LaserUser.com Forum (formerly RDWorksLab)

I got this weird problem happening with my cutter, co2 80w reci, ruida 6645g controller.

When i cut something, the line is crisp going up towards 0 but is blurry / weird looking when going down on the Y axis. X is fine. Mirrors are aligned perfectly and hitting dead center of the tube. everything tight, nothing loose.

Wow, that's REALLY weird. A few questions, if you don't mind:

What happens when you cut a circle?

Does this happen at all powers and speeds?

If you turn off the power to the machine, and move the Y-axis by hand, does it feel any different when moving towards you vs moving away from you?

Does this happen at all parts of the bed?

Hi Jeffrey,

I tried the power off thing, feels the same pulling or pushing. As for all powers and speeds thats a yes. circles are the same, basically whenever the move involves a down it's blurry like the photo above. so only half the circle is clean, other half not so.

Almost like you have bad / loose Y rail bearings, or your belt tension is too tight.

Dave Vigness wrote: Tue Sep 08, 2020 Almost like you have bad / loose Y rail bearings, or your belt tension is too tight.

wouldn't that cause the line to be bad going both directions though ?

Depends on your belt geometry. If your belt is going up or down between pulley and gantry, it will add stress the closer it gets to a drive axle or front pulley. If it's been tight and wearing on the bearing block for a while it would start to wear. If you are crisp up but blurry down, I would expect that your attachment point on the gantry is lower on the close side than the top of the pulley. As it pulls forward, it torques the gantry enough to take it off the sweet spot on your lens.

Put a long piece of something, a piece of wire, a long twist tie, something you can visually watch, then move your gantry in Y and see what you have. Like I said, I would suspect the bearing blocks, and most likely the front edge. If it was the rail you'd get it in both directions. I have the same problem. The adjusting block on the back has it's top edge parallel with the top of the pulley, but the hard point in the front has the belt coming up from underneath. If they had put the belt over the top and down it would at least have been close. Just another one of those things I need to fix in the excellent Chinese engineering.

Dave's hint that the axis may be skewing is a good one.

I just noticed that your bed size is 1200 x 600 - considerably larger than my 500 x 300.

How is your Y-axis gantry driven? A double-shafted stepper motor, or some other method? Describe in detail (or provide pics) how it's driven. I'm starting to think the gantry is off-square (let's say 89 degrees to the side rails) when moving up, and skews to be 91 degrees on the way down.

Is there enough "play" that you can skew the gantry by hand?

Skew or torque, either one will throw it off. The first test of that is to do a line on either edge. If it's skew, it will happen on one side and not the other, unless you did you #3 alignment dead center of the travel. If it's torque it will most likely show up on both sides. The caveat to that being if you have something loose in your Y axis drive, you could still get something screwy going on. Another test you could do is to turn your Y spees way down, like 5mm/s. Move the gantry a bit, then set blocks of some sort against the gantry, then put it into step mode and jog it one or two steps. See if the gap is equal side to side. Do that both directions.

Dave Vigness wrote: Tue Sep 08, 2020 Skew or torque, either one will throw it off. The first test of that is to do a line on either edge. If it's skew, it will happen on one side and not the other, unless you did you #3 alignment dead center of the travel. If it's torque it will most likely show up on both sides. The caveat to that being if you have something loose in your Y axis drive, you could still get something screwy going on. Another test you could do is to turn your Y spees way down, like 5mm/s. Move the gantry a bit, then set blocks of some sort against the gantry, then put it into step mode and jog it one or two steps. See if the gap is equal side to side. Do that both directions.

wouldn't skew or torque affect both lines ? that pattern happens anywhere on the bed. As for the layout of the printer, it's kinda sideways, Y is the engraving axis (the one with the issue) X is the long axis driven by a nema 23 with a shaft connecting both sides. The gantry (one having issues) is driven by a nema 23

For me, skew is when one side is not parallel with the other, or one side is not being driven the same as the other. Torque is a twisting motion.

If you have a solid Y drive shaft, and it's being driven from one side then skew is pretty much out the window. If you have one that has the drive motor in the center, or it has a coupler of some sort in the middle, you could have an issue with one side not being driven the same as the other. Some slop in one coupler or the other.That would allow one side to advance slightly more than the other throwing your beam off between #2 and #3.

Torque, the twisting of the gantry, has roughly the same effect between the #2 and #3, but in a different motion. Instead of the ends of the gantry not traveling together, the gantry is lifting on one edge as it moves forward. To me, from your description, the front edge of the gantry is being lifted slightly as it is pulled forward. But...

The skew problem could produce the same effect.

Don't know if it will tell you anything, but do you get the same thing if you do a series of lines, all moving forward? Like a dashed line kind of thing. Don't remember if you said if speed made a difference. Is a line at 10mm/s just as bad as one at 25mm/s? That kind of thing.

Do not assume all the couplers are tight. I found one in my Z drive that had me thinking the gear reduction was going out. It was slowly getting worse. After a bit of investigating, found a pair of loose set screws. A bit of back and forth to get the set screw centered on the flat, and every body is now happy. Me included.