMM.

[fred everett]

Sorry, I've been offline for awhile.

If you are getting a divot at the end of the cut, it is most likely that the bit or fence is not set properly. Possible, but less likely, you have too large of table opening and pushing down too hard on the trailing edge of the stock causing it to dip into the opening. By the way, divots caused by improper settings can start at almost any time but most often once at least half of the board has passed the cutter. It happens because the bit is too high or the fence is too far back which allows the cutter to remove too much material. That leaves the knife edge edge unsupported on the outfeed side of the cutter. It allows the outfeed end of the board to pivot slightly down toward the table or towards the fence so the cutter removes even more material for the remainder of the cutting pass.

Even using the Lock Miter Master (LMM), there are still a number of things that can cause a bad lock miter joint. I have addressed most, if not all, of them in the downloadable manual on Infinity's website. I'll try to list some of the more common ones here:

1. First, I don't remember if the website mentions it, but I am almost certain I included it in the manual- the LMM can not be used with a few lock miter router and shaper bits whose carbide is a chord of the cutter and not a radius. It should work with any bits where the cutter is, or close to being, a radius (see drawings below) The large Infinity Lock Miter shaper cutter falls in this category.

Carbide is a radius:

http://www.ncwoodworker.net/pp/data/2580/medium/Shaper_New.jpg

Carbide is a chord:

http://www.ncwoodworker.net/pp/data/2580/medium/ShaperB-1.jpg

2. Be as precise as possible marking the centerline of your stock and aligning your marks the reference lines on the LMM. Get the stock close to the jig (but don't bump the jig) and get up close and personal to avoid parallax.

3. When setting the fence, rotate the bit so the LMM is perpendicular with the fence. In their video, Infinity uses a small square to check this, but I have found using your eyeball is typically good enough.

4. It is easier if they are, but the boards DO NOT need to be the same thickness for a good joint. But, you must follow the procedures I put in the manual for this situation- the process is different than cutting standard lock miters in boards that are the same thickness. Cutting lock miters in boards of unequal thickness is a two step process- you cut all thinner boards first, readjust the fence/bit (depends on what profile you plan to use with each board), then cut all thicker boards.

5. If you are using the LMM you should ensure it is seated so the diagonal ledge is always against the 45° carbide and the center point is against the underside of the diagonal of the profile. If you are not careful you can bump the jig and cause it to slip from the correct position. If the recess for the carbide in the bit's body is too shallow or improperly machined the LMM may not sit on the carbide properly.

6. Unless you are using the LMM, if you have a bad test joint, NEVER, NEVER attempt to adjust just the bit height or fence position alone. Always check both settings. Since the main cutting edge of a lock miter bit is at 45° to both the table and the fence, if you change the position of one it will affect the other! You will be forever chasing the proper settings. This is one procedure that all the magazines articles I have read about setting lock miter bits get wrong!! Once you have set the bit height and fence position, if you go back and raise or lower the bit to adjust the joint for a board cut in the horizontal position you have likely screwed up the fence position for any boards you will cut in the vertical position! Both settings must be done and verified at the same time!

7. Feed rate, and pressure on the stock, and position of the stock can affect the cut of the joint. If you press too hard, you can deflect the knife edge and remove too much stock. If you attempt to move the stock too quickly, you may end up removing too little stock, especially if you allow the stock to separate from the bit in either the horizontal or vertical direction.

8. Since the knife edge can be fragile, especially with soft wood, consider, as someone has already mentioned, leaving blunt tips on both boards- that is easy to do with the LMM.

Thanks Alan. I read the LLM instructions AND read everything else you posted online before I bought the bit/LLM...lol. I am leaning toward your thought the issue is related to the table opening. There's really not much space between the bit & insert but I remember feeling the end of the vertical cuts pulling downward on many of the vertical test cuts. I tried switching pressure toward the front of the piece like you would on a jointer. In the end I was able to get 1 or 2 good vertical cuts out of 20 or so. Considering this would you agree a sled that immobilizes the work piece is worth a shot?

[Alan Schaffter]

Fred, an easy way to check if the bit is removing too much material-

Set the bit and fence as best as you can- now start cutting a lock miter but stop after cutting .5" - 1" or so (in either horizontal or vertical position), turn off the router, and inspect the knife edge. Lay a straightedge across both 90° corner and the short 45°knife edge of the board or put the edge of the board up against a flat surface. There should be no gap between the knife edge and the straight edge (or flat surface)- the 90° corner should be collinear with the 45° knife edge. If there is the slightest gap, too much stock is being removed.

Put the LMM back on the bit and check your technique by verifying that the jig's reference lines align perfectly with the centerline marks on the boards. Using the LMM is easy but like a lot of things in woodworking the results require careful use, i.e. a box joint jig not set within just a few .001" will yield a lousy joint- close is typically not good enough.

