[Nipels]

So, now that I have finally pulled the trigger on some Chinese Linear Rails and Ballscrews, I thought it was time to start a thread. I mostly want to dedicate it to the questions I will have leading up to the final build. So here goes.

This is what I have ordered as of now.

3 x RM1605-950mm Ballscrews with Ballnuts and ends machined
1 x RM1605-400mm Ballscrew with Ballnut and end machined
4 x BK12 with locknuts
4 x BF12
4 x Ballnuts housings/brackets
4 x SBR20-950mm Supported Rails
2 x SBR20-400mm Supported Rails
12 x SBR20UU Bearing Blocks

I am waiting for the parts to get here before I begin the final designs. I plan on following much of this design by 2rdesigns.

My first question is, what should I build the machine out of? I am ruling out steel. As of right now, I am looking at 80/20 extrusion of simply aluminum flat stock. I am leaning towards the aluminum flat stock. I have a drill press in my garage and access to a large band saw at work that cuts steel like butter. It should handle aluminum no problem. So...

Would Mill finish Aluminum plate be a good material to build the machine out of? I don't need super crazy tight tolerances. Was thinking 1" x 4" flat stock.

I will probably incorporate some extrusion into the build to mount the linear rails to. I believe the distance between the mounting holes is 30-32mm. I know the 30 series extrusion has the right slots to mount the rails, but it is kind of light. Any ideas on mounting the rails?

I was thinking about the 380oz stepper / g540 over at cncrouterparts. Any feedback on this setup? Are the 320oz low inductance a better choice?

What kind of computer should I get for the setup?

I plan on buying the chinese water cooled Spindle for 300$ off ebay that comes with the VFD. I have searched this forum and read the feedback but any extra would be much appreciated.

Software! I have solidworks already and plan on getting mach3.

Couplers! Oldham?

My overall budget is 2-3k. I would like to stay closer to 2k.


That is all for now, I will keep a notepad document and periodically come back here to ask questions. Any and all feedback is appreciated! Thank you in advance.

OH! I plan to cut plastics and wood, mostly MDF to start with. I dont have any definite plans to cut aluminum, but it would be nice if the machine could handle it.

[Walky]

Check the Misumi 60x60mm "heavy" extrusion (30mm slot separation), they should be a good match for the rails, and are thicker than other extrusion such as Bosch Rexroth and 8020. 60x30 would not be strong enough for a gantry that long IMHO, unless maybe you reinforce it. I'd still reinforce the 60x60 Misumi extrusion with thick plate if you want to cut aluminum, though (I'm thinking about one separate extrusion for each rail, and a back plate joining them). 1" aluminum is more than enough for the kind of machine you are aiming for, assuming the design is good. I use a depth micrometer to make sure that the rails (most especifically, the aluminum supports) are parallel to the extrusion, but a simple homemade guide will work too if you don't have one, this is because the slots and the mounting holes will allow for some adjustment play.

Compare the motors' torque curve, since the ratio is different from R&P to 1605 ballscrews the performance might vary. In this particular case I'd say it's most likely that the lower inductance motors will perform better anyway as they'll keep torque at a higher speed and the low lead ballscrews need that. On the other hand, because of the length of the screws you won't be able to go as fast as the motors can be, because they will whip badly if you do. This will affect your rapids. For this kind of build if would be better to get 1610 for X and Y (not for Z as it would have a tendency to backdrive), in case you're still at time and are willing to pay the difference (they are a little more expensive).

This build should be good for inspiration too:
http://www.cnczone.com/forums/spanis...imera_cnc.html

The chinese VFD's seem to be hit or miss (at least the huanyang, which is the most common of them), a lot of people get an Hitachi X200 and say it's far better while still affordable. While some people have has problem with the chinese spindles, actually most seem to have had a good experience with them. Even if you have problems with a spindle, a USA made one could easily cost you the same as three chinese ones, so for the same money you can get a couple of spares and having all of them fail seems quite unlikely. It's a good idea to check the spindle's internal connections as many people have reported problems like ground not being internally connected for safety).

[Nipels]

Check the Misumi 60x60mm "heavy" extrusion (30mm slot separation), they should be a good match for the rails, and are thicker than other extrusion such as Bosch Rexroth and 8020. 60x30 would not be strong enough for a gantry that long IMHO, unless maybe you reinforce it. I'd still reinforce the 60x60 Misumi extrusion with thick plate if you want to cut aluminum, though (I'm thinking about one separate extrusion for each rail, and a back plate joining them). 1" aluminum is more than enough for the kind of machine you are aiming for, assuming the design is good. I use a depth micrometer to make sure that the rails (most especifically, the aluminum supports) are parallel to the extrusion, but a simple homemade guide will work too if you don't have one, this is because the slots and the mounting holes will allow for some adjustment play.

So, for the most part, a piece of extrusion reinforced with a 1" piece of aluminum on one side should be good? I mean, until I design something in CAD, this is all just spit balling. I am not quite sure what you mean about making the rails parallel with the supports. I can kind of picture it, just not 100%.

Compare the motors' torque curve, since the ratio is different from R&P to 1605 ballscrews the performance might vary. In this particular case I'd say it's most likely that the lower inductance motors will perform better anyway as they'll keep torque at a higher speed and the low lead ballscrews need that. On the other hand, because of the length of the screws you won't be able to go as fast as the motors can be, because they will whip badly if you do. This will affect your rapids. For this kind of build if would be better to get 1610 for X and Y (not for Z as it would have a tendency to backdrive), in case you're still at time and are willing to pay the difference (they are a little more expensive).

