To summarize: Anything under .05 degrees of error is super square. If you get an error of .02 over a fence greater than 24” long you are super square. To calculate how many degrees your 5 cut method answers equates to use this formula (or Google rise over run degree calculator): Degrees= tan(-1) X (rise/run)
To be smart enough to plan a launch trajectory to some far away rock in space is one thing. To share that knowledge with the rest of us, whether you're like me and not fully understand it, or the next time these methods are used to build a sled with the tolerances mentioned, is a whole different level of mentoring. To provide that knowledge in writing for free to beginners and pros is a level of generosity I've never come across in the 2 dozen or so woodworking channels that I watch regularly. Very admirable and appreciated. Now I just have to watch this a few more times to fully digest it.
I recently made a 45° miter sled for boxes. Doing the 5 cut method on a sled when the blade is at an angle gets tricky. The bottom of the sled looks like Swiss cheese as it took 10 tries before reaching a number I was satisfied with. The calc I used suggested .001 as the target, when I finally hit .002 I stopped. Today I'm building a 60° miter sled for my hexagon boxes. After watching this, I'll relax my hunt for accuracy! Thank you, Johnathan.
Check out William Ng's video on how to make a 45° miter sled. He claims you can get it in two tries with two cuts each try. You can, IF you use good quality wood that doesn't leave raggedy corners.
That was my problem on the first test. After a couple more failed tests. I realized the issue was how you measure the length of the off cut - I had been measuring the long side (the top of the cut), but I should have been measuring the short side (the bottom of the cut, against the base of the sled).
I think there are a couple things to remember: We're working with wood, not precision steels. A joiner told me years ago...it's a natural material and it'll do what it wants to anyway. I think the point is, no matter what you are making, you have to allow for all sorts of movement. Also always told to measure the parallel accuracy of the grooves to the blade on a table saw from the same tooth - but you might choose a mis-ground/distorted tooth and although it cuts on the feed side, it can also cut on the opposite side if everything is slightly out of alignment, presumably giving a slightly wider kerf (although my own saw seems to cut a strange very shallow curve?). I have found the best way to achieve alignment accuracy is to use the longest "true" straight edge against all of the reference surfaces, whether you measure with digital test indicators or your thumb!
I've been in the Quality industry for 13 years checking machined parts to, in some cases, .0002 in. This video checks out. Great job explaining this. Because I worked in the industry, I was hunting that accuracy in my wood working. Because it is wood, I could never maintain that accuracy. I could measure something like thickness to a thousandths of an inch, but give that freshly cut side a little time, and the whole piece would be ever so warped, bent, or changed in length. I do tolerance studies of the drawings we make for the shop now as a drafter. I think in terms of what is acceptable to the end assembly all the time now. Super tiny nip pick... Your joe block that you use to check the calibration of your calipers and mics is not perfectly 1in. However, for the level of accuracy needed this video it is way more than close enough (tolerance on a joe block of 1in is something like ±.000005 at 68° F). Overall I freaking love this video. I can nerd out all day about math!
I'm a Quality Engineer & agree 100% I often make my own study coupons & machine 1 off parts/fixtures. I was lucky to be able to call it "good enough" when it comes to woodworking projects early on. I actually do a lot less measuring woodworking... fit/form/function is more applicable I think. The only time in woodworking I try to hit accuracy to a few tenths is setting up a piece of equipment. I did, however, buy a some cheaper items for construction (framing, landscaping, etc) I used our Vision System to verify accuracy. My aluminum $9 speed square was something like 90.075°, 45.something° & straightness was less than .003. I did not expect that at all.
9:56… I’m a fan of plastic runners… I buy cheap plastic cutting boards from Amazon and cut them up for my runners. Can’t remember the exact type of plastic to use but there supposedly is one you should use (but it’s the plastic that cutting boards are made from anyways). 2 things for my personal shop set up. First is my shop is heated for winter (I’m in Minnesota), but I often am to cheap to fill the 500 gallon propane tank for it lol, and my shop is not cooked at all in the summer. So I have a GIANT delta of both temps and humidity during the different seasons of the year. Second is that my table saw miter slots are slightly different than each other (which I would think is fairly common), so I need two different sized runners to fit the specific miter slot. But this is not a big deal at all. Basically the plastic runners don’t move with the seasons AT ALL. My first few sleds have 1/4 sawn hardwood. But I must have made them in the winter, because come summer time there’s no chance those sleds fit my saw anymore lol.
Great video. I watch people chase these incredibly small numbers and think we are not machinists, we are woodworkers. My first sled had soooo many screw holes in it from adjustments. My last sled I adjusted once and called it good. I measure corner to corner and it is always dead on. Great advise, stop chasing numbers, make sawdust.
Great video - especially as my sled is horribly wobbly and I need to build a new one. Square is a pipe dream with my current sled. Can you please reshoot the entire 14:21 video again....but in metric 🙂
Another awesome video. I'm getting ready to make my first sled, so this was very timely. I was concerned about getting my sled spot on, and knowing what spot on was so now I know. Your tip on how to make the runners tight was fantastic!!
Great topic. I once used a Grizzly contractor tablesaw and just dragging it a foot across the shop would throw it out for fine work. I was able to dial it in and make a 3x4 ft. picture frame that was 3 inch wide oak and had tight miters. I eschew the use of filler to hide miter errors. I did not realize you don't do sponsors. Just ordered a few items to help support you. More of this stuff, please.