Any sled would need to ensure the stock stays put. One technique that some like but I find too time consuming, a pain, and in reality not needed, is to use a few spots of glue to attach a carrier board to the face of each board. That ensures the edge of the board is always supported.

[fred everett]

Fred, an easy way to check if the bit is removing too much material-

Set the bit and fence as best as you can- now start cutting a lock miter but stop after cutting .5" - 1" or so (in either horizontal or vertical position), turn off the router, and inspect the knife edge. Lay a straightedge across both 90° corner and the short 45°knife edge of the board or put the edge of the board up against a flat surface. There should be no gap between the knife edge and the straight edge (or flat surface)- the 90° corner should be collinear with the 45° knife edge. If there is the slightest gap, too much stock is being removed.

Put the LMM back on the bit and check your technique by verifying that the jig's reference lines align perfectly with the centerline marks on the boards. Using the LMM is easy but like a lot of things in woodworking the results require careful use, i.e. a box joint jig not set within just a few .001" will yield a lousy joint- close is typically not good enough.

Any sled would need to ensure the stock stays put. One technique that some like but I find too time consuming, a pain, and in reality not needed, is to use a few spots of glue to attach a carrier board to the face of each board. That ensures the edge of the board is always supported.

Thanks. I'll do a reset to ensure a perfect setup then test a partial cut as you described.

[Mike Schuch]

Ditto......

Before you give up try this method. It works well for me and I use lock miter cutters often. A sixteenth flat left is plenty and an 1/8 or 3/16 roundover will take care of it.

I agree. I use lock miters a lot and really like them. The sharp corner on your board is collapsing and the whole board is moving closer to the bit in that last inch causing that indentation. As soon as the broad face of the unmachined portion of the board is no longer in contact with the infeed portion of the fence the whole board is moving in towards the fence causing this issue. If you lower the bit a touch the machined portion of the board will be just a touch blunt instead of that super sharp 45 degree edge. I am talking about 1/64" or less of a blunt edge instead of your perfectly pointed edge. You can easily round out the 1/64" notch with a touch of sanding.


P.S. to get optimal results with a lock miter bit the spindle needs to be exactly perpindicular to the table in all directions (front and back as well as side to side). The first time I did lock miters on my shaper I had a heck of a time and could never get them right until I figured out the shaper spindle was just s little touch out of perpendicular to the table. My shaper will actually deflect the spindle a couple minutes of a degree when the height adjustment on the spindle is unlocked... a good lesson to always lock the height adjustment on the shaper.

[ian maybury]

That's a good point Mike - that too much downward pressure could be causing a sharp corner to deform. That's another possibility in addition to potential table or fence flaness issues...

[fred everett]

I made a step or 2 toward getting my ML setup dialed in today. Based on photos Jim J sent me I made a sliding vertical jig that does an excellent job of holding down the work piece....made 6 test cuts with no issues....many thanks Jim! I also made a stationary fence from the same material to keep the distance from the fence the same on both vertical & horizontal cuts. This completely resolved the previous issue. However, I don't think I'm there yet as I didn't get a knife edge on the horizontal cuts like I did on the verticals. When dry fit with clamps I was able to sand over the edge and it looked really nice. I felt good about everything UNTIL I glued up. Unfortunately I didn't get a uniform squeeze out....not even close. The outer edges did not completely close-up like in the dry fit.

Definitely good progress, but not sure where I'm at right now. Should I mix sawdust/glue, fill the gaps and sand smooth? My gut say no, but not sure how tight one can get a lock miter joint? Considering the margin for error is .001, with glue is it possible to completely close the joint? Is it normal to fill and sand these joints smooth? Here's a few photos. I'm missing something....or several somethings....lol

303009303010303011303012303013303014

[Alan Schaffter]

A good set-up and two good lock miters are only half the job. Now you need to do a good job assembling the joint and that requires flat boards and good clamping. If one of the boards is not flat it won't mate properly to the other one. Even flat boards can be a problem- remember you are dealing with sharp corners which can deflect when the glue is forced from the joint. The only solution is good clamping. My suggestion is to put a board on each side of the joint so clamping pressure is applied to both sides for the entire length of the joint. Position the boards as close as possible to the corner, then use two or more band clamps to pull everything tight. That should yield a tight joint with nearly invisible glue line (see pics below). You should avoid placing the clamp in the middle of the sides of your box (F clamp in bottom center pic)- that will cause the side to bow inward and cause the joint to open at the outside corner. The joint in your last pic looks pretty good to me.

Below are some close up photos of a bookcase I built a few years ago. I decided to use lock miter joints which I had never attempted before. After researching and testing various setup techniques I wasn't happy with the results so before I started the project I spent some time developing the Lock Miter Master. I was really happy with the results. Everything, except the shelves, was joined together with lock miters- the birch ply sides, top, bottom, AND maple face frames were all joined together with lock miters. If you look very carefully you can see a few tiny gaps- I probably could have added filler, but you really can't see them from a foot or two away.

