Dual Z stepper motor sync with linear rail carriages
October 09, 2019 07:08AM
I have fit dual Z steppers to my printer, well aware of the conflicting opinions regarding dedicated stepper drivers (in my case) for each motor vs. Single stepper motor driving 2 lead screws via timing belt..

I have dabbled in the idea of using a belt but it's not aesthetically pleasing. Initially, I purchased THAT kit - but the supplied belt was too short, I just needed another stepper motor for the alternative setup so I went down that pathway.

So with the linear rails I mounted along both Z axis, it maintains alignment across the length of the X axis (measuring from the bed) with a single lead screw, regardless of what side it is on, thanks to the tolerance and sturdy frame prohibiting deviation as it moves up away from the bed (not to mention requiring minimal effort).

I selected pancake steppers for the dual Z actuation - they are low profile and I redesigned the frame to basically eliminate any loss of Z that would have resulted from introducing linear rails, a mains powered bed (and mic6 plate etc etc).

If this Z axis can maintain a level X axis throughout its entire range, does that negate any of the associated issues typically referenced during design stages in relation to the use of separate stepper motors and drivers (going out of step and whatnot)?

Perhaps along the lines of remaining out of sync and subject to drag, causing an eventual fail of the stepper motor and or driver?

Edited 1 time(s). Last edit at 10/09/2019 07:14AM by imotion.
Re: Dual Z stepper motor sync with linear rail carriages
October 09, 2019 09:56AM
A problem with using multiple Z stepper motors is that when you first power them up, they jump to the nearest multiple of 4 full steps that matches the current. If the two motors are between positions, they may jump in opposite directions.

Another problem is that if the leadscrews are course, the bed might drop a little when powered off, and it might drop unevenly on both sides.

In theory it shouldn't make any difference whether you use one stepper driver or two. However, some Trinamic drivers are very sensitive to noise on the step inputs during power up. This can result in the driver(s) advancing a semi-random number of microsteps at power up. If dual drivers are used, they may move by different numbers of microsteps.

There are various automatic ways of levelling the motors if they get out of step and you use separate drivers:

- Reduce motor current and stall-home them against individual Z max physical endstops
- Use a separate endstop switch for each motor, then home them simultaneously but independently
- Use the Z probe to probe 2 or more points on the bed, then the firmware computes and applies the corrections required to individual motors

HTH David



Large delta printer [miscsolutions.wordpress.com], Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].
Re: Dual Z stepper motor sync with linear rail carriages
October 09, 2019 10:44AM
There's another option- use one motor and belt the two Z axis screws together. They will never get out of sync because they can't (at least as long as the set screws in the pulleys are tight).


Son of MegaMax 3D printer: [www.instructables.com]
Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Re: Dual Z stepper motor sync with linear rail carriages
October 11, 2019 05:04AM
Quote
dc42
There are various automatic ways of levelling the motors if they get out of step and you use separate drivers:

- Reduce motor current and stall-home them against individual Z max physical endstops
- Use a separate endstop switch for each motor, then home them simultaneously but independently
- Use the Z probe to probe 2 or more points on the bed, then the firmware computes and applies the corrections required to individual motors

HTH David

Does combining the first suggestion - which is what the i3 carries out when calibrating (if I am not mistaken); with the auto bed levelling probe option (which I have already implemented) provide long-term reliability with respect to maintaining a level X axis gantry after it has been homed in that manner?

I remember watching a prusa stall-home against the top of its frame once and it must have been a fair bit out of square because it was VERY visible. The assembly I have currently is rigid enough to not permit that, to the same severity but I would be happy with less than half a millimeter difference in a worst case scenario.

While I will always consider a belt driven dual Z axis, I have the same confidence in a grub screw being able to maintain the grip required to clamp a pulley to a leadscrew (unless it has been ground flat in that spot) as I do when it comes to the use of 2 seperate steppers.

Something about a motor on one side just doesnt give me the feeling that the gantry is being raised in equilibrium. I am pondering whether to throw a belt across the top and retain the motors on the bottom (both sides) to force it in sync six ways from sunday. Remembering these are pancake steppers, a bigger motor placed on one-side would most likely foul on the gantry before the nozzle was able to reach the bed - which was placed as low to the bottom of the frame as possible. Not that any loss of Z is a big problem, but the low profile bed is an important aspect of the design intended to minimise negative effects associated with non-coreXY designs (where the bed does not have to move back and forth whatsoever)

Edited 1 time(s). Last edit at 10/11/2019 05:09AM by imotion.
Re: Dual Z stepper motor sync with linear rail carriages
October 11, 2019 05:12AM
Quote
the_digital_dentist
There's another option- use one motor and belt the two Z axis screws together. They will never get out of sync because they can't (at least as long as the set screws in the pulleys are tight).