The first thing to check for is that the blade is parallel to the miter slot. Don't ask how I know. The second is to use a 1/4 inch bolt in a 3/8 hole opposite the pivot screw to hold the fence in place and avoid turning base into swiss cheese. Just set the fence to approximately the right spot and drill a 1/4 hole through the base and fence. Then drill a 3/8 hole just in the fence. Use a bolt or carriage bolt inserted through the base connected to a large washer and nut on top. Make your adjustments by loosening and tightening the nut. Much faster than screwing around.
Thanks for this Jonathan, your comments on the human induced errors were gold. I will be building a sled and I'm looking for 'practical accuracy', not perfection. Those folks looking for machinist perfection might do well to remember that wood isn't steel, it's squishy and it moves a wee bit. Your comments on the sled runners were interesting. I think I'll try using some pieces of phenolic offcuts I found for that.
I started using a dial indicator and a long square. Hold the square against the fence of sled, mount dial indicator to table and touch it to square along the cutting edge and move sled back and forth. Adjust fence until most run out is gone. Done
I love this Channel bc when I build my next sled I’ll play this video once more to go over the math. I can’t thank you enough for putting out informative content. It’s not flashy but it helps me make a lot of money bc accuracy saves me on the back end. Thanks again.
Refreshing to hear a reality-based approach to what accuracy ACTUALLY needs to be so that people can stop chasing their tails pursuing unnecessary "perfection" and more time doing actual woodworking projects. Thanks, JKM!
Thanks! I’ve been chasing perfection on my first shed for the last 3 weeks! Yes, I’ve had to put it aside a couple of times. The last time I walked away was specifically because of the “screw bump” I was really getting frustrated because I’m dialing in my fence and then my error from .0095 to .015! I don’t think my fence is flat enough so I think I’m going to rebuild that and start over. Yes, the blue tape works - but I needed to make sure I sanded paste wax off. Glad to hear that it shouldn’t slide like a bobsled as well!
The timing was perfect. Literally just yesterday, was trying to dial in a new sled. The original five cut method video, he got it to within .001 and that's what I thought I needed to shoot for. Nice to know that there's a big more play to work with.
The New Yankee Workshop has a video where Norm was making shop helpers, one of which was a panel sled. Even though a panel sled is a bit different than a crosscut, their end goal is basically the same. All he used was a carpenter's framing square, and from what he said, and from what I could see, it worked perfectly. When I made mine, I used a Woodpeckers framing square. It was as accurate as it ever needed to be. Any panel or single board I cut, was right on the money. But, years before I made that sled, I had purchased and still use the Incra 5000 which is excellent. Especially for angles. The only issue is, it seems to be getting heavier every year.
Good video as usual, just a couple of points to make. When you are making your five cuts use a piece of material that is as big as you can cut in the sled, it makes measuring the error more accurate. Sorry, using an 8in square on a sled that looks like you could have used a 20in square is nuts. Also, expressing error in degrees is a bit hard to understand more meaningful is what is the error over the largest piece you can cut. I've been making and using sleds for nearly 50 years and I think an hour or two spent finessing is well worth it. One additional tip is always make sleds with two runners where at all possible it removes the potential for racking. As for runners I have recently gone to using the shop bought adjustable variety, they are great, just be aware that if you have to remove slop from any sled it needs re calibrating. Although woodworking does require engineering levels of accuracy there is no reason not strive for perfection, as errors in a big project can compound themselves.
As for the runners play, I would use a few spring plungers along the length of the runner/runners to eliminate any play and to keep the runners moving smooth. 3/8 thick runner you could use 1/4-20 or 6 mm plungers. Thanks for the videos Jonathan.
I assume _fixed_ squares are available much cheaper at tighter tolerances than adjustable ones simply because the adjustment mechanism is the hardest place to minimize slop.
Great video! I lost my mind on the first sled I made 10 years ago. Solid tips on the runners, blue tap, or packaging work wonders! A New saw means a new sled..I sure don't feel like I have to chase a Nat's hair as I did on the first 😮. Damn mathematics. # I still have a feeler gauge that size!😂
Excellent approach to this topic. From and engineers perspective it is excellent. Thank you, I have a great deal of respect for you and your approach to many things.
i really enjoy the videos, however, I'd like to see you build a corner china hutch. I've been looking for plans online, with very few results. I love your approach to woodworking, and machine setup. Thanks for all you do Jonathan.
I used a dial indicator to do my 5 cut method and one issue I had was not pulling the dial indicator back and "resetting" the measurement. There was a surprising amount of play in the measurement by just leaving the indicator touching the fence and moving it. I chased that issue for a couple hours until I realized what was happening.
Good information, I'm with u on the plywood runners. Although I di think its worth getting aftermarket miter bars for adjustability and the ability to reuse them.
Great piece! I’m wondering if there is a Part II coming, in which you investigate the influence of the fact that we’re working with materials that in nearly all cases are subject to all sorts of environmental and other influences? For example, some materials were better-dried than others when cut or manufactured; some materials will be drier than others when we start cutting and assembling them; some materials are inherently less subject to environmental effects than others; and (this is a guess/assumption) some wood grows in ways that result in greater or lesser amounts of internal tension that can release and cause dimension changes over time, either while stored before use or after cutting/assembly. I’m wondering if the results of variables like these are, or can be in some cases, so great that they make chasing dimensional accuracy that has to be measured in single thousandths of an inch (or less 😅) a challenge that may be diverting and satisfying but honestly isn’t all that important? 🎉
Great information Jonathan, thank you. What do you think if one uses a 24 inch piece of extruded aluminum with one end swiveling relative to the base and the other adjusted with a micro adjustment set screw like the one in the Rockler sled to fine tune it accurately and quickly?
