[V1 - 61] Strange Bug in CamBam V1

[pixelmaker]

hello
I get these crazy toolpaths in CamBam V1 and V1 64bit if the object is a circle, I set lead in via spiral with angle 0 and the starting point is exactly below a holding tab.
The bug does not occur in V0.9.8. If I move the holding tab away from the starting point, the bug also does not occur. Since the error occurs only in V1 it is for me a definite bug

ralf

[dh42]

Hello

Same here, added to buglist

++
David

[Dragonfly]

Same here. But if I convert the circle to a poly then the problem disappears.
It is not the first time I notice that CB produces different results when working with built-in entities like circle or rectangle. Converting them the poly usually solves the problem.
I suspect it lies in the floating point math during tool path calculation.

[pixelmaker]

I have written that the error only occurs with the circle. But there's more.
I turn the circle into a polyline, change the ID number. The error is gone.
Now I take two steps back. The circle is circle again and the ID number is reset. Nevertheless the error does not occur any more and I see no difference in the values.

Even if I create the file in 0.9.8 and open it in V1 the error is not there.
See attached file
Maybe this is a .NET4 problem

ralf

[lloydsp]

Well, to follow up in 2025... It happens in V1.0 on polylines, also.  I was cutting some 'time-critical' components of which there were four identical copies on the stock, and I let CamBam decide the start points.  All four shapes had holding tabs.  They were 'irregular" shapes; not circles, nor rectangles.  I was unaware of this bug.

One of the starting points landed almost (not quite) at the peak of a triangular holding tab's position.  I was using an outside profile, spiral lead-in with angle 0,  and a depth increment of 0.015" per pass around the shape.

When it reached the top of the tabs, it began to move above the clearance plane a distance about the same as it should have been moving below the stock surface... giving very much the same toolpath as shown in the circle error below.

So it does happen on polylines, as well.

Lloyd

[lloydsp]

And following up on that -- it seems to be well-ensconced in 1.0.

Today, I tried it with everything different:  Different shaped polyline figures.  Different distance between them.  Different thickness of material.  Different cutter diameter.  Different holding tab size in both dimensions.  Different number of holding tabs.

The first generation of toolpaths created a start-point on the second shape that landed JUST inside the edge of one tab.  Viewing the toolpaths in the orthogonal view, it plainly created the same anomaly.

Watch your start points on multiple profiles with holding tabs!!!

Lloyd