Bookcase- intersection of 3 lock miter joints:

http://www.ncwoodworker.net/pp/data/2580/medium/IMG_4439.JPG

Bookcase- face frame-to-top lock miter joint:

http://www.ncwoodworker.net/pp/data/2580/medium/IMG_4435.JPG

Bookcase- side-to-top lock miter joint:

http://www.ncwoodworker.net/pp/data/2580/medium/IMG_4434.JPG

A few pics of other lock miter joints. Setup is easy with the LMM (joints are dry fit, no glue, no clamps):

Lock miter with different thickness boards. (dark areas are shadows, not voids):

http://www.ncwoodworker.net/pp/data/2580/medium/IMG_4472.JPG

Off-set lock miter for when you don't want the miter line at the corner:

http://www.ncwoodworker.net/pp/data/2580/medium/IMG_4470.JPG

[Larry Edgerton]

Another trick I use, especially in the winter months is to keep a spritz bottle of water around and dampen the pieces [gluing surfaces]before you glue up. It gives you a touch more time to get clamped up and has no adverse effects on joint strength. Not standing water, just damp.

[ian maybury]

Great to see good progress Fred. The bottom line seems to be just that this joint is sensitive to minor inaccuracies.

If the knife edge is consistent but different for the horizontal and vertical cuts that may suggest just that the set up isn't quite 100% right - depending on which is off either the cutter height or the fence position.

If the joint isn't consistent all along its length and from piece to piece that might as berfore suggests some out of straightness in the stock, or something in the fence/table/handling of the stock.

It's possible to end up in a space where there's a certain amount of minor variability that won't consistently tune out which probably is down to hard to identify (never mind predict the effect of) inaccuracies/flexibilities in the machining rig or and/or workpiece. A quick touch with a sanding block can disguise a multitude.

It's probably advisable to brush the glue on to the jointing surfaces to get an even coating, and to assemble before any of it thickens (use a slow cure) - chances are that blob of glue enclosed in one of the grooves in the joint might not squeeze out so easily and prevent it closing properly.

Tensioned tape might well do a good job of closing the corners if the gap is small and a bit irregular - if nothing else wetting the inside of the lip (especially on softwood) might cause it to expand/curl outwards a little and open a small gap.

These joints close only in the line of the splines - from one side only/not from both sides. So it helps to know which side is which before trying to tap or clamp them shut - and to apply the force/pressure accordingly.

[Mel Fulks]

That is a strong joint by its geometry. I usually put the glue on those with a squeeze bottle . If I'm doing a bunch at one
time I put heavy paper or plastic between them and cauls on outer pieces ,clamp for about half hour ,then take out and
stand them against wall to let excess glue harden. Sometimes I clean off some glue then,but if not,belt sander will get it
later.

[fred everett]

Thanks again for all the follow ups. After studying my test leg with fresh eyes today, it seems the bit MAY not have been set high enough as I have a flat edge a full 1/16" high on the horizontal pieces. You can see this in the close up pic below. That said, I know I had it set perfectly against the LMM when I started. Here's a few of the factors I believe I have ruled out:

-Thickness variance: The 4 pieces measure .749 to .753 so I have to assume thickness is NOT the issue.
-Flatness: Test piece came from HD which I edge jointed but did not face joint. I assume this is NOT an issue because the 1/16" flat edge is consistent across the cut.
-Bit Speed: PC 3hp router tested 16k to 21k & 21k cut was smoother in tests. However, a 21k test cut (same feed rate) in the Sapele burned - jury still out on which Sapele speed.

Assumption: Something must have shifted after setup OR something in my setup dictates that I must set the bit height a hash mark higher than dead center on the LMM. Does it seem that I'm on the correct course with my troubleshooting, or am I down a "bunny" hole?

One other question....do you typically hog off all the materiel in one pass when working with 3/4" hardwood?

303041

[Alan Schaffter]

Wow, that looks like a 1/8" or larger offset lock miter (see my earlier post)! Are you sure you used the main reference line on the LMM?

Is it possible your bit slipping in the collet? I had a problem with the collet on one of my routers and it drove me crazy until I discovered what was happening! Or could it be your router or lift is slipping down.

[Pat Barry]

Fred, that picture doesn't tell the whole story for me. The reason being that for the board on the left, we are seeing the end of the cut but we are looking at the beginning of the cut for the board on the right. Are you sure the cut is consistent along the entire length of each cut? Take a picture of the other end of the same joint.

[fred everett]

Wow, that looks like a 1/8" or larger offset lock miter (see my earlier post)! Are you sure you used the main reference line on the LMM?

Is it possible your bit slipping in the collet? I had a problem with the collet on one of my routers and it drove me crazy until I discovered what was happening! Or could it be your router or lift is slipping down.

