Gantry CNC - Dual Y's - Missing Steps

I hope your right about the problem being mechanical. Here are some pics.

I have marks on the stepper shafts, ball screw shafts, next to the couplers. I'm not seeing any movement

Ok seems ok, just be wary some of those cheaper fixed end bearing blocks come with std bearings rather than angular contact bearings.
Are the lead screws parallel to each other and in the same plane ? If they aren’t it may cause binding.
Also if the mounts that attach the ballscrews to the gantry are not square to the ballscrew you may get some binding.
One thing you may want to try is back off the screws that hold the ballnuts in place, just slightly, and run the gantry back and forth and allow them to find “their place”. If you tighten every part home as it’s installed it can cause a slight misalignment in each part which can add up.
When I converted my mill that is what I did for each axis, and it is a quite smooth runner.
It may also be an idea to check that there is no twist in the frame as well, it need not be “level” as long as it is flat and in the same plane. Although picking a side and levelling it can make life a little easier, as that can be your “base reference”.

those flexible plum / alignment couplings can slip as they only work through clamping force. no key / flat spot that an gun screw can be tightened against such as these type

i.ebayimg.com/images/g/e~YAAOSw5JBe~2ii/s-l300.

try adding a marker pen line on the motor shaft the coupling and the ball screw....

I like a flat spot on my drives / or screw mechanism so that there is some positive engagement.

or a keyway on bigger drives or shafts.

I’ve been using the clamp only type Oldham couplers for 5 years on my mill with no slippage. 99% of my stuff is with steel and it’s been pushed hard. Mind tho they aren’t the ones of eBay and they were damn pricey.

BeagleBrainz wrote:

Mind tho they aren’t the ones of eBay and they were damn pricey.

X 2 I paid AUD $70 or so for one.

Try finding torque ratings on eBay parts so you can make sure you don't exceed the design forces. If in doubt, remove the grub screws an spot with a small drill bit then replace the screw.

those are not Oldham couplings, which are a sliding center arrangement which is either metal or hard plastic, those have a rubber spider. in my limited experience with them they are not very good.

the Oldham coupling is for parallel alignment and is a very rigid coupling and the plum coupling is for small angular alignment or a sliding drive that has float between the ballscrew shaft end and the fixed motor shaft. they also provide a degree of shock absorbing being a mildly compressible rubber...... not something in my opinion you want for a precision drive ..... but I'm sure there are millions sold and in service

you could try the addition of a shaft locking compound such as loctite 638 (or similar alternative in your locality) to help the clamping effect and prevent stripping the threads in the coupling form overtightening a high tensile cap screw in an aluminium treaded hole.

www.henkel-adhesives.com/uk/en/product/r...nds/loctite_638.html

What I was trying to say is that the clamp type couplers do work.

Yes I have pushed the Z to it's max limit and the steppers (540 oz/in 48v@4.2A ) have missed steps but the leadscrew & stepper shafts show no sign of having slipped. Having had to remove the Z stepper as I tried to put the mill under shelf that was too low and almost dropped the mill on myself, it wasn't on it's usual stand as I'm rearranging the garage. No damage done to the machine or myself. It's just a little BF20.

So, I put set screws in the couplers today. Thinking, maybe this is the problem. I tightened everything down, marked them and ran a job. My Y's are not staying in sync.

I'm going to check calibration between the motors and see if this is not a missed step issue, but calibration issues.

The Calibration is perfect between all axis's. Even checking both Y's separately.

When I home this machine after running a gcode job, the gantry hits the home switches skewed. The Y2 axis always lagging. But that doesn't mean that is the axis with the issue. I tried to get help with the coding on the miss_step routine above, but haven't gotten any answers. At least then I would know which axis has the issue.

In that case it might be good to swap the cable, motor or amplifier with each other. If the error then follows you will know what the problem is.