best method for interfacing linear square rail to 80/20 framing?

[louieatienza]

I used the milled extrusion from Misumi on a mini router build a while back.... nice stuff, and the price was reasonable..

[wizard]

I'd be interested in seeing a price comparison between regular 8020 T slot and Misumi ground flat T slot.

Have you seen any builds using it Wizard?

Somebody on this forum suggested that they where going to use it. I haven't seen such personally. At work almost all linear rails sit on tooling plate that is then bolted to the T-Slotted extrusion. This is done mainly because it is far easier to get the tooling plate drilled precisely on a CNC than it is to hand assemble a T-slotted extrusion. If done right (not always the case) you can have everything drilled and tapped in one go. At the moment I don't know of any linear rails (at work) being bolted directly to T-slotted extrusions.

The other option is to use a welded steel frame, with welded pads on the frame work, for your linear rails. The big problem with this approach is that anything of size needs a well equipped machine shop as the pads are usually milled flat and coplanar. Thankful there are work rounds for linear rail mountings on steel tube for the DIY world. The biggest reason I can see for going steel for the frames is that it is actually cheaper to weld one up and machine it, especially if you are looking for reliability. The biggest problem with T-slotted extrusions in my mind is that they do not stay together for the duration unless you get really excessive with gusseting and tie plates.

A side note:

Mush of the extrusions we use at work are for special machine application or sometimes called automation. One of the reasons I call it T-slotted above is because we have all sorts coming into he plant over the years. This includes stuff from 80/20, Misumi, Bosch, Fastek, Some German company (don't have a name) and a bunch of others. For whatever reason our engineering department doesn't seem to care though it does cause a lot of grief for the rest of us. Right now I'd have to say that we are roughly 60% 80/20 style and maybe 30% Bosch. That unnamed German company (best guess to be German) actually makes a really nice extrusion with heavy walls and flat surfaces.

Not much of this is used in the primary structures of the machines though.

Cutting steel isn't too hard, you can do it with an angle grinder, but it is much easier if you have access to a large bandsaw. Getting things flat is where steel can be tricky.

I'm not sure where this idea that working with steel is so bad got started. Ee are talking mild steel here or structural steels which beyond an occasional hard spot isn't that difficult to drill. Generally I find steel tubing to be rather easy to work with even compared to rolled structural steel. I'm not sure if that has to do with being formed from sheetmetal stock or what but it is generally easy to drill, cut and form. In other words we aren't working with tool steels or work hardening steels. The biggest problem a person might run into is designs that call for a milling machine, the way around that is to design to minimize the need for such machines.

Not too hard to drill and tap some holes, but as a minimum you need a good variable speed drill press. A simple hand drill isn't good enough. I use a small mill.

With practice an hand drill, possibly with some guides can be passable but yeah a drill press is a smart choice. The good thing here is that you can get by with a pretty pathetic drill press if you can support the work, this mostly because you will be drilling small holes for tapping.

The other option is to use aluminum extrusion, and I'm not talking about T-slot. I'm talking about square aluminum tube, perhaps 3/8" wall thickness, not to hard to drill and tap some holes for your rails, and you could fill it with something afterwards like high density polyurethane or epoxy granite, you could even press in so ribs made of plywood. IMO, performance wise, that would be superior to T-Slot, and cost wise it might be on par. You'd have to make some inquiries to see what is available, and visit the metal merchant with a long straight edge to see if it's flat. IMO, aluminum products are typically flat enough for our purposes.

The biggest problems with such extrusions is the costs. This especially when I see drops for so little at the steel supplier. The difference can often cover the cost of a good right angle grinder and a drill press and probably a few more tools.

If you do this, please report back, I'd like to hear the results for cost and flatness. Also, look at the price for a full length from the wholesaler, and you'd use the rest for other parts of the machine, otherwise the cost may be double for what you get.

Aligning the rails and drilling / tapping some holes is really a joke - not hard to do, but like I said, you need at least a good variable speed drill press.

I might add that it pays to make up a few good tap guides. A broken tap can be more frustrating than anything that would happen at the drill press. If nothing else somebody not familiar with metal working probably should practice on some scrap metal first. At the drill press using techniques form the wood rocking world, such as fences and hold downs can lead to excellence repeatability in one direction. so your layout lines only need to be placed for one axis.

I think part of our goal is to get people to relax about steel tubing. Yes it is a bit harder than aluminum but it isn't unobtainium. Steel tubing might not be the right choice for everybody and every situation but it does have its place at the DIY table. Also unless a person wants a light weight transportable machine, weight isn't a bad thing in a router design.