I'm sure I used the main reference line. The collet was definitely tight on bit but I will investigate and swap out as I have a 2nd collet. I don't have a "lift" per say, I simply have the base of my PC 7518 attached to a phenolic base plate....bit height is adjusted by loosening the router base thumb screw and turning the body of the router. As you can imagine adjusting the bit height in this fashion is not a precision operation. Do you think lack of a precision lift makes miter lock joinery a non-starter for me?


Fred, that picture doesn't tell the whole story for me. The reason being that for the board on the left, we are seeing the end of the cut but we are looking at the beginning of the cut for the board on the right. Are you sure the cut is consistent along the entire length of each cut? Take a picture of the other end of the same joint.

Thanks Pat. Here's a photo of the entire cut.

303124

[Alan Schaffter]

Do you think lack of a precision lift makes miter lock joinery a non-starter for me?

I don't know about a non-starter, but it sure could make it extremely difficult, considering the tolerances involved. A true lift isn't required, but a router that can be adjusted from the top of the table would be much better than having one you must adjust from below the table.

Speaking of tolerences, if prior to setup using a LMM, you mark your board centerline with a pencil having a thick or wide tip, you can end up with a bad setup and bad joint because of bad alignment. Also, if you don't align your mark with the center of the LMM's white reference line you can be a tad off. To reduce the possibility of this, I recommended that the LMM cursor lines be thinner. Engraving and coloring limitations on such a small part and concern over the ability to see the cursors were overriding factors. Setup is so sensitive, it is even possible to get different results by aligning a finely-drawn board centerline with opposite sides, instead of the center, of the LMM reference line!!!

[fred everett]

I don't know about a non-starter, but it sure could make it extremely difficult, considering the tolerances involved. A true lift isn't required, but a router that can be adjusted from the top of the table would be much better than having one you must adjust from below the table.

Speaking of tolerences, if prior to setup using a LMM, you mark your board centerline with a pencil having a thick or wide tip, you can end up with a bad setup and bad joint because of bad alignment. Also, if you don't align your mark with the center of the LMM's white reference line you can be a tad off. To reduce the possibility of this, I recommended that the LMM cursor lines be thinner. Engraving and coloring limitations on such a small part and concern over the ability to see the cursors were overriding factors. Setup is so sensitive, it is even possible to get different results by aligning a finely-drawn board centerline with opposite sides, instead of the center, of the LMM reference line!!!

I have serious doubts I'll ever get consistent results considering my lift and the fact that I using two different setups to make the vertical & horizontal cuts. Seems like a moving target due to the variables. On the upside, I learned quite a bit and I finally rebuilt my old router fence during this process. Unless I have some type of epiphany tomorrow, I going to build a tablesaw sled and make these traditional miter cuts. Having said that, I believe your Lock Miter Master is sheer brilliance, but as they say, one must have a good "foundation" before building a house.

[Alan Schaffter]

Though it can be confusing, using two separate setups is certainly an option to get the joint right. In fact, that is what is required to make the off-set lock miter and lock miters in boards of different thickness I showed at the end of my earlier post, but that technique is best used in only those cases.

If you cut just one test board in the horizontal position, cut another in the vertical position, but then change the bit, fence, or both settings in an attempt to "fix" the fit of the board cut in the vertical position, any boards subsequently cut in the horizontal position will be incorrectly cut and the joint will still be bad.

In order to use two setups, before you can adjust the setup to cut the second group of edges (in the second position), you MUST use the original setup to cut ALL edges in the first group (in first position). Then you can adjust the setup to get a good joint before cutting ALL edges in the second group (second position). You can not go back and cut any more edges in the first position because you have already changed the setup for those and will not get a good joint. Only ONE lock miter bit/fence setup will allow you to cut boards in any position in any order. The LMM should allow you to do that.

The beauty (advantage?) of the current lock miter bit design is so that a correct setup will allow you to cut boards in either the horizontal or vertical positions at any time, in any order without fiddling with the setup.

[ian maybury]

That's (as Alan) definitely a rather thick corner Fred, and because it's parallel (and given the spline direction) it may be the result as you seem to suspect of an inaccurate cutter height setting rather than a problem with table flatness/bent work or similar. That setting has to be precisely dialled in to cut horizontally presented narrow test pieces dead right (see early parts of the first two links below) before moving on to setting up the fence to cut the mating part in the vertical - as Alan any height adjustment made to the cutter height after the fence is set up will mess with the fence setting and require revisiting it.

If you are getting perfect horizontal test cuts on narrow pieces as shown (using narrow test pieces minimises any possibility of bridging as a result of an out of flat table, or a bent workpiece), and your subsequent work pieces are coming out different but consistent/parallel along the length of the cut then it'd seem like the cutter may be moving. Repeating the horizontal test with some more narrow pieces should show what has happened - if the error repeats having previously been OK on the first test then it pretty much has to be a cutter movement issue. Backlash (always ease up to the required setting from one direction) or unreliable locking of the router height adjustment might also be a cause of this.