I don't think I am going to change screws. I will make due with what I have got. I have seen successful machines with longer screws of the same type. I thought for a 1610 would lose precision?
What is whip? I have been all over these forums without a clear answer. It is always talked about as if the reader already understands what it is! I am sorry if this is a ridiculous question!

What are rapids? Again, same situation. Is this when you are roughing? Or jogging? I understand that a lot of CNCers will make a roughing pass, and than a more precise finish pass. Or are rapids just the jog speed between cuts?

If you don't think the G540/320oz steppers are a good fit, do you have any suggestions for different ones? I appreciate you taking the time to answer my questions.

This build should be good for inspiration too:
http://www.cnczone.com/forums/spanis...imera_cnc.html

I WISH IT WASN'T IN SPANISH! LOL. I love the build though! It might be a little bigger/bulkier than I was hoping, but I will try to incorporate it into the design. It is an amazing machine.

The chinese VFD's seem to be hit or miss (at least the huanyang, which is the most common of them), a lot of people get an Hitachi X200 and say it's far better while still affordable. While some people have has problem with the chinese spindles, actually most seem to have had a good experience with them. Even if you have problems with a spindle, a USA made one could easily cost you the same as three chinese ones, so for the same money you can get a couple of spares and having all of them fail seems quite unlikely. It's a good idea to check the spindle's internal connections as many people have reported problems like ground not being internally connected for safety).

Thank you. I will definitely check the ground when I purchase one.

Edit: What kind of computer specs should I be looking for? I want a dedicated computer for the machine.

[PaulRowntree]

What is whip? I have been all over these forums without a clear answer. It is always talked about as if the reader already understands what it is! I am sorry if this is a ridiculous question!

What are rapids? Again, same situation. Is this when you are roughing? Or jogging? I understand that a lot of CNCers will make a roughing pass, and than a more precise finish pass. Or are rapids just the jog speed between cuts?

You can only spin a shaft so fast before the centre-of-mass will start to wobble off the shaft axis, and it starts to bend. The farther it bends, the more the mass is off-axis, and the more it will bend... you get the idea. This is whip, and the max rotation speed is set by the shaft material, diameter and the way that you have fixed the ends. This in turn sets an upper limit on how fast you can move the machine, and is why higher speed machines have low- turns per inch screws. It is also why people experiment with rotating nut designs, where the shaft is fixed.

Rapids are for bouncing around the table between cuts, and for bragging rights. It is a measure of speed, not power, nor precision.

[awerby]

My first question is, what should I build the machine out of? I am ruling out steel. As of right now, I am looking at 80/20 extrusion of simply aluminum flat stock. I am leaning towards the aluminum flat stock. I have a drill press in my garage and access to a large band saw at work that cuts steel like butter. It should handle aluminum no problem. So...

[Steel is good for building the table, and any parts that don't have to move. It's cheaper than aluminum and more rigid; the only problems are that it's heavier and not quite as flat (unless it's flattened by some machine process.) I don't see the point of using 80/20 extrusion of flat stock; it's more expensive for no particular gain. The reasons you'd use an extrusion is to get a wider cross-section with little extra weight, and perhaps to use the t-slots.

Would Mill finish Aluminum plate be a good material to build the machine out of? I don't need super crazy tight tolerances. Was thinking 1" x 4" flat stock.

[That seems too heavy for most purposes. Box beams would be stiffer for the same weight.]

I will probably incorporate some extrusion into the build to mount the linear rails to. I believe the distance between the mounting holes is 30-32mm. I know the 30 series extrusion has the right slots to mount the rails, but it is kind of light. Any ideas on mounting the rails?

[Most extruded aluminum is flat enough. I've used angle stock for that.]

I was thinking about the 380oz stepper / g540 over at cncrouterparts. Any feedback on this setup? Are the 320oz low inductance a better choice?

[Probably, if your gantry isn't too heavy.]

What kind of computer should I get for the setup?

[An XP Pro machine with a parallel port and separate graphics card, and 1+ gz processor would work for not much money.]

I plan on buying the chinese water cooled Spindle for 300$ off ebay that comes with the VFD. I have searched this forum and read the feedback but any extra would be much appreciated.

[Most people like those spindles much more than the VFDs that come with them.]

Software! I have solidworks already and plan on getting mach3.

[You'll still need a CAM program of some sort, the choice depends on what kind of parts you plan to make.]

Couplers! Oldham?

[They should work, but you need to keep the shaft from backing off them somehow.]

My overall budget is 2-3k. I would like to stay closer to 2k.

[Good luck with that...]

Andrew Werby
www.computersculpture.com

[Nipels]

Mistake, didn't see your responses in the quote.

[Nipels]

You can only spin a shaft so fast before the centre-of-mass will start to wobble off the shaft axis, and it starts to bend. The farther it bends, the more the mass is off-axis, and the more it will bend... you get the idea. This is whip, and the max rotation speed is set by the shaft material, diameter and the way that you have fixed the ends. This in turn sets an upper limit on how fast you can move the machine, and is why higher speed machines have low- turns per inch screws. It is also why people experiment with rotating nut designs, where the shaft is fixed.

Rapids are for bouncing around the table between cuts, and for bragging rights. It is a measure of speed, not power, nor precision.

Oh, well i don't care about bragging rights. Thank you for answering that question. It has been nagging on me for a long time.

Sent from my DROID RAZR using Tapatalk 2