Help! Losing “Home” during carves

Update,

I believe the issue to be a few small things that added up to the right side Y locking up.

First, there are no loose bolts anywhere on the right side Y. I have not checked the wires inside the rail but will if the trouble continues.

Second, while my right Y screw rotates the end opposite the motor has a pronounced wobble whereas the left side does not.

Third, while I was squaring the rails after replacing the X rail, I noticed something. I had one pivot screw in the front foot of my right Y (left side was still screwed down to the table). I sent X to the back and saw the rear foot move as the X rail would travel. The foot would move left and right as the X would travel front to back. I marked the extremes of the movement and got about 6mm of movement. I decided to split the difference and screwed it down at the 3mm mark.

It seems to be working better now. Backlash is tolerable and it has not locked up with the joypad yet.

I am thinking of replacing the thrust washers with thrust bearings.

I will keep posting if more issues come up.

Thanks again to the forum and OF support for the help.

1 Like

When I first installed mine with drag chains I used the stock router power cable. I experienced some similar events while jogging high speed. I have since replaced with a shielded power cord of longer length which helps in dressing things up. I have not experienced any issues aside from homing without lifting z and really nailing a locating pin for a jig. EMI is almost as much fun as a leaky capacitor.

Joe,

Can EMI be the cause if nothing else is plugged into an outlet besides the controller and the screen?

I finished a carve last night with no issues. This morning, X would not home. In fact, after false homing, X would travel in both directions when pushing the joystick the same way! One time it would move left while pressing left, next time it would move right while pressing left.

I checked for loose hardware and connections and found none. I moved X by hand and felt no obstructions at all. I also checked the washer behind the take-up nut on the X ball screw. It would spin free.

Only because of my previous Y issue, I ever so slightly loosened the nut. I the. Tried homing again and it worked perfectly.

I believe the surface between the punched washer and the milled bearing base are not parallel. That coupled with the fact that there seems to be some wobble to the threaded portion of the ball screw leads me to believe this causes binding.

I will attempt to put a witness mark on the nut and all screw to indicat if anything has moved.

In the meantime, I am ordering thrust bearings from McMaster today.

I will keep you posted.

Dear John, I’m no expert, but have you checked your “square” of all your mechanical ball screws ? Parallelogram versus square screwed down to your work table that your OneFinity is sitting on, that would explain some binding issues on your ball screw

1 Like

Sam,

I don’t believe this is the issue. I had the X axis off and just ran the Y. When I would run Y, the right side would lock up. Also, while running this test, the Y control wires as well as all other wires were out/ off of the machine.

I have not measured yet but I suspect bent components.

Thanks for the thoughts,

Eric

Well,

After stripping the machine down to OEM original, I decided to run the machine some more to test it out and hope that it doesn’t lock up or lose steps. It now has no after market extension cables, no accessory switches, no extra thrust bearings, every power cord is plugged into its own separate circuit on different outlets, no drag chains, no 3d printed drag chain parts, and no aluminum rails for drag chains.

I had removed the X axis completely to test the Y axis and discovered that the right side Y was the culprit and was causing the machine to lock up as in the video I posted earlier. I surmised that the take-up nut was self tightening, but this does not seem to be the case.

From the very beginning, I posted here (in a different thread) that the movement of my machine was loud, but having no frame of reference, I couldn’t say that it was unusual. I am beginning to think that the right side ball nut is messed up. That is the only thing I can think of causing the loudness from the right side Y.

That being said, I have been working with this OF for the last 3 days solid, to the dismay of my wife and child. In that time, I have broken two bits, ruined every piece of wood that I placed on the spoil board, and spent hours staring at Fusion 360 trying to determine if I did something wrong in CAM. NOPE!

The latest issue I have been having is probing XYZ. I probe XYZ with a .250" steel dowel, when done, I change to my bit of choice and probe Z with the touch plate upside down. Press play and the router/bit/Z drops below the work surface by about .200" and tries to rapid to its start position.?!? I pushed on and kept trying to get it to work. I have a name sign for a client to do, so I tried it. Fail!

After the last loss of steps/home, I calmly removed the clamps and work from the board, poured a cool and refreshing beverage, reassessed my situation, and sent OF Tech Support a note stating that I would like a replacement controller, a replacement machine, or a refund.