Why make it simpler, cheaper, better ? So much fun trying to fix issues one creates smiling smiley

Edited 1 time(s). Last edit at 10/11/2019 05:14AM by MKSA.


"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)
Re: Dual Z stepper motor sync with linear rail carriages
October 11, 2019 05:51AM
Quote
MKSA
Quote
the_digital_dentist
There's another option- use one motor and belt the two Z axis screws together. They will never get out of sync because they can't (at least as long as the set screws in the pulleys are tight).

Why make it simpler, cheaper, better ? So much fun trying to fix issues one creates smiling smiley

The idea is that the parts purchased will be re-purposed in any future builds. For that reason, certain selections were made (eg. the length of linear rails and motor sizes etc.) which may have exceeded the immediate requirements to get only what I was working on into an operational capacity. Had I not done so, likely half of the purchases made would be unsuitable (or insufficient in some capacity) for reuse, based on equipment requirements listed for the builds I would potentially go on to make. That includes a belt-driven Z axis, which WAS purchased - but like I said, the belt supplied was too short and another stepper motor was able to make use of the assembly and provide the dual-Z action I was seeking. Done.

"Simpler, cheaper and better" is debatable when a kit is completely dependant on a single component that isnt included. Generally, people just dont assume that coordinated design considerations were made in the first place - they assume only that poor design considerations were made. But I dont blame people for not thinking outside the box when they live in one.

I agree with the fun part, though. But only because I dig mechanical crap so much I have a trade and engineering degree in the field.

Edited 2 time(s). Last edit at 10/11/2019 05:58AM by imotion.
Re: Dual Z stepper motor sync with linear rail carriages
October 11, 2019 08:49AM
Quote
imotion

While I will always consider a belt driven dual Z axis, I have the same confidence in a grub screw being able to maintain the grip required to clamp a pulley to a leadscrew (unless it has been ground flat in that spot) as I do when it comes to the use of 2 seperate steppers.

Something about a motor on one side just doesnt give me the feeling that the gantry is being raised in equilibrium. I am pondering whether to throw a belt across the top and retain the motors on the bottom (both sides) to force it in sync six ways from sunday. Remembering these are pancake steppers, a bigger motor placed on one-side would most likely foul on the gantry before the nozzle was able to reach the bed - which was placed as low to the bottom of the frame as possible. Not that any loss of Z is a big problem, but the low profile bed is an important aspect of the design intended to minimise negative effects associated with non-coreXY designs (where the bed does not have to move back and forth whatsoever)

I don't know why a bigger motor placed on one side would cause a problem for the X/Z axis. This is the Z axis in one of the printers I built- the single NEMA-23 motor drives the two screws using a belt. It's about as asymmetric as you can get, but it doesn't matter - it works fine and I never have to make any adjustments to the X axis unless I take it apart which I've probably done twice in the last 5 years. The motor I used is a small NEMA-23 part- you can get NEMA-17 motors with the same holding torque. I used the NEMA-23 motor because it was already on-hand. I don't know why you'd want to use pancake motors unless you really have no space available. I like to eliminate potential problems in a design by overbuilding (yes, my printers tend to be heavy and even ugly, but they work reliably). If you're not sure how much torque you'll need to move the Z axis, use a big enough motor so that you don't have to worry about it and move on to the next problem. Refinement can come later, after things are working well. Building a machine to be duplicated 1000 times might get you thinking about minimizing parts cost by using just-right sized motors, but building a one-off machine doesn't require that sort of effort.



Here' the 695 mm Z axis in my corexy printer- entirely belt lifted. If you're into symmetry, the motor/worm gear is located at the center of an 8mm shaft that turns the two drive pulleys. This has been working well with minor changes for the last 3 years. The small, 2A, NEMA-23 motor came mounted on the 30:1 worm gear reducer that holds the bed position even when power to the Z motor is off. I operate it at just over 1A and it barely rises above ambient temperature and still has plenty of torque to lift the bed assembly and about 4 kg of print (which will never happen).





