Unlock Incredible Joinery With Your CNC Router!

Gaze Custom Creations is supported by Timber King Sawmills. For more information on my 1220CRZ sawmill or many other fully hydraulic sawmills, go to TimberKing.com. If you have a CNC machine, I highly... Highly recommend trying to figure out a way to put a vertical table on it. My machine is an Avid CNC and these machines are basically an infinitely configurable platform. You can make them literally as long as you want. Pretty darn customizable in width. I'm sure there's some limitations there by the gantry size, but you can make them almost any size that you want. You can cut wood with these with a spindle. You can cut metal with these with this plasma. You can laser engrave wood. With the laser attachment you can do so much with this machine. This is my second Avid machine on my first machine I had a vertical table and now on this machine. I'm gonna add one right here right in the front So why would you want to add a vertical table the easiest and simplest answer and the most exciting answer for me is joinery This is one of my favorite pieces I've ever made. It's a pecan crotch slab coffee table Bar inclusion area and this this one's kind of like cheating with some black epoxy infill in the... when you when you get a slab like this all you have to do is flatten it and throw some finish on it and bam you have something that's an eye catcher and something that's gorgeous just because of what it is. There's so much character here but what makes it one of my favorite pieces is there's a concealed drawer down below there's mortise and tenon joinery there's no mechanical fasteners in this entire The entire piece, the... Legs are mortise and tenon directly into the bottom of the slab, so let's go down here and I'll show you The maybe we can see it. Maybe we can't I'm not sure They can so for the mortises. I simply cut three slots. There's the radius of the slots poking out And then on the tenons, on the end of these boards, on the vertical table, I cut three fingers, I guess you could say. So it looks like this going into the slab. And that's it. Just glue them together and you have a table. So on this particular piece, I didn't want the table to, or the slab to, warp over time or have a cup in it. So you can do c-channel, but what I chose This is a sliding dovetail. So here's the cap on the end and this dovetail slot was cut on the CNC. It goes all the way through to the other side. Well, not all the way through. It stops close to the end of the other side. And this rail right here is a sliding dovetail. It's glued on the far end and there's a little bit of a gap right here. So as this slab expands and contracts, which won't be anywhere near the size of this gap. The sliding dovetail will keep it nice and flat. Well, if I'm doing a sliding dovetail rail right here, it allows an opportunity for a concealed drawer that slides out nice and easily. And the joinery on this drawer is box joints. That was cut on the vertical table of the CNC as well. For the tenons the vertical table is used. First a shoulder is established and then a spiral profile cut is made for each of the three tenons. Rinse and repeat for all four legs. A test fit confirms the fit and you can see the mortises are a little longer than the tenons are wide. Now I did this just because it's easier to do and it also allows a little bit of room for the epoxy to squeeze out. You're of course not limited to mortise and tenon joints though. If you have a table Where you can clamp material vertically like this. What other joints could you cut on the end of this board? Well, you can cut dovetail joints. You can cut integral dowels to have through dowel holes where two pieces meet. You can cut a waterfall joint with some decorative inlay. You can do all kinds of fun stuff with the vertical table right here. Now, this machine is five feet wide. So, I think my vertical table is going to be about three inches less than that. I'm going to have an inch and a half gap on either side. That way I can use those gaps on the side to slide a clamp in and clamp close to the edge if necessary. On my last machine I had an array of dog holes so that way I can use the top one in the middle as a pivot and say put one over here in one of these dog holes and that would be a specified angle if I wanted to do an angled tenon. And I used that angle feature exactly zero times. So I'm not going to implement this, the angles on this particular one. Although if I needed to do an angle at any given time, I will have a vertical array of dog holes or a vertical column of dog holes. And I can simply use the flat work up here to cut a desired size pie shaped wedge or desired angle and clamp that angle. Right to the vertical table and then put my piece on top of that so there's no need for me to integrate angles into this What I do want to integrate, or what I do want to have, is a bunch of slots for clamps. I'm using clamps, no need to reinvent the wheel. Clamps are going to be fine to just, well, clamp your material all the way wherever you want, rather than relying on what's typically used in my workflow on the CNC, and that is these hold down clamps that go into T-track. For this, It's kind of sort of cumbersome and redundant because A lot of the vertical stuff that I want to cut is thicker than, say, 3 quarters of an inch, which is where you would kind of sort of be the majority of time on the horizontal table. So if I'm clamping in a 4x4 beam right here, then those types of clamps won't work. So no need to reinvent the wheel. I'm using F-style clamps. I need to have a lot of adjustments built into this design for some very fine tuning. There's a lot of angles that we have to make sure meet up perfectly for the CNC to be precise. So as the machine is built, it is calibrated so that way it cuts square. The gantry is squared up so that way after the machine is homed, we know that we're... We are cutting exactly perpendicular in the X direction as opposed to the Y direction. Well, I need to have that exact perpendicular from the horizontal surface to the vertical surface. So I need to be able to adjust the bottom in or out, adjust the top in or out to calibrate that. I also need to be able to calibrate the right side in or out versus the left side in or out