PDA

View Full Version : Confused about stepper sizing



Jeff
22-09-2017, 01:46 PM
Hi

My mechanical build is almost complete 20mm HiWin rails and bearings on X & Y axes with 2010 lead screws X axis is 1500mm and Y axes are 1000mm. Total weight of my gantry including the spindle is around 40Kg and I've been trying to calculate what size motors I need for my axes, but both my own calculations, and the output from several online motor calculators are giving me results that I wasn't expecting.

For example on the Y axis with a total load of 40Kg and a cutting force of 5N my total required torque @ 80 rpm (800mm/min) is coming out with a figure of around 0.18Nm, this would easily be achievable with the 175oz Nema 23 stepper on the openbuilds store which should put out around 0.7 Nm @ 48v and 80 rpm.

My gut instinct was to go for motors in the 400+ oz range, which seems to be what everybody recommends for this sort of size machine. Are my torque calculations way off or is there another reason that I need to massively overspec my motors?

Clive S
22-09-2017, 02:22 PM
Hi Jeff and welcome to the forum. In my opinion I would go with nema 23 3.1Nm from cnc4you and get AM882 drives from China and run them at 68V with a toroidal transformer power supply.

Jeff
22-09-2017, 02:49 PM
Hi Clive, I had looked at the motors from cnc4you before I confused myself with my calculations. I think I know where I stuffed it up I was getting confused about feed rates inches/m vs mm/m the speeds I used in my calculations are way too low. I'll do them again at around 10000mm/min and see where that leads me :)

Jeff
22-09-2017, 03:28 PM
Yeah that was where I was going wrong :). The 4Nm motors from cnc4you will give me cutting speeds of around 3000 mm/min and rapids around 10000 mm/min (which is as high as I can go due to the critical speed of the ballscrew), on a 48v supply with a 3x margin. 68V is way above the ideal voltage for these motors (3.0mH inductance) so I think I'll stick with 48V rather than turn my motors into workshop heaters :)

I'll probably get the CWD556 DSP drivers from there too.

Clive S
22-09-2017, 04:23 PM
Yeah that was where I was going wrong :). The 4Nm motors from cnc4you will give me cutting speeds of around 3000 mm/min and rapids around 10000 mm/min (which is as high as I can go due to the critical speed of the ballscrew), on a 48v supply with a 3x margin. 68V is way above the ideal voltage for these motors (3.0mH inductance) so I think I'll stick with 48V rather than turn my motors into workshop heaters :)

I'll probably get the CWD556 DSP drivers from there too.

Jeff It is the norm to use 68V with the motors from cn4you 3.1Nm you won't have any trouble with them. You will stiffen the machine with only 48V. Your choice though>

Ger21
22-09-2017, 05:01 PM
68V is way above the ideal voltage for these motors

The "ideal" voltage is the voltage that makes them perform the way you need them to.

If you'll be cutting wood, you'll quickly find out that 3m/min is quite slow.

Jeff
22-09-2017, 07:27 PM
I was basing the "ideal" voltage on the 32 x Square Root of inductance which, in the case of the cnc4you steppers would be 55v. The information on the Gecko website seems to infer that voltages above this "ideal" will just produce more heat through iron losses with negligeable torque gains. If you guys are using 68v without issue that would suggest the mathematical "ideal" and the real "idea" are not one and the same thing :)

Neale
22-09-2017, 07:50 PM
It's more important to get the max current right and not worry too much about the voltage. The higher voltage gets the motor current up quicker on each pulse which helps keep the torque up. I'm running 3Nm motors on 68V with no problems, like a lot of people. I'm running 2 X axis motors, 3Nm, each driving 2005 ballscrews. I'm getting critical speed on the ballscrews, very similar size to yours, which is around 5m/min rapids. My gantry is probably heavier than yours as well. When you do the sums, don't forget the inertia of the ballscrew as well - when you get to this size, that value can be as much as the gantry.

Clive'a rule of thumb advice is on the money - do the sums, analyse as much as you like, and the answer's pretty well always 3Nm (4Nm), 68V! No point in going smaller (doesn't save much money anyway) and bigger gives worse performance.

Ger21
22-09-2017, 08:33 PM
5-10 years ago, the owner of Gecko used to post on the forums quite a bit. He used to say that the ideal voltage, was the lowest voltage required to provide the performance that you need. And any excess voltage is wasted energy and heat.

They then came up with a formula that was the theoretical maximum (not "ideal") voltage. As all motors are different, this formula isn't a firm rule, even though their website will lead you to believe that.

Here's the main issue.
In an industrial application, a stepper motor would be sized to have 30%-50% more power than the application required, to insure the motor will never lose steps.

In the DIY CNC world, a motor is selected, and then the user tries to run that motor as fast as he possibly can. This requires throwing more and more voltage at it. With better drives, like AM882's or EM806's, the current reduction, and other features, can allow you to run at much higher voltages, without the motor overheating. Mounting your motor to a big block of aluminum also helps keep the motors cool, by acting like a large heatsink.

Jeff
22-09-2017, 09:09 PM
Thanks guys, really great insights.

I guess that the lazy ass part of me thought that 48v was good because I could get away with an off the shelf PSU instead of building a 70v 12A linear :)

Clive S
22-09-2017, 09:32 PM
Thanks guys, really great insights.

I guess that the lazy ass part of me thought that 48v was good because I could get away with an off the shelf PSU instead of building a 70v 12A linear :)

Jeff here is the stock answer

https://www.rapidonline.com/50a-bridge-rectifier-metal-case-519318

https://www.rapidonline.com/samwha-hc2a478m35040ha-4700uf-100v-85deg-hc-snap-in-capacitor-11-3121

https://www.rapidonline.com/lcr-ep0882-p-35mm-nylon-capacitor-mounting-clamp-11-3004

https://www.rapidonline.com/single-primary?attributes={%22Secondary%20Voltage%22:[%222%20x%2024V%22],%22Power%22:[%22500VA%22]}

the above is 500va but 625Va will give you a bit more upth.

Jeff
22-09-2017, 10:54 PM
Thanks Clive, ordered the 25v-25v 625Va... Bigger is always better :)

Clive S
22-09-2017, 11:01 PM
Thanks Clive, ordered the 25v-25v 625Va... Bigger is always better :)

I take it you ordered three caps:beer:

Jeff
22-09-2017, 11:25 PM
yeah, I realised it would be 3 x 4700uF in parallel, thanks again.

Desertboy
23-09-2017, 10:37 AM
"Selecting Supply Voltage

The power MOSFETS inside the AM882 can actually operate within +20 ~ +80VDC, including
power input fluctuation and back EMF voltage generated by motor coils during motor shaft
deceleration. Higher supply voltage can increase motor torque at higher speeds, thus helpful for
avoiding losing steps. However, higher voltage may cause bigger motor vibration at lower speed, and
it may also cause over-voltage protection or even drive damage. Therefore, it is suggested to choose
only sufficiently high supply voltage for intended applications, and it is suggested to use power
supplies with theoretical output voltage of +20 ~ +72VDC, leaving room for power fluctuation and
back-EMF."

from the AM882 manual obviously check your stepper driver specs first.

Personally I went with 68v on advice from the forums the PSU build was very easy with Clive S help and I spent longer worrying about it than it took to build which was about 2 hours with 1 hour of pottering, smoking cig and drinking tea.

Good luck with your build ;)

AlexDoran
29-09-2017, 08:14 PM
I'm using the same motors, am882 drivers and a 68v psu. Got the motors running at 12500mm/min at first testing.