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

[QuinnSjoblom]

changed to show all the extrude. added full back plate on gantry. still missing alot of brackets and all fasteners, just havent drawn them in

[QuinnSjoblom]

received all my linear motion guides and ballscrews yesterday. For the 500 bucks i spent, im very happy. All 4 ballscrews feel really good. No slop, at least that i can feel. wont know for sure until i actually build the machine and check for backlash, but it feels good. All bearing journals are perfect fit, all rails and screws are perfectly straight. Just one slight concern, out of the 12 linear carriages, 6 of them feel perfect, slight preload, zero slop, 3 of them basically no preload, if i twist the carriage against the rail i can kind of barely feel a bit of a tic bac and forth. The remaining 3 have a noticeable amount of slop when twisting the carriage back and forth, not a lot, but i can definately feel it moving slightly. Im thinking i could probably give some preload myself just by adjusting the distance between rails to remove any slop, but i contacted the seller to see if they can be swapped out. So far theyve been great, so i dont doubt they will send some new ones. These are labeled as hiwin hgh20ca carriages, but i would assume they are clones because of the price. worse case scenario, if they dont want to replace them, i can replace at least the 3 bad ones with genuine ones.

[ger21]

worse case scenario, if they dont want to replace them, i can replace at least the 3 bad ones with genuine ones.

There's no guarantee that genuine Hiwins would fit the rails you have. Hopefully they replace them.

[NIC 77]

We've never talked about what end supports you're going to use (for the ballscrews).

You mentioned the bearings, this is for your ballscrews?, so what did you buy FK, FF series or BK, BF series or a combination of the two?

You won't be able to tell if they are straight and machined on center unless you mount them in the bearings and spin them (perhaps you've done this?), and also, when mounted, you can feel if there is any slop / play in the fixed end, which is a common complaint with the less expensive bearing housings.

[QuinnSjoblom]

They came with bk/bf support bearing blocks. I mounted them up and tightened the nut on the fixed end. There is absolutely no end slop. It looks like they have the proper back to back angular bearings in the fixed end. With both ends of screws in the blocks, i put the blocks on the counter and spun the screw. Looks to spin true and centered.

[NIC 77]

They came with bk/bf support bearing blocks. I mounted them up and tightened the nut on the fixed end. There is absolutely no end slop. It looks like they have the proper back to back angular bearings in the fixed end. With both ends of screws in the blocks, i put the blocks on the counter and spun the screw. Looks to spin true and centered.

nice!

[QuinnSjoblom]

heres what i plan on doing for the fixed end of the screw. The 1/2" plate that the block is mounted flat against overlaps onto the frame 100mm, then a 1/2" plate mounted perpendicular to that plate and bolted to side of frame, both braced together with the top and bottom 1/4" plate that extends back to hold the plate for the stepper.



and heres a view from the back. the 1/4" plate that the linear rail sits on will also be bolted down into the edge of the 1/2" plate that overlaps onto the back

[louieatienza]

Take the three loose fitting ones, and put them on the rails that you feel a preload, and then do the opposite...

[QuinnSjoblom]

not sure what you mean. All 6 rails are very consistent, the difference is in the carriages. The loose ones are just as loose no matter what rail i put them on

[QuinnSjoblom]

heres y and z. just need to draw the screws in. z is pretty simple, i needed a 15mm spacer under z carriages to get correct mounting of ball nut mount and bearing blocks. Y gets a little trickier, just need to draw it. let me know if you guys see any issues. The spindle mount could be raised a bit to overlap the bottom z carriages for better stiffness if you think the 1/2" aluminum plate will flex. Spindle would just need to be slid down further in its mount.

[ger21]

Move the spindle mount up so that the bottom of the mount aligns with the bottom bearing, and just clamp the spindle higher on the body. This eliminates the chance of the plate flexing where it's hanging down.

My preference would be side mounted rails, tucked under the table, which will keep them a LOT cleaner.

[QuinnSjoblom]

i had exactly the same thought. Ill move the spindle mount up. Just wasnt sure if it was optimal to have most of the spindle sticking out below the mount and being held onto towards the top. Sounds like thats not an issue so ill do it that way.

About the side mounting of the rails, i thought about this alot. Definately easier to keep them clean over there, but then there was a lot of other things i considered. Getting the rails mounted properly on the side, perfectly parallel, on the same plane, correct distance apart, just seems a LOT harder to do on the side vs the top. With the rails on the top, i can start by making sure my aluminum mounting plates are on a perfect plane, which wont be difficult to check, then start by tightening one rail down paralell to the frame, install other rail with screws left loose, install gantry and slide it back and forth to position the second rail perfectly parallel, then tighten the bolts. For side mount, this whole process seems to get a lot more complicated. If the outside edges of the frame arent precisely parallel, i would have to shim under the rail to adjust. A solution would be having the end supports of the frame run long with the side supports running into them, then i could adjust width and parallel by adjusting position of the side frames on the end frames. But this creates alot of other inconveniences, like all cross members under the frame would need a slight gap left at the end so they are all adjustable with their 90 degree brackets, also i cant connect the t slot table to the side frames with the method i used. Even after addressing the issue of width being adjustable, i still have to deal with adjustment for rails running parallel to the table surface. If minor adjustments need to be made, i have to deal with the weight of the whole gantry pulling them down once i loosen the rail bolts Just sounds like a nightmare in terms of getting everything dialed in perfectly when all these things become very simple to do with the rails on top. I do plan on having vertical plates coming up a couple inches next to the carriages on the inside. I could even have detachable plexiglass pieces that attach to those plates to come up even higher. They could come up almost as high as the lower gantry cross beam, just has to come right up to the lower gusset edge. Obviously limits y travel when i have the plexiglass on, but my travel is much larger than needed for most of what im working on. For me it just seems like a good trade to put the rails on top for ease of building and adjustment.

