[slow-poke]

I'm more of an electrical type guy than a mechanical type guy so fairly clueless on selecting a suitable belt for this application.

I have the x and Y axis doing their thing with ac servos and LinuxCNC, really happy with that aspect. I'm now thinking about how I'm going to add CNC capability to the knee joint of my A1S mill.

I want to use a similar 2.4Nm continuous torque 7.2Nm peak ac servo, except this one has an integrated brake. I have a beefy double nut ballscrew waiting in the corner.

From my measurements and reading about knee conversions it seems I will need a 3:1 torque multiplication to be on the safe side. Aim is to have < 0.001" error, so driving through the bevel gears is not an option. There appears to be plenty of room to mount this servo just inside the front edge of the knee and then drive the new ballscrew via a suitable timing belt as shown in the concept drawing below. The servo moves with the knee and the upper end nub of the ballscrew simply pushes up against the existing bearing (more likely on a spacer that fits in the existing bearing).

Can anyone spot any obvious problems with this concept?

Suggestions for suitable timing belt that will accommodate about 20+ Nm of torque?

[peteeng]

Hi SP - I'd suggest a 25mm belt but its miles stronger then required. I'd choose it for its stiffness. See attached calc. The 25mm wide AT10 belt has a rated load with 15t engaged at 1930N (well over 4x SF). Using a 3:1 with a 49T pinion the tension is 286N plus preload at 20Nm. So all good. Peter

I've attached the belt catalogues FYI... I think the torque calc is incorrect. I'll check... before this times out. so belt tension is 286N << 1930N

[slow-poke]

Peter,

Much appreciated, I will pick something suitable from the catalog now that I know what to look for.

One step closer, have a great day!

[peteeng]

Hi SP - I did correct the calculation so the attached prior is correct.. Cheers Peter

[slow-poke]

Peter,

I found some suitable pulleys on McMaster-Carr. so I'm going to take a crack at printing them for trial fit and who knows perhaps they will be strong enough to try out the concept.

Much appreciated.

Jeff

[joeavaerage]

Hi,
I use 750W Delta B2 servos on my mill. They are 2.4Nm (rated), 7.1Nm (overload), 3000rpm (rated). They are direct coupled to 32mm C5 BNFN double nut THK 5mm pitch ballscrews.
They can go up to 25m/min, and at standard tune do 0.27g with a 150kg axis. Why gear them down? Is there any need?

If you have 10mm pitch screws or higher....then yes.....you might want more torque, but with 5mm pitch screws you don't need any reduction to my way of thinking.

Craig

[slow-poke]

Hi,
I use 750W Delta B2 servos on my mill. They are 2.4Nm (rated), 7.1Nm (overload), 3000rpm (rated). They are direct coupled to 32mm C5 BNFN double nut THK 5mm pitch ballscrews.
They can go up to 25m/min, and at standard tune do 0.27g with a 150kg axis. Why gear them down? Is there any need?

If you have 10mm pitch screws or higher....then yes.....you might want more torque, but with 5mm pitch screws you don't need any reduction to my way of thinking.

Craig

Craig,

I appreciate the feedback, my A1S knee mill is fair bit larger/heavier than your mill. I measured the torque required to lift the knee with the existing leadscrew (no weight other than a Kurt vise) @ 7-8 Nm. With the ballscrew it will be less, however who knows how much weight could be on the table in the future, and extra 100-200 lbs is entirely possible. I want to design to the continuous rating of the servo not the peak rating, because even at 100IPM full knee travel would likely exceed the peak duty cycle.

Thanks,
Jeff

[joeavaerage]

Hi,
you need to do the calculations, you are as you said electrically inclined not mechanically inclined.....me too as it turns out, but if you're going to do this then you
need to put numbers to it.

Just as an aside my mill weighs 800kg. The cast iron axis beds alone weigh 115kg each.

What pitch is your ballscrew?

A 5mm pitch ballscrew can support 1250N (125kgf) with 1Nm applied. Note that this is NOT diameter dependent, it depends on pitch alone.

With that in mind your rated servo torque of 2.4Nm can support 3kN or 300kgf.

So now actually work out or measure the combined weight of the X and Y axes....are they 300kg ....or more....or less?