to be exactly parallel with the X direction. And I also need to make sure that the top of the vertical table is exactly parallel with the x-axis travel, so the specified z-height. So if I drop my spindle over here and jog it all the way over to here, the top of the spoilboard needs to be exactly flush with that plane. So a lot of adjustment needs to be built in and the way I'm going to get this adjustment is with some I'm going to have some T-nuts and... Set screws. So these are three quarter inch long set screws, quarter 20 and then a quarter 20 T-nut. That'll give me a lot of adjustment in many directions. And I believe I haven't thought out in my head, I haven't designed it quite yet, to have four on each one of these beams. I also need to make sure that each one of these beams is the exact height. Left and right, so that way all the stuff lines up. And I also need to have a way to clamp this to the machine, to secure this to the machine. There's a few different ways that you can go about this. And one of the ways is these roll-in t-nuts that come with the machine. So this machine is basically a bunch of extruded aluminum bolted together with roll-in t-nuts and bolts of various different sizes and types. Here's one right here. So there's those same two roll-in t-nuts and two bolts right here allow you to mount anything to the rail. I thought about and I initially wanted to use this setup to bolt it to the machine but the downside of that is then I have a bolt sticking out of the front of the vertical table and that may or may not get in the way of actually putting material uh well the full width if I was ever to use the full width or maybe uh from one side to the next bolt you know it basically it's limiting So I'm not going to go that route. And instead, for this top rail, I'm actually going to drill through the rail. Just this top slot right here, four times, one, two, three, four, and that'll allow me to get the top secured exactly where I want to, really close to where the cutting is going to happen, and then for every one of these securing points all the way along these bottom two rails, I'm going to put the hole in the vertical table slightly above it, and then run a bolt through and have a piece on the back to allow me to tighten that bolt. And clamp it down on the front side of the bolt. I'm going to have a T nut on the well on the front side of the table. I'll have a T nut and the head of the bolt will be on the back of the table or the inside of the table. So that way I can clamp it down. And if the bolt is sticking through the T nut too far, I can add washers to the other side. So that way nothing is sticking out the front and I should have a nice flat reference surface all the way across. So that I have the most amount of options. Now on my horizontal table, I use these bench dog, well dog holes and all of these bench dogs to reference and to have repeatability. To know exactly where my material is when I start my job. And I want to have that same feature when I make the vertical table. So now I know exactly where this corner is in my software and I can start my job from from the exact same spot every single time. I want to have a bunch of these dogs in a vertical column, a few different vertical columns, so I know exactly where the new corner is up against the dog and up against the front of the table right there. Vertical, let's see, let's see, bench dogs, slots, I think that's pretty much it. So now I need to jump into the computer and take all of my measurements with me and get this designed out real quick. The first tool path is a scoring tool path. So this goes in by 0.1 inch all the way around the perimeter and this is just to establish a very crisp and perfect shoulder with no tear out. This is a spiral down climb cut. Then the actual pocketing toolpath that creates the shape of the tenon which is a dog bone shape. This is just an offset profile. So I took the profile of the tenon itself and then offset it large enough to cover the rest of the material. So in one sense it is kind of an inefficient toolpath because there's a lot of wasted air cutting around the side of it. But on the other side on the other side of that coin I guess is the fact that this is just a prototype kind of a project. mass producing these. If I was mass producing it, I think I would take a little bit of time to really optimize this particular toolpath to save this, to cut this time down quite a bit. The tenon creation for the three and one eighth of an inch tenons, the entire toolpath took exactly six minutes. And for the three and a half or 3.625 inch long tenons, that took exactly six All of the half-inch diameter work is done and That bit I know is is it's really sharp. I used that half inch diameter bit to cut the slots and the spoil board for the t-nuts or for t-track rather as well as all the dog holes and flawless results. On this MDF not so flawless results. Got a lot of fuzzies everywhere and that's because this MDF just doesn't feel like that good of quality. It's the first time I've used this product. Normally I drive 45 minutes one way to Tupelo, Mississippi to get my MDF from Home Depot. But I found this 15 minutes away for $10 cheaper a sheet and I thought you know what MDF is MDF doesn't matter. Well my opinion on that has changed a little bit. This just doesn't feel like it's as compressed as much as the stuff that I'm used to getting I guess. This feels more like powder whereas the top layer of the other stuff feels more like a pressed layer of it's not resin but pressed layer of resin, essentially. It feels harder. I think this is going to be just fine for what I'm doing with it, but because this is a little bit lesser quality and I noticed it after having it in the shop, I stayed away from the perimeter quite a bit. I'm gonna have a decent amount of waste all the way around these two sides, or three sides over here, and that's because this front edge was sticking out in the sun when I had the door open and I could see it, visibly see it, swelling up right at the edges. So you know what? Forget about the inconsistency at the edges and just cut it on the inside. Anyway, all the half inch diameter cuts are done. I'm going to switch over to a quarter inch diameter bit, cut this out, and then repeat the process for