Stepper motors for CoreXY

[T3P3Tony]

Hi deckingman

Its not necessarily a waste of time another use of large nozzles is to print transparent materials which benefit from higher layer heights:
http://taulman3d.com/t-glase-features.html

for example

[deckingman]

Ahh, that's interesting about large nozzles and t-glass. Thanks.

The other thing I kind of wondered about was how the Diamond hot end would compare to an E3D volcano. It is a big old lump of brass. I hear what RCarlyle says about residency time in the hot end but with the Diamond, there are 3 melt zones (one for each filament). It'll be interesting to set the mix ratio to 33% for each filament then see how much can be extruded. Lot's to play with but this is going way off topic so I'll shut up now.

[RCarlyle]

A mixing hot end with equal feed rates should let you print dramatically faster. It has been proposed on a few occasions to do this, although I haven't seen any public reports on how well it works in practice. (Stuff like retraction and heat creep jamming tends to be more problematic in mixing hot ends.)

[deckingman]

@RCarlyle:

A mixing hot end with equal feed rates should let you print dramatically faster. It has been proposed on a few occasions to do this, although I haven't seen any public reports on how well it works in practice. (Stuff like retraction and heat creep jamming tends to be more problematic in mixing hot ends.)

Yes, I've found that heat creep jamming can be a problem but only when I'm using the Diamond as a single filament hot end over a prolonged period of time and the unused inputs get affected. Even then, it only happened when I increased the temperature from 195 to 205 deg C. That's with PLA - I haven't tried any other filaments. You do need a seriously efficient cooling fan to blast air over the heat sinks. I've never had a problem when using it as a mixing hot end.

The trick to retraction is that you have to retract all 3 filaments, even if you have a mix ratio of 1:0:0. If you only retract one filament, it simply tries to draw filament from the other "unused" inputs and basically does nothing. DC42 very kindly incorporated firmware retraction into his firmware. Since I implemented that, it's been fine. I'm using about 5mm on 400mm long Bowden tubes but I've found that it needs to be varied depending on the size of the object being printed (or more specifically, the length of non-print moves). For non-print moves of about 100mm of more, 6mm retraction works better to reduce stringing. For smaller objects, 4mm works better.

[RCarlyle]

Good to know on the retraction – I figured there would be "crosstalk" issues during retraction but didn't know how specifically to manage it.

[deckingman]

I was a bit concerned initially that retracting and un-retracting all the filaments especially the "unused" ones would just grind away at the filament and cause problems when it came time to actually move the filament forward and extrude some. In practice, this hasn't been found to be an issue at all.

[sga]

Thank you for the spreadsheet!!! the instructions are so clear, great job!
I played with it tonight and simulated this guy:
Nema 23 CNC Stepper Motor 2.8A 1.26Nm(178.5oz.in) 23HS22-2804S and with my 16 teeth pulleys went up to 260mm/s with 77% of the low speed torque still available.. So shall be plenty even for my preconstrained rails and and too heavy carriages.
Then I considered the distortion too large and stopped there, this seems pretty fair to me so thank you so much for the nice tool.

I guess the next step is to validate those predictions through practice. Just waiting for some piece of electronics… to arrive in about 6 weeks...;)

[educa]

Hi,

I read the above calculations with a lot of interest and have a question. I am busy building a large scale (wel, 325x325 buildplate is not considered large by a lot of people, but I am happy with it) coreXY printer.

For the X/Y I am considering 2 possible motors which are quite different from eachother

Model A is the very popular and widely used Kysan 1124090 Nema 17 1,8° stepper with 5,5Kg,cm holding torque

Model B is the Lin Engineering 5704M-03 Nema 23 0.45° stepper with 10,1Kg.cm holding torque

I did some calculations based on each motor using GT2 20tooth pulleys

Model A:

U = 24V
R = 2.8Ohm per phase
I = 1.5 Amps
Inductance = 4.8mH/Phase

Time to fully energize coil = 1.5 x 4.8 / (24 - 0.5 x 1.5 x 2.8) = 0.329ms
Max speed (5steps per mm) -> 1000 / (5 x 0.329) = 607 mm/s

Model B :

U = 24V
R = 1.2Ohm per phase
I = 3 Amps
Inductance = 1.2mH/Phase

Time to fully energize coil = 3 x 1.2 / (24 - 0.5 x 3 x 1.2) = 0.162ms
Max speed (20steps per mm) -> 1000 / (20 x 0.162) = 308 mm/s

Model B @ 36vDC:

U = 36V
R = 1.2Ohm per phase
I = 3 Amps
Inductance = 1.2mH/Phase

Time to fully energize coil = 3 x 1.2 / (36 - 0.5 x 3 x 1.2) = 0.105ms
Max speed (20steps per mm) -> 1000 / (20 x 0.105) = 476 mm/s

I do know that duet board doesn't do more than 2 Amps, but I intend to somehow drive the motors using external drivers. I currently have these nema23 motors (yes, you read well, 800 steps per rotation, I also didn't know they existed before) hooked up to 35vDC on my self built lasercutter.

Now my question.

