Resources
Episode # 63: Electric Actuators for Multi-Axis Configurations
Contact Valin today for more information at (855) 737-4716, or fill out our online form.
First off is understanding the form of your multiple axes. And really the first question, before you get into that, is really to ask yourself these 10 Questions to Consider in Deciding Between Robots and Modular Mechanics. Go to that episode which could be pretty enlightening. But here we are really going to talk about the modular mechanics. Modular mechanics is where we are picking, sizing and selecting one axis at a time and putting them together. Or maybe there is a pre-designed version that comes together.
If you think about an XY application, is that XY like this or is the Y on its side? Or is it upside down where the Y is hanging off of the X? The XYZ has a lot more variations than that. It could be that the Y is overhanging and hanging the Z down. It could be that the Z is sticking straight up off of the XY. Then XX’YZ is even more variable than that. So, when you say you want an XYZ system, what do you have in mind? What configuration does that look like? Those are some things to think about that you are going to get asked about.
Some factors I previously taught on is are in the LOSTPED acronym: load, orientation, speed, travel, precision, environment, and duty cycle. Those are all factors that we have to consider when picking one actuator. There are a couple other things that come into play when we are talking about multiple axes. The dynamic loading is much more important when we are talking about multiple axes then when just talking about one actuator. You could have a load that is sticking straight up off of an X-axis, and as you move, the dynamic loading is the forces caused by the acceleration and deceleration, or that velocity as it is moving as well. If you have an XY system, now you have the Y hanging off of an X and that load out there can cause more dynamic loading than just with a single axis. So that needs to be considered.
Orthogonality: if you say an XY application, or let's talk about an XZ application, the orthogonality is how square that is, so you assume perhaps that they are going to be perfectly square. Well, nothing is perfect, everything is going to have tolerance, right? So maybe it is 120 arcseconds. Maybe it is 30 arcseconds. How important is that orthogonality to you? You do not have to think about that with just one actuator.
Deflection: if you have an X-axis with a Y-axis hanging over it, and that load is out here, those bearings may deflect some. For that matter, the actuator itself may deflect some depending how strong its base is. This is very difficult to actually take into account, so your manufacturer, distributor, or your designers may say they cannot take that into account. Your X-axis and your Y-axis precision can be looked at together, but that does not take everything into account because you have the precision of the X and the precision of the Y, but then you have that deflection, which is a whole different variable. So that comes into play as well.
Cable management: If you have an XYZ system and you have a motor up here with cables, how is that cable going to be managed all the way down through that motion? Is it going to be just hanging up there? Is it going to be hanging loose? Or do you want it to be routed and look nice and neat and not get tangled or cut in the process of your application? If those cables are moving, which invariably is going to be the case with the XYZ system, the cables as they go along are going to bend. If you do not get good quality high-flex cables, that bending is going to cause the insulation to start to breakdown and crack and start to create problems with intermittent errors like hall faults, feedback problems, and bridge faults in the drive.
Same series versus mixing and matching series of actuators: you may have very different needs between your X and your Y-axes. You could go with a really inexpensive actuator for the Y-axis versus a more expensive one for the X-axis. But, do you want to use different actuators? Or would you rather use the same actuator series for those two axes in order to keep them more homogeneous and make it easier to support and use? Two actuators that are vastly different from each other may just be a little messy looking. It may be a little harder to make them mount together because they were not really designed to go together. So those are some things to consider. And that really comes down to time versus money. You may be able to cut a few dollars off the cost of that Y-axis, but maybe it is going to take more time to design the bracketry and use and maybe even get up and going just because it is so different. Maybe it requires a different motor. Time versus money is always a factor in the end. Do you want to spend more money, which is going to reduce your time to get it going, or are you short on your budget therefore you are willing to spend a little more time in order to save that money?
Having done this a lot, there are a few lessons I have learned along the way that I would like to share with you…
Every little design change has a pretty significant ripple effect. If you have sized the X, Y and Z actuators and you change that load, you have to go through and resize each actuator which ripples throughout the whole design. What is more is if you are trying and making a lot of different design changes, every one of those design changes can have a ripple effect throughout the system. As you make all these changes, it is very difficult to keep track of all the variations. So, version control and managing the changes and keeping track of those different concepts can be very challenging. And it can lead to mistakes overlooking things if you are not keeping good track or just are not doing your due diligence up front.
Not including all the cables and tubing: perhaps you asked me for an XYZ system and I know all the cables that are going to go into it because I am supplying them as they go with the motors. But what if you do not tell me about the tubing that goes to your end effector? Your pneumatic tubing, or something that goes to your process? Now there is no room in my cable management to put that tubing. So now you have the cables all nicely routed, but now that tubing is just kind of sticking off to the side. So do not forget to take into account all the tubing, all the cables that have to be routed with the actuators.
Not synchronizing the parallel axes: in an XX’YZ application, you have an X and a slave, an X-prime (X’). Those actuators can be connected together by the Y, but if the Y is too long and therefore the X and X’ are too far apart, you need to either put a motor on each X and X’ and motorized them, or you need to have a drive shaft going across them. You cannot rely always on that Y to be able to drive that slider over here on the X’ because you might cause a ladder effect and it might cause some binding and motion problems. So not taking that into account is a whole different factor as well.
I hope all this helps. This is an area I have spent a lot of time in and have some things that I can share further. If you have any more questions you can reach out to us at this email address here or this website. I am Corey Foster of Valin Corporation. I hope that helped.
If you have any questions or are just looking for some help, we're happy to discuss your application with you. Reach out to us at (855) 737-4716 or fill out our online form.