Recently, I have changed to glass core belts, larger drive pulleys, flipped the bed lifting brackets, and redesigned the belt clamps. A little lock-tite on the pulley set screws (and flattened shaft) ensures that they aren't going anywhere.

Neither of those printers has any sort of bed sensing other than an endstop switch to set the Z=0 position. Both can be transported across town in my car, stood back up, and started printing without adjustment.


Son of MegaMax 3D printer: [www.instructables.com]
Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Re: Dual Z stepper motor sync with linear rail carriages
October 22, 2019 03:08AM
Nice. I am starting to lean more and more towards a belt at the moment. I cant really find a lot of information about how reliable (long term) the prusa i3 slam-the-x-axis-into-the-top-of-the-frame-to-align calibration method is, I would prefer something which didnt degrade over time (if that is the case). That is the current objective at the moment, however (after I tweak the firmware to perform that maneuver), before coming back down to level the bed with a sensor.



As for pancake steppers, my bed arrangement was carried out to keep the bed as low as possible, for a number of reasons. You can see how low it is now with the pancakes before the bracket bottoms out on the motor. If I were to switch to a single motor, its not hard to see where the coupling would sit and where the new, lowest possible Z would be about 1 inch higher and I would need to space the bed that much higher as well, raising that center of gravity I want to keep low.

What I didnt think of until now was mounting a bigger (single) Z axis stepper motor at the top of the frame in the center, between the 2 leadscrews (with belt pulleys on both). Not sure of the best way to do that so it's out of the way of a direct drive extruder and filament roll.

Not a *huge* fan of floating/hanging leadscrews, but cant see that causing too much of a problem. I assume something that can support the screws at the bottom would exist, like a bearing that isnt a through-hole.

But I couldnt agree more with building stuff right the first time. My mistake was not doing 1 thing at a time - mind you, not neccesarily possible in my case - it always feels like a forever wait for the next part to arrive and I wonder how much longer until I can use the thing.



Edited 2 time(s). Last edit at 10/22/2019 06:29AM by imotion.
Re: Dual Z stepper motor sync with linear rail carriages
October 22, 2019 06:56AM
I doubt there will be any long term problems with the Prusa method- you can reduce the motor current when running the X axis to the end stops so it doesn't hit it hard. All you're trying to so is sync the screw positions.

I struggled with where to put the belt drive Z motor for a while. Right now the filament spool sits on top of the machine and feeds filament straight down through the center of the electronics enclosure to the extruder. Putting the Z motor at the top would require moving it to the back of the machine and a couple more pulleys. In the end the belt length between the drive pulleys and the load would be about the same as it is now. I heat the enclosure and had some concerns about the motor in the heat, but the motor runs at about 1/2 rated current and never gets more than a few degrees above ambient temperature so the temperature is still well within the motor's specs.


Son of MegaMax 3D printer: [www.instructables.com]
Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Re: Dual Z stepper motor sync with linear rail carriages
October 26, 2019 08:04AM
You might find some good ideas for a belted Z from this build. I'm intrigued by his 5:1 pulley reduction versus a worm gear box for the Z axis.

[elitemachineworks.com]

Edited 1 time(s). Last edit at 10/26/2019 08:25AM by Rewas.
Re: Dual Z stepper motor sync with linear rail carriages
October 26, 2019 12:08PM
It looks interesting, but I didn't see any mention of something to stop the X axis from dropping when the machine or just the Z motor loses power. Though it appears heavy, the X axis wouldn't be dropping very far, so maybe there's no need.


Son of MegaMax 3D printer: [www.instructables.com]
Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]
Re: Dual Z stepper motor sync with linear rail carriages
October 26, 2019 01:28PM
Quote
Rewas
You might find some good ideas for a belted Z from this build. I'm intrigued by his 5:1 pulley reduction versus a worm gear box for the Z axis.

[elitemachineworks.com]

ONE Z motor OK but WHY TWO Y motors ????
And two linear rails for X gantry etc....

Too complex and costly for a Prusa.

With the same parts, same cost better build something else.


"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)
Re: Dual Z stepper motor sync with linear rail carriages
October 27, 2019 09:13AM
He claims the 5:1 pulley reduction to the drag of the z motor is more than sufficient to hold the x-gantry up. The printer was on display at ERRF ( see videos on youtube) and it indeed was stationary with no power applied.

[youtu.be]

He designed this printer to run at high speeds and admits that it is over-built, but the proof is in the prints. That thing produces some of the nicest prints I have ever seen.
Sorry, only registered users may post in this forum.

Click here to login