Thanks very much Jonathan. What do you think about the metal runners with set screws to adjust for fit? I've seen them on Amazon for not too much. I whole heartedly trust your woodworking opinions! Any input will be greatly appreciated. 😊😊😊❤❤❤
What do you think about single vs dual runner sleds? And, for dual-runner, do you think the runners should actually be the full width of the slots or should they be deliberately cut narrow and _just_ be snug on the inside edges (like a train, like you said)?
I've done the same with chasing error. In my opinion, the fence should aluminium extrusion with a thin mdf layer for zero clearance, and never be screwed down, it should be adjustable with a bolt just like with most metal working machines, and dialled in often as pieces of the jig wear. At the end of the day though, as long as the sled is made from wood, it's going to have moved the next day. Not to mention the human error involved in using the sled itself, for example like with a mitre saw, its easy to unconsciously be applying too much pressure in one direction.
I don't remember what the error numbers I was getting were, but I did at least twelve adjustments when trying to get my sled "perfect". I just gave up and started using it. Never had any problems because of it. I'm pretty sure I was well below your 0.004.
AS regards runners- Most commercial ones have set screws or discs to make up for slop. Problem is they are spaced too far apart. I make mine out of ipe(quite hard and wear well) and and place every 3-4 inches a brass headed screw to adjust for slop. They stay true for the entire length of the runner and easily adjustable.
I’m thinking that if the sled runner (s) are not tight the sled will “skew” - the front of the blade will cut a particular width until the back of the starts to cut the opposite side of the kerf making it wider than intended? I try to push my sled directly over the runners thinking that will minimize skew or at least make it consistent - Somewhat different than out of square but still gives inaccurate cuts
I’m a type A weekend warrior and found my sled splaying a tiny bit when I unscrewed the fence to make the adjustment. I moved my fence the amount dictated by the 5 cut method, so sometimes it was worse than before. I’m down to 0.07 over 24” and can’t see any light, so after 9 tries I’m calling it good enough. Chasing dead nuts is too aggravating.
My guess is you’re not accounting for which side of your blade you’re cutting on. In our five cut method calculator we have a drop down menu for which side your pivot screw is on. If you miss that part you can move your sled the wrong direction.
If you can’t see it, you can’t see it…although as a footnote, I think my fingertips are probably more sensitive than my eyes. In other words, I think I can feel dimensional differences that I can’t see…though I also wonder sometimes if not being really careful about the exact angle I’m looking at something obscures something that would be visible if viewed exactly on-axis. In any case, I do think there are cases where we can measure differences that don’t make any difference in real life.
@@katzmosestools Thanks for your response! I love your channel. I thought about the pivot screw, too, but I have it correct. I cut on the left side of the blade, but accounted for that, too. I was essentially moving it forward and backward the same amount each time. I chalk it up to crappy plywood.
This is a great video, but I have a different view on a couple of your points. 1. For most angles, it is impractical to focus on small angular errors. in engineering drawings, Geometric Dimensioning & Tolerancing detfines an angular error as a linear measurement of the deviation from the ideal angle. You were doing that, but you kept throwing in a meaningless angle value. 2. The 5 cut method does not increase the error after each cut. The angle remains the same. You measure the error over the perimeter of the test board. It still multiplies the error to improve the resolution.
Thank you. There is such a measure as "good enough". Yes, I know what precision is. Father and sister in law worked in angstroms which makes your caliper insensitive by comparison.
Note about accuracy. When talking about the Mitutoyo Micrometers... their accuracy is not .00005 or 5 decimal places, that is their "resolution". The actual accuracy will be closer to .0001 - those particular mics are more "accurate" than that (roughly ±.00004 or .00008 total tol), but can't display it (due to the .00005 resolution) in imperial (inches)... so rounding up to .0001 is your accuracy.
Yes I misspoke. That is so accurate I don’t think most people can picture what that actually means unless you’ve seen what that looks like on something like a scientific grade granite plate etc.
Another thing to account for is that the sled can easily distort by a few thou and even the alignment of the wood cut piece against the wood sled can vary with holding pressure. Wood is squishy :)
I got my sled accurate to .009” over 20”. Not bad for a $99 Canadian tire table saw. It’s more accurate than my grandpa’s set up and he made beautiful stuff
Using calipers--I was taught to measure 3 times. It could reveal technique errors and if all are close -take an average. If one is markedly different-figure out what you're doing incorrectly.
I recently made a crosscut sled and I was wondering if anyone has ever had kickback issues during using a sled. I have a new Skilsaw job site tablesaw and got some nasty kickback using it the first time. It was so bad that I’m convinced my arbor is bent. Thanks, asking for a friend.
I used to worry about this level of accuracy, but now I just remind myself that wood is a hydrophilic, inconsistent, and often volatile organic material and that’s why I love it. I’m not a machinist. I can barely afford to be a woodworker.
You can everything you want but it comes down to execution, some guys have to take 30 shots for a dime size group at a quarter mile and then someone can pick up same rifle and do it in 10 shots- you know that guy at the bar hasn't shot pool all night gets off the stool and runs the table 3 times and then sits back down like it didn't matter but you've practiced all night - sometimes people just can nail it 👌
I love buying stuff from you guys but you're constantly out of stuff. I've been trying to get one of your brass hammers and they are continually out of stock. Can I get on a waiting list or something to get one? Thanks
This is no doubt a stupid question: If after doing the 5 cuts you are off and need to do it again, your board is not square. Can you do it now with the out of square board and get an accurate result? Also, do you go clock wise or counter clock wise? Told you these are stupid questions no doubt. Thank you for all the informative videos and comments.