I read above that motors with more steps per rotation would not do a lot of difference in quality, but thats so hard for me to understand. A normal 200 steps motor normally can achieve its accuracy in positioning with an acceptable rate of 5% of the size of 1 step. So if 1 step is 1.8 degrees, 5% of it is 0,09° accuracy, which on a 20 teeth pulley results in 40 / (360 / 0,09) accuracy of 0.01mm . The nema 23 motor I would like to use is accurate up to 0.017° , which on a 20teeth pulley results in accuracy of 40 / (360 / 0,017) = 0,00188mm

So…. isn't in my case then the nema23 a better choice? It has plenty of power, can go 476mm/s with fully energized coils and accuracy of 0,00188mm

Also. it goes 476mm/s with fully energized coils, but I guess it can go faster than that with less energized coils ? With that amount of torque coils don't have to be fully energized to pull the load guess.
The nema17 motor only has 5,5Kg.cm torque and the nema23 has 10,1 , so I could theoretically run the nema23 at aproximately 5.5/10.1 x 3 = 1,63 Amps and get the same torque, but a lot more accuracy ?

[deckingman]

My twopence worth. Regarding accuracy, we are squirting hot filament out of a small nozzle whilst moving that nozzle too and fro (and around curves which aren't true curves but small segmented moves). There are just so many variable in that process that you are most unlikely to be able to produce an object that is dimensionally accurate to anything like 0.1mm in every axis. So 0.01mm is plenty good enough and anything else is overkill although you could use 0.9 degree steppers or 1:2 gearing to double it.

For info, my Y axis (which includes the X axis rails and a bulky Diamond hot end) weighs is in at around 1,900 gms driven by Nema 17s (2amp driven at 1800mA) with 0,59Nm torque (around 6kg.cm) and I regularly print at 90mm/sec with non-print moves off 320mm/sec on a 400 x 400 bed.

Personally, I'd stick with the Nema 17s driven by the excellent onboard stepper drivers using the default 16X microstepping with interpolation).

[dc42]

It's not a good idea to plan to use high-torque motors and then run them at only a fraction of their rated current. High torque motors also have high rotor inertia, so if you run them at low current then you are restricted to low acceleration. Also the detent torque becomes more significant in relation to the holding torque, so microstepping is less accurate.

When calculating resolution, bear in mind that stepper motors only produce their maximum torque when the rotor position is nearly 1 full step behind the commanded position. So if your design demands a significant proportion of the motor holding torque, you won't get close to single microstep accuracy.

In delta printers, using 0.9deg motors does improve print quality noticeably.

If you decide to choose Nema 23 motors, then I suggest a current rating of about 2.8A to match the drivers in the Duet WiFi and Duet Ethernet. I have recently bought these motors for testing the Duet WiFi at high currents: http://uk.stepperonline.com/3pcs-of-nema-23-cnc-stepper-motor-28a-126nm1785ozin-23hs222804s-p-395.html. But for a CoreXY printer with a bed 325mm square, I think Nema 17 motors will be more than adequate, probably quieter, and perhaps more accurate.

[educa]

Ok thank you for fast answers. I guess it would be better then for me to go nema17 0.9°

Regarding duet wifi. I see the standard board has 5 stepper drivers.

I will need 2 drivers for the xy part, then 2 steppers for the 2 titan aero extruders and then I need my Z

I am planning to let the printbed go up/down on 3 1204 ballscrews. But I planned to let them be moved by 1 stepper for each ballscrew. For that I would need to be able to drive 3 Z steppers. Or would this also work with 1 nema 17 motor with for example a 20t pulley and then a continuous gt2 belt driving the 3 ballscrews which have a 40 or 60tooth gt2 pulley on them?

[kelchm]

For those of you looking for 0.9 degree NEMA 17 stepper motors, I've had my eye on the MS17HA series from Moons Industries.

In particular the 2A / 75 oz-in MS17HA6P4200 looks really appealing. The only downside is that the best pricing I have found for them is around $60 each not including shipping.

[deckingman]

@educa:

…...................I am planning to let the printbed go up/down on 3 1204 ballscrews. But I planned to let them be moved by 1 stepper for each ballscrew. For that I would need to be able to drive 3 Z steppers. Or would this also work with 1 nema 17 motor with for example a 20t pulley and then a continuous gt2 belt driving the 3 ballscrews which have a 40 or 60tooth gt2 pulley on them?

The choice is yours but for info, my bed is 400mm x 400mm x 10mm thick aluminium tooling plate with 12mm insulation under and 6mm glass on top, all fixed to a 2020 aluminium extrusion frame. It weighs in at around 7kg and I drive it with a single Nema 17 and continuous belt. The motor is rated at 2Amp so I normally run it at 1800 mA but I reduce it to 1200 mA for homing and it still works fine. However, I use 1mm pitch single start screws so the lead is 1mm. As you are electing to use course, 4mm "lead" screws, you'll need more torque and/or gearing. I'd have thought 1:2 gearing would be adequate so 20 tooth on the motor and 40 tooth on the shaft (but it depends on the weight of your bed).