Something is wrong with the machine if, when moving the machine with the controller or the touch screen, the movement stalls due to some sort of electrical or mechanical lock-up!

Am I wrong?

Unfortunately this is the problem with buying a tool from a smaller company. You are the repair technician. If the repair is beyond your skills then the last option is to replace parts of the whole tool. Lots of people love Grizzly tools while there’s a large number who will not buy from them. You can find lots of people who got a tool from them and had issues that couldn’t be resolved. Even when they could be repaired the purchaser also played technician to identify the problem and replace the parts.

Hopefully 1F will come up with a solution to resolve this. Maybe in the future they can make an app for the controller to chart the stepper currents. That way they could set a range that’s acceptable and come up with a simple tool path to make sure it’s in that range over the entire path.

I had an issue with my y rail earlier on but not to the point binding up, just some excess noise in the bearings and Support just replaced it and now I haven’t had a problem since. Rather than going through all that diagnosing and ruining your work pieces just have them replace it and you will love your machine and the company.

I am sorry to hear about the issues you are having, and can only imagine how frustrating this is, especially given the time you and Onefinity support have put into troubleshooting. Your guess of a faulty ball screw nut was one of my guesses as well. Did Onefinity have you do anything to confirm or dismiss this as a possible cause of some of your CNC’s issues?

OF support has been responsive but not resolution yet. They got back to me right away today when I let them know of my continued issues.

In the meantime, I have taken the X off again, as well as the Y steppers. The right Y bearing block is much more difficult to move than the left one. I believe this is the main issue. I am guessing a bad ball nut as the machine does not lock up in the same spot every time.

A replacement Right Side Y assembly would be appreciated.

while I’ve been waiting to hear back from OF, I decided to measure my rail diameter. My right side top Y rail is .001” larger than my left side top Y. I used a Starrett outside mic to measure. Curious to know how much tolerance the linear bearings allow for. If anyone wants to compare, my right side top Y rail measures 1.378”

Just doing a quick McMaster check, their 35MM linear motion shaft has a upper tolerance limit of 1.377”. Of course, I don’t know where OF gets their linear shafting from nor do I know their tolerance limits or surface finish requirements.

I was curious so do a really quick search of some related tolerances, and for the precision I am guessing we have on our CNCs, 0.001+" seems to be common for both shaft and linear bearings.

On a hunch, I loosened the two mounting screws that hold the ball nut to my right Y bearing block (the black machined part that the X rail sits on). What a difference! The bearing block is much quieter and way easier to move than ever. Loosened about 1/4 turn and it slides along the rails nicely. Either the ball screw is bent or the ball nut mating faces are not machined parallel to each other. I will investigate more as I try to get my machine reliably running.

If you take the mounting screws off completely, then rotate the ball screw nut out all the way, your eye may be able to spot any ‘wobble’ or clearance differences as it emerges from the hole in the mounting block.

So far, OF wants more video, more typing, more EMI issue. I’ve taken every single thing off the machine and it is bare stock. The X rail has been removed 3 times, I’ve been more than patient.

What I’ve found so far:

X rail is off, steppers on Y are off, slide block is very difficult to move by hand on the right side, left side moves nicely, if I loosen the two bolts that hold the ball nut in, the right slider moves very nicely. When tightened back down, it binds up again.

OF asked for videos of Y running without X and I’ve given that to them. Now they say that both sides look like they run the same?!? That was never the issue! It locks up from time to time and causes carve failures.

Ugh.

Did you have a chance to try my suggestion? Did you see a ‘wobble’ in the face of the mounting plate as you rotated the ball screw nut out of the hole in the offending Y axis mounting block? If you have a set of feeler gauges, you could see if there is a big difference in space between the back of the ball screw nut mounting face and the front of the Y axis block. Linear bearings would need to be loosened and pushed back a bit for this test.

I was going to try your suggestion but could not. The right side ball nut is so tight in its mount that it won’t easily come out. It has to be tapped out. I was able to verify that .015” shim under the outside edge makes it slide better.

I am not a machinist, but that seems like a fair amount, especially since the force is acting over a relatively short ball screw nut.

OF has come through and I believe my issue will be fixed!

OF tech has been very responsive, even on Sunday!

Thanks OF

1 Like

What was your final diagnosis. Having a similar issue. Losing xy Orgin during carve. When carve ends even sits at diffrent position than when started