2 steppers 1 axis. Why not?

[Blackened]

*edit* Damn sorry, posted in servo sub-forum instead of stepper sub

I'm tossing up the pros and cons of using 2 steppers, or 1 stepper with mechanical linkage for my belt driven hobby laser gantry. Running a smoothieboard clone. The gantry spans about 570mm (22 1/2 inches)

I can't decide!

Is missed steps on one of the steppers really a problem to be concerned over, under working conditions?

I'm leaning towards 2 steppers because:

• Individually, they can use separate drivers, so they won't overload any single onboard driver.
• Mechanically simpler.

But on the other hand:

• A single stepper would make it easier to track down any motion related errors

Cheers

Similar Threads:

• Need Help!- using 2 steppers on x axis
• 2 steppers on one axis
• 2 Steppers For one Axis
• Different steppers on each axis?
• 2 steppers on same axis

[A_Camera]

Missing steps is a reality and with two steppers you need to constantly calibrate. With a single stepper you just need to make sure it is levelled and missing steps is never a problem. 570mm is not a huge span, easy to link but yes, two steppers is easier to install.
I am right now in the process of modifying my 3D printer from two steppers to one and a belt in between. When I am ready it will be much better than with two steppers.

[Blackened]

Missing steps is a reality and with two steppers you need to constantly calibrate. With a single stepper you just need to make sure it is levelled and missing steps is never a problem. 570mm is not a huge span, easy to link but yes, two steppers is easier to install.
I am right now in the process of modifying my 3D printer from two steppers to one and a belt in between. When I am ready it will be much better than with two steppers.

Thanks for the input . Did you nail down the reason for your missed steps? I was expecting missed steps in the process of testing, but once I had found a good acceleration rate and max feed that the missed steps (hopefully) wouldn't be a constant problem?

[A_Camera]

Thanks for the input . Did you nail down the reason for your missed steps? I was expecting missed steps in the process of testing, but once I had found a good acceleration rate and max feed that the missed steps (hopefully) wouldn't be a constant problem?

I haven't really experienced missed steps, my printer was just a week old when I started with this upgrade, but it is enough to power off and on a few times or to touch one screw and the screws are out of sync. That can't happen if you have one stepper.

[109jb]

Missed steps for a laser or 3d printer would mean there is a problem in the mechanics of the machine, or the stepper are not appropriately sized. I have a3d printer with 2 motors on the one axis and have never missed a step in many hours of use.

Sent from my Z958 using Tapatalk

[Blackened]

I haven't really experienced missed steps, my printer was just a week old when I started with this upgrade, but it is enough to power off and on a few times or to touch one screw and the screws are out of sync. That can't happen if you have one stepper.

Ah I see. If the two sides of the gantry aren't linked mechanically, and the motors are not powered, they could go out of sync from a stiff breeze lol. I'd have to place a hard end stop on each side of the gantry to bring them both into line before powering on. I've seen multiple steppers on an assortment of designs. I wonder how they all handle this issue? Just rely on gantry being stiff enough that it doesn't go out of whack far enough to matter?

[Blackened]

Missed steps for a laser or 3d printer would mean there is a problem in the mechanics of the machine, or the stepper are not appropriately sized. I have a3d printer with 2 motors on the one axis and have never missed a step in many hours of use.

Sent from my Z958 using Tapatalk

Thanks for the input I guess it depends heavily on the type of machine, but have you had issues with the axis moving when the motors are off? As A_camera has pointed out it's not necessarily a missed step that would cause the 2 sides of the axis to go out of step.

[NIC 77]

*edit* Damn sorry, posted in servo sub-forum instead of stepper sub

I'm tossing up the pros and cons of using 2 steppers, or 1 stepper with mechanical linkage for my belt driven hobby laser gantry. Running a smoothieboard clone. The gantry spans about 570mm (22 1/2 inches)

I can't decide!

The mechanical linkage could introduce more problems than you are trying to solve.

Is missed steps on one of the steppers really a problem to be concerned over, under working conditions?

No, it isn't generally a problem to be concerned with. When you power on and off, you may have a very slight shift between the two motors, over a span of 22 inches, you won't notice it, and there are ways to home the gantry each time you power on which eliminates this. Not really a problem or something to be concerned with for you at all IMO.

So a belt driven laser needs fast speeds and accelerations. I have to think the mechanical linkage would hinder your performance. Can you post an example of one? Have you seen anyone else do this on a belt driven laser? I don't own a laser so excuse my ignorance please about lasers or belt driven systems. I am talking about CNC in general.