the second layer. This MDF, this MDF came from Home Depot 45 minutes away. This MDF came from a local lumber yard 12 minutes away. 51 or 52 dollars a sheet, 41 dollars a sheet. So, I thought, you know what, let's save a little bit of time. I just found this particular source. I'll save some time, I'll save some money, and I'll save some fuel, which is money, to get the cheaper stuff, and everything will be good because it's all MDF, right? Well, it was on this day that I realized MDF is not all created equally. This stuff, I mean, visibly it looks different, but this stuff just feels more porous than the stuff that I'm used to working with. And after cutting it a little bit it does feel a little bit more porous especially in the cuts. And these cuts I'm not getting the best results from. Now I know that bit is nice and sharp because I used it to make all the dog holes on this on the spoilboard as well as the slots for the t-track and it was just flawless cuts. So getting some fuzzy results which are less than desirable for what but for what this is going to be used for. I don't think it's that big of a deal. Got all the dog holes cut, which is more dog holes than I need to actually work with it. But I wanted a bunch of extra holes available so I could store all the extra bench dogs that I have in my way over there on the cart. And then I've got my counterbores for the t-nuts that will be used to secure this to the machine, slots for clamps, and I need two of these layers. But before I can If I can... swap out for the other sheet, I need to switch over to a quarter inch end mill and cut this actual perimeter out. Oh and also all of the quarter 20 mounting holes, I've got to get those cut as well. But yeah, not all MDF is created equally. This will be fine for what it's used for though. The bench dogs that I'm using, the actual shaft of the bench dog that goes in the hole is around an inch long which is convenient because I can use that or those to perfectly align these two panels before securing them together. It goes through the top layer and the bottom layer so with enough of these installed these two panels are perfectly aligned and on the back side that's where the t-nuts will be installed for which is up in this orientation. The t-nuts for the set screws Can be installed and And I didn't worry about a counterbore on the back side because the set screw itself is what's going to be interacting with the machine rails. So it really doesn't matter if there's a counterbore or not. And to get these two panels locked together I'm not using glue. I just didn't want to have a bunch of squeeze out and all these dog holes and all the clamp holes or clamp slots. All right. Squeeze out just sounds like a nightmare here. And then also I don't want to have a situation where the glue is applied a little bit more liberally in one area and not in the other. And then you kind of modify or change the thickness of the panel. Although we're talking thousands of an inch, it's not that big of a deal for a size this big. But the squeeze out is really what wanted me to not use glue here. So I'm using screws and anytime you pass a screw from one layer to the next below it, You risk the the chance of as the screw goes through the back side of the first layer it mushrooms out and then as it enters the first uh the top layer of the second piece it can mushroom up and that little mushroom can cause an inconsistency in thickness or spread these two panels apart. So I'm pre-drilling all of my holes and then using my body weight close to the hole while I drive the screw and that'll keep the panels nice and tight together while they're locked in place by one or 35 or 40 screws with everything locked together the bench dogs can be removed and on the front side all we get is the t-nuts in the counterbores now these t-nuts did have to be counter bored or i do want them to be counterbored below the surface because i don't want the head of the t-nut to interfere with the registration of the workpiece whatever i'm holding onto the vertical table and these t-nuts on the front side are going to accept the threads through the back Let's take a look for the mounting bolts. Every time I use this machine, I'm constantly gathering information. This is the first time that I've maxed out the vertical capacity, and in this case, I can get a 43.5 inch long board clamped to the vertical table and cut a tenon on the other side. And that tenon is, in this case, 1.6 inches in length. I can get a longer piece if I switch out to a smaller diameter bit, like a 1.25 inch diameter I'm going to do a diameter bit with a 1 inch length of cut instead of the one I'm using now. But that's just a, you know, topic for another day. It's good to get that little bit of information and just keep it in the back of my head. All these cuts turned out great, and there was a couple things to learn and tweak along the way, like scoring cut, yes or no on a scoring cut. All of these were cut with an up cut bit. And if I bring this one right here and let the camera focus, you see this, just this little itty bitty fuzz right there, like this is being so nitpicky. That's the only bit of tear out that I had without, oh, there's another little fuzz, without a scoring cut. When you cut in the vertical orientation like this and you use an up cut bit, your cuts are pretty much always flawless. So how much difference does no scoring cut make versus a scoring cut? Well, not much of a difference, but it does make them pretty much flawless. So this... Here is the scoring cut. It's nice and crisp. One thing that you may or may not pick up on is this flat right here at the top of the keyhole. That's not supposed to be flat. I was experimenting with the scoring cut and this scoring cut was actually just a little bit too deep. That one was a full one-eighth of an inch. I had to back it off quite a bit in order to not interfere with this. All right, I'm probably going to repeat myself here. I can't remember what I have and haven't said. Um, and I think I said I was going to skip all the fiddly bit, but I wanted to explain how this is, how this is adjusted and actually how easy it's adjusted. It's going to be a little time consuming. Uh, but anyway, so the top row here, we got five. T-nuts that have a hole drilled