Direct drive with stepper on rack and pinion.

[jerryrigge]

Someone else brought this up but there was not much comment on this issue.

Most rack and pinion drive systems on 4 foot x 4 foot tables use a 3 to 1 reduction for the stepper to rack. I copied the others and built a similar reduction system for my x axis. Works fine.

Yesterday I connected the stepper on my y axis direct to the rack. Less than 1 to 1 ratio. Seems to work fine and moves extremely quick. I tuned the motors in mach2 for about 200 inch per minute. Tried to hold the y axis back with my hand and the motor has plenty of power.

A stepper motor power curve has more torque at lower speeds so the slower the steppers turn the more happy they are.

I am puzzled. Is there any reason that I cannot run the steppers direct to the rack?

JR

270 oz steppers, 50 pound bridge, plasma table 4 by 4, gecko drives, mach2

[Al_The_Man]

The main reason for using reduction on a rack and pinion system is to allow the smaller motors to be used and keep the recommended motor to load inertia ratio below 10:1 and in some cases below 5:1. This of course depends on the weight of the gantry and the type of operation, Plasma, Routing, Milling etc.
If you want to obtain certain feed and acceleration rates for a given gantry mass then these calculations are important. Often the best bet is to use the highest reduction you can and still maintain your target top feed rate.
Al.

[jerryrigge]

I understand that objects in motion tend to stay in motion, objects at rest tend to stay at rest. In order to overcome the mass of the objects and the static friction, force must be applied. A lower gear ration facilitates overcoming this resistance.

If one is willing to accelarate more slowly then a higher ratio is acceptable, taking into account the heat that might be produced in the motor.

Putting the theory aside for a moment, can anyone say I tried it and it worked or I tried it and it did not work?

Thanks,

jr

[fyffe555]

I thought about this and stalled when working out the resolution. If you have a 200spr stepper that's 1.8 degrees as your smallest resolution possible. By the time you include the pinion diameter, say 1", 0ne step becomes 0.015".

Andrew

[jerryrigge]

My understanding is that the Gecko drives have micro stepping. I beleive that they provide 2000 steps per revolution so I do not think resolution would be an issue. Earlier, taking into account gear reduction, I was running at 8500 pulses per inch or 8500 steps per inch. (feel free to correct me if this is not correct)

Even .015 would be OK for the brackets and artwork that I cut.

My concern is about things such as the side bearing load on the stepper which might cause it to miss a step... or a bunch of steps.

I have seen no one running steppers direct to rack and pinion so I assume there must be an overwhelming reason. But what is that overwhelming reason?

[Al_The_Man]

I have seen no one running steppers direct to rack and pinion so I assume there must be an overwhelming reason. But what is that overwhelming reason?

I would say the reasons in the answer I and Andrew gave earlier pretty much sums it up?
Al.

[pmurray]

jerryrigge,

No reason why it should not work if you are satisfied with the Accel/Vel/Decel profile that is optimized for the inertia in your system.

Prior to moving to voice coil actuators, the early 5” and 3” hard drives for small computers used to use a rack & pinion, as you are suggesting, and a simple optical encoder to initially find track zero – the positioning system was all open loop. A little slow for seek access times but, other than that, it worked just fine for track densities less than 1000 TPI.

As mentioned in previous posts, the inertial mass of your system has to be taken into consideration, which is going to be considerably more than that of the actuator of a drive. Just depends on what you are looking for in the way of performance.

plm

[ger21]

My understanding is that the Gecko drives have micro stepping. I beleive that they provide 2000 steps per revolution so I do not think resolution would be an issue.

It's not a good idea to consider the microstepping amount when figuring resolution. When microstepping, each microstep only gets a small percentage of actual motor torque, and the machine may not actually move until a few microsteps are issued. Read this for a beter explanation. http://www.machinedesign.com/ASP/str...tedArticle.asp

[jerryrigge]

Thanks for those who replied.

The machine is set up direct drive now. I will let the machine run for a while then reply to this thread with the results. So far the direct drive axis darts back and forth. I need to slow the acceleration and speed down before it jumps off the table.

It could be that it will miss steps due to the inertia issue already cited.

I will post results in about Monday evening.


jr

[DEEDEE]

Hi jerryrigge,

I was the one that was asking questions on the other forum page about direct driving a pinion on a rack, i am really interested to know how you got on with this yourself, could you give me the pros and cons with this, as i am about to set this up myself and would like to know if it works.

Thanks for any reply.

DEEDEE

[jerryrigge]

To answer DeeDee as to why I wanted to try direct drive on rack and pinion...it is that it is the most simple. I have been building this machine full time for 2 months and know first hand a lot of things that will not work very well. Often times an idea seems great but when tested proves to be impractical. Sometimes an idea gets a lot of scorn in the forums while in fact the idea might work well for some applications. If anyone is interested I'll share what I have learned in another thread.

As to my dual drive rack and pinion axis...seems to work fine. This axis is NOT direct drive, 3 to 1 reduction. They stay in correct timing unless one of them binds or experiences a mishap. Lots of builders use dual motors, Shopbot and Torchmate to mention two. Other expensive units also use two motors on one axis. Having done it, I doubt that I will use two on my next table ( I am trying to develop a plasma table that I can market). Unless someone needs the power or has some other reason, a one motor system is likely a better choice. (Torchmate makes a table that has one high side and one low side raill ...they really needed two motors)

As to the direct drive axis...it is a .75 to 1 ratio. That means it is in overdrive constantly. I worked with the motor tuning in Mach2 and got the acceleration and velocity tuned really nice. My bridge is very light and the motor is only accelerating and decelerating about 10 llbs. As most people know a stepper has its best torque at low rpm so I think this helps the 270 oz stepper to do the job. Acceleration and deceleration have not been a problems although I see potential issues.

Having said all other that, I do not think that I will use it on my next table.

Part II to follow after supper.

jr

[jerryrigge]

Regarding direct drive stepper to rack systems, although workable, present several issues that can become problems. Because they use a high gear ratio (mine is .75 to 1, overdrive) they operate on the margins of needed power. If they cannot accelerate as fast as needed they will miss steps. You might tune the motors using Mach2 or other program and make it workable. I did. But it is not flexible. My acceleration and velocity setting must be set right.

Motors seems to be a little bit warmer than it should. Demands on the motor will be higher, although some motors will certainly provide enough power. Don't expect to use an under powered motor direct drive.

Direct drive will dart back and forth like a rabbit. It will go faster than is practical.

Bearing load is another concern. If the rack has any irregularities in it then the bearing on the motor will get that extra pressure. If the bearing binds a little it might miss steps.

Just my initial thoughts after 3 hours of testing.