Z axis design query

[richard-at-home]

It's not actually. It just started out as that during the initial design concept. It's a moving gantry machine. Look at page 54, and the corresponding YouTube channel.

Ha! Yes, I am on page 27 now and have seen the bonkers welding of a million triangular bits of steel plate!

I really like the design of the machine though, it has a deep Z axis and high Y rails which mean no bendy gantry sides. He's also made the spindle plate into a box as per my sketch!

Lots of internal bracing to keep it all rigid.

Access to a big mill to get the important faces flat and square was clearly vital, but if that was not an option, i guess epoxy beds for the rails might do?

Just another 50 pages to read!

Cheers!

[richard-at-home]

Having read the whole 72 pages of posts on Rider63's build, I though it was interesting that he is using 2.2Nm steppers on a 900kg machine, with a 75kg gantry and 35kg Z.

In fact the steppers, although nema 34, look tiny...

It seems that low inductance was key to getting the fast feed speeds he wanted.

His machine appears to plough through ali and does steel pretty well too!

[NIC 77]

Having read the whole 72 pages of posts on Rider63's build, I though it was interesting that he is using 2.2Nm steppers on a 900kg machine, with a 75kg gantry and 35kg Z.

In fact the steppers, although nema 34, look tiny...

It seems that low inductance was key to getting the fast feed speeds he wanted.

His machine appears to plough through ali and does steel pretty well too!

He started out with Nema 34s but then changed to 400W servos. He wouldn't be able to get those super fast accelerations with steppers. Those weights sound a bit low to me, perhaps that is without all the motors, etc. You could always ask him, he still checks his thread. I asked him a couple questions not too long ago.

On my build, I am using a moving gantry, the gantry will weigh about 800 lbs when all done, but that includes everything, z axis, motors, etc, everything that moves with the gantry. With two Nema 34's 960 oz in low mH motors to drive the gantry using 10mm ballscrews I am aiming for around 0.2G acceleration with a max speed of 300 IPM. 0.15G would be just fine for my purposes probably. With a servo driven belt reduction I could get much faster speeds and acceleration. But you also have to look at the critical speed of the ballscrew.

The weight isn't so much of a problem as you might think. Choosing the right pitch of ballscrew is more important than a few extra pounds. I can easily slide my gantry back and forth with one hand in it's current 400lb configuration (not finished yet). As long as it's level it's not a problem. It's the 100 lb spindle and heavy Z axis that really add the weight to my build.

For the Z axis, if you get too heavy, you'll need a pneumatic counterbalance.

I'm not suggesting you copy exactly what someone else has done, but the two examples I mentioned are finished and have proven that they cut aluminum well. Both examples also use some kind of filling, epoxy granite, or what not, and have internal ribs in the structure. One has bolted aluminum plate and the other has welded steel plates for the gantry. You may choose to do something that is easier to build.

What is your experience level, have you built a CNC before? Do you have welding equipment, etc? What size of a machine are you building?

If you are unsure about what to do, you can always do a design log first to ask for advice. As you have seen, much can change during a design log.

[richard-at-home]

He started out with Nema 34s but then changed to 400W servos. He wouldn't be able to get those super fast accelerations with steppers. Those weights sound a bit low to me, perhaps that is without all the motors, etc. You could always ask him, he still checks his thread. I asked him a couple questions not too long ago.

On my build, I am using a moving gantry, the gantry will weigh about 800 lbs when all done, but that includes everything, z axis, motors, etc, everything that moves with the gantry. With two Nema 34's 960 oz in low mH motors to drive the gantry using 10mm ballscrews I am aiming for around 0.2G acceleration with a max speed of 300 IPM. 0.15G would be just fine for my purposes probably. With a servo driven belt reduction I could get much faster speeds and acceleration. But you also have to look at the critical speed of the ballscrew.

The weight isn't so much of a problem as you might think. Choosing the right pitch of ballscrew is more important than a few extra pounds. I can easily slide my gantry back and forth with one hand in it's current 400lb configuration (not finished yet). As long as it's level it's not a problem. It's the 100 lb spindle and heavy Z axis that really add the weight to my build.