through at that location through that top rail with a bolt installed from the back side with a series of washers and a lock washer so that I get maximum thread engagement on these T-nuts without it sticking through the front. Of course if the bolt sticks through the front of the T-nut then the material as you put it in place will be adjusted inappropriately. It's not going to register flat against the vertical table so Maximum thread engagement through those t-nuts and right next to each one of these t-nuts is a hole for a t-nut on the back side and I installed a set screw through it. That's what this wrench is is engaged with is a set a set screw. Now because I use the machine itself To cut these dog holes perpendicular to the top edge, which was already cut with the machine, I know that if I clamp a straight edge to the top of this or some type of stop block to the top surface of the vertical table, referencing off of the already surfaced spoil board, I know that this line should be absolutely perfect vertical in relation to the whole setup here. So there's no real calibration this way as far as the right side going up or down or left side up or down. First thing I need to calibrate is along this top edge making sure that it is parallel in this direction as far as the left and right side coming forward or backward with the travel in the X direction. So the very first thing that I did is I got these bolts finger tightened to where they were engaged and holding but they weren't compressing the spring washer that I have on the or the lock washer on the back side of the bolt. And then I figured out which one of these sides was sticking out the furthest with none of the set screws engaged at all. It happened to be this side. From there I creeped up as close as I could get using a flashlight which is somewhere around here. It's probably in my pocket. It is in my pocket. Using a flashlight behind this precision rod here shining from the side and as soon as As soon as daylight was removed, I set my Y zero with my pendant here. This particular pendant has three different speeds, 100X, 10X, and 1X. And that's basically thousandths of an inch. At the 100X, you're moving a tenth of an inch per click on the dial here. Let me turn that to off so I don't mess anything up. So each click... Daylight is 1 tenth of an inch on 100x. On 10x, it's 1 one hundredth of an inch. On 1x, it's 1 thousandth of an inch. You can be very, very precise moving this into place by a thousandth of an inch, which is really, really nice. So I got it by the one thousandth of an inch to where daylight is eliminated over here and then set my Y zero. From there, I can move out and over to the far side. And get to Y0, use the set screw to push it away from the table so that the spoil board loses daylight in between the precision rod and the spoil board, the vertical table, and then clamp down my screw. As I clamp down my screw, it's going to pull it a little bit further and tighter into the horizontal table up there, into the machine, and a little bit of daylight. was shown again. So then I switch over to the set screw, crank it just a little bit more, and pull it back. And you're kind of trying to establish the tension on those bolts to where it locks it into place, but it's not so loose that it'll come apart, right? So once I got this one done, I moved over to this, or I moved over to this one, then I moved over to this one, And this is the final one right here in the middle, trying to establish the... This vertical plane along the X direction, that's the only one left to adjust. And what's interesting is when I first set this at Y0, when I went over to here, I had about 8 thousandths of an inch to move this out. So if you think about it, right off the bat, my error of this calibration when I started was 8 thousandths of an inch along 50, I don't know, 55 inches of travel, something like that. So that right there really isn't that bad. You're not going to notice eight thousandths of an inch along that span on a board, say five, six inches wide. You're really not. But this type of micro adjustment essentially allows this setup to be dialed in. Really, really precise. One thing to note is that these top bolt holes are slightly higher in elevation than the set screws. So, without this bottom piece clamped to the bottom rail, it's going to want to, as I tighten the bolts on top, it's going to want to pull the top forward, which would essentially kick the bottom out. And it was doing that just slightly, so I clamped it into place. Okay. You really can't crank on this type of a setup because there will be some flex in this MDF vertical table. It is two pieces screwed together with a bunch of screws, but there will be some type of little flex there and what you don't want to end up with is the top side being pushed in. You don't want a wavy surface. Then it comes back down over the set screw and so on as we go to these next few layers. But that's the process anyway. I'm only referencing off of the top say Quarter of an inch or three-eighths of an inch off of here to dial in this first screw. From there, I'm not going to use the machine at all. I'm going to use the largest square that I have to get the rest of the vertical orientation plumb, I guess you could say, with these top set screws. But it is a little bit fiddly. I figured I'd run through the process here, and I'm gonna run through the rest of this off-camera. This one last screw is ready to go. Oh, what I wanted to show you also while you have you here is how close you can get this very very precisely with this pendant. So let me zoom in. So the pendant, like I said, can be moved in very precise increments. So my safe distance out here to where I can jog in the X direction where these clamps aren't interfering with The spindle at all is four inches. As a matter of fact, Y negative four. So once I'm in a position to where I can go Y positive. And nothing is going to be in the way. I have my pendant set to Y at 100x. Let's actually go back to off. That way I can change the dial to zero without adjusting anything. Now with the dial at zero, I can go to Y at 100x, and it's reading at negative four on the digital readout. From negative four, I can go in 39 clicks. And because this is numbered on the dial, I can quickly go