[adam_m]

How do you determine the correct spring tension for a rack and pinion design similar to CNCRouterparts designs?

Is there a concern and detriment for using too much upward force from the pinion into the rack gear?

Thanks.

Adam,

[wcarrothers1]

I never really could sort out what I considered a good hinge mechanism for the "spring" rack deal. So I just hard mount mine while keeping the alignment as precice as possiable.. Really don't see the point in having spring tension if you kept the tollerance between your linear rail and rack mount precice..

b.

[adam_m]

I never really could sort out what I considered a good hinge mechanism for the "spring" rack deal. So I just hard mount mine while keeping the alignment as precice as possiable.. Really don't see the point in having spring tension if you kept the tollerance between your linear rail and rack mount precice..

b.

So your rack doesn't pivot then, it's a fixed distance from the rack? What about the chance of binding any worries about that?

Adam,

[wcarrothers1]

To me if things are aligned properly in the first place you should not have to worry about binding. My first machine was pretty low on the tollerance on the rack axis and still hard mounted bearing blocks and tension.. ..

First and second machine are hard mounted. Just wanted to say it's not always needed.

b.

[adam_m]

Assuming I were to go with a tensioning system how would one determine the force required?

Adam,

[adam_m]

I can't believe no one else has any ideas on how to determine spring force... thoughts, suggestions, ideas, anyone.???

[ger21]

I'm pretty sure I've seen the formula posted here before. Try the google search.

[ger21]

http://www.cnczone.com/forums/mechan...ar_forces.html

http://www.cnczone.com/forums/72304-post3.html

http://www.cnczone.com/forums/420226-post2.html

[adam_m]

Thanks ger21 exactly what I was looking for except I might need a little help in digesting the information.

So, trying to understand the math and the application here...

From (http://www.cnczone.com/forums/72304-post3.html):

"3) The separating force for 20 degree components is .364 x the drive force applied. Thus if you need 100# drive force, the spring engagement force must be greater than 36.4# in order to keep the pinion engaged."

What determines the driving force that is required? Is that the force of the stepper or the maximum force required before slippage?

How does this apply to a NEMA 380 oz. stepper?

Adam,

[moswald80]

I have been thinking about this lately as I am close to installing the R&P drives on my build. I am using the CNCRouterParts NEMA 23 drives and motors so the way I see it:

380 oz.-in. (motor torque) * 3:1 (reduction [belt from motor to pinion]) = 1140 oz.-in. (neglecting any losses from the belt drive)

1140 oz.-in. (pinion torque) / .5 in. (pinion pitch "radius" [half the pitch diameter]) = 2280 oz. = 142.5 lbs. driving force

142.5 lbs * .364 = 51.87 lbs. max separating force

This seems pretty high to me but it is the theoretical maximum and would likely only happen if the axis was run up against a hard stop with the motor trying to push through it. I would guess that half or less of that force would be required in normal practice. Does anyone have any experience with this?

Thanks,

Mickey

[adam_m]

Mickey,

So does that mean that a spring that has 51.87lbs. force keeps the pinion gear in constant pressure of the rack 100% of the time and would stall the motor before skipping a tooth?

Adam,

[moswald80]

Yes, I believe so.

[adam_m]

If that's the case then that would be bad and highly not recommended.

The concept of the spring is to provide constant pressure but not to allow the pinion/motor/rack to become damaged in case of a bind, it should "slip" and be the relief point, I'm sure this makes sense to everyone no secret there.

So, anyone with some experience that can provide some insight into what an appropriate spring weight would be given the above is correct?

I'm guessing that we might want to take the motor stall into consideration and somehow factor that in... just thinking out loud. That would allow for slippage just before a motor would stall and give some feedback that something is amiss..

Adam,

[adam_m]

I can't believe no one else can't chime in on this...

I'm no engineer and could certainly use some opinions and additional thoughts or suggestions...

Adam,

[jsheerin]

I've never heard that the spring loading was to limit damage - I've always heard it was to reduce backlash. There's no equivalent system in a screw drive machine and you don't hear about people shearing the threads out of their nuts when they crash their machines. While the plastic antibacklash nuts usually have flexible elements, they can't flex far enough to disengage the drive to protect components. Ball screws don't even have that feature.

If you're using steppers, the motors will probably just lose steps if you somehow bind it. If you run it into a hard stop at the end of travel, your limit switches should kill power (really they should stop movement before you hit a hard stop). If you're using servos, I'd think you would get out of position and the system would fault, ideally before you burned up a motor which would be up to you to design for. I think the correct way to go here is design your machine so your gears are shielded and don't get clogged with debris and use limit switches to limit the travel. Then if you're still really concerned, design your gearing to be strong enough to take the force generated by the motor when stalled or design a link into the system somewhere that will break before anything more expensive does, like a shear pin in a shaft somewhere that drives a gear.

