Zero Play Guide Bars for table saw miter slot

by Bwolfey Mar 27, 2018
Download All Files

Thing Apps Enabled

Please Login to Comment

I know you were trying to copy the commercial product as much as possible but I wonder if there would be a reason not to widen the bottoms so that they actually engaged the "T" part of the T-Slot? Also I'm curious why you have the bottom print vertically a it doesn't seem like it needs that. OTOH the top (metric anyway) looks as if it would benefit from printing vertically but is positioned to print horizontally. Just curious...

Well, I mainly developed this thing for selfish reasons as my home 90's ridgid table saw does not have T-profile slots. It makes sense that a company would develop a product to the lowest common denominator so as to be more widely useful instead of specializing in only T-profile saws.

As far as the print orientation goes, I don't really understand why my CAD export comes in vertically like that. I definitely recommend printing flat either on the long edge or the long face (depending on whether or not you have soluble support).

HI, if you import your stl files into cura, it is expecting mm but since you designed in inches somehow this doesn't work. So instead of being 9 inches long it is 9mm long in cura. I think there may be a setting to save stl files in mm which cura can then import correctly. I have not found a way to deal with this on the cura end.


If you open Cura and go to the menu item, Marketplace you will find a plug in called Barbarian Units. Once installed, restart Cura and go to Menu item Extensions, select Barbarian Units snd it will convert your inch.stl to metric.

Hey Hey! Sorry to hear that you're having trouble with the files. From what I've found, you need to specify the units in the import software. So meshmixer/solidworks/fusion 360 (all that I've used) all have an option to set for importing STL files (which are dimensionless). That way it knows the 9 is mm or m or in or ft.

If Cura doesn't have that workaround, if you import a file into Cura, and it's way too small, you can try scaling it up on all axes by a factor of 25.4, as there are roughly 25.4 millimeters in 1 inch.

Thanks for the quick reply. My understanding is Cura doesn't have this option and since STL's are dimensionless it assumes all files are mm (someone correct me if there is a setting in Cura somewhere that I couldn't find). The only workaround that I have seen posted anywhere is to multiply as you said by 25.4 or when exporting the files need to be converted to mm for Cura users. I imported your metric files with no issue and scaled the inch file to be the same mm (which by the was was 228.6 vs the 227 you have in the description ... at least that's per Cura).

Thanks for the great work on this.

You bet! Happy to hear you've found this (somewhat) useful despite the unit confusion. I'll update the description to match the value you used.

Let me know how it works after you print it out!

Any chance of a version that can go down to 15mm to fit craftsman/ryobi table saws?

I'll look into addressing it. 15mm is skinny! I'll have to shift around a few features to get it to work.

What type of guide are you hoping to use? The captured nut or the brass insert approach?

Yeah, it uses non standard miter slots, I scaled a set to 15mm wide and they fit, but I didn't try them with any hardware.

I was hoping to use them with a captured nut since I think it would be easier to find that hardware at a big box store.

This is a great build! Any chance you could do a metric version version using M5 hardware? Down in AUS it's hard to get the imperial hardware without paying through the nose!

Thanks again

M5 is too big when it comes to the brass inserts so I sized it for M4. I've uploaded new files to the listing (along with the specific hardware I used), but I've also attached the STLs here.

Let me know if you have any questions!

You are an absolute champion, very much appreciated.

I appreciate that! Are you good with the metric version being with the heat-set inserts? I think that's a much cleaner design compared to nutting the fastener on the other side.

I'm working on that now :).

Could you make these with a standard hex nut instead of the heatset insert? Those things are ridiculously expensive and difficult to find in my neck of the woods. Even a standard threaded insert would be easier to find. How much force are these things subjected to? Or is the heat set nut more of a nicety than a necessity?

My aim was to replicate the original design from MicroJig, who I noticed used molded-in inserts. I wanted to be on-par with that product. Apart from that, I would absolutely say it's more of a nicety than a necessity.

I like the thought of a tweak where captured hex nuts are used instead of the inserts (as long as they're not falling out when you're trying to assemble things.) Do you think I should make a separate thing page for that upload? I'll have to see how the stacked thickness of the parts work out but initially I'm thinking I'll use a low profile (or "narrow" hex nut - https://www.mcmaster.com/#90760a009/=1ccn0gq) to have enough material underneath.

Let me know your thoughts, as I feel like your view is probably commonly held.

I would suggest adding variants on this page unless you significantly modify the design. Might be worth adding a metric equivalent version too, since those are pretty much the standard outside the US. I can get metric bolts, but I don’t have them on hand, and particularly for smaller sizes they can be a bit scarce.

What size are these? Just wondering if they fit on a 200x200 bed.

I just reworked the model to accommodate a hex nut. Both the top and bottom parts changed geometry. Let me know if you print this, I'd love to get input on this change.

I just finished my first print of these parts, and here's what I found:

Material: PLA

Print Settings:
0.2mm Layer Height
1.2mm Top/Bottom Thickness
1.2mm Wall Thickness
40% Infill (Triangles)

Support Density: 20%
Support Pattern: ZigZag (connected ends)
Support Z Distance: 0.2mm
Support X/Y Distance: 0.4mm (Just enough to leave an air gap, considering even less...)
Horizontal Expansion: 2mm (Fills the holes properly)
Interface Thickness (Top & Bottom): 0.4mm
Interface Density: 50%
Interface Pattern: Lines

First of all, I found that a pronounced dip in one corner of my build platform meant I could not use that diagonal orientation, and so I had to go the other way. One of the downsides of squeezing it in to a 200x200mm platform, would be much nicer on a large format printer.

