Page 1 of 3

Possible Crooked Flange Bearing?

Posted: Fri Jan 31, 2014 3:53 am
by MeanderBolt
Hey all,
I have a frustrating bit of nonsense I have been trying to sort out for nearly the last 50 min or so (not including post). What I have boiled it down to is that the flange bearing on the Z axis has a crooked ring in the center. Here is what is going on. I have the M8 threaded rod. When I insert it into the flange bearing and lock it down with the 2 M8 nuts, it then has a 12.5-13mm wobble. I can see this reflected (smaller diameter wobble) above the flange bearing. I was initially thinking that the rod was bent, but every time I remove it, it appears to be flat. I can roll it on the table and it is as smooth as can be. If it had over a centimeter of bend (if it were the rod) I think one would see it flopping along when rolled on the table. Wanting to verify that it is not perhaps the location of the table, I can flip the rod end over end and roll in again and it is super smooth. I have retried assembling everything thinking that perhaps I didn't get aligned right on the M8 rod, but I get the same results. So frustrating because I was full of steam after work to gain some ground on this.

Re: Possible Crooked Flange Bearing?

Posted: Fri Jan 31, 2014 4:11 am
by MeanderBolt
One more thing to add, if I flip the rod over and lock it down with the M8 nuts in the bearing, it will wobble roughly the same amount.

Re: Possible Crooked Flange Bearing?

Posted: Fri Jan 31, 2014 4:25 am
by jacob32123
I just got my kit and have been trying to fix the same problem. My threaded rod is straight, so it's either the nuts or the bearing. I've tried flipping the rod, the nuts, the bearing, everything. I emailed Inventables. We'll see what they say. I don't really want to wait 5 days to get more parts from them though.

Re: Possible Crooked Flange Bearing?

Posted: Fri Jan 31, 2014 6:27 am
by raj
I had this problem too.. it took a while to fix it.. Here is what I did:

First, I thought the problem was that the nuts were not sitting flat against the bearing, so I put a washer on each side (3/4" washers with 1/4" hole will fit perfectly on the bearing: actual o.d.: ~0.72in, i.d. ~0.31in). After trying to tighten down the nuts on either side of the bearing with washers, I had only improved the situation slightly. However, I left the washers on.

So then I did something I am quite ashamed to admit. I rotated the rod so the bottom of the rod wobbled away from the z-axis makerslide as much as possible.. and then.. well.. I just pushed on it. Seriously. I have no idea why this worked, but I got the wobble down to a negligible amount. Somehow applying pressure on the rod fixed the issue with the bearing. Now that the rod was not wobbling anymore I was able to insert it into the derlin nut without issue, and the z-axis is working fine.

Re: Possible Crooked Flange Bearing?

Posted: Fri Jan 31, 2014 9:38 am
by cvoinescu
The problem is fairly common. Ordinary nuts aren't always threaded straight -- they often sit slightly tilted on the rod. Depending on how the nuts happen to be threaded, when using two of them, the effect could cancel out, or add up. Washers can sometimes help with this, or simply try to flip one nut over and see if it improves things. Also, the surface of the nut isn't flat, the inner race of the bearing is narrow, and the threaded rod is thinner than the inner diameter of the bearing, so there's lots of opportunity for it to sit crooked. Try assembling and disassembling it several times -- with luck, you may get it down to a very small amount of wobble after a few tries. And do try flipping one or both nuts over, and/or swapping them round.

Re: Possible Crooked Flange Bearing?

Posted: Fri Jan 31, 2014 10:56 am
by WillAdams
Added this to the wiki:
**Note** the nuts are commodity / off-the-shelf parts and have a fairly wide manufacturing tolerance. If your threaded rod wobbles unduly after assembly, mark the nuts so as to indicate their orientation and try swapping them around in terms of orientation and positioning, keeping notes on which you have tried --- you should hit upon an arrangement which allows the rod to run fairly true, or at worst case, not hit any other parts. If nothing works, contact Inventables customer service, or try sourcing new nuts (one option is to take the bearing and threaded rod into a hardware store and try different nuts until you find a pair which works).
Hardware comes in different grades, right? Would a pair of chrome plated nuts from the automotive aisle be manufactured to tighter tolerances and be more likely to work?

Re: Possible Crooked Flange Bearing?

Posted: Fri Jan 31, 2014 12:29 pm
by MeanderBolt
Thanks all. I will give a go this morning. Thanks for adding the note to the WIKI Will. I am sure this will help many folks down the way.

Hey Will, slightly off topic, but part of the same moment in the instructions, and coming from the point of view of someone who has never put this together, i.e. no hindsight knowedge... if you could modify the comment about the stepper motor...
"if at all possible, you should attach the stepper motor at this time, doing it later will probably require some disassembly"
Don't make this a suggestion, make it part of the instructions, with the added M3 washers as a part of the instructions before the instead of a note. This is not an if possible sort of thing, this is a necessity because the disassembly is unavoidable with the current instructions. You need the brass stand offs in order to attach the stepper. The standoffs are also needed to hold the 2 plates togetherwhilst fighting the flange bearing and spring. It is a chicken / egg sort of thing, but I would venture to say, that you need the offsets on the stepper, then fight the plates having something to screw into. The way it currently has us going, (with the 'if at all possible' comment) is to fight the plates being held together with the standoffs, then to realize you have no good way to attach the stepper. More, the instruction is given at a point when you are pulling together all the bits you need before the instructions. While it would appear straight forward, this is one of the goofier building moments given the clown circus of trying to get all the bits together without it falling apart.

I am sorry that is totally long winded. Does any of this make sense?

I'll share this... Last night, my son, who is learning to program said "Dad, this is like a program" I asked how he meant the comment. He said "Well they are a set of instructions in a specific order. Follow them out of order and it probably will not work." I thought it was a profound comment for a 12 year old, because that is exactly what it is. Just written in human machine language.

Re: Possible Crooked Flange Bearing?

Posted: Fri Jan 31, 2014 1:42 pm
by WillAdams
It makes perfect sense, and addresses an aspect of the instructions I've been worrying over for a while now --- I re-worked the instructions a bit, hopefully they're clearer now.

Re: Possible Crooked Flange Bearing?

Posted: Fri Jan 31, 2014 5:16 pm
by will1384
I had a bad wobble on my Z-Axis, it would shift back and forth as the the Z-Axis traveled up and down, one problem was that the Z-Axis stepper motor shaft was touching the M8 Threaded Rod, when I put a god size gap between them inside the Flexible Coupler, that got rid of most of the wobble, the other thing was adjusting the V-Wheels, but there is still a slight almost unnoticeable wobble, I may try the washers idea next.

Re: Possible Crooked Flange Bearing?

Posted: Fri Jan 31, 2014 11:55 pm
by jacob32123
I bought 2 "precision" M8 hex jam nuts from a local hardware store, which made no difference. Inventables didn't respond to me email either.