For the Z axis, if you get too heavy, you'll need a pneumatic counterbalance.

I'm not suggesting you copy exactly what someone else has done, but the two examples I mentioned are finished and have proven that they cut aluminum well. Both examples also use some kind of filling, epoxy granite, or what not, and have internal ribs in the structure. One has bolted aluminum plate and the other has welded steel plates for the gantry. You may choose to do something that is easier to build.

What is your experience level, have you built a CNC before? Do you have welding equipment, etc? What size of a machine are you building?

If you are unsure about what to do, you can always do a design log first to ask for advice. As you have seen, much can change during a design log.

Hi Nic,

I'll check that build log again to see where he switches to servos....(Looks like Jan 2016 - even so all the videos prior look like it's chewing through ali no prob and he milled the steel bed with a 0.5mm cut at 10m/min) I did learn some good french sayings by reading it, like 'don't push granny into the nettles!' and the google translation of a face cutter appears to be strawberry cake!

Anyway, you're right its a great build.

I have no experience with diy cnc at all, I own a small lathe and a knee mill and mig gear. I'm a software guy, but I'm not phased by getting my hands dirty!

My aim is for a machine that's bigger than a 6040 and smaller than 8'x4' !!! 1500x750x500mm maybe?

I have not laid hands on a linear guide so i don't have a feel for how easily they move, but those large masses must need a fair bit of torque to accelerate quickly, but of course the ballscrew gives a huge mechanical advantage.

It's an interesting balance of ratings/specifications/forces etc to get the whole machine optimal and I'm looking forward to learning more about it, but I don't want to wast too many $$$$'s making rookie mistakes, especially with the cost of some components.

I'll put a rough design up for comment at some point, when I have a better handle on things!

In the mean time, I'll keep picking brains and reading!

Cheers!

[skrubol]

Hi Gerry, yes that's for a single carriage, so 4 will be a lot stronger, but am I right with my understanding of the datasheet? 200kg at 1 metre cantilever? (0.20k/n-m)

If that is correct, then having a long z axis that does way down past the carriages, so say 500mm between the tool tip and the bottom of the lower carriage, means that the carriage is not the weak point. The spindle plate flex and twist in the gantry will be the weak points.

spindle plate flex can be mitigated by using a c channel or box instead of a plate and the gantry twist can only be mitigated by sheer bulk!

The moment rating on linears (should) only come into play in a case where you are using a single linear rail to support a load. Where you are planning to use a pair of linear rails, the moment of the Z axis is translated into linear force on the 2 rails. In your picture the lower rail would have force into the rail, and the upper would have the force away from the rail. More important than the maximum moment and force ratings for the linear rails and carriages (you should be well below these for anything with good accuracy,) is the rigidity ratings. These allow you to calculate how much deflection you will get due to the rails.

As far as how much force linear rails take to move, it's probably burried in the datasheets somewhere, but depending upon preload and the size of the rail, it's usually between a few oz and a couple pounds per carriage. It's tiny compared to dovetails and generally inconsequential compared to acceleration and cutting forces.

[richard-at-home]

The moment rating on linears (should) only come into play in a case where you are using a single linear rail to support a load. Where you are planning to use a pair of linear rails, the moment of the Z axis is translated into linear force on the 2 rails. In your picture the lower rail would have force into the rail, and the upper would have the force away from the rail. More important than the maximum moment and force ratings for the linear rails and carriages (you should be well below these for anything with good accuracy,) is the rigidity ratings. These allow you to calculate how much deflection you will get due to the rails.

As far as how much force linear rails take to move, it's probably burried in the datasheets somewhere, but depending upon preload and the size of the rail, it's usually between a few oz and a couple pounds per carriage. It's tiny compared to dovetails and generally inconsequential compared to acceleration and cutting forces.

Thanks skrubol.

I may see what fusion 360 says about how much flex there will be with a long z axis...

Are there any figures for cutting forces for a given cutter, DOC and feed in ali or steel? (roughly?) I know it's unlikely to be a constant but ballpark?

Is it reasonable to say that 4 x 20mm carriages are not the weakest point and that flex in the spindle plate and twist in the gantry are, in the layout above?