The overhangs for the bolts take a lot of farting about with to get good support. Ultimately I think I came up with a good recipe, but they are still kind of a pain. Also there is a gap between the top surfaces of the wedges, and their mating piece. Not sure if that's intentional or not.

The end result was ok, I had a problem with the edges lifting off the platform, which is almost never an issue for me. I think this might be because I didn't re-orient the zigzag of the top and bottom surfaces, which meant that the print aligned with the direction of the zigzag. As the long very long lines of filament cooled, they probably contracted and warped the ends up. I will fix that for the next print.

The final piece is a bit flexible, but feels quite strong. I think there is probably sufficient support for the screws by the walls, but time will tell. I'm not 100% sure how it's supposed to be attached to the sled, I feel like one of the sets of hex things is supposed to be screwed in? If this is the case, perhaps instead of a hex hole, that one should be a round counter-bore, or even a countersink for woodscrews.

If you were to revise the design, I have a few suggestions: I recommend that the "full height" edge of each piece be made into a free standing wall at full height, the length of the runner. This way there is a larger bearing/wear surface for the runner, and also it'll act as a rib, and add some more rigidity to the pieces. You'll probably need to salvage a bit of extra material, but it looks like you can probably take that off the opposite side. Also the wedge "holes" can probably be completely removed to simplify the print. They don't really do much so far as I can tell, and are probably just a good trap for dust.

I can't test it myself, but you might want to make a longer version for people with a CR-10 or similar. 9" is pretty short for a runner. A 12" version, and whatever fits on the diagonal of a 300x300mm might also be good, although perhaps leave them until the kinks in this one are worked out.

I kind of hope you do, I think it could potentially be really cool. I guess the nice thing about 3d Printing is you can iterate the designs pretty quick... you could be on to a real winner here.

PS. Extrude the wedge "holes" all the way to the bottom, lift the sliding wedges to full height, and add another row to the other side. Even if you don't do the full height wall thing I said, having another row of wedges on the other side increases the bearing surface on that side, and strengthens the wedges overall.

PPS. Added a hastily done photoshop in attempt to better illustrate my design notes, hope it helps!

This is great input. I really appreciate you taking the time to type out such a detailed response to the design.

That gap you noticed was intentional - I didn't want the contact points competing between the tops of the wedges and the larger contacting surfaces.

As far as attaching to the sled, I'd suggest referring to the microjig instructions: https://www.microjig.com/products/zeroplay#resources
It's a little backwards from anything shop-made as you're going through the top of the sled instead of through the bottom. There's some tricky stuff involved with making sure you're counterboring deep enough for the fasteners you're using without sticking proud of the captured nuts on the bottom of the guide.

The attachment points for the guides are meant to be tightened from above so you do your final lock-down with the jig in the slots while referencing the saw blade (either parallelism or perpendicularity). I absolutely see where you're coming from, but having retrofitted one set on a previous jig, it's well aligned only because of the top-down access.

I like the concepts that you have proposed (especially the full-height stiffening rib). In practice though, after the fasteners draw the guides to the base of the jig, you address a lot of the concerns you raised with flexibility and bearing surface (assuming a U profile slot, not a T profile). I'm choosing to believe the microjig dev team looked at some of the other changes you proposed (removing the wedge recesses) and settled on their final design seeking to maximize contact area with the bottom of the sled. That's an assumption though. Similarly, the wedges are just used for initial positioning and then it's the fasteners that lock the assembly together. I don't think it's necessary to add a second row of them. I don't think the fact that they could accumulate dust will interfere with the guide operation. I need more time with the set I have in rotation though.

I agree with the 9" length seeming a little short. I wouldn't be able to test a 12" version either but make throw one into the collection.

I'd love to know if you put these guides on a sled of yours! Please let me know how that experience is - I've had a lot of people like and save this design but it seems you're the first to actually print one.

I'm also assuming a U shaped channel. I can see not bothering with wedges on both sides, but I think making the wedges full height and a cutout simplifies the design somewhat. Also worth remembering that when Microjig designed and tested their product, it was probably a solid piece of material, and not FDM, with shear lines. Also not sure what the material they use is, but it's probably not PLA or ABS. I really think the full height rib would be a solid improvement for bearing on the sides of the channel, since a sled guide isn't supposed to touch the bottom anyway.

Great input, thanks.

These are 227.0 mm (I'll add the lengths to the description). You'll be able to fit parts for one guide diagonally.

Didn’t get a chance to start the print before I left this morning, but I did try to get them sliced. Kind of a pain the units being inches, had to manually upscale them 2540%. Also the rotation was off, one part on it’s side and one standing on end. Could only fit one part diagonally, rather than both too and bottom. Might consider rescaling your original to mm (1” = 25.4mm) since this is what most things seem to support, it would make the STLs more portable.

PS: Seems like when converted from in. to metric, they wind up 228.6mm long. Not sure if this is correct.
PPS: Why does this thing need a raft? Surely standard supports would be preferable?