Rack and Pinion - Revisted
I think I'm going crazy...(Long post)
I have been looking at the "Rack and Pinion" setup for the X-Axis (long axis). However, most servos and steppers have a pretty high rpm (1000 rpm+). Even those motors running 500 rpm are too fast for a rack-n-pinion.
If the pinion-gear diameter is 1" then that means it is going to travel 3.14" (Pi inches) per motor revolution. Assuming you have a slow 500 rpm motor and want a 100"/min travel then you still need a 15:1 reduction ratio. Most timing belt pully setups will only get you to a maximum of 5:1 or 7:1 reduction.
Ergo, I don't see how you can use a rack-n-pinion with even a slow 500 rpm motor and use a simple timing pully for reduction.
My servos are 2000 rpm. In order to gear it down I would need a minimum of 60:1 reduction (for 100"/minute). Now granted, a 60" diameter timing pully mated to a 1" pully on the motor would be quite impressive! But I think we can rule that out, LOL. I could do this with a worm-gear. They are very compact and it would only cost around $75. But a worm-gear is another source of backlash. Add to that any backlash from the rack-n-pinion system and your asking for trouble (Echoing Balsaman here).
Another thing to think about is this: My motors are 1500 oz-in. Whenever you do a reduction in speed it increases your torque. I calculate my ending torque after a 60:1 reduction to be 94247.77 oz-in. That is not only ridiculous, but a little scary!
So I was looking at plans for routers from MTC and CadCut and it looks like they are both using timing belt reduction on their rack-n-pinion at about 3:1 or 4:1. That is way too small for a 15:1 reduction needed for a 500 rpm motor. I couldn't find the rpm specs on either of the above so I have no idea how fast they run. But what do they know that I don't? Are their motors running a piddly 100 rpm??? I don't think so. So how do they get away with it?
Also, I have been looking at many Torque Graphs for servos and steppers (Just search Yahoo for "Torque Curve" and "Servo" or "Stepper".)
1. It seems that for steppers the lower the speed the more torque you have. It looks like a steep curve that is very high to start then drops quickly and flatens out with increased rpm.
2. For servos the torque is very high till you approach the maximum rpm and which point it falls quickly with increases in rpm. That is, The torque stays high for the first 2/3 of the graph as you increase rpm, then falls rapidly as you approach the maximum rpm.
3. In either case, low rpm means high torque, within reason. So what harm would it do to tell the software to run it at several hundred rpm instead of thousands of rpm?
Questions:
1. So, does anyone still think that I need to run my servos as close to their maximum speed to get the most torque from them? (opposite to what all the servo torque graphs say.) If so why? I'm still trying to understand this.
2. I don't see other rack-n-pinion systems doing big reductions. They are doing very small reductions. How are they getting away with that?
3. What is the best "inches per minute" to aim for? Is 100"/minute really best? Doesn't that limit you so you can never go over that? Why not set it for 150" or even 200", then just tell the software to make it go slower?
4. In light of the above, do you think a 5:1 reduction and then just telling the software to go slow would be enough for a rack-n-pinion?
5. Again, please answer any questions I'm too ignorant to ask!
P.S. I'm probably missing something very big here.
------------------------------------
Brought to you courtesy of Newbies-R-Us.
Similar Threads:
Last edited by samualt; 12-29-2003 at 07:40 AM.
I think the one factor lacking in your figuring is the motor driver. With a chopper drive you get (nearly) full rated torque at any rpm. That's how the rack and pinion machines get away with small reduction ratios. Any reduction they do use is for torque multiplication, not just speed reduction.
robotic regards,
Tom
= = = = =
We may not have it all together, but together we can have it all.
Samalt I have a digital tool machine I bought way back in the stone ages of 91 that uses rack and pinion on all the axis. They did it with a transmission setup. Not one reduction but three if I remember correctly. I ran 450 oz/in steppers at 1000 ipm with a man riding on the y axis gantry. Now admittedly in three weeks he had to send me another stepper motor as that one burned out but I routinely rapid at 300 ipm and cut up to the limit my porter cable router puts on me. I have just upgraded to the gecko 201 drive and the mach2 controller and still using the reduction transmission I can now carve at 300 ipm if the router motor will do it. I did a program comparision between the old controller and the new setup and I have now reduced my carving times by ten times. On one of the programs it used to take me 37 minutes to run because the full step system would set up an vibration in my machine and nearly shake it apart if I ran over 50 ipm cutting. Please understand this was using an interpolated program made up of lines that were about .050-.200 in in length. With the new system 300 ipm takes 3 min47 sec and there is no vibration.
Here is the transmission formula for my machine
1st gear 18 tooth
2nd gear 48 tooth
3rd gear 24 tooth
4th gear 80 tooth
final drive 2 Pi(1.5/2) this final gear is a 1.5" that sets in the rack.
Mike
No greater love can a man have than this, that he give his life for a friend.
There is picture of the set up at http://cnczone.com/showthread.php?s=...ght=drive+belt
actually the picture is at www.axxustech.com
http://www.axxustech.com/Axxus4x8/axxus_4x8.htm is the right picture tread. If I am this bad with tread I can't wait to see how badly I'll screw-up G-code.LOL
>3. What is the best "inches per minute" to aim for? Is >100"/minute really best? Doesn't that limit you so you can never >go over that? Why not set it for 150" or even 200", then just tell >the software to make it go slower?
Nothing wrong with 300 imp....
>I calculate my ending torque after a 60:1 reduction to be >94247.77 oz-in. That is not only ridiculous, but a little scary!
It doesn't matter if you go with a leadscrew or rack and pinion. With your motors the final force will be the same, regardless of how you get there, presuming you use all the available rpm to get there.
I don't think you need to reduce enough to use all the rpm of the motor if you don't want to, but it makes the most efficient use of the motors power to do so. You have lots of power, so use two steps to get 25 to one or so. 5:1 and again 5:1. Even 15:1 will work fine I would guess.
I wish it wouldn't crash.
Posting Permissions
Reply with Quote
