Page MenuHomeAleph Objects Inc

AS-PR0096: Mini 2 X-carriage having too much drag on the x-axis
Closed, ResolvedPublic


Production is saying they're seeing a large fallout of these due to having too much drag on the axis. This is after they have been broached for a second time.

Current corrective procedures being taken:

  • Sending a bag back up to cluster to be re-reamed
    • Noticed that the reamer was hitting the metal plate (corrected and techs informed)
    • Possible time to swap out the reamer for a new one (Will be replaced)

I will be taking some measurements of the parts that are currently sitting in back stock to try and see any trends.

Event Timeline

oliver created this task.Apr 11 2019, 10:32 AM
oliver triaged this task as High priority.
tutley added a subscriber: tutley.Apr 11 2019, 10:36 AM

i have mentioned this previously when this issue came up before. Often just runnning the reamer through the bore once isnt enough
i found running the reamer through 2x or even 3x gave good results.
@david.hall not sure if the techs are just running through once and calling it good, but figured id bring this up again

I will check with the techs to make sure that they are doing them more than once.

oliver added a comment.EditedApr 11 2019, 2:02 PM

@tutley Here is the data set that I gathered from the x-carriages that production was failing for being to tight. The last few are a few samples of ones that I had cluster re-reamed.

I gathered this data by measuring the distance between the 2 rods while they were unrestrained by the x-ends.

Here is the control chart that I created:

I did noticed that if you applied just a bit of pressure to squeeze the rods closer together on both ends, it would increase the amount of drag experienced by the x-carriage.

It would be my hypthesis of where the top is reading on the high side would be due to improper reaming initially. This is because the current jig reams from the bottom of the part first and comes out of the top.

@oliver it looks as if some of them are getting reamed a little crooked. We may need to make sure the fixture isnt getting worn out.

@tutley One of the issues I found is that when the reamer is hitting the metal plate serve as the bottom, it will 'skip' along the bottom and cause parts to come out crooked. I noticed and informed the cluster techs to make sure that they are not hitting the bottom of the plate when they are putting in a new jig. A worn out reamer will cause those issues as well, so the reamer has been swapped out and they should be dating them from now on.

The biggest problem is that all of those are ones that supposed are too tight on the actual axis but for the most part they are with in tolerance. There is currently a ticket for new jigs already, T5985.

oliver added a comment.EditedApr 11 2019, 2:32 PM

I just put one of the ones that production failed out for being too tight on to the mini 2 on my desk and did have a bit of drag there.

we may need to look at the x ends as well

@david.hall @EricNugent @tomc @kent Here is the ticket that I opened to try and diagnose the issues with the x-carriages being too tight.

@oliver have there been any changes to the mfg process?

@tutley As far as MER is concerned we have not heard of any changes, though I was told that they are no longer using the force gauge to check the drag on the parts in cluster.

@oliver that is because of the amount of drag that has been on all of the parts. To get them to be okay on the printer they have needed less drag than the spec, so we have been broaching all parts. They are all checked for drag before they leave Cluster, and only once inserts are put in does it appear that the drag increases. This has been an issue with the parts since we started printing them, but the issue seemed to get worse some time ago (perhaps 1 1/2 months). We should note that the drag is not due to the bushing that is installed. it is due to the orientation of the 2 parts once they have been assembled and tightened down on the printer.

@david.hall These changes should have gone through an MCO.
With proper reaming we shouldnt need to broach all the parts. Also i suspect the rise in binding may actually be coming from broaching all of the parts, as it is very easy to make the bore not aligned with the axis of motion when broaching by hand. (hence the issue with the orientation of the parts once installed, both bores are not parallel)

I know there was a binding issue when we started production, and that's why we went to a 3 point mount with a single bearing on the lower bearing holder of the x carriage. This alleviated some of misalignment that occurred when the parts were assembled.

Another note, I know the reamers were going to be replaced frequently, is this something that is still being regularly replaced? Also reaming should be slow, if the reamer is being forced through the part quickly this could be causing a crooked ream.

I know @samantha was working on getting some of the fixtures replaced, but has been very busy with brizo stuff.

@tutley Thanks, I believe that this is the ticket to figure out what is wrong with the alignment. The broaching is simply to make the parts work on the line right now so that we don't throw away all the printed parts. MER should be trying to resolve why we have the issue. Perhaps they could get some help on it.
The reamers are changed regularly and we are reaming slowly. The last information from MER was to do a 3 reaming pass (slowly) to see if it assists at all with the issue. We will be glad to print with a solid fixture when that is ready and we can see if that will help with the issue at all. Just making things work while the issue is figured out.

tutley added a comment.EditedApr 24 2019, 1:23 PM

@david.hall until the fixture gets updated with the milled one, can we try and reprint the printed fixture to see if that makes an improvement?

@tutley we can set that up. I will need to know where the file is for the fixture.

@david.hall Projects/mini-hibiscus/production_parts/tooling/fixtures/ should have what you need!

MikeR added a subscriber: MikeR.Jun 3 2019, 3:28 PM

@tutley have we gotten the new jig milled out yet?

@MikeR from Kent, when discussing design review of machined fixtures prior to fabricating them "For now, the plan is to leave the process as is, and revisit this issue with our new Manufacturing Engineer."

MikeR added a comment.Jun 4 2019, 9:35 AM

We should add christian as a subscriber to this ticket then.

oliver added a subscriber: cmerot.Jul 15 2019, 12:39 PM

A current issue that the calibrators are having is a visible z line across the calibration print that appears to be stemming from x carriages with bushings with to much side to side play.

@oliver what do you mean by a "z line"?

The one on the left is a little more pronounced than the one of the right. I have found that replacing the x-carriage double bearing with one with less wobble seems to help the issue at least it helps it enough that it can be considered passing for that issue.

tomc removed a subscriber: tomc.Jul 15 2019, 2:10 PM
oliver closed this task as Resolved.Jul 19 2019, 11:00 AM

it seems like the z line may not be completely caused by having loose bearings and only further aggravates the defect. The original issue highlighted on this ticket appears to have been resolved as I have not heard of any other issues. Will do some further investigation into the seam along the z axis and create a new ticket if needed.

@oliver its a line along the z axis? i thought the photos were pointing out the change in color that looks like a line in the xy plane

@tutley No there is a line that goes along the z axis that almost seems like its due to a issue with the cable chain as it is on both sides. Calibrators were having this issue on a few printers and removing the cable chain didn't seem to help.

I recently tightened the x-belt and that seemed to have helped the issue but not remove it entirely. After some more investigation I will open a new ticket if it's still an issue.

is it the z seam?

No I don't believe the mini calibration file has a set z seam. And it appears on both sides of the octopus, it is still seen on other prints and after updating the firmware.

@oliver the z seam setting could be set to "shortest" which may be causing it to appear on both sides