[NIC 77]

Louie, that's interesting, do you recall the cost difference? If the costs are comparable, then definitely, it's worth exploring, and even people who make kits, like CNCRP should look into it.

Wizard, I think that we agree on most things.

steel is easy enough to drill, obviously you need to use a lower RPM than you would for aluminum and you need something variable speed. Perhaps you're better with a hand drill than I am, I just don't see a way around a drill press (minimum) for good results.

For the cost of the aluminum tube, I have to think that if you go to a metal wholesaler, like Russel Metals, that you will get a comparable cost to T-Slot of you buy in full lengths. Not certain, I haven't bought aluminum in a while, but I do know that resellers will often double or triple the price from wholesalers, and in my experience, the minimum order from a wholesaler is one full length, if it is something they regularly stock. And in the past I have bought significant quantities of aluminum from a wholesaler, and this was my experience. Also, the last time I bought T-Slot from 8020 it cost more than a couple hundred for shipping, and when I bought aluminum from the wholesaler, the truck stopped off in my driveway, and I think it only cost me 40 bucks extra for that convenience, and those were 24 foot lengths.

I absolutely agree with you that weight isn't a bad thing in a router design.

The big thing remains that steel is harder to get flat than aluminum which generally comes flat enough, and that is the hardest point to work with. Steel does for certain take way more time for the average person to complete a build with. You don't have to paint T-slot. Buy a kit, assembly over a few week ends. For steel, our builds can span a very long time.

Steel designs often require some welding, which even though it's not a hard skill to lean, requires equipment and space that some people don't have.

Perhaps also the lack of kits and plans with step by step instructions is also something that holds people back from all steel construction of DIY machines. You or I could do it, but also, this isn't our first time.

I do think there may be a middle ground using aluminum construction that isn't T-Slot.

Always nice to hear your points of view, and also, my opinions are simply that, and sometimes, I am mistaken.

[louieatienza]

Weight isn't so bad - then again it can be. It's not like we're taking a 5" shell mill and facing a steel billet with out DIY routers. Even a DIY machine made of steel tubing would not fare too well with that. Heavier moving parts means larger steppers or even servos with gear reduction, larger power requirements would be needed. While a part made of steel would be up to three times stiffer than a part of aluminum of same dimension, it would also be three times as heavy. You couldn't have it shipped by one of our favorite carriers - you need it delivered freight. Practicality of design is not just about using one material because it's "stronger" and "heavier;" the benefits have to be weighed against the downsides (no pun intended.)

The DIY router table builder's wet dream is to make a machine out of thick wall steel tubing, welded, stress relieved, and all bearing surfaces machined in one setup on a large 5-axis. THen use 1.5HP servos on each axis to move it all. It's just not as practical or feasible as been said here. If you weld mild steel tubing it's going to move around, no matter how careful you are. A drill press don't help much unless you buy or build an auxiliary table. It's probably better to rent a mag drill and take the tool to the work.

While steel drops are relatively cheap, unless you're so equipped the costs of everything else adds up. There's no convenient method of connecting tubing at right angles that doesn't involve welding, grinding, machining, epoxy bedding... If you really want to get reduction ad absurdum I can say there's only about $500 worth of raw materials in a Bridgeport, and probably less if I bought it as scrap that I could melt, recast, pull into wire, form, and machine. You could buy David Gingery's books - make a furnace so you can cast parts of patterns so you could build a lathe, then use the lathe to build the parts for a shaper, which then you'll have the machinery to make a mill. Then once you have all three you can make the parts for your DIY router. Point I'm trying to get across is that one must stop and think how far they want to go to save, or how much strength and stiffness is really needed for a DIY router, using small tools to route relatively soft materials.

[louieatienza]

Louie, that's interesting, do you recall the cost difference? If the costs are comparable, then definitely, it's worth exploring, and even people who make kits, like CNCRP should look into it.

Wizard, I think that we agree on most things.

steel is easy enough to drill, obviously you need to use a lower RPM than you would for aluminum and you need something variable speed. Perhaps you're better with a hand drill than I am, I just don't see a way around a drill press (minimum) for good results.

For the cost of the aluminum tube, I have to think that if you go to a metal wholesaler, like Russel Metals, that you will get a comparable cost to T-Slot of you buy in full lengths. Not certain, I haven't bought aluminum in a while, but I do know that resellers will often double or triple the price from wholesalers, and in my experience, the minimum order from a wholesaler is one full length, if it is something they regularly stock. And in the past I have bought significant quantities of aluminum from a wholesaler, and this was my experience. Also, the last time I bought T-Slot from 8020 it cost more than a couple hundred for shipping, and when I bought aluminum from the wholesaler, the truck stopped off in my driveway, and I think it only cost me 40 bucks extra for that convenience, and those were 24 foot lengths.