Screws should not be used for locating parts. @RexKreuger has a really nice solution to getting a fence really square in his advanced shooting board video. Basically, you fasten the back half the fence rock solid to the board/sleed as square as you can, then you bolt the front half to it and shim it to sufficient squareness.
I like this and I will make a decent effort to be accurate, but my motto is and I will have a plaque on my wall that will say, “We are not building a watch”.
“Acceptable error” equates to distance. .004 is acceptable at 6 inches, not at 600 meters. Just something for your followers to keep in reference. Everything is subjective, don’t let them fool you.
The reference block is not exactly 1 inch wide. It is exactly 25.4 mm wide. Exactly. Can you imagine what a world it would have been if the inch was 25.0 mm instead? Harmony, peace, angels singing.
Useless unrelated fact: While it's written 'Mitutoyo', it's actually pronounced 'Mitsutoyo'. The difference in writing vs. pronunciation is because they used a romanization system (i.e. kunrei-shiki) that is more focused on showing morphological structures than accurate representation of sounds (which also happens to be the official romanization system of the Japanese government).
Just refurbish a couple antiques and you’ll quickly discover how to accommodate for errors. This fascination with precision is kind of a joke. When you study those beautifully crafted old pieces, that are incredibly inaccurate yet gorgeous and perfect ’to the eye’. Are we craftsman or machinists?
Yes I agree. Perfectly square is great if achieved, but it can be elusive. Having a perfect sled would be nice, but how big of a sled can you have. Even if you had one 2’ across, if it was out a 1/32” or even a 1/16” in that 2’ are you going to really see it. A sled is just a more controllable and safer miter gauge. Think about how wide of pieces you crosscut with a miter gauge. Most of the time it’s probably less than 10”. And I don’t want to spend half the time checking the sled for square before I use it every time. And as with the antique furniture you mentioned, wood does move, so even if you cut it perfectly, it may not stay that way. I think instead of spending all that time making and fussing over over a sled that I apparently have to treat with kid gloves so it doesn’t maybe get messed up, I’d go with either the Dubby sleds, the Incra miter gauge/sled. They are basically a miter sled that cut 90 degrees and miters. The Dubby has right and left sleds, both with adjustable fences. Buy one or both and have a place for the cutoff to land. The Incra has a separate platform for the other side of the blade. With either, if they get off a bit it’s easier to adjust them. And finally, no matter how square your sled or whatever is, if you don’t measure right, mark the workpiece right, run it through the saw right, it ain’t gonna be right.
Sort of. You ever tried to glue up a box where every joint is half a degree out of square. That is maddening. In this video I think I’m agreeing with you to a point but that you should get things under .05 degrees to avoid seeing errors in your work. Maybe you can’t see it but spending half your time trying to fix wobbles and poorly fit drawers is pretty frustrating.
Five cut is a waste of time. I use several tools I trust when building fine furniture. Those same tools damn well better be capable of building a crosscut sled or what’s the point?
*Support what we do at* KMTools.com
*5 Cut Method Calculator* kmtools.com/pages/five-cut-method-calculator
*KM Universal Box Joint Jig* kmtools.com/products/km-tools-universal-box-joint-jig-for-table-saws
*Make an Ultra Square Crosscut Sled* ua-cam.com/video/OigXgK0Vj4E/v-deo.html
*The Katz-Moses Woodworkers with Disabilities Fund ABOUT* kmtools.com/pages/kmww-charity
To summarize: Anything under .05 degrees of error is super square. If you get an error of .02 over a fence greater than 24” long you are super square. To calculate how many degrees your 5 cut method answers equates to use this formula (or Google rise over run degree calculator):
Degrees= tan(-1) X (rise/run)
Another home run. This channel along with Stumpy nubs has helped me tremendously with my woodworking.
My absolute pleasure
To be smart enough to plan a launch trajectory to some far away rock in space is one thing. To share that knowledge with the rest of us, whether you're like me and not fully understand it, or the next time these methods are used to build a sled with the tolerances mentioned, is a whole different level of mentoring. To provide that knowledge in writing for free to beginners and pros is a level of generosity I've never come across in the 2 dozen or so woodworking channels that I watch regularly. Very admirable and appreciated.
Now I just have to watch this a few more times to fully digest it.
Thanks Jonathan, appreciate your time and effort to help the rest of us.
I recently made a 45° miter sled for boxes. Doing the 5 cut method on a sled when the blade is at an angle gets tricky. The bottom of the sled looks like Swiss cheese as it took 10 tries before reaching a number I was satisfied with. The calc I used suggested .001 as the target, when I finally hit .002 I stopped.
Today I'm building a 60° miter sled for my hexagon boxes. After watching this, I'll relax my hunt for accuracy! Thank you, Johnathan.
Check out William Ng's video on how to make a 45° miter sled. He claims you can get it in two tries with two cuts each try. You can, IF you use good quality wood that doesn't leave raggedy corners.
That was my problem on the first test. After a couple more failed tests. I realized the issue was how you measure the length of the off cut - I had been measuring the long side (the top of the cut), but I should have been measuring the short side (the bottom of the cut, against the base of the sled).
