Dual Spindle CNC Router Build

[tahustvedt]

Nice machine.

If the bearing block moves a couple 1/1000" out of alignment then that could explain it, especially if they are already a little misaligned at zero. If you haven't already tried it, try shimming under the edges of the bearing mount to straighten the bearing.

[ger21]

The current most frustrating issue is that when if I have the bolts holding the ball screw nuts to the blocks loose 200 ipm is no problem on both the x and the y axis. Everything sounds smooth. If I even just finger tighten those bolts everything goes to hell. Ballscrews get harder to turn, they sounds bad going one direction (not too bad the other direction), and steppers start stalling at anything over about 20-50 ipm. Any ideas why? I thought it was an alignment thing, but with a dial gauge on the block while I tighten those bolts nothing moves more than a couple thou. Anybody else have any issues like this before???

Most likely it's an alignment issue. Alignment is extremely critical.

I seem to be able to do 200ipm rapids without any problems, which I am fine with. I would like to go higher not for practical purposes but just for show. If I go any higher than that I seem to start stalling the steppers. Would you say that the steppers are just torquing out or is it possibly resonance issues?

Not likely to be resonance. What screws and motors are you using, and what voltage?

[bakon_boy]

570oz in motors from AutomationTechnology, screws are RM2010 from Linearmotionbearings2008, and PSU is 60V.

[ger21]

Most likely running out of torque, but it could also be alignment issues. A small misalignment can cause a lot of binding. If you remove the motors, do the screws turn very easily with your fingers?

[JerryBurks]

Not so different from my setup with Gecko203V and 465 in-oz steppers (very low 1.6mH inductance, 2 steppers on X and one on Y), 72V analog supply and 15mm pitch screws.

If my machine is set up properly I can run rapids up to 1000ipm (about 1700rpm). But the steppers become really weak at that speed and it takes very little to make them stall. To be reliable I have set the max. jog and rapid speed to 280ipm and mostly machine between 80ipm and 250ipm, depending on the bit.

What I must watch most closely is the pre-load of the angular contact ballscrew bearings. It takes only a tiny amount of over-tightening and the bearing friction makes the stall speed drop seriously. The other is the proper synchronization of both X-steppers. I have them coupled with a timing belt to avoid racking and the pulleys must of course be tightened in a position that the steppers don't work against each other. But that is easy: loosen one pulley, turn the machine off and on again to get thesteppers into the same full step start position, then tighten the pulley set screw.

[louieatienza]

Looking good...

As to alignment and tightening - it's very possible your gantry uprights are not 100% parallel to each other, or at least the surfaces where the end supports mount to. The way to check would be to loosen one end support, jog the axis to that end, and see how much the block itself shifts, or if it is still normal to the upright. Once that is adjusted, loosen the other side and do the same thing. You may have to repeat this a couple of times to get both sides to line up properly. If not, then you might have to relocate the ballnut. Or, the ballnut may not be perpendicular to the screw. Also you can loosen the bolts holding the upright on one side to the bearing blocks, and see if the upright moves away from the block; you may need a shim to align it.

200ipm equals about 508rpm for your setup, and I think you might be able to coax more out of it. Also recheck all your motor connections and make sure the current settings are correct, as well as the motor leads correct.

[bakon_boy]

Couple of questions concerning programming.

1. I am wondering if there is anyway to record the expected vs actual coordinates after homing into a log file? I would like to keep record everytime the machine is homed so I can track if the motors are skipping steps.

2. I am writing a macro to handle switching between spindles and I need to store the tool number and tool height offset of the spindle that is not being used. I wrote it using Set/GetVar(XXX) and used variable 600 and 601. I couldn't find anything that says that those variables can be used but it seems to work. Any comments on that?

[ger21]

1) I think the only way to do something like this is to use the Mach3 verify function, which I believe tells you if you've lost steps from the previous homing. I've never used it, so I don't know exactly how it works.

2) There's little to no documentation on what variables can be used in Mach3 by the user. You might want to do some searching on the Machsupport forum to see if you can find any more info.

[mmoe]

Couple of questions concerning programming.


2. I am writing a macro to handle switching between spindles and I need to store the tool number and tool height offset of the spindle that is not being used. I wrote it using Set/GetVar(XXX) and used variable 600 and 601. I couldn't find anything that says that those variables can be used but it seems to work. Any comments on that?

My advice, as it was earlier in the thread, is to use the work coordinates (G54, G55, etc.) to create the tool height offset. This may not work for an automatic tool length calculation, but in practice that hasn't been an issue for me as it's rare that I need more than 2 tools. You're going to need to change work coordinates no matter what you do in order to align the heads, so may as well add the Z offset there IMHO.

[ger21]

It might be safer to add a User DRO to the screen and store the number there, rather than use the vars.

[Frankmali17]

Hey,
Great looking machine and concept...
Where did you find the PC case? Do you know where one could find one?
Also... I am very curious about you dual spindle setup.
Do you have any updates for us?
How is the dual spindle working out?
Would you make any recommendations.

[bakon_boy]

Thanks, overall I am pretty happy with how it turned out and the performance. I have been meaning to do an update but I have been to lazy so thanks for the motivation.

As far as the PC case it was kind of a stroke of luck. I picked it up pretty cheap off of Kijiji. I don't know how common they are but it worked very nicely, lots of room to fit everything.

The dual spindles work very well for kitchen cabinets which was the main driver for the doing it that way. ATC would be nice but seemed too expensive when I really only needed 2 tools at a time. I haven't tested it too much for other projects that would require more accuracy between the 2 spindles but I have no reason to think it wouldn't do well. The way that I have it programmed in Mach3 is if a M6 code comes up it checks to see what tool is being requested and compares that to what is loaded in the spindles. The outcomes are either the tool is already loaded in the active spindle, it switches spindles using an axisswap() command, or runs a manual tool change sequence.

I am planning on posting a separate update to prevent this post from getting to long.