Use good components / drivers if you want to eliminate problems.

I haven't really experienced missed steps, my printer was just a week old when I started with this upgrade, but it is enough to power off and on a few times or to touch one screw and the screws are out of sync. That can't happen if you have one stepper.

So you're getting a shift between them, have you considered trying a homing routine to square the gantry each time you power on instead? If the linear rails and frame aren't sufficient to hold the square of the gantry, then IMO, you are better to try a homing routing or something, because at least that way the two motors are always helping to keep it square once aligned from the homing.

Are you doing anything interesting with your 3D printer, like multiple printing heads, etc?

I plan to get into 3D printing this year sometime, that's why I ask.

[Blackened]

The mechanical linkage could introduce more problems than you are trying to solve.

Thanks NIC 77. Do tell? What sort of issues might arise from the mechanical linkage? I think I've got a fair handle on it, but I haven't modelled anything yet due to my uncertainty of method. As you mention below, there will be more mass to accelerate/decelerate for one. If I drive the mechanical linkage with a dual shaft stepper, then I don't think I can get the same drive ratio as the gantry so accuracy will suffer slightly. So I'd need to use yet more gubbins to get my desired ratio.

No, it isn't generally a problem to be concerned with. When you power on and off, you may have a very slight shift between the two motors, over a span of 22 inches, you won't notice it, and there are ways to home the gantry each time you power on which eliminates this. Not really a problem or something to be concerned with for you at all IMO.

So a belt driven laser needs fast speeds and accelerations. I have to think the mechanical linkage would hinder your performance. Can you post an example of one? Have you seen anyone else do this on a belt driven laser? I don't own a laser so excuse my ignorance please about lasers or belt driven systems. I am talking about CNC in general.

So how is the squaring of the gantry achieved during homing? I was imagining a removable spacer, manually push the gantry against it to square it up, then turn on the machine to home. But if there's an electronic solution that would be preferred. I'm still on the fence as to which solution to choose.

Thanks EVERYONE for your input so far. Really appreciate the suggestions and help.

[NIC 77]

Thanks NIC 77. Do tell? What sort of issues might arise from the mechanical linkage? I think I've got a fair handle on it, but I haven't modelled anything yet due to my uncertainty of method. As you mention below, there will be more mass to accelerate/decelerate for one. If I drive the mechanical linkage with a dual shaft stepper, then I don't think I can get the same drive ratio as the gantry so accuracy will suffer slightly. So I'd need to use yet more gubbins to get my desired ratio.

Backlash, binding, extra torque required, and only one motor.

So how is the squaring of the gantry achieved during homing? I was imagining a removable spacer, manually push the gantry against it to square it up, then turn on the machine to home. But if there's an electronic solution that would be preferred. I'm still on the fence as to which solution to choose.

Thanks EVERYONE for your input so far. Really appreciate the suggestions and help.

It depends on the electronics you have. I'm not an expert on it, I only know that people do it. So better to ask specifically about the electronics you plan to use if it can do it.

It's electronic, it will run each side until both limit switches are hit.

Some more advanced solutions, like the clearpath servos will home to the hard stops on both sides. There are ways to do it. For a first build, try not to reinvent the wheel, copy something someone else has done that has worked well.

A_Camera, if you do decide to do a linkage, I'd be interested to see what you come up with. I shot you down perhaps a bit prematurely. I just don't see it working well for a belt driven laser. No offense was intended.

[Blackened]

It depends on the electronics you have. I'm not an expert on it, I only know that people do it. So better to ask specifically about the electronics you plan to use if it can do it.

It's electronic, it will run each side until both limit switches are hit.

My MKS BASE V1.3, smoothieboard clone/compatible/whatever has to have jumpers on the board to slave a second driver/stepper to the same axis. So the software is completely unaware that it's driving 2 separate steppers. AFAIK I can't home them individually. As you suggest, I might make that a separate thread.

[NIC 77]

I spent some time trying to figure this out. I couldn't find any answer for you on this.

Here's a video I found that describes in general what I'm talking about:



If this feature is important to you then you may have to dump the smootieboard. I don't know.

If you already have a smoothieboard, and others are using it for lasers like you want to build, well then, you could always start out with that and see if you have problems.

