Mendal Nema 17s torque question

Looking on the list it looks like the stepper motors need about 1400 g/cm of torque. Converting to oz/in I came up with 125 oz/in. However I am having a fair dificulty finding a Nema 17 frame stepper with that kind of holding torque.

Do I really need that much or would 80oz/in holding be enough?

What stepper motors are your using on your mendal?

Thanks,
David

ok, i guess goolge didnt do the calculation correct or something. using the calc on lin Enginerring 1400 g-cm ends up 19.432 oz/in. That opens up the possibilitys alot. Sorry for the confusion.

David

Yes you multiplied by 2.54 instead of dividing.

I only have a feeling for Nm so 1400g.cm is 1.4 kg.cm = 14 N.cm = 0.14 Nm. That is very weak for a NEMA17, most are between 0.22 and 0.44 Nm. That is why you can use NEMA14s on the axes but not on the extruder where you need a high torque NEMA17 unless you have gearing.

---

[www.hydraraptor.blogspot.com]

i am thinking about trying a geared stepper in an extruder, i think it should work good, dose any one else have any other thoughts.?

what about a geared DC motor has any one used them with the mendle yet.?

Yes I think a geared stepper is the only sensible way to go. A DC motor cannot instantaneously reverse at 8 times the speed it was running forward. A stepper can.

A stepper without gearing does not have a small enough step size to match the filament resolution to the axis resolution. Even with gearing you probably want microstepping. Otherwise the gear ratio has to be higher, so you cannot reverse as fast.

Since I moved from a DC motor with a shaft encoder to a geared stepper my build quality is much better. No strings to remove

---

[www.hydraraptor.blogspot.com]

Oh I was going with the standard Mendel block design and a NEMA 17 with 0.45NM Torque.

Do you mean the only way top go with a NEMA 14 not a NEMA 17 ?

I have a few Mechano gears to play with no worm gear thou..

---

Bodge It [reprap.org]
=======================================

---

BIQ Sanguinololu SD LCD board	BIQ Stepcon	BIQ Opto Endstop
BIQ Heater Block PCB	BIQ Extruder Peek clamp replacement	BIQ Huxley Seedling
BIQ Sanguinololu mounting	BIQ standalone Sanguinololu or Ramps mounting	Print It Stick It Cut it

My rep strap: [repstrapbertha.blogspot.com]

Buy the bits from B&Q pipestrap [diyrepstrap.blogspot.com]
How to Build a Darwin without any Rep Rap Parts [repstrapdarwin.blogspot.com]
Web Site [www.takeaway3dtech.com]

No the problem is a 200 step motor does not have enough resolution, even with microstepping.

Assume the smallest pinch wheel is the motor's shaft, i.e. 5mm. One rev feeds ~16mm of 3mm feedstock. With 8th microstepping you have 1600 steps, so one step is 0.01mm of feed. If you are extruding 0.5mm filament that is 0.36mm, so does not meet the 0.1mm target of RepRap. I also extrude 0.3mm filament sometimes, so one micro step would be 1mm per step in that case.

I don't know how Mendel aims to work with no gearing and only half stepping. Each half step is 1.44mm of 0.5mm filament!

I use a 13mm pinch wheel that gives ~ 40mm feed per rev. That is geared down 40:1 so I feed 1mm per rev, with 1600 microsteps that is 0.0625mm of 0.3mm filament so I exceed the RepRap aim of 0.1 with 0.5mm filament.

---

[www.hydraraptor.blogspot.com]

nophead, never looked at it like that before. my drivers are 16x microstepping, so I'm a bit closer to that goal- will see how things go I suppose

Ok thank you your comment now makes sence to me.

So if you could microstep at 16th microstepping or 3200 steps that would be considerably better than a standard Mendel extruder without any gearing ~ though still not meeting the 0.1mm target.

I will now look into this in more detail I only have so much Acrilic material to make my Extruder block out of and dont want to waste any.

---

Bodge It [reprap.org]
=======================================

---

BIQ Sanguinololu SD LCD board	BIQ Stepcon	BIQ Opto Endstop
BIQ Heater Block PCB	BIQ Extruder Peek clamp replacement	BIQ Huxley Seedling
BIQ Sanguinololu mounting	BIQ standalone Sanguinololu or Ramps mounting	Print It Stick It Cut it

My rep strap: [repstrapbertha.blogspot.com]

Buy the bits from B&Q pipestrap [diyrepstrap.blogspot.com]
How to Build a Darwin without any Rep Rap Parts [repstrapdarwin.blogspot.com]
Web Site [www.takeaway3dtech.com]

Yes 16th steps would be 8 times better than Mendel, but still nearly half the required resolution with 0.5mm filament.

I expect it would still give results way better than any of the DC motor versions because it will be able to instantaneously reverse at many times the extrusion speed. That is something a DC motor cannot do because it is spinning so fast and geared down so much that it can't reverse much faster than it can go forward.

---

[www.hydraraptor.blogspot.com]

Preaching to the priests of RepRap.
Yeah, don't forget though your going to need a lot more energy to compensate the
loss in controlling those micro steps to achieve similar torque of full step

I can't see why there are any more loses involved compared to a full step chopping driver. You just have to increase the peak sine wave current to be 1.4 times the current on each coil of full step mode to get the same torque. The total current in the "both coils on position" will be 2 x 1.4 * sin(45) = 2, the same when full stepping. In other positions the total current will be less. I.e. in the "one coil on position" it will 0.7 times less for the same torque.

Full stepping is not very energy efficient because the torque contribution of both coils is sin(45)+cos(45), so only 1.4 times the single coil value, but using twice the current.

---

[www.hydraraptor.blogspot.com]

Hi nophead

You mention a Reprap target of 0.1mm of extrusion per step. Where is that data on the website?

---

Regards

Ian

It isn't explicitly stated and I think it has been overlooked, but the positioning system aims for 0.1mm resolution and the extruded object 0.2mm, so it follows that you need the same on the extruder, otherwise the resulting objects will not achieve the desired resolution.

---

[www.hydraraptor.blogspot.com]

Thanks

That makes sense. I guess that is one of the many reasons that you get better results. I noticed you are using rapid reversal and rapid return on your extruder before you actually extrude any material. Have you built this in as a function of your equivalent of the firmware? Or are you adjusting it in the control code so you can adjust it on the fly from skeinforge?

---

Regards

Ian

Neither actually, all my machine control strategy is the Python script in between. I only use Skeinforge for the filament path and my firmware only moves the motors and controls the heaters from binary commands that are in firmware friendly units.

It means I don't have to regenerate the g-code when changing materials or my machine, only if I want to change the filament diameter or number of shells.

---

[www.hydraraptor.blogspot.com]