Craig

[slow-poke]

Hi,
you need to do the calculations, you are as you said electrically inclined not mechanically inclined.....me too as it turns out, but if you're going to do this then you
need to put numbers to it.

Just as an aside my mill weighs 800kg. The cast iron axis beds alone weigh 115kg each.

What pitch is your ballscrew?

A 5mm pitch ballscrew can support 1250N (125kgf) with 1Nm applied. Note that this is NOT diameter dependent, it depends on pitch alone.

With that in mind your rated servo torque of 2.4Nm can support 3kN or 300kgf.

So now actually work out or measure the combined weight of the X and Y axes....are they 300kg ....or more....or less?

Craig

I don't have an easy way to weigh my mill, the shipping paperwork stated 2000lbs, so a bit under that best guess. 300kg would be a good estimate for everything on the knee. Ballscrew is 5mm pitch.

All that being said, the configuration of this style of mill does not lend itself to direct drive of the knee joint because the ballscrew moves up and down to lift the knee, and there is not enough room at either end.

[joeavaerage]

Hi,
OK, I would guess your estimate of the weight of the X and Y axis combo of 300kg is about right. That in turn suggests that your servo could hold it aloft without reduction. That would leave little or no torque, without
dipping into the overload capacity, for accelerating the load. That in turn suggests you need some reduction, but I would suggest 2:1 would be fine. That would mean that 2.4Nm is to accommodate the weight of the axis
and another 2.4Nm to accelerate the axis. You may think that does not sound adequate but I can assure you it is, and can show you the calculations that govern the acceleration.

As you say you need a belt drive in order to make everything fit, so a belt reduction of 2:1 or thereabouts would be perfect.

I use a belt drive on my fifth axis platter with a 750W Delta B2 servo of 2.4Nm (rated). The gear reducer is 6.75:1 so in truth I'm really only using a small fraction of the servos available torque mostly whereas you will be using
the lions share of the servos torque continuously. My belt is only 15mm wide and is perfectly adequate. In your situation I would think that 20mm or 25mm width would be appropriate as you can expect that continuous load.
Wider is not required.

What you don't quite seem to grasp is that a 5mm pitch ballscrew offers a great GREAT deal of force multiplication and that only a modest servo will in fact cause the knee assembly to move very comfortably indeed. Don't sweat the
belt, as I have proven even a 15mm belt is more than sufficient to transit 1 hp (750W).

One thing you do need to consider is the diameter of the pulley. It is very tempting to reduce the size of the servo pulley, but there is a distinct limit. The belt is likely to have a tooth pitch of 5mm, or maybe 8mm in the wider
belt sizes. Thus with a small pulley very few teeth are engaged, increasing the strain on each tooth with consequent accelerated wear. As you can see from my pic the pulleys I chose are 47mm diameter and therefore there are
15 teeth engaged (5mm pitch). This I have found to be perfectly adequate but for continuous service at moderate loads I would not go lower, but would in fact prefer more engagement.

I would suggest with an 8mm pitch belt that the min diameter of the servo pulley be 76mm or thereabouts. If you adopt a 2:1 reduction then the ballscrew pulley will be 152mm or thereabouts.

If you want to do the acceleration calculation I need the diameter, pitch and length of the ballscrew, your best estimate of the mass of the knee assembly, and the rotational inertia of servo from the spec sheet.
What may surprise you is that the acceleration is limited NOT by the mass of the knee assembly, but rather the rotational inertia of the ballscrew. My guess is that you'll find that counter-intuitive but is none-the-less
correct.

As an example my machine has a servo very similar to yours. It is direct coupled to a 32mm diameter 5mm pitch ballcsrew 750mm long. The assumed axis mass is 150kg. My servo will accelerate the axis
at 2.7m/s² or 0.27g using rated torque only. The momentum is 80% in the ballscrew, 12.5% in the servo and only 7.5% the linear accelerating mass. Yes that is correct, despite the axis weighing 150kg
the ballscrew has ten times the momentum!!!

I have done this same calculation for many others on the forum, and that basic idea that the ballscrew either dominates or at least is a large contributor to the total momentum occurs again and again, to the surprise of many.

Craig