So I'm still having issues with my Ox router offsetting the outside cut of

So I’m still having issues with my Ox router offsetting the outside cut of the part from the holes in it. At first it was shifting over on the Y axis and now it is doing it on the X axis always about the same amount but just enough to ruin the part all together. Any help would be greatly appreciated, strange that it’s only doing it between operations, I wish I could cut the whole part in one operation and just choose the order so I can make it cut the outside last I think that would roundabout fix it.

I am not exactly sure what you are describing. But if I do get this… I think you are talking about “drift” or aka backlash. Belt tension is the most likely suspect in that case.

My original guess was that the collets that lock your lead screw in place were loose. Now I’m thinking you are missing steps somehow. Slow feed rate down just a tad or try a different USB cable?

Grab an axis and see if there is play in it?

Ok, my belts are pretty tight now, very little slop if at all between the L&R sides of the machine. My rapid speed is set to the default 2540mmpm maybe I should slow that down but with the belts tight I know it’s not slipping teeth so how would a faster rapid hurt it?

So not about slipping belt teeth. If your motor is running too fast for the torq it provides, the motor gives out and skips a little; this is called missing steps. Usually you can hear motors make a high pitch wine for a second or two .Your program keeps running as though nothing happened, except now your out registration is off. Often happens between moves where motors are accelerating too quickly. Slow the acceleration down a tad.

Unrelated to missing steps or belt tighness. Grab your router when machine is off and see if you can move x axis side to side. Collets might be tight but not snug against your pillow blocks.

Are you using tinyg? If so, which version?

Do you have two belts for an axis? If so, they need to have the same tightness.
I agree with Todd Hewitt. You can’t have the router rapid between cutting positions too fast. Just doing a series of rapids in mid air should help you test your accuracy during rapids.

Yes, TinyG old version, not the newest one. I think it might be the rapid between operations. I’m gonna go try it and I’m gonna slow the rapid down from now on and see if that fixes it. Is there a way to change the default rapid to the new lower speed? Thanks you guys! you help me a lot and I need it!

My rapids are set to 24000 mm/min. My current resistors are set to about the 50% point. I know that there was a bug in 438.02 where it would turn off an axis that wasn’t moving when it was supposed to leave it powered. This was fixed in 440.14. I did a quick blog post about it here:
http://blog.davehylands.com/2015/04/ox-more-clamps-and-tinyg-firmware-issue.html

In my case, the z axis motor was being turned off. I have my timeout $mt set to 10 seconds. So as soon as one of the axis stopped moving for 10 seconds, even if other axis were moving, it would turn off the idle axis, which allows that axis to be moved (by say cutting forces) when it shouldn’t.

My complete settings can be found here: https://github.com/dhylands/TinyG-Utils/blob/master/Config/TinyG-20150423-135312.config.show

Hooked up my TinyG to a motor for first time tonight. You will adjust max velocity and or jerk. Using the Hillbilly CNC Bible I’ve always tuned as follows…300oz motor divided by 16oz in a pound gives 18.75 pounds resistance or force your motors should be able to apply. I used my hand to resist a rapid movement until it skips steps; should feel about 18 pounds = or -. Adjust max vel. accordingly to get maximum resistance. Adjust acceleration or “jerk” by jogging + and then - doing the same thing. This should mimic a direction change.

Dave Hylands…24000mm/min. That almost 1000 inches per minute! What motors and steps per revolution are you running. Are you microstepping? Thank for the link.

Here’s a video showing the rapids: https://youtu.be/W1im1LBqW5Y I’m using the openbuildparts store nema 23 with 8 microsteps, 200 step/rev 40 mm travel/rev. So that’s 600 rpm or about 16000 microsteps/sec.

That what I’m using too, I slowed my rapid down and have had success so far.

that was pretty quick. I have a custom build and everything is pretty beefy heavy and bulky. I’m lucky if I’m moving any faster than 50 IPM.

It sounds like it might be acceleration/jerk settings. There could also be some binding (if the eccentrics are too tight) or the gantry isn’t square.

Could it be I need to turn up the current to the stepper motors? It worked so I just ran with it.

I’d make sure that the pots are at least half way. The steppers are rated for more current than the board can give. So using less makes the steppers run cooler. Using too much less will cause missed steps

Steve, if you want to experiment and have a 30V supply available, you could try that as well, or anything in the 24-30V range(30V is tinyG max). Stepper are ultimately driven by power (voltage x amps), so if voltage is higher tinyG can deliver more power to hold position, accelerate, decelerate, etc. The actual math is more complex, as these are all pulsed inductors, but in general more Vmot, up to the limit, provides “snappier” response from the motors. This is one reason why you see folks using external drivers.