Connecting the Threaded Rod to the Motor

Hello!

I don't think theres documentation on how to do this yet, so I'm wondering what other people have done. For connecting the threaded rod to the stepper motor I have some plastic tubing that I screw into the threaded rod (which holds very tight) and then I have a little ring I use to clamp the tube to the stepper motor. I'm not sure how well this will work on the z-axis since there will be a lot of weight. I'll find out in a few days if this is an issue or not, but what have other people done?

Thanks

Hey,

I do pretty much the same thing. I dont have a ring, but I just force the tubing onto the motor shaft. This works great for the X and Y axes, but occasionally if the Z motor gets hot, it will fall off and the Z stage will fall. What type of ring do you use?

There are 2 things I'm going to try to get around this:

1. use a counterweight on the Z axis to take the force of gravity away or lessen it.
2. use a spring to pull the Z stage towards the top.

Any other ideas?

mellery,

I used zip-ties on mine. Works great, is cheap, and can be easily removed for the frequent tweaking I'm doing. Also, think about some crazy glue on the motor shaft end. You won't be needing to take that portion off--just the rod-coupling end--so it is okay for it to stay permanent IMO. Gorilla glue works well too.

I have been using windshield washer tubing for my coupling. It is very rubbery and provides a nice tight fit on the 1/4 shaft motors. You can get it by the foot at most auto supply stores.

Demented

There a little metal rings in the same section as the tubing at Lowes. I didn't keep the little bag they came in so I dont know the part number but they were about 50 cents for 2. They seem to work pretty well.

ps: my arduino gcode parser is almost done!

Endfloat

I was messing about with a motor that was kicking about here and toying with the McWire repstrap lead screw idea but using M6 stud instead of M8 (M6= 1 turn per mm, M8 = 1.2mm per turn) Also because the motor end shaft was 6mm.

Something I did notice is that the motor had a degree of endfloat that was greater then the required 0.1mm of precision I wanted from the mechanism.

endfloat is where the shaft can be moved in and out of the motor along it's length. Motors were never designed for loading and precision in this direction. It normally has to be taken out in the bearings and supports etc. (Plane bearings have no way of preventing this movement)

On a Z axis gravity effectively preloads the mechanism and the slop won;t be as apparent.

On the X and Y axis though the situation is problematic.

How are the motors you guys are using for endfloat ?????

cheers

aka47

---

Necessity hopefully becomes the absentee parent of successfully invented children.

i'm not sure... but i imagine there is some float there. i'm not sure if its enough for me to worry about yet.

what are some options to counteract it? (preloading, etc?) could we implement them in an easy/cheap mcwire style fashion?

Guys - this kind of thing is very common in the manufacturing world. Any kind of screw will have backlash in its movement because if the nut is tight enough that there is no backlash then the nut will lock into the screw and seize. The way to fix this is if you must move in the reverse direction then you must move further than the about of backlash is. (IE. if there is 1.5 mm of backlash in a screw/nut combination the you need to move about 2 mm in distance back so that the movement forward will be the same everytime). This is very common with using a metal lathe.

You also could preload the movement distance and have the nut alway on the same side of the thread. One method for a mcwire is to use 2 coupling nuts with a spring between them that forces them apart thus preloading the nuts in each direction. But this does not fix the backlash problem in the motor. Maybe a spring pushing against the end shaft of the motor to remove the backlash in it.

---

Bob Teeter
"What Box?"

As bob suggests backlash is common and a sprung nut on the lead screw is a common solution.

However the issue I am describing would still be present if the nut on the lead screw were preloaded (like the z axis assembly on the Darwin this is spring preloaded, although arguably it doesn't need to be because gravity is a pre-load in itself).

All motors are deigned to take a load that is perpendicular to the shaft not in line with the shaft.

In the least expensive motors they use plain bearings ie sintered bronze or some such.

More expensive motors may use ball bearings. (SOme steppers i.e. from floppy drives etc I think use ball bearings)

Consider a car wheel which must resist forces of a high magnitude that bear along the hub as well as across it.

A car wheel uses roller bearings that run on a conical surface and these are used in pairs with the cones arranged to cope with a force in each direction ie <> These are known as taper bearings. When you tighten up the wheel nut it nips along the bearing pair and results in a pretty tough combination.

Skate Board bearings have balls that run in a deep groove and similarly (but cannot resist as much end load) resist the side to side movement or end load.

A plain bearing or bush has nothing to stop this movement at all.

OK first things first, take your motor and measure how much end float there actually is, be aware that this will increase with wear. A Dial gauge if you have one can be used to measure this. A less sensitive method is to use some Vernier Calipers. Some motor specifications will specify the amount of end float and how much you may/may-not end load the motor shaft.

Basically push the shaft as far in as it will go (use reasonable force ie by hand) and measure the case to end of shaft length. then pull the shaft as far out as it will go measure again and see what the difference is. that figure is the end float.

If your end float is suitably less than the precision you want then you can safely ignore it if the spec for your motors says they will cope with the amount of end force that you intend to apply.

