[chrisinestes]

Short story: I want to build a lightweight, stiff, and fast gantry for my existing machine. My goal is 200 IPM carves that don't show oscillation marks due to machine flex and/or backlash.


More details: I have a 25 year old 4' x 8' Vytek Rebel router that's painfully slow on 3D carves. Because the gantry alone is around 400lbs, I get oscillation marks in the carve if I have the speed up much past 60-70 IPM. This is due to flex in the Z drive carriage, a 45lb spindle, and 25 years of wear on the machine. So, after seeing a lightweight, and fast machine at the Vectric group meeting in Denver last week, I was inspired to build a light, stiff, and fast gantry for my bulldog of a base & bed.


The gantry (X-axis) is essentially a rack & pinion, which drive from both sides of the gantry. The rack is a 75mm wide steel reinforced 5 TPI precision timing/power-transmission belt. The pinion is a matching toothed pully driven by a stepper with a drive shaft to the other side of the gantry. There's no measurable backlash in the gantry, and it runs within .001" - ,002" from where it's supposed to be anywhere on the bed. The gantry runs on 3/4" round linear rails with 2 linear bearings per side.

The Y and Z axes are ball screw. I have about .001" backlash in the Y axis, The Z axis is much worse at .008", which is accommodated for in Mach3.

I'd like to use my current linear rails. They're strong, and already tuned to the base. The rest I figure will be new. I'm thinking of a dual motor rack & pinion drive for the gantry, and single rack & pinion for Y, and Z.

Is my 45lb 5 HP Perske spindle too heavy to use to get to my desired non-floppy 200IPM?

This seams possible, but having not built a machine, I'm not sure where to start. I need help with parts selection and design.

Thanks for reading, and where should I start?

Chris

[peteeng]

Hi Chris - Read the Maximus thread it will answer some of your questions about the size of the gantry needed. Your 75mm belt rack sounds interesting, any photos? Peter

[chrisinestes]

Hi Chris - Read the Maximus thread it will answer some of your questions about the size of the gantry needed. Your 75mm belt rack sounds interesting, any photos? Peter

Thanks Pete. A search for "Maximus" resulted in about 1300 results... Can you point me to the specific thread you're talking about?

Here are a few pics of my belt rack & pinion with the gantry cover off. The upper shaft is dual purpose... it's the drive shaft to the other side, and it's the upper support for the Y & Z axis carriage. I suspect my new gantry will have to be dual motor drive in order to minimize weight and complexity.

Chris

[peteeng]

Hi Chris - Here's the thread https://www.cnczone.com/forums/austr...76822-cad.html

The drive is called a U drive or an omega drive. Nothing wrong in principle with the mechanics (on review not so true) , perhaps a good overhaul and review of the machine may get what you want? 7"/min is slow, how fast can it go on straights. Maybe its just your cam software. What do you use for CAD CAM and velocity planning? If your CAD/CAM for instance is breaking lines into very small pieces and your velocity planner has various settings then this is limiting your speed. 200"/min (5000mm/min) is not fast either so I think we can get there with what you have.

It has a 75mm belt on both sides? Don't forget your glasses I have mine tied to my head...Peter

rethinking and rereading, its the dual purpose shaft that maybe the weak link. These things whip and the shaft is not really stiff enough for a machine of that size as a support... hmmmm interesting... and the shafts run on bronze bearings? These need running clearance and its 25 years old... yes there are fundamental issues here...

Maybe you maintain the belt drive but place a new gantry beam and linears for the gantry transport? So the belts only do one thing...Do you have a CAD system? So you can develop the ideas? i like the 75mm belts and you say that the X works fine so keep it? There are issues with R&P as well so keep everything that works.

[chrisinestes]

Hi Chris - Here's the thread https://www.cnczone.com/forums/austr...76822-cad.html

The drive is called a U drive or an omega drive. Nothing wrong in principle with the mechanics (on review not so true) , perhaps a good overhaul and review of the machine may get what you want? 7"/min is slow, how fast can it go on straights. Maybe its just your cam software. What do you use for CAD CAM and velocity planning? If your CAD/CAM for instance is breaking lines into very small pieces and your velocity planner has various settings then this is limiting your speed. 200"/min (5000mm/min) is not fast either so I think we can get there with what you have.

It has a 75mm belt on both sides? Don't forget your glasses I have mine tied to my head...Peter

rethinking and rereading, its the dual purpose shaft that maybe the weak link. These things whip and the shaft is not really stiff enough for a machine of that size as a support... hmmmm interesting... and the shafts run on bronze bearings? These need running clearance and its 25 years old... yes there are fundamental issues here...

Maybe you maintain the belt drive but place a new gantry beam and linears for the gantry transport? So the belts only do one thing...Do you have a CAD system? So you can develop the ideas? i like the 75mm belts and you say that the X works fine so keep it? There are issues with R&P as well so keep everything that works.

