steel frame gantry flex

Hi all,

After a long time away from all things CNC i've finally got back into it and have become so pre-occupied that my wife reckons i'm having an affair CNC.

Any way I was drilling a grid of holes in the bed today to take hold down bolts and hit a few issues:

1. I was getting few random limit switch faults. I have used shielded cable for all steppers and for the VFD, but not for the limit switches. Could this be the cause???

2. I was doing simple plunge movements into the MDF bed in a 100mm x 100mm grid with an 8mm router bit fitted to the spindle (its all i've got at the moment). There is enough flex in the gantry that you can see it move and it results in an elongated hole instead of a round one.

Does anyone have any tips on how to stiffen the gantry. it is 150mm by 50mm by 2mm RHS tube

Attachment 190904

2mm wall thickness is considered thin wall for CNC machine construction purposes. Most of us building steel tubing CNC machines use 3mm (minimum) to 6mm wall thickness. For your gantry beam I would recommend substituting thicker steel tubing or the nearest equivalent of 150 mm x 50mm aluminum extrusion.

Your frame could be made stiffer by adding triangular gusset plates at the appropriate corner joints, and/or additional tubing to tie the ends of the frame together at the top and near the bottom of the legs.

Using shielded cable for the limit switches is a common practice for eliminating glitches. It doesn't work unless the shield is properly grounded. There is a setting in Mach3 that can be used to reduce the sensitivity to the glitches. First, try moving your existing wiring away from all other wiring as much as you can and twist the pair of wires to each switch in a tighter spiral to see if it helps enough to get rid of the glitches. This includes inside your control box where the limit switch wiring comes near other electrically noisy wiring that may be cross coupling to the limit switch wiring.

You could try increasing the debounce setting in Mach3 to maybe 5000 and see if that helps with your limit switch issue. Shielding is a better fix, but the debounce should work unless you have really bad noise issues.

As for the flexing? Drilling puts enormous stress on a machine.
Is the tube itself flexing, or is the entire gantry moving? (Flexing at the bearings) Ideally, you'd want to use a much larger and heavier tube, and use THK type rails rather than the supported round rails. Pinpointing the exact location of the flexing should dictate the best course of action.
FWIW, I've seen a 1000lb gantry move a substantial amount when drilling 1/2" holes on a $150K router.

Quote:

---

Originally Posted by wentyfalls

Does anyone have any tips on how to stiffen the gantry. it is 150mm by 50mm by 2mm RHS tube

---

Barring starting over with thicker tubing, welding steel angle to the backside might help cut down on the torsion you might be getting.

Quote:

---

Originally Posted by wentyfalls

Hi all,

After a long time away from all things CNC i've finally got back into it and have become so pre-occupied that my wife reckons i'm having an affair CNC.

Any way I was drilling a grid of holes in the bed today to take hold down bolts and hit a few issues:

1. I was getting few random limit switch faults. I have used shielded cable for all steppers and for the VFD, but not for the limit switches. Could this be the cause???

2. I was doing simple plunge movements into the MDF bed in a 100mm x 100mm grid with an 8mm router bit fitted to the spindle (its all i've got at the moment). There is enough flex in the gantry that you can see it move and it results in an elongated hole instead of a round one.

Does anyone have any tips on how to stiffen the gantry. it is 150mm by 50mm by 2mm RHS tube

Attachment 190904

---

Looking at your setup, and you mentioning elongation...is the elongation parallel to the y axis? Kinda bet it is due to the x bridge torquing back when drill thrust applied... Stiffening with heavier tube is a lot of rework, suggest putting a couple indicators from top of bridge to tops of your side rails to see if the whole bridge is tipping away from the spindle...if so, you might find a easier rework would be to extend your y/bridge bearing mount, and move the spindle side linear bearing outward, ideally a little past your spindle centerline so the thrust is directly trying to lift the bearing, instead of twisting it with the spindle/bridge/z axis stackup height becoming a moment arm... But just a guess, never played with a tube frame router. As to stiffening tubing, i have seen some cheep tube framed machines that were filled with cement to add some rididity and mostly some damping mass.
Lastly, how is that 8mm cutter ground? Is it center cutting on just one flute, etc... A chipped center cutting tool will create huge thrust loads... Try doing helical interpolation plunge, even just a little can greatly reduce thrust loads, and improve hole roundness...