I absolutely agree with you that weight isn't a bad thing in a router design.

The big thing remains that steel is harder to get flat than aluminum which generally comes flat enough, and that is the hardest point to work with. Steel does for certain take way more time for the average person to complete a build with. You don't have to paint T-slot. Buy a kit, assembly over a few week ends. For steel, our builds can span a very long time.

Steel designs often require some welding, which even though it's not a hard skill to lean, requires equipment and space that some people don't have.

Perhaps also the lack of kits and plans with step by step instructions is also something that holds people back from all steel construction of DIY machines. You or I could do it, but also, this isn't our first time.

I do think there may be a middle ground using aluminum construction that isn't T-Slot.

Always nice to hear your points of view, and also, my opinions are simply that, and sometimes, I am mistaken.

Don't remember... but it was cheaper than buying the extrusion and sending it out to get machined. To note the surface is not ground but fly cut, but the machining was very well done. Really though, the straightness of the thick wall extrusions are pretty darn close to the tolerances needed for the linear rails, at this relatively small size, that it's probably overkill. Correction - it's overkill. We're not making machines that micro-dispense fluids or such.

As to the welding - no one is going to build a frame suitable for machinery with a $149 welder that plugs into the 110V outlet. Then there's the accessories needed, masks, welding rod, gas... Even storing welding rod properly is out of the scope of many DIYers.

I think a good compromise, for those with access, a steel weldment for a base with the gantry and Z carriage out of aluminum extrusion. You have a stiff, heavy base, along with an easier-to-assemble gantry and carriage that's lighter, and use motors that won't break the bank but still give good performance. For a small machine, if moving the machine was not a problem, the Wade-O design would be a good option if one had access to a mill.

[A_Camera]

He ride the short bus.

Not heard that expression since I left South Africa and moved to Sweden in 1980... Didn't know it was still in use. The only difference is that we said "he rides the short bus to the school".

[NIC 77]

Not heard that expression since I left South Africa and moved to Sweden in 1980... Didn't know it was still in use. The only difference is that we said "he rides the short bus to the school".

I have to admit it was a 5 beer impulse post.

Nevertheless, I really hate that guy "corvetteguy50" after watching several of his YouTube videos. I hate the condescending tone, I hate the bullying, and I hate the making fun of people who post their DIY machine vids, only to have them stolen and reposted by him, with mocking comments, and as far as I know, he's never built anything of his own and doesn't really know what he's talking about. A real pet peeve.

So, in some cases I might regret a 5 beer post, but in this case, I'll let it ride.

[A_Camera]

I have to admit it was a 5 beer impulse post.

Nevertheless, I really hate that guy "corvetteguy50" after watching several of his YouTube videos. I hate the condescending tone, I hate the bullying, and I hate the making fun of people who post their DIY machine vids, only to have them stolen and reposted by him, with mocking comments, and as far as I know, he's never built anything of his own and doesn't really know what he's talking about. A real pet peeve.

So, in some cases I might regret a 5 beer post, but in this case, I'll let it ride.

I don't know that guy and never seen any of his vids. After reading your post I tried to watch one, but it was too long for my taste and too boring. Fast forwarding didn't help... He had some valid points in that video, but that's no reason to speak condescendingly about others or mistakes others are making.

[NIC 77]

I don't know that guy and never seen any of his vids. After reading your post I tried to watch one, but it was too long for my taste and too boring. Fast forwarding didn't help...

I know your pain. I prefer 20 words or less to make a point.

[QuinnSjoblom]

Great stuff guys. It sounds like steel has huge advantages over aluminum t slot if you can get around the drawbacks. I'll look into it further, but at this point I just don't think it's feasible for me. First of all I have no welder and very minimal experience welding, not since high school 15 years ago. Also it sounds like it's pretty much required to build a frame, weld rail mount pads on, then get it surfaced flat and parallel. Unless you can somehow build a frame out of steel tube that is already flat enough for bolting rails on. A couple more things I should mention, this machine will be assembled and operated in my second story apartment. Parts will be made at work (cabinet shop) and then brought home for assembly, in my subaru legacy I should add ha! So the idea of welding a heavy steel frame, putting the whole thing in my car (probably not possible) then lugging the thing up the stairs, doesn't really sound all that convenient. I'm thinking for now, with this size machine and what I want to get out of it, it makes sense to go with aluminum. Footprint is 1000mm by 700mm, and travel around 650mm X 450mm. Really not a big machine. For cutting carbon fiber and some aluminum at the tolerances and speeds I need, I think aluminum will suffice considering the convenience and amount of time. I plan on having this thing functional within a couple months. Might not happen, but going with steel would surely lengthen that time. Like I mentioned earlier, this probably won't be the last machine I build. I'll gain some experience with this one, learn some lessons, then maybe in the future build a second much nicer machine. At that point I might be moved into a house with a garage where a steel machine would make more sense.