I think there are a couple things to remember:
We're working with wood, not precision steels. A joiner told me years ago...it's a natural material and it'll do what it wants to anyway. I think the point is, no matter what you are making, you have to allow for all sorts of movement.
Also always told to measure the parallel accuracy of the grooves to the blade on a table saw from the same tooth - but you might choose a mis-ground/distorted tooth and although it cuts on the feed side, it can also cut on the opposite side if everything is slightly out of alignment, presumably giving a slightly wider kerf (although my own saw seems to cut a strange very shallow curve?). I have found the best way to achieve alignment accuracy is to use the longest "true" straight edge against all of the reference surfaces, whether you measure with digital test indicators or your thumb!
I've been in the Quality industry for 13 years checking machined parts to, in some cases, .0002 in. This video checks out. Great job explaining this. Because I worked in the industry, I was hunting that accuracy in my wood working. Because it is wood, I could never maintain that accuracy. I could measure something like thickness to a thousandths of an inch, but give that freshly cut side a little time, and the whole piece would be ever so warped, bent, or changed in length. I do tolerance studies of the drawings we make for the shop now as a drafter. I think in terms of what is acceptable to the end assembly all the time now. Super tiny nip pick... Your joe block that you use to check the calibration of your calipers and mics is not perfectly 1in. However, for the level of accuracy needed this video it is way more than close enough (tolerance on a joe block of 1in is something like ±.000005 at 68° F). Overall I freaking love this video. I can nerd out all day about math!
I'm a Quality Engineer & agree 100%
I often make my own study coupons & machine 1 off parts/fixtures.
I was lucky to be able to call it "good enough" when it comes to woodworking projects early on. I actually do a lot less measuring woodworking... fit/form/function is more applicable I think.
The only time in woodworking I try to hit accuracy to a few tenths is setting up a piece of equipment.
I did, however, buy a some cheaper items for construction (framing, landscaping, etc) I used our Vision System to verify accuracy. My aluminum $9 speed square was something like 90.075°, 45.something° & straightness was less than .003. I did not expect that at all.
9:56… I’m a fan of plastic runners… I buy cheap plastic cutting boards from Amazon and cut them up for my runners. Can’t remember the exact type of plastic to use but there supposedly is one you should use (but it’s the plastic that cutting boards are made from anyways).
2 things for my personal shop set up.
First is my shop is heated for winter (I’m in Minnesota), but I often am to cheap to fill the 500 gallon propane tank for it lol, and my shop is not cooked at all in the summer. So I have a GIANT delta of both temps and humidity during the different seasons of the year.
Second is that my table saw miter slots are slightly different than each other (which I would think is fairly common), so I need two different sized runners to fit the specific miter slot. But this is not a big deal at all.
Basically the plastic runners don’t move with the seasons AT ALL.
My first few sleds have 1/4 sawn hardwood. But I must have made them in the winter, because come summer time there’s no chance those sleds fit my saw anymore lol.
Great video. I watch people chase these incredibly small numbers and think we are not machinists, we are woodworkers. My first sled had soooo many screw holes in it from adjustments. My last sled I adjusted once and called it good. I measure corner to corner and it is always dead on. Great advise, stop chasing numbers, make sawdust.
Great video - especially as my sled is horribly wobbly and I need to build a new one. Square is a pipe dream with my current sled.
Can you please reshoot the entire 14:21 video again....but in metric 🙂
Another awesome video. I'm getting ready to make my first sled, so this was very timely. I was concerned about getting my sled spot on, and knowing what spot on was so now I know. Your tip on how to make the runners tight was fantastic!!
Thank you my friend!
Great topic. I once used a Grizzly contractor tablesaw and just dragging it a foot across the shop would throw it out for fine work. I was able to dial it in and make a 3x4 ft. picture frame that was 3 inch wide oak and had tight miters. I eschew the use of filler to hide miter errors. I did not realize you don't do sponsors. Just ordered a few items to help support you. More of this stuff, please.
The first thing to check for is that the blade is parallel to the miter slot. Don't ask how I know. The second is to use a 1/4 inch bolt in a 3/8 hole opposite the pivot screw to hold the fence in place and avoid turning base into swiss cheese. Just set the fence to approximately the right spot and drill a 1/4 hole through the base and fence. Then drill a 3/8 hole just in the fence. Use a bolt or carriage bolt inserted through the base connected to a large washer and nut on top. Make your adjustments by loosening and tightening the nut. Much faster than screwing around.
Another gold mine here! Thanks Jonathan for the great info. Seeing that 5 cut method drawn out like that really helped seeing how it works.
Thanks for this Jonathan, your comments on the human induced errors were gold. I will be building a sled and I'm looking for 'practical accuracy', not perfection. Those folks looking for machinist perfection might do well to remember that wood isn't steel, it's squishy and it moves a wee bit. Your comments on the sled runners were interesting. I think I'll try using some pieces of phenolic offcuts I found for that.
The tip about wedging the runners before attaching just solved a two-year-old problem of mine. Thanks.
I started using a dial indicator and a long square. Hold the square against the fence of sled, mount dial indicator to table and touch it to square along the cutting edge and move sled back and forth. Adjust fence until most run out is gone. Done
best video on the subject ever---no ifs-ands-or buts about it thanx rick
I love this Channel bc when I build my next sled I’ll play this video once more to go over the math. I can’t thank you enough for putting out informative content. It’s not flashy but it helps me make a lot of money bc accuracy saves me on the back end. Thanks again.