into 39. And I know at 39... I am 1 tenth of an inch away from where the spoil board should be. Let's go back to off, reset our dial to zero, go back to Y, and change it from 100X to 10X. Now from here, I can go in 9 clicks, and that'll leave me at 10 thousandths of an inch away from where I need to be. And of course, while I'm doing this, I'm making sure I'm not going to hit anyway. Anyway, Y, 10X, I'm going down to... That is nine. That's nine right there. And if I shine my flashlight on the back side, I can still see a little bit of a gap on the back side. So I should be, as the digital readout reads, ten thousandths of an inch away from where I need to be. So now let's go back to off. Reset our dial to zero just for easier math. Go back to Y. Change it to 1X. Now I'm moving one thousandth of 1,000th of an inch, and I can go down. Very carefully to 9, so I should be 1,000th of an inch away. Now, of course, I haven't calibrated this yet, but we're pretty darn close to begin with. And from here, I still see daylight, and I am 2,000th of an inch away. So let's go one more. According to Y0, I'm 1,000th of an inch away. And right there on the digital readout it reads Y zero. So I'm at the exact zero point that I established over here. Already verified there, there, and there. And now at this location it looks like I'm like a thousandth of an inch off or something like that. I need to use the set screw to push it away from the table. So I'm not going to mess with this. Go back to off so I don't accidentally bump it. The spindle is in the perfect location for where it needs to be. Where's my wrench? Here's my wrench. I'm an idiot. Right here I'm looking for daylight on the back side. I can't zoom in and show you super good with this. Maybe I can. Maybe that other camera is picking it up. But I'm looking for daylight on the other side and I'm going to probably get in the way. I'm going to tighten this set screw so it's going to push the spoil board into the spindle. Let me go this side real quick. Push the spoil board into the spindle and right there I lost daylight. So now we are contacting, we're touching. Now I'm going to switch over to the back side and tighten this bolt real quick, which will take just a second. I'm not going to crank on it, I'm just going to give it just an ever so slight adjustment, compress the lock washer, and then just a tiny bit more. Not much at all. Let's see what type of a change that made over here. It looks like I can see just a smidgen of daylight once again so I'm going to crank the set screw a smidge just to put some tension on the whole system. Daylight is lost and just an ever so slight adjustment on the bottom side once again. It's like an eighth of a turn with my ratchet. I'm looking at the lock washer and it looks like we have a smidgen more of compression. So from here that bolt on the back side should have pulled this away from the spindle a smidge and it did not so i need to make a slight slight adjustment on the bottom side once again adjust this bottom one that was another eighth of a turn or so maybe a quarter of a turn and then from here looking for daylight ever so slight daylight so now i'm going to make a slight adjustment on the set screw to push it back to the spindle and that looks pretty darn good now i have tension On the set screw it's tight and the lock washer on the back side is completely compressed so that shouldn't wiggle out. Although I'm going to give it one more crank just just to verify that lock washer is nice and secure. I don't want to go too crazy with this like I said because if I adjust it too much then what I'll have is a wavy pattern in the vertical table and that's not what we want. But we're talking about... thousandths of an inch here and it's all going to be distributed through the length of whatever piece that you have. so right there. that's my final adjustment. that should be enough pressure on both of these screws so that it's locked into place. now this plane is perfectly parallel to the y travel. this top plane is perfectly parallel to the horizontal spoil board. The only thing left to do is to make sure that those stay that way while the bottom of it is either kicked out or pushed in to get the correct position so that we are perpendicular to the vertical table. All that to say a bunch more adjustments to be made. That's the process. A lot of back and forth and it's probably gonna take me a little bit. So I'll see you later. With the mortises cut on the legs, it's time for the tenons on the rails. And one of my main reasons for pursuing this particular CNC machine was its open frame concept. You can make just about anything you want for it and mount it just about anywhere you want. I made a vertical table which allows for the joinery to be cut on the end of pieces. And just like with the horizontal table, The dog holes in the vertical table allow for repeat positioning as their exact location is calibrated and stored in my machine file on the computer. This makes batching out tenons, dovetails, dowels, anything similar to that to be very quick and easy. The tenon cuts are broken into three tool paths and there's room for improvement in each one. First, a shoulder scoring cut is made and this is a climb cut all the way around the To ensure that there is absolutely zero tear out on the shoulder of the tenon. An improvement here would be to make this toolpath a full depth cut. You know, a single pass full depth cut. Because the amount of material being removed, it's less than a tenth of an inch, so it's not that much at all. The second toolpath is a climb pocket toolpath to remove the waste down to the exact size of the tenon. As you can see, The bit unnecessarily raises and lowers between passes and this is an inefficient use of time and travel. Because there wasn't much material to remove, I don't think it was ever the same diameter as the router bit, a faster method would have been to use a profile toolpath set to the outside of the tenon vector with a spiral ramp. This would allow the bit to Spiral around the tenon all the way down to its final depth without lifting a single time. That would have been much faster All right. Here's what I've settled on for the final calibration vertically. This is the largest precision square that I have. It's a 