[A_Camera]

Missed steps for a laser or 3d printer would mean there is a problem in the mechanics of the machine, or the stepper are not appropriately sized. I have a3d printer with 2 motors on the one axis and have never missed a step in many hours of use.

Missed steps is a fact if you have two steppers connected in parallel. The only way you can avoid that from from causing issues is by having TWO home switches, one for EACH screw and have a proper homing routine to sync the screws, or if your motors use some other electronic synchronisation. Steppers ALWAYS start from full step position when energized and you never actually know which way they jump into position, one can rotate to the left, the other to the right and next time you start can be another combination. Another thing which can cause issues is if any of the screws are touched. You don't need to be violent to cause a slight twist of a screw, if the steppers are disabled the screws are easy to turn. This will obviously result in misalignment. In 3D printers this misalignment is hidden through using auto levellers or manually levelling the bed. This process is only hiding the actual problem because the bed should not need to be levelled every time and the extruder should not get misaligned if everything is right from the start.

Missing or extra steps as per definition caused by electrical signal problems may be rare, especially with 3D printers, but I think lasers (which is the subject here) are completely different. These can cause huge issues due to electrical noise unless properly handled. Anyway, most cheap 3D printers use a simple parallel connection with one single driver for two motors. This is an extremely bad solution, but cheap and simple. Of course, this is pretty easy to fix by adding an extra diver but the other issues mentioned above are not going away just because another driver is added.

BTW, when you say you "never missed a step"... what does that mean? You do level your bed every time or at least often, don't you? Have you measured accuracy between first use and now? Because unless you level the bed often I bet you have serious differences between first use and now. Of course, the differences may not matter to you, all that depends on your expectations and requirements, but I am pretty certain that unless your printer is a high grade high quality printer with synchronized screws you actually have issues you don't really see.

[A_Camera]

I spent some time trying to figure this out. I couldn't find any answer for you on this.

Here's a video I found that describes in general what I'm talking about:



If this feature is important to you then you may have to dump the smootieboard. I don't know.

If you already have a smoothieboard, and others are using it for lasers like you want to build, well then, you could always start out with that and see if you have problems.

Yes, that's the only way you can make sure your screws are aligned if you are using two motors. You MUST have two home switches and have a routine to handle those properly. As far as I know, Mach3 can handle slave axis, as well as UCCNC or like in thes video, LinuxCNC, but I haven't seen a 3D printer doing that, I don't think Marlin can handle it, perhaps others can.

[A_Camera]

My MKS BASE V1.3, smoothieboard clone/compatible/whatever has to have jumpers on the board to slave a second driver/stepper to the same axis. So the software is completely unaware that it's driving 2 separate steppers. AFAIK I can't home them individually. As you suggest, I might make that a separate thread.

In that case single motor and a belt between the screws is the only solution.

[A_Camera]

The mechanical linkage could introduce more problems than you are trying to solve.

I think you misunderstood his plans. He wants to use one motor and a belt in between to drive two screws. That solution is not going to introduce problems at all, but sure, if he would use two motors and a belt in between... that's really bad.

No, it isn't generally a problem to be concerned with. When you power on and off, you may have a very slight shift between the two motors, over a span of 22 inches, you won't notice it, and there are ways to home the gantry each time you power on which eliminates this. Not really a problem or something to be concerned with for you at all IMO.

I'd be very much concerned, because you have no control. One time... sure no problems, but what about hundreds of times? Also, like I have said, steppers start in full step position and you never know which way the start, so I think it does not take long before this causes binding and serious misalignment.

So a belt driven laser needs fast speeds and accelerations. I have to think the mechanical linkage would hinder your performance. Can you post an example of one? Have you seen anyone else do this on a belt driven laser? I don't own a laser so excuse my ignorance please about lasers or belt driven systems. I am talking about CNC in general.

I also don't have a laser, but have real about a lot of issues people are having with noise. Normally the Y which is slaved and I have seen several with belt in between but have no link. Of course, two stepper motors can also be used but ONLY if those have individual home switches.

Use good components / drivers if you want to eliminate problems.

Yes, sure, but none of that solves the issues I am talking about if you have two motors. That can only be solved with separate homing and automatic squaring.

So you're getting a shift between them, have you considered trying a homing routine to square the gantry each time you power on instead? If the linear rails and frame aren't sufficient to hold the square of the gantry, then IMO, you are better to try a homing routing or something, because at least that way the two motors are always helping to keep it square once aligned from the homing.