“It’s not flashy….” might be one of my favorite comments ever. So true and the whole basis behind the “no bs woodworking” series.
Refreshing to hear a reality-based approach to what accuracy ACTUALLY needs to be so that people can stop chasing their tails pursuing unnecessary "perfection" and more time doing actual woodworking projects. Thanks, JKM!
Thanks! I’ve been chasing perfection on my first shed for the last 3 weeks! Yes, I’ve had to put it aside a couple of times. The last time I walked away was specifically because of the “screw bump” I was really getting frustrated because I’m dialing in my fence and then my error from .0095 to .015! I don’t think my fence is flat enough so I think I’m going to rebuild that and start over. Yes, the blue tape works - but I needed to make sure I sanded paste wax off. Glad to hear that it shouldn’t slide like a bobsled as well!
The timing was perfect. Literally just yesterday, was trying to dial in a new sled. The original five cut method video, he got it to within .001 and that's what I thought I needed to shoot for. Nice to know that there's a big more play to work with.
The New Yankee Workshop has a video where Norm was making shop helpers, one of which was a panel sled. Even though a panel sled is a bit different than a crosscut, their end goal is basically the same. All he used was a carpenter's framing square, and from what he said, and from what I could see, it worked perfectly. When I made mine, I used a Woodpeckers framing square. It was as accurate as it ever needed to be. Any panel or single board I cut, was right on the money. But, years before I made that sled, I had purchased and still use the Incra 5000 which is excellent. Especially for angles. The only issue is, it seems to be getting heavier every year.
Thank you for all this valuable information!!
Good video as usual, just a couple of points to make. When you are making your five cuts use a piece of material that is as big as you can cut in the sled, it makes measuring the error more accurate. Sorry, using an 8in square on a sled that looks like you could have used a 20in square is nuts. Also, expressing error in degrees is a bit hard to understand more meaningful is what is the error over the largest piece you can cut.
I've been making and using sleds for nearly 50 years and I think an hour or two spent finessing is well worth it. One additional tip is always make sleds with two runners where at all possible it removes the potential for racking. As for runners I have recently gone to using the shop bought adjustable variety, they are great, just be aware that if you have to remove slop from any sled it needs re calibrating. Although woodworking does require engineering levels of accuracy there is no reason not strive for perfection, as errors in a big project can compound themselves.
As for the runners play, I would use a few spring plungers along the length of the runner/runners to eliminate any play and to keep the runners moving smooth. 3/8 thick runner you could use 1/4-20 or 6 mm plungers. Thanks for the videos Jonathan.
Thank you so much for all you help Jonathan your videos are so amazingly helpful to us new woodworkers. Keep up the great work.
I assume _fixed_ squares are available much cheaper at tighter tolerances than adjustable ones simply because the adjustment mechanism is the hardest place to minimize slop.
Great math and helps us improve our joinery!
Great video! I lost my mind on the first sled I made 10 years ago. Solid tips on the runners, blue tap, or packaging work wonders! A New saw means a new sled..I sure don't feel like I have to chase a Nat's hair as I did on the first 😮. Damn mathematics. # I still have a feeler gauge that size!😂
Excellent approach to this topic. From and engineers perspective it is excellent. Thank you, I have a great deal of respect for you and your approach to many things.
Awesome, Jonathan! Thanks for all the info! 😃
Stay safe there with your family! 🖖😊
Very good video had to watch a couple of times to understand it all. Thanks your are the man.
How about the alignment of the blade to the runners?
i really enjoy the videos, however, I'd like to see you build a corner china hutch. I've been looking for plans online, with very few results. I love your approach to woodworking, and machine setup. Thanks for all you do Jonathan.
I used a dial indicator to do my 5 cut method and one issue I had was not pulling the dial indicator back and "resetting" the measurement. There was a surprising amount of play in the measurement by just leaving the indicator touching the fence and moving it. I chased that issue for a couple hours until I realized what was happening.
Honestly though, how many times a day are you startled by that Mr Bean cutout?
Once or twice. It needs to get moved. My daughter used to be terrified of it
Seriously! My eyes went to that cutout before anything else in the video.
I didn't notice it. Now I can't unsee it. Thanks for that. 😬😉
rip danny
Good information, I'm with u on the plywood runners. Although I di think its worth getting aftermarket miter bars for adjustability and the ability to reuse them.
What is the pivot point? The screw on the left? And do you measure from the screw on the right to the pivot point or the right end of the sled fence?
A lot of great tips and tricks thanks mate
Man, I wish he had done this one a few years ago before I had to learn all these lessons the hard way!
how did you check the accuracy of those combination squares. or did you just go by the name?
Incredibly helpful!!
Well thought out and presented.
Great piece! I’m wondering if there is a Part II coming, in which you investigate the influence of the fact that we’re working with materials that in nearly all cases are subject to all sorts of environmental and other influences? For example, some materials were better-dried than others when cut or manufactured; some materials will be drier than others when we start cutting and assembling them; some materials are inherently less subject to environmental effects than others; and (this is a guess/assumption) some wood grows in ways that result in greater or lesser amounts of internal tension that can release and cause dimension changes over time, either while stored before use or after cutting/assembly. I’m wondering if the results of variables like these are, or can be in some cases, so great that they make chasing dimensional accuracy that has to be measured in single thousandths of an inch (or less 😅) a challenge that may be diverting and satisfying but honestly isn’t all that important? 🎉
Great information Jonathan, thank you. What do you think if one uses a 24 inch piece of extruded aluminum with one end swiveling relative to the base and the other adjusted with a micro adjustment set screw like the one in the Rockler sled to fine tune it accurately and quickly?
