Symmetrical file cuts asymmetrical wood

[jules_mccann]

Hello all, and thanks for whatever input you may have with this. I am pretty new to the world of the Shopbot, but I know enough to be pretty sure my 'Bot shouldn't be doing what it's doing.

The ingredients: PRT96: With rack & pinion drive, porter-cable router (typically at 21,000 rpm), dust vacuum, & metal table. Vector 9. SB v2.25.

The Problem: We cut a countertop out of 3/4" MDF at .35"/sec - which was basically an arc about 7 feet long and about 31" wide. The inside and outside are arc sections of concentric circles, and the sides are radii of those circles, so the thing should be perfectly symmetrical. This should allow two pieces to butt against each other to form a larger section of the circle. It doesn't work.

The cutfile was made properly, and I checked the NC Code to be sure it was instructing the 'Bot properly, which it was.
When cutting, the endpoints are matching perfectly with the starting points, but the shape is slightly warped, I think towards the Y axis (but Im not exactly sure, since it wasn't square on the blank)

We've had some problems with repeatability on the Y Axis - cutting the same file at different depths, from the same XY-zero point, will align perfectly on the x-axis, but *may* be off by 1/32"-1/16".

The gear ratios are set correctly. The table seems to be level and square. I suspect my Y-Axis may not be calibrated properly, but I havent read
a recommended procedure for calibrating a rack-and-pinion table.

Ok, that's enough outta me for now. Thanks for your help.

[dave@shopbottools.com]

Jules,
If your cutfile is correct then the asymmetry in your cut could be due to cut direction (i.e., deflection of bit or wood during conventional cutting vs. climb cutting). But, since you suggest that the problem might be in your Y axis, the unit value for your Y axis may be incorrect or the set screw in the pinion gear on the Y axis may be loose or off of the shaft flat-- either of these problems would cause the distance moved by the Y axis to be incorrect. Tighten all pinion gear set screws, then, since there is no "calibration" per se of the Y axis, just use the VU command to set the unit values correctly for all axes. Unit values determine the number of steps per inch and should be the same number for X and Y axes. The correct values are on the info sheet shipped with your tool.

[bill.young]

Jules,

The "X run's big?!" thread in the ShopBotter Message Board section has some good troubleshooting tips to help you figure out if it's a movement problem or a cutting problem.

Bill

[Gerald D]

Aside from what David and Bill mentioned above, you may have one dead X motor, causing the gantry to go out of square with the x-axis. One X-motor may have a loose pinion. With motors "On", you shouldn't be able to move either end of the gantry by hand.

Alternatively, your gantry may be permanently out of square from an assembly error.

Suggest you lay a big rectangular board on the table and lick the edges clean with the SB. Then use a good tape measure to check parallel and diagonal dimensions.

[support@shopbottools.com]

Hi Jules,

Let me throw my 2 bits in ...

The Y-axis is more likely to be drawn out of position by the force of cutting because there is only a motor on one side of it. This means you need to be careful to make sure the carriage operaton is tight.

Start by checking to make sure the pinions are not slipping (as Dave suggested). Just try moving the car with the motor on. It should be rigidly locked.

Now, lower the Z-axis and then push the router firmly back and forth in the Y direction? Is the Z carriage tight? If not, slightly tighten the eccentric bushing on the Z bearings to tighten it up. Or, does the car flex back and forth? If so, tighten up the hold-down bearings, and make sure the router is mounted as close to the Y motor as it can be (i.e. keep it close to the pinion).

I'm not sure I fully understand the file problem, you might send it along to us to have a look (support@shopbottools.com (mailto:support@shopbottools.com)).

[jules_mccann]

Thanks all, I will investigate this and report back after I can investigate some of the possible causes you've suggested.