New Shark, but have a serious problem

[saramos]

I just got a CNC Shark about 3 weeks ago. I ran the machine for the first time this past Monday, Memorial Day. I had to work through a couple of problems that were a matter of learning how to use the machine. In the end, I managed to cut one reasonable sign on Tuesday. I decided to run the program a second time to clean up the sign a bit after lowering the z by .010". When done, I noticed that the cut was shifted in the Y axis about .1" to the + side. At the time, I figured that it was again an issue of learning how to properly use the machine.

Yesterday, I drew up a rectangle the size of the work area, and setup a pocket cut to a depth of -.003" to plane the surface so it would be parallel to the router head movement. It took several passes and some slight adjustments to the size of the rectangle to use a 3/4 bit and run to the work area specs to cut a flat that was 3/8" larger in each direction than the work area. It was while doing this that I came across what looks to be a serious problem.

On 4 different runs, while traveling in the -Y direction, the gantry would stop moving, but the program was continuing to run and the controller was still commanding the stepper to turn, but the stepper was stalled. The first time it happened, I was caught off guard and did not press the E-stop before the program had changed Y direction. When it moved in the +Y, it ran the full length of the +Y command, overshooting the expected end by an equivilent to the period of time the Y stepper was stalled. I finally hit the E-Stop and then used the fast jog to run back down in the -Y direction and it traveled past the area where the stall had occured with no probem. I then had to find my zero point in the Y and re-ran the program with no problem. I was taking off very thin layers of wood with each pass because I wanted to take off only the minimum amout of wood. As a result, I had to run the program multiple times. At various times, the same issue occured. It always happened when traveling in the -Y direction, but not always at the same spot.

I'll be contacting Next Wave for support. I'll try tomorrow, but they probably won't be available unitl Monday.

Any thoughts?

Scott

[jeb2cav]

I would check the lead screw/motor coupler for that axis. It sounds like it could be slipping for you. 2 good posts on this coupler topic and also some other slipping components are:

http://www.cncsharktalk.com/viewtopic.php?f=2&t=684

and http://www.cncsharktalk.com/viewtopic.php?f=9&t=722

I hope you'll let us know what resolved this bug for you in the end.

[saramos]

On another forum (rcuniverse) someone suggested checking wire connections as they had experienced the same issue on a different brand machine. I looked at the connector from the motor to the controller for the Y axis, and thought the zip tie closest to the connector was causing stress on one pin. I removed the zip tie, wiggled the connector, and ran through the same program where I had experienced the problem. I was unable to duplicate the problem again. Time will tell if this was the cause of the problem. If it does reoccure, I'll take a look at the lead screw coupler.

I ran a program for a sign this evening which used two bits. One for large pocket areas, and a 60 v bit for the sign.The results were rougher than I expected, but still not too bad. Here's a pic of it, and a pic of my setup. I added a couple of hairline indicators for finding the center.

Scott

[saramos]

Just an update, I've not had any more problems with the Y axis since I cut the zip tie next to the connector.

I still have an issue with tool marks, particularly where plunge cuts with the vee bit. I've cut the plunge rate down to 15 ipm, and the feed rate to 80. The problem is not as bad, but still there. I'm pretty sure the issue is with spring-back of the table when the z comes to a stop.

Scott

[saramos]

I thought I would post a followup. The issue with loosing steps in the y axis has been resolved by the afore mentioned treatment of the driver connector for the y motor, and also a light lubrication of the rails.

As for tool marks, I still have an issue there, but I no longer think it is a problem with the table top rebounding, but overall flexing of the gantry. I got a Fowler Depth Setter which is designed to set mill ends on a metal mill. I was able to modify it to work with the Shark, even though there is no conductivity from the table to the bit by adding a wire and clip that attaches to the collet nut. I was using the technique of setting the Z axis by manually turning the Z axis lead screw to lower the bit to the depth setter's surfase. I found that it was very easy to flex the gantry down with light pressure when very close to the setter. A bit more pressure would flex even more. I also notice that some of the tool marks were readily explained by this flexing. I've cut a number of signs in which I have used a mill end for flat area clearance, the a second pass with the Vee bit for final cutting of the sign. Some of the tool paths of the Vee bit travel through uncut material, and edge flat cleared areas as it mostly traverses in the x axis. When the bit is cutting through uncut areas, there is equal pressure on the left and right side of the bit. But when the bit is adding a beveled edge to the side of an area that has already been cleared by the mill end, there is only pressure from the one side, and there is a shift in the y axis. One thing I would like to try is to make the Vee tool cuts before doing the flat area clearance and see if I get a better overall result.

Any thoughts on how to improve the rigidity of the Shark, or how to lessen the effectes, would be appreciated.

Scott