Ridges in curved cuts ever diagnosed?

[scottgus]

Greetings, all!

I have a PRT Alpha (2006) 5x8x12", running 3.5.20 control software. I get the ruffled-potato-chip ridges on the edges of my curved cuts. I see them in any materials where I cut a curved edge.

I remember much discussion here on the forums about these rippled curved edges some time ago, but not much now. Has this condition been diagnosed & cured? If so, can someone point me to the cure? (I searched but was unsuccessful in choosing my search words.)

[cr2]

I have found that the more vertical the model, the more pronounced the ridges will be. So if a model curves off near the edges, that's where the ridges will be. Try reducing your stepover and see what difference it makes. It helps with my work, but I still have to sand some ridges out.

[scottgus]

Hey there Charlie!
It sounds like you might be refering to 3D carving? Smaller stepover will make for a smoother 3D surface, true.

The ridges I'm inquiring about are only in the XY movement of the cutter as I cut, say, a piece of plywood or mdf. As I go around a curved edge I get ripples in the edge, as though the bit were not being dragged exactly along the edge, but rather as if the cut were pixellated, like a curve shown on a computer screen would be.

I usually cut double passes in 3/4 ply, for example, and I get the ripples showing up in the first pass, then the ripples show in exactly the same places on the second pass. There's something either electronic or mechanical causing them...

[steve_g]

Hi Scott

Sounds like you have the 1:1 ASM911 motors from that era... either upgrading to the ASM98AA T7.2 motors or adding belt reduction boxes cures the problem! See this thread or search for other similar ones... http://www.talkshopbot.com/forum/showthread.php?t=15760 (http://www.talkshopbot.com/forum/showthread.php?t=15760)

Happy routing!

SG

[khaos]

I have found that the more vertical the model, the more pronounced the ridges will be. So if a model curves off near the edges, that's where the ridges will be. Try reducing your stepover and see what difference it makes. It helps with my work, but I still have to sand some ridges out.

This has been my experience as well. I try to have a slope if I can, if not I regign myself to sanding. On very rare occasion a profile cut can clean it up.

[feinddj]

Actually, the culprit may be the line you are cutting. I get files to cut from many sources, and a lot gets "lost in translation" with curves. Go into node edit and see if your curves aren't a series of straight lines linked in a curve. If they are and you still have problems, check the speed at which you are cutting and the chip load. you may be getting deflection in the bit. A light pass on the line will clean it up.

David

[larry_k]

is there anyway we can still get the upgade price since there are still motors & drivers out there.

[Brady Watson]

is there anyway we can still get the upgade price since there are still motors & drivers out there.

Call ShopBot: 1-888-680-4466 to inquire about current pricing.

-B

[scottgus]

Thanks, everyone, for the responses!

I read the thread Steve_g referred to - yes I do have exactly the same motors you do, Steve - it sounds like a simple solution. (And it's just three weeks old - must've missed the correct search words)

re the curves I'm cutting, good thought, David. I've had pixellated curves on files from PartWizard and Alphacam. I checked by running a smooth-curved sbp file that came out ruffled edges thru a spreadsheet to get a distance-between-points reading. I had not more than 1/8" between points on the curve, usually 1/16". That could be part of the problem, I think there's a tolerance setting in PartWizard.

But the ruffling looks more like waves in the cut than facets, like a little sine wave was superimposed on the smooth curve, not like the profile of a diamond.

It sounds like the consensus is that the geared-down or belted-down motors might help this ruffling-edge problem. Does the reduction make the problem go away or just reduce the peaks of the ruffles in size?

A question on the increase of pulses required to the controller, too. I have an 85.5% on the speed test, and I have cut at 7 ips usually. If I do a 3-1 belt reduction I assume the controller will have to put out 3 times more pulses to cover the same distance. So I'd have to get 21 ips' worth of pulses to achieve the 7 ips cutting speed I have been using. Is this reasonable to expect, or am I going to have to roll back my expected cutting speed? Or am I looking at this all the wrong way?

[scottgus]

Did some more searching last night ("Belt Drive") and got a lot of info, sounds like the extra resolution reduces the amount of pixelization of the curve to acceptable levels, and i might have to cut slower, tests will show...

Earmarking some of the income tax refund...