If you can get the horizontal test pieces dialled in as in the video then chances are your have something that can work, but need to figure out why its moving.

I have a Router Raizer on a De Walt DW 625 in my router table - while it can have some limits in terms of available vertical travel (its useful to get well above the table for easy bit changing) it's presuming there is one for your router a very low cost method of gaining precision height adjustment. Much cheaper than a full on lift.

Here's some material on setting up mitre locks. The method shown in the first link is very close to what I have used - right down to the principle of his sled. (although i was on a shaper) The horizontal test cut is not far down the page. Note that none of these guys are keen on using exclusively hand feeding - the implication is that it doesn't hold the work down consistently enough/leaves scope for it to lift off.
http://legacy.woodshopdemos.com/cmt-lm10.htm

Here's another take on lock mitre set up (for shaper and then router - same principle) for added enjoyment. :) Notice how all emphasise table and fence set up accuracy.
http://www.garymkatz.com/trimtechniques/lock_miters.html

Another form of locking corner joint:
https://www.youtube.com/watch?v=d148cyISkR8

On the alternative of using a plain mitre cut off the saw instead. It seems likely that for a plain mitre to be accurate the cut has to be placed just as accurately, and be just as straight off the saw or whatever as it does with a lock mitre bit for it to be dead right. It's going to be subject to most of the same issues to do with table flatness, fence spring and alignment. Yet many seem to regard it as much easier.

My guess is that the perceived difference is possibly because (a) the piece of stock typically starts wider than required, so if the knife edge of the bevel cut wanders a little it may not matter so much - it'll still leave a sharp edge. i.e. it won't necessarily be dead straight, but that isn't necessarily obvious, and it's possible to go back later and place the second cut as necessary relative to it during assembly if e.g. a leg casing of a given width is required. It might be very different if it was necessary to place the 45 deg saw cut so that it precisely split the 90 deg corner right along the full length of one edge of the piece of stock…

(b) is that when joining two 45 deg mitred edges there's a fair amount of freedom to slip the two joint faces on each other to get the corner perfectly lined up. Biscuits and the like are likewise referenced off the cut edge - which while not necessarily perfectly straight may not matter so much...

[Mel Fulks]

I think the set up gage would be more useful with instructions that focus on the standard 'all pieces exactly the same thickness ' set up than how to work with different thicknesses when there is no practical reason to do so. After I posted that
because the material for lock mitres is run back to table rather than face to table the material must be exactly the
same thickness, Alan posts in bold print 'pieces do not have to be same thickness' ,well , for a frustrated beginner they need to be SAME THICKNESS.

[Al Launier]

Just thinking that perhaps you could make a "plunge cut" first from the tail end of your work piece to establish the finished profile, then make the cut from the leading edge of the work piece as you normally do. If the board tens to "sink" on you as you pass the trailing end by the cutter there will probably be enough "clearance" to eliminate the divot. Just a thought.

[Alan Schaffter]

After I posted that because the material for lock mitres is run back to table rather than face to table the material must be exactly the same thickness, Alan posts in bold print 'pieces do not have to be same thickness' ,well , for a frustrated beginner they need to be SAME THICKNESS.

Don't sell yourself short. For most projects and if you are a beginner, it is best to use boards of the EXACT same thickness. It is just easier. I would NEVER attempt to cut lock miters in different thickness stock to compensate for poor milling. In addition to requiring a two step process, it is totally possible that poor milling has resulted in the thickness varying from board to board.

While the LMM allows you to do the setup quite simply, the only time I would ever want to put lock miters in boards of a different thickness is for a special project, e.g. between casework with 3/4" thick faceframe and 1/2" thick sides/bottoms. (Note: the profiles are reversed in the first drawing- the dark wood faceframe can't be assembled to the box :rolleyes: Faceframe pieces should always be cut in the vertical position. The profiles are correct in the second drawing).

FYI, the first drawing illustrates a combo joint- a lock miter joint in stock of different thicknesses + an offset lock miter joint (uses the short reference lines on the LMM). An offset lock miter joint is useful if you want to put a radius on the corner but don't want to expose the laminations at the edge of the ply (see second drawing). Also note that an offset lock miter joint in stock of equal thickness results in a stepped inside corner.

http://www.ncwoodworker.net/pp/data/2580/medium/Offset_LM.jpg

http://www.ncwoodworker.net/pp/data/2580/medium/Offset.jpg

The key to successfully cutting lock miters in different thicknesses of stock is you mark the boards differently (see drawing below). First, you mark a centerline on the thinner boards, do a normal bit/fence setup, and cut all of of them. Then, you re-adjust the setting using the thicker board's line- it will not be a centerline. Step by step instructions are in the LMM manual.