Thanks very much Jonathan. What do you think about the metal runners with set screws to adjust for fit? I've seen them on Amazon for not too much. I whole heartedly trust your woodworking opinions! Any input will be greatly appreciated. 😊😊😊❤❤❤
Great production values, logic irrefutable……and I was a precision machinist. BTW your router is pretty amazing, for the tariff.
What do you think about single vs dual runner sleds? And, for dual-runner, do you think the runners should actually be the full width of the slots or should they be deliberately cut narrow and _just_ be snug on the inside edges (like a train, like you said)?
I've done the same with chasing error. In my opinion, the fence should aluminium extrusion with a thin mdf layer for zero clearance, and never be screwed down, it should be adjustable with a bolt just like with most metal working machines, and dialled in often as pieces of the jig wear. At the end of the day though, as long as the sled is made from wood, it's going to have moved the next day. Not to mention the human error involved in using the sled itself, for example like with a mitre saw, its easy to unconsciously be applying too much pressure in one direction.
How do I determine the feeler gauge adjustment point on the fence? Or do I just pick a random point? Thanks!
At the end of the fence. That’s the distance between your pivot screw and adjustment point in the five cut calculations.
Thank you!!!@@katzmosestools
I don't remember what the error numbers I was getting were, but I did at least twelve adjustments when trying to get my sled "perfect". I just gave up and started using it. Never had any problems because of it. I'm pretty sure I was well below your 0.004.
AS regards runners- Most commercial ones have set screws or discs to make up for slop. Problem is they are spaced too far apart. I make mine out of ipe(quite hard and wear well) and and place every 3-4 inches a brass headed screw to adjust for slop. They stay true for the entire length of the runner and easily adjustable.
To fix sloppy runners, try UHWM tape instead of blue tape. Dual benefits. Also, UHWM tape comes in different thicknesses.
I’m thinking that if the sled runner (s) are not tight the sled will “skew” - the front of the blade will cut a particular width until the back of the starts to cut the opposite side of the kerf making it wider than intended?
I try to push my sled directly over the runners thinking that will minimize skew or at least make it consistent -
Somewhat different than out of square but still gives inaccurate cuts
Great information here. :) Enjoyed the video. Anyways, back to my🍦🍦👯♀️
How important is this? For cross cuts how often is that the finished surface?
I’m a type A weekend warrior and found my sled splaying a tiny bit when I unscrewed the fence to make the adjustment. I moved my fence the amount dictated by the 5 cut method, so sometimes it was worse than before. I’m down to 0.07 over 24” and can’t see any light, so after 9 tries I’m calling it good enough. Chasing dead nuts is too aggravating.
My guess is you’re not accounting for which side of your blade you’re cutting on. In our five cut method calculator we have a drop down menu for which side your pivot screw is on. If you miss that part you can move your sled the wrong direction.
If you can’t see it, you can’t see it…although as a footnote, I think my fingertips are probably more sensitive than my eyes. In other words, I think I can feel dimensional differences that I can’t see…though I also wonder sometimes if not being really careful about the exact angle I’m looking at something obscures something that would be visible if viewed exactly on-axis. In any case, I do think there are cases where we can measure differences that don’t make any difference in real life.
@@katzmosestools Thanks for your response! I love your channel. I thought about the pivot screw, too, but I have it correct. I cut on the left side of the blade, but accounted for that, too. I was essentially moving it forward and backward the same amount each time. I chalk it up to crappy plywood.
Good intel > Thanks!!
Ty for this video
This is a great video, but I have a different view on a couple of your points.
1. For most angles, it is impractical to focus on small angular errors. in engineering drawings, Geometric Dimensioning & Tolerancing detfines an angular error as a linear measurement of the deviation from the ideal angle. You were doing that, but you kept throwing in a meaningless angle value.
2. The 5 cut method does not increase the error after each cut. The angle remains the same. You measure the error over the perimeter of the test board. It still multiplies the error to improve the resolution.
Thank you. There is such a measure as "good enough". Yes, I know what precision is. Father and sister in law worked in angstroms which makes your caliper insensitive by comparison.
Note about accuracy.
When talking about the Mitutoyo Micrometers... their accuracy is not .00005 or 5 decimal places, that is their "resolution". The actual accuracy will be closer to .0001 - those particular mics are more "accurate" than that (roughly ±.00004 or .00008 total tol), but can't display it (due to the .00005 resolution) in imperial (inches)... so rounding up to .0001 is your accuracy.
Yes I misspoke. That is so accurate I don’t think most people can picture what that actually means unless you’ve seen what that looks like on something like a scientific grade granite plate etc.
Another thing to account for is that the sled can easily distort by a few thou and even the alignment of the wood cut piece against the wood sled can vary with holding pressure. Wood is squishy :)
Off topic, but are these white airducts made of plastic? Don't you have any problem with electric charge on it?
Ive realized this a while ago, I put my focus on perfect HDPE runners and I only use ½ phenolic birch ply now.
Yeah I ditched the hardwood a couple years ago. I should do more hdpe stuff.
@@katzmosestools the phenolic birch ply is a real game changer, down side it's super slippery
I got my sled accurate to .009” over 20”.
Not bad for a $99 Canadian tire table saw.
It’s more accurate than my grandpa’s set up and he made beautiful stuff
This video is one of those you come back to every few months to remind yourself you don’t have to be perfect lol
Using calipers--I was taught to measure 3 times. It could reveal technique errors and if all are close -take an average. If one is markedly different-figure out what you're doing incorrectly.
