Minimum torque calc. for conversion

Hello all.
Im current undergoing a full rebuild of my little Optimum BF16V benchtop mill,Im thinking I might take the opportunity to convert it to CNC while im at it.

I tend to like to plan everything through beforehand, especially in reguards to budget, I also tend to like to "buy once cry once"

Im hoping someone here can either help me calculate or even offer there experience with same/similar models, as to what is the minimum torque and other metrics I can get away with for each axis.

Not because Im cheap or anything, but because the machine is so small that there isnt necisarrily alot of free space to work with and I assume generally smaller torque = smaller motor. But also because I'm cheap

Working on the assumption that I will use closed loop steppers. Probably uses things like the Nemas as a basis for comparison. But obviously if someone would like to refer a model that's awesome too!

Im a general machinist by trade but most of my work ends up being machine rebuilding, repairing and toolmaking. Ive tidied the mill up alot and made some modifications but functionally and it is mostly the same with the glaring exception of a new saddle, made to be a bit bigger. I dont have a mass for you just yet but if it will make the difference between one size and the next i will add when I have the number.


Thanks in advance

Hi NP - At a guess a medium size N23 will do it. A large nema23 will be heaps. Are you doing Z? or only X,Y. What is the pitch of the lead screws and I can calculate for you. Not sure you will get any benefit from closed loop steppers, better to put the extra dollars into something else. I say this because closed loop can run slightly faster then open but in your case speed won't be too important being a small machine and not using ballscrews so your a little limited in top speed anyway. I expect your keeping the lead screws?

So effectively you plan to replace the handles with motors? Peter

Hi,

Quote:

---

I say this because closed loop can run slightly faster then open

---

Where do you get that from?....its incorrect. The manufacturers claim that is the case.....and is pure BS. A closed loop stepper is still a stepper, it has no more power, speed or torque
than an open loop stepper.

Closed loop steppers do offer a couple of advantages, small that they are, and so I think the premium paid is unjustified. If you want genuine closed loop performance then get AC servos,
don't mess with closed loop steppers. Downside is that servos cost bigtime.

I agree with Peteeng, good low inductance open loop steppers will be fine. You should look at 23 or 24 size steppers of 1mH-2mH inductance with 1 mH preferred and reject anything
over 2mH. You'll want a high voltage driver, I would suggest 80VDC. The high voltage is required to maintain stepper torque at speed.

Really if you want a genuine CNC type machine then you'll have to use ballscrews. The existing lead screws will have lash, and even small amounts of lash will screw a CNC toolpath.
Looking at photos of your machine I would say 16mm ballscrews would be ideal.

Craig

Hi Craig - I've looked at lots of torque curves and voltage to voltage the closed loops are a bit faster, I don't know why maybe they are doing it just to show its a bit better. (say 1200rpm vs 1000rpm but at these speeds there is small torque so in reality no help) . But I don't think they are worth the cost addition. Agreed the higher the voltage of the system the better. Peter

Hi,
the manufacturer uses EXACTLY the same stepper as the open model, its just that the closed loop one has an encoder grafted onto it.

If the open loop stepper has 30 oz.in at 1000 rpm, the closed loop stepper has 30 oz.in at 1000rpm, the power, torque and speed limitations are properties of the physics
of the steppers, they do not change by virtue of a closed loop control.

Quote:

---

Agreed the higher the voltage of the system the better. Peter

---

That is better use of OP's money.

Craig

For a stepper in this setting, I don't think minimum torque calculation is particularly necessary or helpful.

Steppers have heaps of torque at low speeds, poor torque at high speeds.

Cost of steppers is low, and marginal difference between different models etc.

As advised by others: Get low inductance steppers. Get decent digital drivers. Run the steppers at higher voltages - 48V minimum, higher is better for torque at higher speed.

I would go closed loop steppers, not because they perform better, but because they will fault and signal an error once you lose a certain number of steps.
Yes, a properly designed stepper system should not lose steps, but let's face it, our hobby systems are rarely optimally designed.
I have had a couple of projects with serious fails (over years) due to stepper lost steps / stall. A single one of these stuff ups cost enough in time +/- materials to justify the cost difference of closed loop for the prices they are these days.
(noting closed loop won't save you from every stuff up, particularly in tight tolerance parts)

Hi,
aside from the control technology, ie open loop vs closed loop, there is still the question about torque.

Lets imagine you have 16mm diammeter 5mm pitch ballscrews fitted.

Lets also imagine you have a stepper capable of 1Nm torque, then the max thrust is:

Thrust= (16 x PI)/5 x 1000/(16/2) x 1
=1256.4N or 125kg force!

So your machine would have plenty of thrust even with a low torque stepper.

But what about acceleration and momentum?. That is a longer calculation but I'll skim through it to give you an idea.
Assuming a 16mm diameter screw of 5mm pitch 500mm long, and an axis of 25kg:

J_screw=0.008²x PI x 8000 x 0.5 x 0.5 x 0.008²
=0.5 x 10^-4kg.m²

J_armature=0.25 x10^-4kg.m² (estimated from spec sheet)

J_linear= 25 x 0.005²/(2 x PI)²
=0.995 x10^-4 kg.m²

J_total =1.75 x 10^-4kg.m²

With 1Nm applied:
d²w/dt²= 1/1.75 x 10^-4
=6666 rad/s²
or converted to linear acceleration:
accel = 6666/(2 x PI) x 0.005
=5.3 m/s²

or 0.5g....which is very good indeed.

Thus a stepper or servo of 1Nm torque would be adequate for your machine. The trick is that a stepper loses torque the faster it goes, so it requires 1Nm at the desired max speed.
If you want a max traverse of 5000mm/min then the stepper must do 1000rpm.

If you get really low inductance steppers ( approx 1mH) and run them at 80V then you might expect a 4Nm (holding) stepper to still have 1Nm at 1000rpm.
That is my recommendation, a low inductance stepper of about 4Nm (approx 500oz.in).

Craig

Hi,
just as a matter of interest note that in the calculation the linear momentum is 1/1.75 or 57% of the total momentum and the armature and ballscrew, which both
rotate, are the remaining 43% of the momentum.

Note also that the ballscrew momentum is highly (to the fourth power) dependent on the ballscrew diameter. Thus if you thought that you'd like to use 20mm ballscrews and gain
some rigidity as a result then the ballscrew momentum increases by (20/16)⁴= 2.44. So the momentum increases by 244% for such a small increase in diameter!
The momentum balance would then be 32% linear momentum (ie the axis mass) and the rotational momentum of 68%.

Clearly the choice of ballscrew diameter is far from trivial. Don't go overboard and specify outsize ballscrews, say 25mm or 32mm, the rotational inertia of large diameter screws
will defeat you.

Craig