[QuinnSjoblom]

I also have a question on the spindle choice. Earlier it was mentioned that the er20 collet size on the 2.2kw is much more convenient, but heres the issue, Ill be running this in my apartment so to begin with 220v is not an option. On top of that, a 20 amp breaker is all i have available for where ill have the machine. I can reach another outlet on a separate circuit so the spindle will have the 20a breaker to itself. For the 1.5kw, it sounds like 20a is plenty with some overhead. 1.5kw at 110v is about 14 amps. If i was to run a 2.2kw, thats exactly 20 amps at full load. Will i even be getting close to pulling that many amps? im not gonna be hogging plywood or running huge cutters at high feed through aluminum. Would i most likely stay well bellow the max 2.2kw rating? Should i just stick with the 1.5kw to be safe? I really think its plenty enough, but is the er11 collet gonna be a big limiting factor for what im doing? Keep in mind ill mostly be cutting 2mm carbon sheets and some small aluminum parts so i cant image ill need to run huge shank cutters.

[ger21]

The 1.5Kw will probably be fine.

[NIC 77]

I also have a question on the spindle choice. Earlier it was mentioned that the er20 collet size on the 2.2kw is much more convenient, but heres the issue, Ill be running this in my apartment so to begin with 220v is not an option. On top of that, a 20 amp breaker is all i have available for where ill have the machine. I can reach another outlet on a separate circuit so the spindle will have the 20a breaker to itself. For the 1.5kw, it sounds like 20a is plenty with some overhead. 1.5kw at 110v is about 14 amps. If i was to run a 2.2kw, thats exactly 20 amps at full load. Will i even be getting close to pulling that many amps? im not gonna be hogging plywood or running huge cutters at high feed through aluminum. Would i most likely stay well bellow the max 2.2kw rating? Should i just stick with the 1.5kw to be safe? I really think its plenty enough, but is the er11 collet gonna be a big limiting factor for what im doing? Keep in mind ill mostly be cutting 2mm carbon sheets and some small aluminum parts so i cant image ill need to run huge shank cutters.

Do you have a stove or dryer running on 220V in your apartment?

IMO, a 2.2 kW spindle should output 2.2kW of power, not have 2.2kW as an input. ie:110 x 20. But perhaps that's the way the Chinese have done it to make it sound better than it is?

More power is better, but your machine needs to be more rigid to take advantage of it. ger21 may be right in this case, perhaps the 1.5kw is the best choice for you. Personally I'd look for a way to use 220V.

If you are mostly planning to cut carbon sheets, well, CNC makes alot of mess, dust, chips, etc, I am worried for your health breathing carbon dust, especially from a machine that will be throwing it out in the apartment where you live. I don't know the best way to handle this.

[QuinnSjoblom]

yes, i mentioned it a couple times. Im very aware of the hazards of carbon fiber dust. I will use 2 levels of dust collection. I will have vacuum collection right over the spindle, and my whole cnc will also be enclosed in a plexiglass box. The box will also have air being pulled through it with filtration for the exhaust. After a cut is done, i will leave the box sealed for a while as it cycles air through and captures the carbon dust. The enclosure will also be important for noise. I cant have it too loud since i have neighbors on both sides and below me. Also going to somehow dampen the machine from the floor to not transfer vibration.

As for the spindle, im thinking most likely ill go with 1.5kw. After looking into some tooling ill need, max 7mm shank will be fine.

[QuinnSjoblom]

Heard back from eBay seller. They are replacing all the carriages that have a little slop. Probably would have been OK if I added preload by spacing the rails properly, but might as well get them swapped. Last night I mocked up all 4 screws in their blocks and checked for spinning true at both ends. All are perfect. Very happy with everything I got

[QuinnSjoblom]

So I also need some advice on cutting out all my 1/2" and 1/4" aluminum plate into gussets. I have a couple ideas in mind. Tools at my disposal are a table saw, chop saw, 23,000 rpm, 1.2kw porter cable router with 1/4" and 1/2" collets, drill press, and a cnc at work that I could cut mdf templates out of. One idea is to basically cut the plates the same way a cnc would, but by hand with a router. The router base has inserts that fit in the bottom for use with patterns. I could draw my gussets and inset the whole outside edge inward by the distance the insert holds the cutter away, then cut these patterns out of mdf on our cnc. I would just need to choose a proper end mill to put in the router that would do well cutting aluminum at 23k rpm. The other option is cutting the aluminum pieces slightly large with table saw and chop saw, then cleaning the edges with pattern and router.

The obvious best way would be to just cut my plates on our cnc which I'm sure would do just fine on our cnc, but I would never talk my boss into letting me do that. So what kind of success do you think I would have with an end mill in a router following a pattern? If this will work well, what type of end mill do I want for running 23k rpm?

[ger21]

The other option is cutting the aluminum pieces slightly large with table saw and chop saw, then cleaning the edges with pattern and router.

I'd use this method. Does the shop you work at have any larger, variable speed routers you can borrow? I'd make the templates actual size, and use pattern router bit with a bottom bearing on it. If you can cut about 1/16" oversize, it's very easy to route to size.

[QuinnSjoblom]

I'd use this method. Does the shop you work at have any larger, variable speed routers you can borrow? I'd make the templates actual size, and use pattern router bit with a bottom bearing on it. If you can cut about 1/16" oversize, it's very easy to route to size.

Ok, I can do it that way, but no, we don't have any lower rpm routers. It will be a little trickier doing it this way with a couple of my parts that won't be easy to cut with chop saw and table saw. What if I do most of it this way and just the more complex parts with the end mill and insert in router with undersized pattern? Will that other method still work? It would just he nice if I could do it that way for a few specific peices.