Just because you dont find it necessary to strive for perfection, does that make it wrong for those who do?
My OCD thanks you.
thanks
I recently made a crosscut sled and I was wondering if anyone has ever had kickback issues during using a sled. I have a new Skilsaw job site tablesaw and got some nasty kickback using it the first time. It was so bad that I’m convinced my arbor is bent. Thanks, asking for a friend.
I used to worry about this level of accuracy, but now I just remind myself that wood is a hydrophilic, inconsistent, and often volatile organic material and that’s why I love it.
I’m not a machinist. I can barely afford to be a woodworker.
You can everything you want but it comes down to execution, some guys have to take 30 shots for a dime size group at a quarter mile and then someone can pick up same rifle and do it in 10 shots- you know that guy at the bar hasn't shot pool all night gets off the stool and runs the table 3 times and then sits back down like it didn't matter but you've practiced all night - sometimes people just can nail it 👌
I love buying stuff from you guys but you're constantly out of stuff. I've been trying to get one of your brass hammers and they are continually out of stock. Can I get on a waiting list or something to get one? Thanks
Literally just put them on preorder today. They’ll ship in a week. Black Friday did so well that we’ve been playing catch up for a couple months
Are there not companies that make precison built crosscut sleds? Wouldn't that be easier than making one?
Awesome video with well thought out explanations. I'm going back to my Barbies and ice cream now.
This is no doubt a stupid question: If after doing the 5 cuts you are off and need to do it again, your board is not square. Can you do it now with the out of square board and get an accurate result? Also, do you go clock wise or counter clock wise? Told you these are stupid questions no doubt. Thank you for all the informative videos and comments.
Throwing some of hose numbers on the screen would have made it easier to follow, especially for the metric people
Screws should not be used for locating parts. @RexKreuger has a really nice solution to getting a fence really square in his advanced shooting board video. Basically, you fasten the back half the fence rock solid to the board/sleed as square as you can, then you bolt the front half to it and shim it to sufficient squareness.
Screw bulge! Wow. Was that with a pilot hole? Displace less wood.
I like this and I will make a decent effort to be accurate, but my motto is and I will have a plaque on my wall that will say, “We are not building a watch”.
“Acceptable error” equates to distance. .004 is acceptable at 6 inches, not at 600 meters. Just something for your followers to keep in reference. Everything is subjective, don’t let them fool you.
Thanks for being square. Now it’s Barbie ice cream time baby!
Where was this when I decided to make the GOAT sled. The underside looks like Swiss cheese!😂
Makes me feel better that mines ok thanks.
The reference block is not exactly 1 inch wide. It is exactly 25.4 mm wide. Exactly. Can you imagine what a world it would have been if the inch was 25.0 mm instead? Harmony, peace, angels singing.
❤
Those guys in the background are standing very very still.
The wood is not stable enough to bother about such errors. The wood changes shape with humidity.
Useless unrelated fact:
While it's written 'Mitutoyo', it's actually pronounced 'Mitsutoyo'.
The difference in writing vs. pronunciation is because they used a romanization system (i.e. kunrei-shiki) that is more focused on showing morphological structures than accurate representation of sounds (which also happens to be the official romanization system of the Japanese government).
Great fact!
When carpenter working as a machinist 😂
Just refurbish a couple antiques and you’ll quickly discover how to accommodate for errors. This fascination with precision is kind of a joke. When you study those beautifully crafted old pieces, that are incredibly inaccurate yet gorgeous and perfect ’to the eye’.
Are we craftsman or machinists?
Exactly!
Yes I agree. Perfectly square is great if achieved, but it can be elusive. Having a perfect sled would be nice, but how big of a sled can you have. Even if you had one 2’ across, if it was out a 1/32” or even a 1/16” in that 2’ are you going to really see it. A sled is just a more controllable and safer miter gauge. Think about how wide of pieces you crosscut with a miter gauge. Most of the time it’s probably less than 10”.
And I don’t want to spend half the time checking the sled for square before I use it every time. And as with the antique furniture you mentioned, wood does move, so even if you cut it perfectly, it may not stay that way. I think instead of spending all that time making and fussing over over a sled that I apparently have to treat with kid gloves so it doesn’t maybe get messed up, I’d go with either the Dubby sleds, the Incra miter gauge/sled. They are basically a miter sled that cut 90 degrees and miters. The Dubby has right and left sleds, both with adjustable fences. Buy one or both and have a place for the cutoff to land. The Incra has a separate platform for the other side of the blade. With either, if they get off a bit it’s easier to adjust them.
And finally, no matter how square your sled or whatever is, if you don’t measure right, mark the workpiece right, run it through the saw right, it ain’t gonna be right.
Sort of. You ever tried to glue up a box where every joint is half a degree out of square. That is maddening. In this video I think I’m agreeing with you to a point but that you should get things under .05 degrees to avoid seeing errors in your work. Maybe you can’t see it but spending half your time trying to fix wobbles and poorly fit drawers is pretty frustrating.
Just a real world reference, .004 is cheap copy paper thickness. Natural wood movement is more.
How square is square? Square is square. Thats how square square is. Square.
Five cut is a waste of time. I use several tools I trust when building fine furniture. Those same tools damn well better be capable of building a crosscut sled or what’s the point?
"How square is square?" Well, it's either square or it's not square. If it looks square, it probably is good enough.
I was told there would be no math.