Solved Z axis getting out of level

I have been getting several days of good prints followed by hours of extreme troubleshooting pain, followed by days of good prints. I thank the group for the help to the couple questions I couldn't figure out so far.

The last few days have been fine and the prints have been fine. Last night's print was normal and I shut down normally. This morning, I powered up the machine, and the auto-level function started normally on a print. It sampled the front left side, normally, then the front middle, then on the front right, it went up, but not down. It slid back to the right/middle then went up again, but not down. At mid/mid, it went down all the way to the bed and at left/mid it went down to the bed normally. Same at rear/left and rear/mid. But at rear/right, it went up again and did not go down.

As a result, I killed the print and restarted everything including the computer, to the same effect.

After several retries, it "just worked". I got a small print out of it (30 minutes), and then it went through the whole failure thing again. I restarted, rebooted and since I could SEE the carriage not sitting on the limit switch when the head was at the right side, I manually adjusting the left Z screw up just until the block was sitting fully on the switch. It succeeded in two more small prints (15 minutes each) and then failed again.

If the carriage is at mid or to the left, the weight of the carriage presses down on the limit switch, but as it moves further and further to the right, I can watch the block on the left side of the carriage lift and lift until the switch is open. Then, instead of any down movement, all I get is up.

Every time I manually adjust the left screw to "balance" the carriage, the next cycle throws it off again.

Is the left screw losing steps somehow? Is it true that the weight of the head is "warping" the long rods of the carriage enough to lift the block off the switch?

I might understand if I had changed anything or done something to the machine, but the one thing I keep finding with this printer is that a lot of times, it "just flakes out". But then most computers do too, so....

I've spent a couple hours searching the forum today, and I'm sure the answer is here somewhere, but I can't seem to find the right key words to get anything useful.

Can anyone offer any insight?

Thanks.

 

So far, the solution here appears to have been that the left threaded rod was not tight in the mounting coupling from the factory.

I just happened to notice it slip once and check the pinch bolts. The top one that holds the threaded rod was loose and the rod was working through simple friction.

I've run the G29 sequence a dozen times now, with complete success.

I'm hoping some of my other print "inconsistencies" go away now as a result!

I figured I'd just post a follow up for others that may have the same problem.

Thanks

 

I had a similar issue to this, but the fix was to lube the rods with a light machine oil and repair a broken Z axis limit switch wire.

 

OK you guys, here's some of my experience from working on 2D copiers for many years.
We had flex couplings in the machines used in an area of rapid start/stop motion. They kept coming loose.
The solution was proper tightening. There were two screws on each shaft, but they are different. One is a cup point and the other is a lock , or pointed point. You had to adjust the coupling then tightening the cup point first, then the lock point. I don't know if they are the same on the 3D machines, but it is worth a look.
Here's another fix we used. The couplings set up a harmonic vibration that causes them to loosen.
We filled the grooved out slots with RTV, a silicone rubber that stays flexible. It still compresses and expands to take up variations in allignment, but kills the vibration.
Just some thoughts from my past experience. Frank

 

Those are great things to keep in mind Frank, and I sincerely thank you for the pointers.