26 by 16 square It's a Woodpecker square. So they're supposed to be really precise That's the best thing that I have to measure my actual finished product in woodworking or measure anything in the shop So if it needs any more precision than this well at that point you're getting into tool and die and not woodworking. So anyway up against the table here just using gravity it's not clamped down at all. Up against the table here we have no light coming through this top edge and down near the bottom we have quite a bit of light shining through right here. So that tells me the bottom of the table needs to come out and I'm going to do that. On a per column basis, get that first column done, go to the second one, get this second column done, and so on. I went with box joints for the drawer joinery, and this was the first time that I tried making vertical table joint cuts without first running a shoulder scoring toolpath, and the results were great. A quick test fit of the joint confirms the fit, and the rest of the joints are knocked out. A couple stop blocks in the vertical table direction make batching out all of these cuts quick and easy. Now I'm not sure how much of the clean cuts were due to not being aggressive with the cut versus the great cutting properties of ash. Ash is probably my favorite hardwood to use with a CNC machine because it just cuts so cleanly. Outro take two. I had to scrap the last outro. It's a few days later and I've already calibrated the table and Houston we have some problems. We have a few different problems. I think we have solutions to those problems but we've got some problems. So everything is installed and calibrated and I've made a bunch of test cuts and whatnot and uh well first off let me show you what I did to the back side here because I didn't cover any of this. So back here you can see how the bottom part is attached to the machine. The bolts go through. This is a piece of Ash, it is about an inch thick, inch and an eighth thick, inch and a half wide, something like that. Just scrap wood, biggest size I can get it milled down to. So that's how it's attached to the bottom two rails. Like I said, top five bolts up there. And I did not want the vertical weight, I didn't want the whole weight of it to be held on by just those bolts. So before I took the stop blocks off and I had it all calibrated, I came back and added... Four of these blocks, two on each side. It's just a big piece of scrap wood milled flat and then glued to the spoil board. And it's touching the rail here. And I did this on the sides because that's where the strongest part of the rail is, where it connects to the leg. Rather than in the middle, I didn't want the rail to sag over time. And I did not want, like I said, all of the weight to be held on by those bolts up there. And then the pressure of squeezing against the rail right Right here, so. Four stop blocks, two on each side. Remember me complaining about the quality? It just felt like this was a low quality piece of MDF. That box has been checked. This MDF is nothing like any MDF I've ever experienced. I've made a lot of flat surface joinery cuts to assemble stuff like this, this cart right here. This is all... CNC box joints essentially, mortise and tenon joints, cut in the flat orientation to build a structure. Those boxes up there are the same thing, built or cut flat with interlocking joinery. And every time you do that you have to measure the thickness of your material because the thickness of your material determines the length of your joint where they interact and form a 90 degree angle So I measure material constantly and thickness also for just general CNC use and MDF is incredibly consistent from piece to piece to piece to piece. I've never been burned by MDF. It is always within like one or two thousandths of 0.750. And with this stuff, this is wild. I started to realize like some of my problems are probably not due to execution, but also due to this material. So I measured with my calipers all through the slots, around the perimeter, and there's a wild discrepancy in thickness. The swing that I have is 28 thousandths of an inch, maximum to minimum. And that's uncharacteristic of good quality MDF. This MDF is two layers not glued together, but they are screwed nice and tight and I can't see any visible gaps in any one of the seams that's exposed. So I'm not thinking that the inconsistency in thickness is due to that mushroom in between layers when a screw passes through, plus I pre-drilled for all that. So what's going on with this stuff? Well, it's just wildly inconsistent, which makes it a pain in the butt to try and get this stuff, get all the adjustments dialed in nice and perfectly. Now, I did it the best that I could. And... I'm still seeing some bad results. Another problem that I found with the execution of my design is I did not put enough of these mounting locations, these screws, bolts rather, bolts and set screws along that top rail. So really I should have at least I think one more in between each one of these and the reason for that is this top rail right here, it's behind here but I'll point to I'll point to this one, this top rail. I'm wanting to come behind it with clamps and clamp my material down. And that will allow me to clamp way up here with the clamps. Unfortunately, the highest I can actually get with the clamps is this right here. That's as high as I can clamp on this board. And the reason being is if I clamp to the rail, now the rail is back here. And you see this gap in between the horizontal and vertical table? If I clamp back here, what I noticed is in areas like the left side of this dog, it's in the middle of these two Thank pull-down bolts, and it deflected the front rail, or the vertical table rather, into the front rail. So that changes my angle. Therefore, when my board is up in here, and it's supposed to be exactly flush with the x-axis, it's off on a little bit of an angle. I'm exaggerating there, but it's off on a little bit of an angle, you. and that throws off the results. May or may not pick up on this piece right here but the left tail where my thumb is is sticking out the appropriate distance and the right tail is Pretty much flush. So that means that this is Not cutting properly and it was cut right here. So hmm