If you go for 3 motors, then you have to consider that they might get out of sync every time you cycle the power but David (DC42) has it in mind to fix this in firmware (if he hasn't already done so). You could also take advantage of the automatic bed levelling feature that is soon to be introduced in firmware. Whether you need it depends on how well you construct the printer. If you think that you might have to frequently adjust the levelling of the bed, then I'd say go for the 3 motor option but if you build the printer so that the bed stays level and doesn't need frequent adjustment, then the single motor option will be cheaper, because you won't need to buy a Duex 2 or 5 expansion board as well as the extra motors.

HTH

[educa]

Well, its not really leadscrews but ballscrews and 4 mm is the lowest I could find at affordable pricing. On the other hand it does indeed sound a very good idea to sync them all on 1 motor with continuous belt. I could als use 20tooth pulley on motor and 60 on the ballscrews. My z table is a lot lighter than yours since I only have 325x325x8 mm PEI coated aluplate supported on 2020 extrusions. I don't think total weight would go over 3.5 to 4 kg + with the 3:1 reduction of the pulleys it might not be a big problem. Even if it would be a problem, I could still use a planetary geared motor if power would not be sufficent, that way I get 5.18 : 1 reduction of the planetary gearbox and 1:3 of the pulley setup. One of these combinations should be powerfull enough then.

How do you let your 7kg z bed go up and down? Do you have images of your setup?

[deckingman]

Any particular reason why you want to use ball screws? The main advantage of ball screws over "conventional" trapezoidal lead screws is lower friction but this is hardly an issue with our slow and limited moving Z axes.

Anyway, with your bed I'd say a single motor and 1:2 gearing would be fine.

My 3 screws are 8mm diameter but they only provide lift so they are more than adequate. I use linear guides to prevent any rotation of the bed so the screws sit on thrust bearings to take the downward force and are constrained with roller bearings at the bottom. The tops are fully floating because the screws are only being used to lift the bed. Any tendency to wobble or rotate is taken care of by the linear guides. There are a few pictures on my blog https://somei3deas.wordpress.com/my-corexy-printer-build/ and some videos of it in action on my YouTube channel - e.g. this one which is a 300mm tall (1,000 layer) object https://www.youtube.com/watch?v=vG1WqijJ634. I can supply more pics if you are interested.

[educa]

Ian, it will not be ballscrews anymore. I will indeed also use the T8 screws with a 2mm pitch, 2mm lead and anti backlash nuts + linear guides to protect against bed movement (rotation/wobble,…)

I wanted to use ballscrews, but they were "chinese" quality and didn't get past quality checking (hey, what would you do if you discovered that 1 1204 ballscrew wasn't even 12mm, but 11,6mm wide. I would call it rubish)

I throwed the 3 ballscrews in the trash

[bilsch]

Not sure if anyone is maintaining that doc. Found this thread so went ahead and plugged in the values on my steppers.

The non-calculated fields:

brand motor torque /Ncm angle rated current /A inductance /mH resistance /Ohm
wantai 42BYGHM208P4 20.5 0.9 1.2 5 2.1

full row with copy/paste of the formulas ( which should update to apply to the row )

brand motor torque /Ncm angle rated current /A inductance /mH resistance /Ohm max speed rot/s time/step (ms) step time calculation according to DC42 steps/rot steps/mm max speed mm/s actual torque N/cm Force on belt N acceleration mm/s2 link
wantai 42BYGHM208P4 20.5 0.9 1.2 5 2.1 10.00 0.25 0.264 400 10.00 379.00 4.041438333 6.348 6910.3 http://www.wantmotor.com/product/42byghm.html#sthash.TXwVDXha.dpuf

I've been running these on my corexy bot at 1amp on a duet wifi. I picked the steppers up from e3d a while back. Running at 1a these motors get pretty hot but I wanted to keep the torque - I do rapid movements at 150mm/sec. If it matters I'm running them on a cbot with a titan extruder / e3d hotend though I don't know the combined weight of the carriage

@deckingman:

Thing is that, I got to different values too if you check the 3rd line of the table below, it's almost exactly double time and half the speed.
https://docs.google.com/spreadsheets/d/1w60BUzpiZ6Gjqt8P8pSoqq4r4j0qxhxQVSbI5c0SoWw/edit?usp=sharing

[deckingman]

The above post prompted me to report back on what I ended up using. I went for these http://uk.stepperonline.com/3d-printer-nema-17-stepper-motor-59ncm84ozin-2a-17hs192004s1-p-18.html which I run as 1.8 amps and which happily throw my 1.9 kg Y axis around at non print speeds of up to 350mm/sec. I do however limit short print moves by setting a lowish acceleration of 1200 m/s^2 but the machine is capable of much higher accelerations (just doesn't print well).

[bilsch]

Also spec'd out the openbuilds steppers

openbuilds nema17 54 1.8 1.68 2.8 1.65 25.51
openbuilds nema23 123.58 1.8 2.8 3 1.1 14.29
openbuilds high torque nema23 243.62 1.8 3 4 1.2 10.00

I use those on a different bot ( well, the 17 and regular 23 ) and like 'em.