http://www.ncwoodworker.net/pp/data/2580/medium/Unequal_Stock.jpg

(Same problem with the profiles in these pics- the faceframe would need to be the lighter colored wood since its profile is cut with the board vertical against the fence. :mad: )

http://www.ncwoodworker.net/pp/data/2580/medium/Unequal_Stock-2.jpg

http://www.ncwoodworker.net/pp/data/2580/medium/Unequal_Stock-3.jpg

[Alan Schaffter]

I just found this drawing which illustrates my earlier post about what can happen when cutting a lock miter joint if you have an incorrect setup. It is possible to remove too much of the stock so the just-cut knife edge isn't collinear with the original edge of the board. Once enough of the board has passed the bit it won't be properly supported and will move inwards (or downwards). The bit will remove too much material creating a deeper divot and the profile will be uneven over its length.

You can verify your setup to ensure this won't happen- take any board of the exact same thickness as your project stock and make a short 1/2" to 1" cut. There should be no gap between the just-cut section and the out-feed fence (or out-feed half of the table for a board cut in the vertical position). You can also use a straight edge to check for a gap.

If you have a split fence with independently adjustable halves, do not attempt to adjust the out-feed half to prevent this- you will create a divot in stock being cut in the vertical position and won't be able to feed it properly!

The gap created by an improperly set bit/fence is exaggerated here, but you get the idea:

http://www.ncwoodworker.net/pp/data/2580/medium/setting_fence.jpg

[Mike Schuch]

Thanks again for all the follow ups. After studying my test leg with fresh eyes today, it seems the bit MAY not have been set high enough as I have a flat edge a full 1/16" high on the horizontal pieces. You can see this in the close up pic below. That said, I know I had it set perfectly against the LMM when I started. Here's a few of the factors I believe I have ruled out:

-Thickness variance: The 4 pieces measure .749 to .753 so I have to assume thickness is NOT the issue.
-Flatness: Test piece came from HD which I edge jointed but did not face joint. I assume this is NOT an issue because the 1/16" flat edge is consistent across the cut.
-Bit Speed: PC 3hp router tested 16k to 21k & 21k cut was smoother in tests. However, a 21k test cut (same feed rate) in the Sapele burned - jury still out on which Sapele speed.

Assumption: Something must have shifted after setup OR something in my setup dictates that I must set the bit height a hash mark higher than dead center on the LMM. Does it seem that I'm on the correct course with my troubleshooting, or am I down a "bunny" hole?

One other question....do you typically hog off all the materiel in one pass when working with 3/4" hardwood?

303041

I remember having joints that looked like this and it drove me crazy until I got my shaper dialed in. Is your spindle perpendicular to the table in all directions?

I clamped a metal bar between 2 nuts on my shaper spindle so the bar was parallel to the shaper table top. I then mounted my dial gauge to one end of bar with the tip of the dial gauge touching the shaper table surfacce. I then turned the spindle 360 degrees until the variance measured by the dial gauge from the bar to the table was as small as I could make it by shimming the shaper spindle supports underneath the table. This was years ago so I don't remember the measured runout but it was very small. After this my gaps disappeared. I don't know how you would accomplish the same thing with a router?

[Alan Schaffter]

Thanks again for all the follow ups. After studying my test leg with fresh eyes today, it seems the bit MAY not have been set high enough as I have a flat edge a full 1/16" high on the horizontal pieces.

Fred, after re-reading your post just quoted by Mike, specifically the text above that I made bold red, I want to make sure you understand- While the bit may not have been high enough, which will cause a blunt tip on horizontal pieces, it could have just as easily been that the fence was too far forward which will also cause a blunt tip on horizontal pieces!!! (Not to mention what could be happening to vertical pieces)

It gets back to one of the points I tried to make in a previous post- the bit height and fence position are inter-related for horizontal (and vertical) pieces- if you change one (bit height or fence position) you'll need to change the other. That is why both bit height and fence position should be set correctly, at the same time, to begin with. You can't just assume only one part of the setup is wrong, and correct only that one- you'll just end up chasing the correct setups.

[fred everett]

Many thanks to everyone for the advice. I definitely plan on revisiting the ML bit, but I made a 45 degree TS sled last night. The cuts are done....not perfect but very close.

303388303389

[ian maybury]

Well done Fred. Don't give up on the lock mitre though!

To Alan. Just wondering on a point. For sure both the cutter height above the table and the fence position need to end up simultaneously and exactly correct. Unless i've got it wrong though (it seemed to work for me) does the A side and then B side up cutting of two narrow test pieces off the table - dialling in the cutter height until the flush joint shown early in the first how to video linked in (47) above 'legacy wood shop demos' is created not deliver the vertical height required? So that while leaving the tool height undisturbed from here onwards the fence position can now be dialled in by the similar second test shown in the video - only this time cut with the test pieces in the vertical against the fence?