The Motion Control Show
When we come to multi-axis applications, applications where two or more actuators are connected together, also often called gantry systems, we are not necessarily customizing the actuators as we are customizing how they go together and all the bits and pieces of how they are used together. I am Corey Foster at Valin Corporation. Reach out to us at this email address or website here. We are always happy to help. Let's see what we can learn.First off is understanding the form of your multiple axes. And really the first question, before you get into that, is really to ask yourself these 10 Questions to Consider in Deciding Between Robots and Modular Mechanics. Go to that episode which could be pretty enlightening. But here we are really going to talk about the modular mechanics. Modular mechanics is where we are picking, sizing and selecting one axis at a time and putting them together. Or maybe there is a pre-designed version that comes together.
If you think about an XY application, is that XY like this or is the Y on its side? Or is it upside down where the Y is hanging off of the X? The XYZ has a lot more variations than that. It could be that the Y is overhanging and hanging the Z down. It could be that the Z is sticking straight up off of the XY. Then XX’YZ is even more variable than that. So, when you say you want an XYZ system, what do you have in mind? What configuration does that look like? Those are some things to think about that you are going to get asked about.
Some factors I previously taught on is are in the LOSTPED acronym: load, orientation, speed, travel, precision, environment, and duty cycle. Those are all factors that we have to consider when picking one actuator. There are a couple other things that come into play when we are talking about multiple axes. The dynamic loading is much more important when we are talking about multiple axes then when just talking about one actuator. You could have a load that is sticking straight up off of an X-axis, and as you move, the dynamic loading is the forces caused by the acceleration and deceleration, or that velocity as it is moving as well. If you have an XY system, now you have the Y hanging off of an X and that load out there can cause more dynamic loading than just with a single axis. So that needs to be considered.
Orthogonality: if you say an XY application, or let's talk about an XZ application, the orthogonality is how square that is, so you assume perhaps that they are going to be perfectly square. Well, nothing is perfect, everything is going to have tolerance, right? So maybe it is 120 arcseconds. Maybe it is 30 arcseconds. How important is that orthogonality to you? You do not have to think about that with just one actuator.
Deflection: if you have an X-axis with a Y-axis hanging over it, and that load is out here, those bearings may deflect some. For that matter, the actuator itself may deflect some depending how strong its base is. This is very difficult to actually take into account, so your manufacturer, distributor, or your designers may say they cannot take that into account. Your X-axis and your Y-axis precision can be looked at together, but that does not take everything into account because you have the precision of the X and the precision of the Y, but then you have that deflection, which is a whole different variable. So that comes into play as well.
Cable management: If you have an XYZ system and you have a motor up here with cables, how is that cable going to be managed all the way down through that motion? Is it going to be just hanging up there? Is it going to be hanging loose? Or do you want it to be routed and look nice and neat and not get tangled or cut in the process of your application? If those cables are moving, which invariably is going to be the case with the XYZ system, the cables as they go along are going to bend. If you do not get good quality high-flex cables, that bending is going to cause the insulation to start to breakdown and crack and start to create problems with intermittent errors like hall faults, feedback problems, and bridge faults in the drive.
Same series versus mixing and matching series of actuators: you may have very different needs between your X and your Y-axes. You could go with a really inexpensive actuator for the Y-axis versus a more expensive one for the X-axis. But, do you want to use different actuators? Or would you rather use the same actuator series for those two axes in order to keep them more homogeneous and make it easier to support and use? Two actuators that are vastly different from each other may just be a little messy looking. It may be a little harder to make them mount together because they were not really designed to go together. So those are some things to consider. And that really comes down to time versus money. You may be able to cut a few dollars off the cost of that Y-axis, but maybe it is going to take more time to design the bracketry and use and maybe even get up and going just because it is so different. Maybe it requires a different motor. Time versus money is always a factor in the end. Do you want to spend more money, which is going to reduce your time to get it going, or are you short on your budget therefore you are willing to spend a little more time in order to save that money?
Having done this a lot, there are a few lessons I have learned along the way that I would like to share with you…
Every little design change has a pretty significant ripple effect. If you have sized the X, Y and Z actuators and you change that load, you have to go through and resize each actuator which ripples throughout the whole design. What is more is if you are trying and making a lot of different design changes, every one of those design changes can have a ripple effect throughout the system. As you make all these changes, it is very difficult to keep track of all the variations. So, version control and managing the changes and keeping track of those different concepts can be very challenging. And it can lead to mistakes overlooking things if you are not keeping good track or just are not doing your due diligence up front.
Not including all the cables and tubing: perhaps you asked me for an XYZ system and I know all the cables that are going to go into it because I am supplying them as they go with the motors. But what if you do not tell me about the tubing that goes to your end effector? Your pneumatic tubing, or something that goes to your process? Now there is no room in my cable management to put that tubing. So now you have the cables all nicely routed, but now that tubing is just kind of sticking off to the side. So do not forget to take into account all the tubing, all the cables that have to be routed with the actuators.
Not synchronizing the parallel axes: in an XX’YZ application, you have an X and a slave, an X-prime (X’). Those actuators can be connected together by the Y, but if the Y is too long and therefore the X and X’ are too far apart, you need to either put a motor on each X and X’ and motorized them, or you need to have a drive shaft going across them. You cannot rely always on that Y to be able to drive that slider over here on the X’ because you might cause a ladder effect and it might cause some binding and motion problems. So not taking that into account is a whole different factor as well.
I hope all this helps. This is an area I have spent a lot of time in and have some things that I can share further. If you have any more questions you can reach out to us at this email address here or this website. I am Corey Foster of Valin Corporation. I hope that helped.
If you have any questions or are just looking for some help, we're happy to discuss your application with you. Reach out to us at (855) 737-4716 or fill out our online form.
A lesson for me is that I need to involve you earlier in the program.
You were tireless in your support and it will not be forgotten!