Of course, homing before each print, levelling bed in all four corners, printing on aluminium bed with glass sheet and all that... but still, the issues mentioned are just hidden by this process, not eliminated. Homing switch is only possible to have on one screw, at least in the Marlin firmware I know of and all 3D printers I have seen. But I have just started, so I am not an expert, however one motor and two screws with a belt is much better than the solution commonly used in 3D printers.

Are you doing anything interesting with your 3D printer, like multiple printing heads, etc?

I plan to get into 3D printing this year sometime, that's why I ask.

Currently I have a single extruder on the printer but I have three more extruders. I don't think I will build with more than two extruders because my aim is not to build something fancy, but something useful with and high quality, just like my CNC is. Two extruders allow me to use two colours or two different type of material and currently I don't feel I'll need more.

[ger21]

Missing steps is a reality and with two steppers you need to constantly calibrate.

More than 95% of all the routers you see on this forum are using two steppers, most without issues.
Missing steps is not a reality, unless you are trying to run your machine at the limits of it's capabilities, or you have other issues. A properly designed and setup stepper machine should never lose steps.

Yes, that's the only way you can make sure your screws are aligned if you are using two motors.

There's a much simpler method. Just slowly jog the machine into hard stops to square it up, and you basically get the same results. I used to do this with my router before I added home switches.

[A_Camera]

Backlash, binding, extra torque required, and only one motor.

None of that is true.

• Backlash depends mainly on your screws, not the belt, so two motors have just as much backlash as one motor with a belt.
• Binding depends on the alignment you start with. A sloppy build and misaligned screws will cause binding but is easily fixed. Binding is DEFINITELY going to be an issue with two motors and only one home switch, even if you start with perfect alignment.
• Extra torque required for what? Driving two screws yes, but you still handle very little weight so there is no need for anything extra. A small NEMA17 can drive my Z up and down very fast. My CNC has a 17kg Z driven by a single NEMA23 motor and at a speed of 9m/min maximum with an acceleration of 700mm/s/s, and that is more than enough. In a laser or 3D printer you don't have that huge mass, so it really is not an issue and that is definitely not a good motive for two motors.
• Only one motor is not a disadvantage if two is not needed. One motor means less current, so that's positive.

The only disadvantage I can see is that it is more complicated to build because you must figure out the belt and a way to have the free route required between the screws. The 3D printer I am building will have the belt under the bed, but the current one has it on top, which is not nice but is good enough to test the concept.

A_Camera, if you do decide to do a linkage, I'd be interested to see what you come up with. I shot you down perhaps a bit prematurely. I just don't see it working well for a belt driven laser. No offense was intended.

Don't worry, it is good to be questioned because it gives me an opportunity to see problems I may have not thought about. I am waiting for two idlers before it is ready. Perhaps I'll make a temporary solution and print my own idlers if I get tired of waiting, otherwise everything is ready, my plastic fantastic Geeetech is upgraded and ready.

I have no idea about lasers, but I would definitely NOT connect two steppers in parallel without having two separate home switches and proper routine to handle homing.

[A_Camera]

More than 95% of all the routers you see on this forum are using two steppers, most without issues.
Missing steps is not a reality, unless you are trying to run your machine at the limits of it's capabilities, or you have other issues. A properly designed and setup stepper machine should never lose steps.

There's a much simpler method. Just slowly jog the machine into hard stops to square it up, and you basically get the same results. I used to do this with my router before I added home switches.

Yes, all that is true, but perhaps you have not read the whole thread... we are talking about two motors with one home switch only versus one motor, two screws and a belt in between. Two motors are going to be misaligned in no time if you don't align them on regular basis. One motor, two screws and a belt will never get misaligned once it is aligned when you start it up for the first time.

You can't do what you are describing with that solution, you need two switches. Try and you will quickly discover the issues I am talking about. Also lasers generate a lot of noise, so considering all the issues people seem to have with spindles, laser cutters/engravers must be even worse. But yes, build a properly designed machine, don't save on parts, have a large budget, use the right software/firmware/driver all the other parts solves many problems. Never the less, my comments are for the alternatives the OP brought up, not that there are no other solutions than mine.

Yes, slowly jogging into hard stops may work, but it is really not a nice solution. Personally I'd not even consider it.

[ger21]

Yes, slowly jogging into hard stops may work, but it is really not a nice solution.

It's an incredibly simple solution that works, and requires no switches or additional components. For this application, it would work perfectly fine.