Thanks for all the replies guys! Learning alot

One thing to add, I will definitely look into the idea of using aluminum square tubing (non t slot) as an option. I can work with that no problem. The tools I have available at work make it no problem to work with aluminum. I can make custom brackets and whatever is needed. Even if using t slot, I was still planning on buying aluminum plate to make my own gussets. Much cheaper than buying them premade. I do have a drill press (not variable speed) and I'm assuming I can get good enough accuracy. Holes will be drill slightly larger than bolts to allow minor adjustment before torquing down

[ger21]

I'm not sure where this idea that working with steel is so bad got started.

Most guys building routers tend to have woodworking tools, but don't have the tools for working with steel. Steel is heavy, dirty, and leaves razor sharp splinters and chips everywhere. I HATE working with steel.

[QuinnSjoblom]

Most guys building routers tend to have woodworking tools, but don't have the tools for working with steel. Steel is heavy, dirty, and leaves razor sharp splinters and chips everywhere. I HATE working with steel.

Yep, and that's exactly where I'm at. I have a whole cabinet shop of tools at my disposable, pretty much none of which are suitable for working steel, all of which good for aluminum. Building from steel isn't impossible for me, but at this time I don't think it's worth the effort. Although I do fully understand a steel machine can be much more capable. I'll leave that endeavor for the future when it's more within my reach (garage, welder, proper drill press, grinding tools, etc.)

All that being said, I am considering going with steel plates instead of aluminum for sandwiching in between my linear rails and 80/20 framing. I plan on making the plate wide enough to span 2 t slots, and the rail will be centered over one of the slots with bolts passing through the plate, into t nuts. The plate will also be bolted down to the adjacent t slot next to the rail. I think this method will ensure a nice flat plane for mounting the rail, and going with steel should stiffen the frame itself. I will have to buy sized stock, probably around 3" X 1/4", and I will just have to cut to length and drill. Shouldn't be too hard to cut stock that small, even if I had to use a hack saw. My drill press isnt variable, but it has 3 positions for the drive belt. I'll have to see if rpm is low enough, otherwise I'll have to find another way to drill the holes. Just an idea at this point. May still go aluminum for these plates.

Another thing I'm curious about, is the difference in thermal expansion of aluminum vs steel something I need to worry about? Let's say I use these steel plates bolted to my aluminum frame, and my cnc is in my full enclosure. During cutting, temperature of the machine will rise. Will the aluminum expanding differently from steel be enough to slightly curl my table?

actually one more thought, if my drill press at its lowest rpm is sufficient for drilling steel, and if i can fit a metal cutting wheel to my chop saw, i may consider making my gussets from steel plate. could save some money there compared to aluminum plate and also improve rigidity a bit. With the current design im drawing now, the bottom gussets on my gantry that join the horizontal extrution to the vertical one will extend down next to the frame side and the ball nut will mount to that downward extended gusset, so needs to be really rigid right there. ill have some pics in cad soon

looks like wood miter saw doesnt work great for metal blades. same issue as the drill press, too much rpm

[louieatienza]

Freud makes a Diablo blade for cutting steel with a circular saw. If you have a 5/8" arbor on the table saw you can use it there as well. You just have to be super careful with sparks if you're doing it in a cabinet shop! The blade is a bit heavier duty with with either zero or negative rake angle for strength.

[QuinnSjoblom]

Looks like I have my work cut out for me this weekend. My mechanics show up Friday and 80/20 is doing a site wide 15% off sale Friday through monday. I need to get this thing designed by Sunday so I can place the order. I think I can do it. Pretty much nailed down my concept today so I think I can get it all finalized by sunday. I'll try to post something tonight for you guys to give input on. I really don't need the mechanics to design it, it's just that the specs provided are a bit unclear in areas so I want to see it all before I nail down dimensions. Also the fact that I'll be fabricating my own gussets from plate will make it easier to place the order. Basically just need all the extrude and a big pile of t nuts from them. Estimating about 500 bucks there so the 15% off is worth hurrying for.