what can we do about that particular problem? Well in that case I can use the machine to cut a dado all the way through here right on that seam. Probably over or right at an inch deep because I've got two layers of MDF on the horizontal table. So let's just say we cut a slot or cut a dado, put a piece of hardwood in there that is is consistently thick along its entire length. So mill down a piece through the planer that is several inches longer and then cut off the planer snipe, stick it in there. That will stop this from compressing in and get better results. So there's a solution to that particular problem. But as it is, I can't clamp higher than that clamp right there. So therefore, the top of this board, I don't know if it's going to show up on camera because I'm probably going to wiggle the camera too, but this board is wiggling a considerable amount. Is that amount of wiggle with the clamp down there as high as it can be going to be a problem? Well, I did cut another test joint, and disregard the sides not lining up, one of my vectors shifted when I was designing it and didn't realize it, but disregarding this not lining up, this actually is a pretty darn good joint. It's an end mill only dovetail joint, so the pins are rounded on the interior side right there, and I would call that acceptable, if the vectors lined up appropriately, but that's my error. So I'd call that acceptable, but I don't like it. You know what I'm saying? Like this just. Drives me bonkers and then I'm limiting how much I can clamp in there because I can't get all the way up with the clamps. So that's kind of a concern when you're cutting stuff like this. The mass of the leg, if I'm doing like a tenon on the end of the leg, it's not going to deflect enough to matter at all. When I was adjusting this, I was talking about, I believe I mentioned this, due to my mounting locations with the uh securing bolt on top of the set screw and then the set screw on top of the securing bolt as i clamp this one down right here it's going to push the top of the table in create a belly out right here and then push the bottom of this in that's just the you know the physics behind this right well i thought i was experiencing that when i put my square up here and my square was what 16 by 24 i think is what it was something like that so it didn't go down Full length of the table and I was experiencing what I thought was a belly right where I thought it would be. So okay I'm getting kind of frustrated and tinkering with all this and getting it the best that I can get it and it just still has what feels like a little bit of a belly right here. All right well I continued got it all done and then it dawned on me my square's not going all the way down the table so let's put a straight edge I got a 50 something inch Veritas straight edge bought that many years ago and by the way that's one of the greatest things Things that you can keep in your shop for dialing stuff in. And I put it up against the whole vertical run of the table and just, uh, I'm sorry, unlike what I thought, there was no belly here. And if there was a belly here, well then the opposite should hold true down here where it should be sunk in because the mounting bolts are lower than the set screws, right? Well it wasn't. It was like perfectly flat. It is perfectly flat from like right here down. And I'm thinking, what in the world? So it's just this top five inches or so that's pushed in on an angle. Well that's kind of strange. Maybe I've got these bolts too tight. So as it sits right now, these bolts are actually backed out to where there's no tension on these bolts at all. It's literally just sitting there and held in place by the clamps down here and the vertical stop blocks that I have on the back side. So these these bolts are not doing anything. They're not deflecting the top surface in at all and I've got a considerable amount of upward lean on this this entire panel. I'm thinking, what in the world? How is... How is that? How is that going on? Well, it's this low quality MDF over here. Here we go. I have the offcuts just sitting in place on the floor, just sitting on the floor. And this corner over here is lifted by a good half inch off the floor just as it sits right there. It's not sitting flat at all. Now, of course, the concrete floor is probably not Not perfectly flat, but it's lifting up over here. So is this MDF good? Out of the box, or from the store rather. Well, what are the odds of both of these pieces of MDF, both of them having a little bit of a bend in the panel, at the exact same location? Well, if you think about it, both sheets came off of the stack at the store in the same exact orientation and were slid into my truck at the same orientation. Then I unloaded them and set them down in the same exact orientation. Then I lifted them up and put them on the table the exact same way and referenced off of the same corner to cut each one of these pieces. So, if this bunk of MDF was sitting long-term at the store and it had a little bit of a wave in it, which sometimes you do see, and that wave telegraphed into the material, It's completely plausible that... I included a little bit of a wave in the MDF from the very beginning because like I said that those that entire top row of mounting bolts there's there's no tension on them that's just essentially free floating as it is. So um how much is that lean? That's a considerable amount. I can get my fingernail uh my my fingernail in between the material and the spoil board right there on both of these. You can probably see it on this one. It's a considerable amount. Uh, which is, is it affecting everything? Here's the thing. When you're dialing in a machine, like let's just say that jointer and the manual gives you specifications of over this particular distance, you need to have an accuracy of say four thousandths of an inch. And for whatever reason you're, you're tinkering with it and tinkering with it. And you can't seem to get it dialed in perfectly. Like say it needs to be four thousandths and you're at six thousandths, right? There comes a point in woodworking where you just simply say, you know what, let's see what it'll actually produce. And if you joint a flat board where you're not technically within specs, but if you joint the board