The effect of this test method seems to be to find a setting that centres the spline cutters in the mitre - for whatever the thickness of the stock? i.e. it's a way of one setting at a time getting to an overall set up where both the cutter height and the fence position aer simultaneously correct? The only real constraint seems to be that both pieces of stock must be the same thickness - i haven't figured out if it can be adapted for unequal thicknesses yet...

[Alan Schaffter]

I'm a little slow tonight so not sure I am following your description, but I think the answer is no. I'm not sure if this addresses your point, but using a board in the horizontal position, you can set the bit height almost anywhere you want (the geometric center of the profile may, but likely will not, be in the center of the miter) and adjust the fence for a sharp point (eliminate the gap shown in my drawing or ensure there is no overcut)- you will get what "looks like" a correctly cut horizontal board.

But, unless you were lucky enough to get the "correct setting," with the geometric center of the profile is at the centerline of the board, boards cut in the vertical position won't be cut correctly- they will either have a blunt tip or you will remove too much material and create a divot along the edge. Sure, you can readjust the bit and fence to avoid the divot in vertically cut boards, but you better not try to cut any more boards in the horizontal position, because you just changed the setting for those.

Now, if you are talking about adjusting the bit height so two boards cut in the horizontal position will mate, with flush surfaces, when one is inverted, that is different. Depending on care taken, that should ensure the geometric center of the profile is in the center of the board (where it should be). Then you can adjust the fence so the corner is a point as I discussed earlier. That should result in good miters on boards cut vertically and horizontally. There are two problems with that approach, the first and foremost is that it may require many test cuts to get it right. Close enough, is not close enough! :D Second, I don't believe that method can be used with boards of unequal thickness.

The answer is use the LMM or some other means to get everything right the first time, or cut ALL horizontally cut boards with your first setting before adjusting settings to cut ALL vertically cut boards. If this method is used, be sure you do not attempt cut any more boards in the horizontal position.

Ian, Where 'over there' are you? LOML and I are headed to the Emerald Isle next fall for a two-week tour of many of the usual tourist areas. We start in Shannon/Ennis, then among other places travel to the Cliffs of Moher, south to Kilarney, Ring of Kerry, Blarney Castle, east to Waterford, up to Kilkenny, back across to Galway and Connemara, wind our way north to Kyemore Abbey, Ashford Castle, Belleek, Belfast and end in Dublin. So, could you please arrange some nice weather for mid-September? :)

[ian maybury]

:) My 'American' is probably not so good Alan. Sounds like we're on the same wavelength. I meant as your third paragraph - when 'two boards cut in the horizontal position will mate, with flush surfaces, when one is inverted' as indicating that the vertical setting has hit 'the geometric centre of the profile'. It seemed to work just fine - provided the surface is flat and the test pieces the same thickness its a reasonably precise indication. The procedure is described in rows 5,6,7 and 8 on the page the link above opens up - again here: http://legacy.woodshopdemos.com/cmt-lm10.htm

The Hammer F3 shaper has a sliding table, I set up as above (no pic) and made the horizontal cuts off it (the slider in my case is set effectively flush with the main shaper table - stock they are set quite high, but any lift will mess with the lock mitre set up as the vertical cut references off the table/the flat jointer table in the pic used to load the sled in my case) using the shaper fence as a stop to exactly locate the end of the workpiece so that the cutter just created a nice knife edge at the previously cut end of the piece - as i was making 4 shop cabinet cabinet sides each of which needed to be exactly the correct length/had previously been cut to exactly those lengths. Having made the horizontal cuts I then made the vertical cuts off the fence (can't remember how much if any dialling in of the fence position was needed - but the cutter height must not be touched) using the sled in the pics below. I could have gone back and cut more horizontals (since the cutter height was untouched), but might have needed to locate the workpiece end by a different means to do so.

The need to precisely locate the mitre lock cuts to create a cabinet side or whatever of a particular length adds another twist to the whole thing. The sled is useful in that it very securely and safely holds the long workpiece during the vertical cuts, and also because it doesn't require a high fence to hit an accurate vertical - it just uses the fence as a guide. Plus it permits if required the controlled shimming of the workpiece up a little above the table using sheets of paper or similar, and shields the cutter. The slider gives very similar capability in the horizontal - the piece is clamped down on the slider, and kept square by the 'cross cut fence' (no pic), and its possible to fine tune the work location by shimming from the main fence.

One complication with a slider is that as well as hitting the other settings the main shaper fence must be accurately flat and aligned in the direction of travel of the slider if it's to be used as a stop/to shim from - it will otherwise give different results at different points. This is why i fitted the DIY fine positioners/threaded stops at both ends of the fence as just visible in the first pic.

The finished corners of the cabinet are not micron perfect, but it's for workshop storage and is in cheap ply - i didn't dial it in the last bit that was possible.