Well, I thought I replied to this 3 days ago... I guess I screwed up!

I use Vectric Aspire for CAD/CAM, and Mach 3, Gecko G540, and a Warp 9 USB smoothstepper for machine control, I use a Mach 3 arcs post processor, and don't think it's jerking because of short line segments instead of smooth arcs.

Yes, belts on both sides... Not a good place to rest a tape measure!

The shafts definitely deflect when I pull or push on the spindle. I don't think the drive shaft spins fast enough to whip, but maybe if I could run as fast as I want, they would. The bronze bushings are warn... I get the little click of clearance being taken up when I pull on the spindle, before the shafts flex. Another issue is that the spindle weighs around 45lbs, so that's a lot to get moving and stopped again.

It's be cool if I could get mine ridged enough to run 200 IPM without noticeable deflection, but I have my doubts. I'd love to hear your ideas,s though.

Thanks!
Chris

[wizard]

Are you certain thAt your issues are weight in the gantry? Frankly it could be a lot of things from detuned servos to backlash and a bunch of other possibilities. 200 pounds really isn’t a lot, lighter weight will help but making something truly light weight and strength RF is non trivial. With the right controls and linear components you should be able to move that gantry around pretty fast.

[ger21]

This is due to flex in the Z drive carriage

A new gantry won't make this go away, unless you'll be using a new Z axis as well?

[peteeng]

Hi Chris - I'd investigate the possibility of using the belt system as is and make a new gantry and z axis. If you have CAD you can model options up and evaluate them. If you can't make a good go at that it, then it's a new gantry and drive from scratch. As wizard says don't get hung up on weight the motors will deal with that...Cheers Peter

[chrisinestes]

Wizard, I guess it's not so much simply the weight of the gantry as it is how it's built for the weight that it is. It weighs about 450lbs, not 200lbs. The spindle alone is around 45lbs. The center of the spindle/bit is a full 7" from the center of the rails it runs on. So a little flex there = a lot at the tool tip. The rails that the Y & Z carriage run on are 2" in diameter, and the centers are 14" apart. The center of mass of the spindle is, on average, only about 2-3" above the lower rail center, so acceleration/deceleration twists that carriage, which shows up in the cut. You're correct, the bushings that the Y & Z carriage run on are warn, so there's some backlash there. I also just discovered that the top rail, that's also the drive shaft for the other side of the gantry doesn't run quite true. There's a little wobble in it. I'll try to measure to see how much it's off... a couple thousandths there translates to a lot more at the tool tip. It's a stepper system, not servo. It'll run 240 IPM, not cutting, without skipping steps.

Gerry, You're correct... I'd also build a new Y & Z carriage. That's where most of my troubles are coming from, anyway.

Peter, I think maybe using my current drive system, and side plates, and stuffing new rails between them for the Y & Z carriage to run on may be the way to go. I have about 5" between the bed and the bottom of the gantry, but would like a couple inches more. But.... If I could get enough Z travel to use the whole 5", it's something I can live with. I have Fusion 360, but find it difficult for my old brain to learn. It's just enough different than Vectric Aspire and my other sign making CAD, Flexi, that I give up. I recently found Tinkercad, so maybe I'll give that a try. I may end up drawing it in 2D, and making it 3D in my head. I also thought about a lighter spindle, but if I can get stiff rails for the new carriage to run on, then maybe it's OK.

Thanks for all of the input, it really does help. I think now, I don't need a full new gantry, just new rails between my side plates, and a new Y & carriage. That'd be a lot easier to do.

Chris

[chrisinestes]

Due to a foot injury, and pending surgery, I haven't made much progress. I did measure the runout on the top driveshaft/ Y & Z axis support, and to my surprise, it's got about 0.015" of wobble measured in the middle of the shaft. That wobble is transferred directly to the bit. That's got to be some part of my not-great carves.

If I put new cross rails between my side plates, then the new Z carriage can run on those, and that shaft will only have to drive both sides of the gantry. I don't think it spins fast enough for that wobble to make a big difference in the performance of the machine if it's only being used for power transmission.

It's little progress, but it's not zero.

Chris

[ironelephant]

Hi Chris, Do you happen to know the belt sizes for the Vytek Rebel 2? john(at)johnmcgall(dot)com THANKS!

[chrisinestes]

Hi Chris, Do you happen to know the belt sizes for the Vytek Rebel 2? john(at)johnmcgall(dot)com THANKS!

Hi John, I only know what my gantry drive belt size is... 75mm wide, 5 TPI.

[Dock7676]

@chrisinestes, I know this is an old thread, but I just picked a rebel up myself and was wondering if you have a source for the 75mm belts? Before I begin refurbishing this machine, I just want to make sure they are available. A google search did not reveal any products.

How did your conversion x/z axis conversion go? If you did it, any pictures?

Thanks in advance!

Andrew