[QuinnSjoblom]

for those of you that have fusion 360, heres my first draft. its not complete but at least shows my plan. i havent drawn the y or z plate, but thats pretty straight forward. i drew the x ball nut holder and end brackets for mounting the ballscrew bearing blocks. didnt take specific measurements for those yet, just drew them in to show how it works. I didnt bother adding in cad drawings of the extrude pieces yet, just drew them as outer dimension blocks. I had everything as seperate bodies in the drawing to show the different pieces but when i share it, it joins everything together. you can explode it to see the different pieces. gussets will be added on the back like i have on the front of the gantry. side gussets that extend down for the nut mount are half inch aluminum, as well as the plates sticking out at either end of the frame for ballscrew bearing block mounts. upright and horizontal gantry extrude is 50mm x 100mm, horizontal feet of the gantry are 50 x 50 with a 12mm thick aluminum plate underneath that the rail carriages attach to, side frame extrusions are 50 x 100, table is pieced together with 8 pieces of 75 x 25, end frame pieces and the 3 horizontal frame supports are also 25 x 75. Im probably gonna have to add in all the actual extrude files for the drawing to make more sense.

A360

[A_Camera]

Most guys building routers tend to have woodworking tools, but don't have the tools for working with steel. Steel is heavy, dirty, and leaves razor sharp splinters and chips everywhere. I HATE working with steel.

That´s why I like aluminium more than steel.

[QuinnSjoblom]

heres an image showing the separated pieces to make it a bit easier to understand since the link renders it as a single solid object



and heres the link again A360

like i mentioned before, its dual drive x, just showing the general idea on one side for now. The plates shown for holding the ballscrew endblocks will be braced with 45 degree brackets as well. One thing i realized is the single piece gussets on the front face of the gantry will need to be split to allow full travel of the y ball nut, but ill probably use singe peices like that on the back side. if its hard to tell from the pic, the horizontal beams have 2" x 1/4" plate covering the whole face and are flush with the gussets which are also 1/4" to allow the y linear rails to overlay both. the big side gussets and plate in between x carriages and extrude are all 1/2" plate. the plates on top of the x carriages will be drilled and tapped for the side gussets to bolt directly to them.

[NIC 77]

I think any advice I could give you would be very similar to the advice I gave to Nikdan in his thread:

http://www.cnczone.com/forums/diy-cn...-software.html

Are you planning to run the ballscrew in between the gantry extrusions? If so, a full back plate and internal ribs that are cut out to allow passage of the ball nut.

Are you planning to run the ballscrew in front of the gantry extrusions? If so, full front and back plates and internal ribs, with the possibility of filling the voids with epoxy granite, of course, you could build it and see how it works first. Or use a larger T-slot tube.

I wouldn't use T-slot for the verticals of your gantry (is that what you have drawn?) Also, I wouldn't use it to mount your bearing blocks to (is that what you have drawn or is that a solid block?).

I'd use solid blocks to make the parts that bolt to the bearing blocks, and extend the side plates upwards, eliminating the T-slot verticals. I'd make the side plates as thick as possible and reinforce them with heavy duty steel angle iron bolted at 90 degrees throughout to add stiffness with the largest gussets you can use.

To be honest, with this design, I personally would be looking at using solid aluminum profiles bolted together, and not T-slot, however this may require more equipment than you have access to and take more time to build.

I think what you have drawn could work though, not saying it's horrible or anything, but as a minimum, a full back plate and think about some ribs (or use a single larger extrusion), and I'd prefer to mount the bearing blocks to something other than T-slot.

[QuinnSjoblom]

right now im working on the model and inserting the actual extrude cad files to show a better representation. The y ballscrew will pass between the horizontal gantry beams. the bearing blocks will be mounted to thick aluminum plate, which will then be mounted to the insides of the vertical extrusions. I agree making the gussets come up higher would be better. also smaller gussets will be mounted on the inside. I would say my thick 1/2" thick gussets will be doing alot more work than the extrude itself. But to answer youre question, all linear rails, carriages, and ballscrew blocks are mounted to aluminum plate rather than the extrude. Its kind of required anyway since holes dont line up. Ill keep drawing and include all the gussets i plan on using

[louieatienza]

I will say, before you go and spend money on 80/20, look at Misumi's GFS series extrusions... I have built with both, and frankly there's no contest. You couldn't tell once assembled (I had end caps on the extrusions). But the machines I built with the Misumi GFS extrusions were a bit quieter and were a lot stiffer.