If your end float is greater than the precision you want to achieve then the best way is to mount the lead screw in bearing/s with collars etc to make the bearings take the end load. As the motor then has no end loading it will last longer and precision due to end float is then as good as the bearings.

Skate bearings are not great (they were not originally intended to provide precision in an end loaded way) but are likely to be able to take an end force well in excess of what you are planning on using them for.

I have some ABEC 3 bearings and the end float on them looks to be .1mm or less.

I don't have much in the way of precision measuring gear and the Verniers aren't great at resolutions of .1 or less.

So in answer to your question Zach sir, my bestest idea is to mount a higher precision skate bearing to something solid via the outer race (Plumbing Blocks are usually used for this)and put a nut/washer either side of the bearing and clamp your lead screw to the bearings inner race, mount your motor to the same solid thing and couple it directly to your lead screw.

The bearing then should take the end load whilst at the same time reducing the end float to that of the bearing.

Bob's suggestion of spring pre-loading the drive nut should also reduce/remove the backlash from the lead-screw/drive-nut combination.

BTW I haven't seen plumbing blocks for 608 bearings but these should be eminently RepRapable. If not a plumbing block then at least a mounting bush. I have been suggesting we may need some of these as a printable item in the "What else can we print" section of the forum and have described them there.

hope this helps

Cheers

aka47

---

Necessity hopefully becomes the absentee parent of successfully invented children.

OK, so this might sound goofy, but to make it so that you could neglect the forces providing the backlash, why not just use two like gears, on on the motor and one on the drive screw? This may add 4 more parts, but I'm thinking that the gears could just be plastic ones out of a knex set, cheap, sturdy, and manufactured to incredible specifications.

---

Jay

i looked at the datasheet for the RRRF motors: [www.kelinginc.net]

it says a Radial Play of 0.025mm, but i'm not sure if that is the right measurement.

yes you can take the drive out of line.

At the cost of increasing frictional losses and increasing the amount of backlash by adding in two more gears and the backlash between their mating teeth, you still need a bearing to support the lead screw because the motor isn't doing it anymore.

Overall no gain and a more losses including precision.

In drive trains the simple rule applies, the more items that are chained together (in series) the more backlash and loss of precision is introduced, also the more frictional loss is introduced. It cannot be avoided.

The direct drive is a clever way to reduce all of this to the barest minimum.

Adding a sprung backlash reducing technique to a drive nut is adding it in parallel.

Adding a bearing to reduce the end float and increase end load bearing is adding the technique in parallel.

Yes driving more results in greater losses somewhere but they are somewhat less than series addition and often the loss is aceptable for the increase in precision it brings.

Hope this explains, I can see where you were coming from though, good try.

cheers

aka47

---

Necessity hopefully becomes the absentee parent of successfully invented children.

BTW

I think Radial is at 90 degree to the shaft so is not what you need to be looking at for end float.

Many manufacturers don't quote end float as the motor isn't intended to be used in this way particularly not budget models with plain bearings.

aka47

---

Necessity hopefully becomes the absentee parent of successfully invented children.

Hi Andy,
Excuse my ignorance, but what is a plumbing block?
Chris

---

[www.hydraraptor.blogspot.com]

Ok Guys - I have 9 stepper motors from Zach. 3 from the 1st batch and 6 from the last batch. I have included a picture of my test setup so that you can see what I did and how I did it. Steel Plate is 3/8" by 3" by 12". The holder for the measuring device is a magnetic base that when turned on I cannot move short of a hand sledge (5 pound hammer). The stepper motor was held in place with a C clamp and also could not be moved.

The largest deviation that I found on all stepper motors was 1/10,000 of an inch. What I mean is that the needle wiggled but just barely. I could visually see it but the dial just came off the mark. This is about ZERO in all practical terms for what we can do with a reprap unit.
So I don't think that the end play (end float) of the stepper motor will cause any problems with what we create.

Edited 1 time(s). Last edit at 01/24/2008 11:07PM by Robert Teeter.

---

Bob Teeter
"What Box?"

Bob, good news excellent. I hadn't got any to test so am relieved.

I will still need to use a bearing though because I intend to build up something for a test bed for the Servo Controller I am working on and this does use cheep DC Brushed motors. which have more end float than we want.

nophead sir, a plumbing block incorrectly named because of too much rum is actually a plumber block aka pillow block.

Here' some pictures cos my waffle above does'nt realy explain too much more.

[www.nbcgroup.co.uk]

Hope this helps and sorry for the diversion.

cheers aka47

PS nophead I will be part over your way this afternoon, I usually exchange the kids at Birch.

Edited 1 time(s). Last edit at 01/25/2008 05:50AM by Andy Kirby.

---

Necessity hopefully becomes the absentee parent of successfully invented children.

Ah, I know them as pillow blocks. Shame, I thought you had found something from the plumbing trade as a cheap alternative.

---

[www.hydraraptor.blogspot.com]