[jsheerin]

Also, if you're concerned with jamming the gear on an obstruction, I'm not sure any of this applies anymore. You would potentially be decelerating the system at a higher rate than the motor's torque could produce, so the separation force on the gearing might be higher than the spring you've spec'd in which case you would get some compliance although not necessarily (or maybe even likely) complete disengagement. You could calculate all this out if you wanted, but personally I'd just do what I said above - design it to keep debris out and build it.

[CarveOne]

My personal opinion of this is that you aren't doing anything useful by doing all of these calculations if you have no accurate means to measure it to insure that your numbers are the actual forces being applied.

Wood, aluminum, plastics, and MDF all have differing resistance to cutting forces. Are you going to re-calculate, re-measure, and reset the spring forces each time you change materials? Your calculated numbers won't be the same numbers that work correctly for another machine either.

Just start with a light spring tension and gradually increase the tension until backlash is gone or won't improve anymore. Watch the pinion drive plate to see how much it is moving while cutting a project and increase the spring tension if it is moving more than you think it should be moving. Put a reference line on a fixed part of the machine that shows how far the edge of the pinion drive plate is moving. No math is needed. If it moves as much as a tooth height (~3/32") it will jump to the next tooth engagement position and make a loud pop. Your project just got hit with a serious case of lost steps.

There will probably be some small movement as the direction of travel is reversed. It should be momentary, and go back to the same fully engaged position while traveling in one direction. The higher the accel number the more it will move momentarily.

CarveOne

[adam_m]

"I've never heard that the spring loading was to limit damage - I've always heard it was to reduce backlash."

So why use a spring at all, would it be just as effective to simply mount the pinion in a fixed position?

Why then would ahren at cncrouter parts design his pinions with spring tension? I also notice that his newer pinions have an adjustable screw as opposed to his older ones that used a turn buckle for adjusting the tension.

Carveone, I'm not opposed to a "trial and error" approach just looking for a good base to start from and a maximum exposure so that can order some parts for it.

Thanks.

Adam,

[CarveOne]

It is not my intention to slam anyone viewing this thread, or you. It's just that I don't personally think all the calculations will make any real difference in the end. Trial and error is a lot quicker and is a lot less mental anguish. (That's me in a nutshell. )

If the tension is a little too much it's not a serious problem. If it's way too much it will cause a little more wear on the pinion gear and rack and it will load down the motors a little more.

If your rack is fully supported over its full length, and is perfectly straight, there should be no reason to need a tension spring. Most installations probably don't fit that description though. So, use a tension spring with enough tension range to follow the perturbations in the rack, with extra range being allowed for DIY builders without the ability do precision rack installations. I'm using Ahrens R&P drives and glad that he made it so simple for me to install and use. Straight racks or not. He already did all the math.

CarveOne

[jsheerin]

Wood, aluminum, plastics, and MDF all have differing resistance to cutting forces. Are you going to re-calculate, re-measure, and reset the spring forces each time you change materials? Your calculated numbers won't be the same numbers that work correctly for another machine either.

This is incorrect. The cutting force required has no bearing on the spring force used at least in the way you suggest. And anyway, the cutting force will vary all over the place even in a single material as you accelerate and decelerate and take varying types of cuts. What the spring force is doing is preventing the pinion from disengaging from the rack up to a certain force. So you pick your max force and then calculate the spring force you need. If you exceed your designed max force, the pinion will move away from the rack. So if you design for a force that lets you cut aluminum with a 1/4" bit, full engagement at 1/4" deep with an acceleration of 20in/s^2 and then you cut at 3/4" deep instead, you shouldn't expect your pinion to stay completely engaged (assuming that your motors are strong enough to supply the extra force required for the cut in the first place). But for that designed for cut and all lighter loads, the pinion would stay pressed into the rack.

It would not be as effective to just fix the pinion in place although some people do that. Inevitably there will be tolerances on the dimensions of the rack. If you set it so it doesn't bind anywhere, at some points it will be a little loose and you'll have a bit of backlash. Additionally if you make tons of small parts in one section of the machine, you might wear the rack more there. Then you'd develop more backlash in that area, but you couldn't adjust the pinion closer without causing binding in other places. Spring loading the pinion will allow the pinion to ride over these variations while still not having backlash.

A different way to do this is to use a split pinion where the two halves are spring loaded against each other. You fix the pinion relative to the rack, but the two halves of the pinion press against the opposing teeth of the rack, eliminating backlash. You're trying to accomplish the same thing in a way that's easier to build.