flat and that board is suction cupped to the table, it feels like it's dead flat. And then you measure it with the best tools that you have. And as far as you can tell it's flat, then you're chasing diminishing returns and you're wasting your time trying to get it even. Even further because the results speak for themselves. The results are acceptable. In this case, I would probably consider that acceptable. You put glue in there and those little gaps from the tooling will kind of disappear. And I see more problems with my feeds and speeds making this perfect than I do from the actual setup on the machine. But the shortcomings with this This top edge of it being leaned too far and then I can't clamp to the table. It's quite annoying. So I've got a few different options. Option number one is what I'm going to start with and that is I'm just going to roll with it. I know that there's problems and it's one of those things that's probably going to bug me more than anything rather than having bad results. So I'm going to roll with it. But as I start making stuff, if I'm seeing some problems that are too much for the standard of Then at that point, I'll consider making some changes. Change number one is probably going to be taking this off, taking all the metal out of it, not the screws on the back side, and putting it on the machine and flattening the front surface. Just enough to get rid of pencil lines that I'll put around the low spot, which will be right here. That's option number one. Option number two is I may get frustrated with the material to begin with because I know this material is not really quality stuff. And I may pull a mulligan and go get some better quality MDF at the same place that I've always gotten MDF from, but the better stuff. And by the way, you can hear. You can hear and feel a difference in the consistency or the, I guess the density of this material when you interact with it versus the stuff that I've been used to. So I may get some better MDF and completely redo this with a couple minor changes. And those minor changes will be to, no matter what, if I take this off and put it back on, I am going to do this seam right down here to stop this from compressing and kind of sort of lock it all into place. That way I can use that front rail. Something else that will help me use that front rail more. Efficiently, easily, whatever is I think I'm going to do a little bit differently with the set screws here. This is staying pretty flat down here, so I think I'm going to keep that. But if I was to do this over again, I would have three holes through the rail instead of one. The first rail, the first hole is already there. I'm going to use that hole. The second two holes. The other two holes on either side, I'll probably just drill a hole and tap it for quarter 20 bolts and put a bolt through the back side as my set screws rather than using the set screw from the front. What that'll do is put the clamping pressure even and horizontal and then one on either side to counteract the other one so you don't have that twisting or bending that I'm getting with this. Theoretically speaking, that little twisting and bending. I don't think it's that big of a deal, but if I'm going to redo it, then there's that. As far as the work holding and design of the table, it's fine. I don't think I'll make any structural changes to this or any capacity or feature changes to this. It's just, I mean, it's basic dog holes and slots, right? I'm not sure if I mentioned this, but due to the slot spacing and the dog hole orientation, Thank there are a couple of these that I can't get clamps on both sides of. For example, the distance between the clamp bar and the clamp head right here is not enough you for me to come in on the left side in this orientation, very come in from a clamp right here, and get to this. So maybe shorten this a smidgen. Anyway, I designed all this so that way you can still get up much. to a 55 inch wide panel all the way across here with a combination of different locations too. Uh, put your vertical piece. I didn't want to just put a set of dogs on one side and just keep clamping from that same corner all the way across, uh, because you run into, there are a couple different dimensions where you run into situation where you don't, if the side of this material right here is say right there, that gap isn't great enough to put a clamp in and then you don't have that distance right here to put a clamp in from this side over there. So there's a couple different Very odd dimensions that you may run into if you simply just start at one side and go this way So that's why I've got a couple different Vertical locations and then each slot width on the row is different It's more narrow over here a little bit wider a little bit wider and then the widest So with a different combination of where you mount it you can get up to a 55 inch wide panel. Now how often are you going to do like a 55 inch wide dovetail joint or something crazy like that? I don't know but in the time and age of doing things that are just super cool and big and over the top and awesome that basically everyone's starting to do. I think it would be pretty cool to do like a big conference table or dining table or some type of big table where on one side over there you have a set of legs and maybe a traditional apron over there but on the other side you basically have a continuation of the top that goes vertical and you have a really long connected joint to connect them. That would be really cool. So if that's even in the thought process I'll take that into consideration as far as maximum capacity. So technically it's done. Technically, it's operational. Technically, it's driving me bonkers on the inside because of the mistakes and errors that I found. If you were to redo this, I got a question for you. Knowing everything that I've said in this video, would you use two sheets of good quality MDF that I know is out there? Or would you just say, you know what, forget about it. I'm going to use two sheets of 18 millimeter Baltic birch. Baltic birch for quality MDF. Interesting choice. That's it for this one. I'm, like I said, just going to keep using it until I get too annoyed with it and have to replace it. Hopefully you learned something from this video, and onward.