I haven't had time to work on my G0704 since I have been out of town for work so my attention is diverting toward my ultimate goal of getting a RF-45 size mill down the road. Since I am always strapped for hobby funds I am contemplating accumulating the conversion parts a little at a time before even getting the mill (ie: axis drive motors, servo drivers, breakout board......). Along those lines, I have decided to go with a servo driven system with the RF-45. I will most likely go with a DC servo system and I have been trying to research motors that people are using successfully, but there are so many posts and such a wide varience of motors and drive ratios used that I thought I would ask. It also seems that many people go total overkill on the motor sizing. I'm basing some of my decision making on a correlation between my stepper driven G0704 and its performance compared to what I want to be able to on the RF-45. Check my thinking here.
1 - The G0704 is using the 570 oz-in steppers and runs the mill at 200 IPM rapid in direct drive through a 5TPI ball screw. I will likely use 5TPI screws on the RF-45 too. Based on that the G0704 steppers are running 1000 rpm and the torque/rpm curves for this stepper show that to equate to 70 oz-in @ 1000 RPM. The RF-45 is a lot heavier, but will also use reduction drives on the servos. Additionally the torque/RPM curve for the servos is pretty flat. So, if I assume I need 2X the torque for the rf-45 that means 150ish oz-in continuous to the screws. Motor torque depends on the drive ratio between the motor and screw which will also affect rpm based on the rpm capability of the motor. Is all this sounding sort of OK?
2 - Option #1 - Based on the above, if I want 200 IPM rapids and I run a 4:1 reduction, I would need about 37 oz-in continuous at 4000 rpm on the motor. Automation Technologies sells a NEMA 23 motor that has a peak torque = 350 oz-in, constant torque = 50 oz-in, no load rpm = 4700 rpm, and voltage at 60V. Seems like this will just fit the bill. Obviously for the Z I may need more due to lifting the head, but I am figuring on gas springs or a counterweight system for the Z. This might be marginal, but does my thinking sound correct?
3 - Option #2 - To provide a bit of margin for things like drive losses, acceleration inertia, etc. , Automation Technologies also sells a NEMA 34 servo with 600 oz-in peak, 119 oz-in continuous, 6000 rpm, but it is at 90V and a peak of 38A. I will probably use Gecko drivers which limit to 80V and 20A. Doing a little math assuming that the rpm is going to be proportional to V, and torque is going to be proportional to A, gives 4800 rpm at 72V, and peak torque of 315 oz-in. Constant torque should remain the same since the constant current rating which should correspond to the constant torque rating is 7.6A and is well within the Gecko's capabilities. So I wouldn't get a "true" 600 oz-in peak motor but it would be better in the sense of constant torque than the NEMA 23 mentioned earlier. I could run this motor at 2:1 for more speed (400+ IPM), or up to 4:1 for more torque (476 oz-in at screw discounting drive losses, etc).
4 - Option #3 - This one would be a combination of options #1 and #2 using the 350 oz-in NEMA 23 motors on the X and Y and the 600 oz-in NEMA 34 on the Z. The thing here though is that the NEMA 23's are 60V motors, so the powers supply would need to be limited to that. this would put the NEMA 34 motor down to 4000 rpm which would still be Ok as I see it.
5 - Option #4 - This one is a possibility since it will be some time before I actually have the actual mill and will have time to play with some stuff. I am contemplating making my own DC servos by getting some DC motors and adding encoders to them. Looking at the NEMA 34 option above it is 119 oz-in continuous and would be 4000 rpm at 72V. Doing the math, that is 0.5 Hp. The NEMA 23 at 50 oz-in and 4700 rpm is 0.2 Hp I could hunt e-bay for some cheap DC motors in the 1/4 to 1/2 Hp range and try that too. If I went this route I would probably buy a gecko, and play with it on the bench long before getting the mill. Can anyone relay their experience with going this route?
Also, I am inclined to go the Gecko driver route at least to begin with, but am open to other options for servo control. I am particularly interested in options that would allow adding glass scales for closed loop feedback eventually. I will start with just the rotary encoders on the motors, but eventually would like to add the glass scales.
Thanks in advance
1 - The G0704 is using the 570 oz-in steppers and runs the mill at 200 IPM rapid in direct drive through a 5TPI ball screw. I will likely use 5TPI screws on the RF-45 too. Based on that the G0704 steppers are running 1000 rpm and the torque/rpm curves for this stepper show that to equate to 70 oz-in @ 1000 RPM. The RF-45 is a lot heavier, but will also use reduction drives on the servos. Additionally the torque/RPM curve for the servos is pretty flat. So, if I assume I need 2X the torque for the rf-45 that means 150ish oz-in continuous to the screws. Motor torque depends on the drive ratio between the motor and screw which will also affect rpm based on the rpm capability of the motor. Is all this sounding sort of OK?
2 - Option #1 - Based on the above, if I want 200 IPM rapids and I run a 4:1 reduction, I would need about 37 oz-in continuous at 4000 rpm on the motor. Automation Technologies sells a NEMA 23 motor that has a peak torque = 350 oz-in, constant torque = 50 oz-in, no load rpm = 4700 rpm, and voltage at 60V. Seems like this will just fit the bill. Obviously for the Z I may need more due to lifting the head, but I am figuring on gas springs or a counterweight system for the Z. This might be marginal, but does my thinking sound correct?
3 - Option #2 - To provide a bit of margin for things like drive losses, acceleration inertia, etc. , Automation Technologies also sells a NEMA 34 servo with 600 oz-in peak, 119 oz-in continuous, 6000 rpm, but it is at 90V and a peak of 38A. I will probably use Gecko drivers which limit to 80V and 20A. Doing a little math assuming that the rpm is going to be proportional to V, and torque is going to be proportional to A, gives 4800 rpm at 72V, and peak torque of 315 oz-in. Constant torque should remain the same since the constant current rating which should correspond to the constant torque rating is 7.6A and is well within the Gecko's capabilities. So I wouldn't get a "true" 600 oz-in peak motor but it would be better in the sense of constant torque than the NEMA 23 mentioned earlier. I could run this motor at 2:1 for more speed (400+ IPM), or up to 4:1 for more torque (476 oz-in at screw discounting drive losses, etc).
4 - Option #3 - This one would be a combination of options #1 and #2 using the 350 oz-in NEMA 23 motors on the X and Y and the 600 oz-in NEMA 34 on the Z. The thing here though is that the NEMA 23's are 60V motors, so the powers supply would need to be limited to that. this would put the NEMA 34 motor down to 4000 rpm which would still be Ok as I see it.
5 - Option #4 - This one is a possibility since it will be some time before I actually have the actual mill and will have time to play with some stuff. I am contemplating making my own DC servos by getting some DC motors and adding encoders to them. Looking at the NEMA 34 option above it is 119 oz-in continuous and would be 4000 rpm at 72V. Doing the math, that is 0.5 Hp. The NEMA 23 at 50 oz-in and 4700 rpm is 0.2 Hp I could hunt e-bay for some cheap DC motors in the 1/4 to 1/2 Hp range and try that too. If I went this route I would probably buy a gecko, and play with it on the bench long before getting the mill. Can anyone relay their experience with going this route?
Also, I am inclined to go the Gecko driver route at least to begin with, but am open to other options for servo control. I am particularly interested in options that would allow adding glass scales for closed loop feedback eventually. I will start with just the rotary encoders on the motors, but eventually would like to add the glass scales.
Thanks in advance