Ox skipping steps. What could the cause be?

Hi, I'm new to the Ox but I have got it to work. However when I do more complicated projects the machine appears to skip steps. For example, when I was cutting a matrix of dimples (concave shapes), it was working and then all of a sudden the machine started cutting them much deeper than programmed! I have had similar problems in the Y direction too.
I'm using VCarve Aspire to design the pieces and produce the Gcode. Could the file format (Gcode) be at fault?
I have tried running the machine with a slow feed rate of 200mm/min but the problem is still there.

Any ideas as to the cause of the problem and possible solutions would be greatly appreciated.

 

Thanks for the reply. In the case I mentioned, of machining a matrix of dimples, I successfully did two sets of them and on the third, things went wrong, that is all of a sudden the bit was cutting in deeper. The three sets were all on one piece of wood so the problem occurred well away from the Y home position.
I have also been trying to machine some Celtic weave patterns using an add-on available in Vectrics Aspire. It was a fairly complex pattern and so the program would have been long. I was using a good quality American bit with a feed rate of only 200mm/min and I was machining teak; fine grained and oily so a good material. What was happening with the weave, was that a few passes seemed ok and then there appeared to be some error in the Y direction, that is missed steps. This started to destroy the good work of the previous passes. I have attempted Celtic weaves a few times without success. I got a mess each time.
All the mechanism appears to be in good working order. There are no loose bolts and the router is firmly clamped.
I'm using a xPro V1 GRBL V0.8 controller with Grbl_Panel. I have saved the code in Grbl (mm) *.gcode format. I assumed that was suitable.
If you or anyone else could suggest something to try I would greatly appreciate it.
Regards

 

I wanted to make 4 dimple matrix panels in order to fabricate a 4 sided box. I only got to the 3rd panel on one long piece of wood. I will try it again because as you say, I've got to establish if it is random or regular, or whether it relates to a particular part of the bed. The position on the bed "may" roughly match where I was having problems with the the Celtic Weave project. As the router ran backwards and forward it was some how losing its position perhaps at the same spot, but I'm not really sure yet. (I just had a thought that I could cut a small and simple Celtic pattern and then if it worked, cut a slightly more complicated one, until I ran into problems. That would give me some information to diagnose the problem.) I'm going straight to a complicated project which in hind sight wasn't such a good idea.
As for heat, it has been warm here (about 30C) but I was working outside under a veranda and the controller has its own fan so I doubt that it was over heating. On the other hand I have been running the machine at what seems to be a painfully slow feed rate of 200mm/min which elongates the job time. (The reason for the slow feed rate is that I thought it may solve the slippage problem between passes.)
I've got heaps of soft and workable Scotts Pine so I will experiment with that to try to produce some clearer evidence of the problem.
Could the motor current settings be a possible culprit?
Anyway I'll do some more testing and get back to you.
Many thanks.

 

I decided to do a very simple test which consisted of routing a 20mm deep circular hole, rectangular hole and triangular hole into some soft wood. The circular hole was 75mm in diameter and it was to be done in 10 passes. I was using a new good quality down-cut end mill bit.
The circular hole was cut perfectly and so was the rectangular hole. The triangular hole started off ok but then just stopped cutting. (See photo, sorry it's a bit blurred.) All the machine did then was just jog in the Y direction perhaps 2 steps backwards and forward. The Grbl_Panel software was still showed the program running. Finally I stopped the program. The ox still jogged. The machine would have been running for an hour or more so your idea of over heating sounds feasible. Also when the router moves in the Y direction there is a definite "chugging" sound. It is certainly not a smooth sound. I'll have a look at the Y motor current controls.
Any thoughts would be much appreciated.

 

I found I was able to tighten the router clamp so it may well have moved a bit. That may explain the problem I had with dimples being routed deeper all of a sudden.
The hanging of the system when machining the triangle recess is still a mystery but I did check the current control pot for the Y motors and found that it was set fairly low. I turned it up and then did another lengthy test with about 8 passes. I routed an oval recess into some soft wood. It was 200mm x 100mm and 20mm deep and took well over an hour to do. It did not skip steps and the system didn't hang up this time!
Maybe it is time to try another Celtic Weave; that would be the ultimate test.

 

I just read Scotty's comment. Earthing is something that I'd like to look at. When I was adjusting the Y motor current a few hours ago I noticed that the shielded cables connecting to the driver board didn't have their shields connected to anything. Maybe the motor ends of the cables are grounded, I'll have to check but I know that it is best to ground one end rather than both ends. However if they are not grounded at all then there is no shielding and so more susceptible to noise.
I'll check the grounding of the router too and let you know.

 

I've just had a chance to try another Celtic Weave using Aspire software. When it was designed and simulated, all seemed well but when I came to run it, I soon ran into problems.

What happened was:
1. It started machining a matrix of holes for the weave. (OK)
2. Then it machined the top border of the piece. (OK)
3. Then it "went crazy" and started routing slots over the top of the good work that it had just done.
4. It then went into a left right oscillation of a few steps over and over again.

This fault is similar to the one I had recently when I machined a rectangle and square successfully, only to have the system stop and go into oscillation half way through machining a triangle (see photo in a previous post.)

Originally I talked about the machine "skipping steps" but it now appears that the problem is the control system shutting down for some as yet unknown reason.

I will have a look at the controller and power supply but any ideas as to what I should do would be greatly appreciated.

 

To confirm the problem I ran the Celtic Weave project again. This time the result was worse with only a groove routed left to right! However this told me that the problem was that the Y motors weren't running! (I say this because the router moved in the Z and X direction but not in the Y.) So I turned up the current for the Y motors a bit more and ran yet another test piece. This time it almost finished when it halted but I eventually worked out the halt was not an error but simply the energy saving timeout in my lap top's operating system. I have since greatly increased the timeout period.
I'm hopeful that my next attempt will be successful. I'll keep you posted.

 

I've looked up the data.
The Motech MT-2303HS280AW steppers in my OX are rated up to 2.8Amps. The driver chips in my xPro controller are rated at 2.5Amps peak.
The current control trim pot was set about half way which must have been too low, so turning it up higher should be beneficial.
I'll keep testing.