I can't guarantee the weather for your trip Alan, but mid September is normally pretty dry although it's a bit cooler than mid Summer. I'm a bit South of Dublin, PM me as it'd be fun to hook up.

303426303429303428303427

[Alan Schaffter]

Sliding tables and carrier sleds can work wonders, especially with very small and even large pieces. I recall an old episode of New Yankee Workshop where they used a sled to capture narrow window sash pieces so they could safely cut the muntin profiles on both edges using a shaper.

To control the width of your stock, you can start with them a little over-sized and sneak up on the right width (length?), but I suppose it is better to use some sort of stop so the pieces (or the sled) are positioned the same from piece to piece.

I recently had a similar problem. I've been working on a new design for non-45° lock miter bits (for 5, 6, 8, 10, 12- sided boxes), that require only one bit instead of two like the current versions of the 22.5° lock miter bit on the market. For tight joints all the way around, the angles need to be perfect since there are so many of them, BUT the sides need to be exactly the same length, too. Since the test boxes I was making were relatively small, I couldn't jig them properly to ensure each side was the same length, so my solution was to put the miters on the edges of a long board (ends of wide boards?). Once the profiles were cut, I ripped/crosscut the individual sides from it. That technique worked well.

As you know, due to the 45° lock miter bit profile and because the angles are complementary (they add up to 90°), the joint is cut with one board horizontal and one vertical using a single setup. You can't do that with other angles because they are not complementary. The industry's solution to date has been to use TWO BITS for other angles. As far as I know the only set commonly available is for 22.5° lock miters. With this bit set, you do a setup for the first bit, cut one edge(end) of all sides, then install the second bit AND MAKE ANOTHER PRECISION SETUP before cutting the opposing edges(ends).

With my bits you set the height of the bit by eye so the profile is "somewhere near the center of the edge of the board." A centerline is not needed and MK-1 eyeball is close enough :D. Then you adjust the fence for a good point, cut the first batch of edges(ends), then, instead of swapping bits, you raise (or lower) the bit a unique amount, pre-determined by geometry for each angle (and etched into the bit?), so you'll need a Wixey or other height gauge. Set the fence, then cut the opposing edges(ends)- pretty simple and fool-proof. If you need to recut a first board, just lower (or raise) the bit the designated amount and reset the fence. I had two bits (one for 5 sided boxes and one for 6 sided boxes) custom ground to verify my design- it works!!! It is now being evaluated by a manufacturer- 50-50 chance the design will be licensed.

[Mel Fulks]

Couple million dollars worth of robots can also be useful. Only advantage I see to sleds is that might allow the cut to come
to a full point and by doing so make the centering function of the set up gauge be better used. But might cause more tear
out. My preference is to use power feeder and run the vertical pieces first ,leaving a little of flat and making sure that the outside 45 flat is shorter than that of inside corner. That assures that when the horizontal pieces are cut the cutter will have to be raised and the fence slightly moved. My opinion is that the most frustrating thing for beginners is finding they
have made an adjustment in the wrong direction. Yes ,that means you must be able to count the number of pieces needed for the job to avoid repeatedly going back and forth between vertical and horizontal set ups.....that's where the
robots really help.

[Alan Schaffter]

My opinion is that the most frustrating thing for beginners is finding they have made an adjustment in the wrong direction. Yes ,that means you must be able to count the number of pieces needed for the job to avoid repeatedly going back and forth between vertical and horizontal set ups.

You got that right! You don't need to be a beginner either! http://www.ncwoodworker.net/forums/images/smilies/BangHead.gif

[ian maybury]

Cutting these joints certainly requires keeping your wits about you, and being very methodical, marking stuff up etc. It's one task where every move must be considered, as what's required is not always obviously intuitive. It's probably the case too that it's one of the few scenarios where it's necessary to be right on the nail with multiple dimensions.

The sled is a little extra trouble, but actually not that much - and the point is that the basic 90 deg chassis has potentially all sorts of uses when fitted with different work holding and securing bits. Even cutting for tenons on the table saw etc.

I can't imagine what odd angle mitre lock cutters might look like Alan - it'd take sitting down with a pencil and paper to figure it out. Perhaps two cutters one above the other, but in one piece or something like that?

I was pleasantly surprised Mel at just how tear out resistant the shaper cutter is compared to a router table - the ply i was working in above was actually pretty poor quality Eastern stuff. It was able to cross grain cut right to a sharp knife edge without problems.

Just as well as it turned out. I started out thinking (but not knowing if it was important or not) that it might be an option to get the cutter set up with zero clearance in both the ply insert ring i made for the shaper table, and the ply false fence. It's possible, but difficult - it requires sneaking up to and arriving at the horizontal and vertical settings from one side only. Overshoot and have to go back and you immediately have clearance. Luckily it (or at least the minor amount i ended up with) didn't seem to matter a lot...