Another indexer/rotation problem

[bobmann]

See the attached pictures.
Desktop Max
Aspire 10.5

I created a rotary part file to verify rotation of the indexer.
Screenshots are below.

I created a rounding toolpath with the gadget program and it worked fine.

The 2 vertical lines created ring grooves, I created these with the quick engrave function. Both grooves cut the full 360 degrees.

The deep groove that was created with a 2 rail sweep did not cut the entire 360 degrees, but the readout for the B value read 359 degrees. Note the location of the reference line on the end of the part. It started at B zero a vertical line, but when the program stopped it was not vertical, but the B value reads 359 degrees. The line should be much closer to vertical at 359 degrees.

The Value for the stepper is 83.3333
I verified full rotation my scribing a line on an aluminum bar and using a pointer in the collet. Rotation at 360 degrees was as close as my eyes could register using a magnifier glass.
Note that the vector box I drew should have been big enough for the tool to overlap.

The puzzling part is why does the B value not match actual rotation in some parts of the program and not others?
34384
34383
34385
34386
34387

[bobmann]

See the attached pictures.
Desktop Max
Aspire 10.5

I created a rotary part file to verify rotation of the indexer.
Screenshots are below.

I created a rounding toolpath with the gadget program and it worked fine.

The 2 vertical lines created ring grooves, I created these with the quick engrave function. Both grooves cut the full 360 degrees.

The deep groove that was created with a 2 rail sweep did not cut the entire 360 degrees, but the readout for the B value read 359 degrees. Note the location of the reference line on the end of the part. It started at B zero a vertical line, but when the program stopped it was not vertical, but the B value reads 359 degrees. The line should be much closer to vertical at 359 degrees.

The Value for the stepper is 83.3333
I verified full rotation my scribing a line on an aluminum bar and using a pointer in the collet. Rotation at 360 degrees was as close as my eyes could register using a magnifier glass.
Note that the vector box I drew should have been big enough for the tool to overlap.

The puzzling part is why does the B value not match actual rotation in some parts of the program and not others?
34384
34383
34385
34386
34387

The last picture
34388

Top pics are the Aspire screenshots.
3&4 are before initiating the program
5&6 are after the program stopped.

[bill.young]

My guess is that the B-axis lost steps. I'd bet the depth of cut is different for that feature or the feed speed.

You could try running that feature without a blank in it and a reference mark on the headstock for B=0...air cut it..and see if it runs correctly?

Bill

[bobmann]

My guess is that the B-axis lost steps. I'd bet the depth of cut is different for that feature or the feed speed.

You could try running that feature without a blank in it and a reference mark on the headstock for B=0...air cut it..and see if it runs correctly?

Bill

Thanks for the response.

As I said, I did verify that the rotation is accurate at 360 degrees.
The curious thing is that this error seems to only occur when doing a 2 rail sweep.
The grooves I cut as a V Varve rotated the proper amount.
Also using the gadget to create a round from a square worked as well.

I did recut the part after re-zero the B, since the part was already cut there was absolutely no resistance from cutting, and the result was the same.
For some reason, when running a program the B value is changing...don't know how that could happen, but that is what it appears to be.

Tool, feed and speed is all the same for both the VCarve and the 2 rail sweep.

Seems like something I am missing in creating the feature in Aspire with a 2 rail sweep, but I cannot figure that out. As shown in the screenshot, the vector boundary I drew should be more than enough overlap to make the machine rotate more than 360 degrees to eliminate the uncut portion.

I will send this to the Shopbot tech on Monday, it is not critical so I don't want to bother anyone on a weekend.
This may also be something to ask the Vectric people.

[srwtlc]

I'm thinking that your 2 rail sweep is actually stopping at the job dimensions of 6.6602" and not 7.5". Try setting up your job dimensions as 7.5 and redo your sweep and try it again. Any components that are out of the job area are not able to be toolpathed. Or put this way, components (your sweep), are not read if they're outside of the job dimensions.

There should be some rotary training videos on Vectric's site.

[bobmann]

I'm thinking that your 2 rail sweep is actually stopping at the job dimensions of 6.6602" and not 7.5". Try setting up your job dimensions as 7.5 and redo your sweep and try it again. Any components that are out of the job area are not able to be toolpathed. Or put this way, components (your sweep), are not read if they're outside of the job dimensions.

There should be some rotary training videos on Vectric's site.

34389

I did study the videos before I started down this path.

I created a new part with a single groove.
I measured the diameter of the blank with calipers, it is exact.
The diameter of the end mill is indeed 1/4"
Zero points for XY are spot on, the Z was set with the touch plate to material surface.

I followed along with the video, step by step by step.
According to the video, a 3D carve will allow for the cutter diameter, so a boundary box outside the job setup size is not needed, and there should be no seam.
The vector box is drawn to the exact size of the job setup, that is the white box as instructed in the video.
The cut ended at B reading 359 degrees, but there is a large seam left, and the reference mark I drew is about 20 degrees off. If I manually rotate the indexer so the mark is about at the start point, the B reads 388 degrees.

I do not know where the issue is, the control software, the PP, something I am doing wrong in Aspire, or an indexer hardware issue.

The puzzling part is the mismatch of the B value when cutting, but it is spot on when rotated manually.

I keep checking the rotation, and 360 degrees on the B value is 360 degrees on the indexer.
I tried running the same program over again so there was absolutely no load on the indexer; no change, it still stopped at 359 degrees on the B and a large seam.

I guess I will have to try SB tech support and Vectric support and see if they can direct me to a solution.

[coryatjohn]

To be absolutely sure you have the right settings for 360 degrees, rotate the indexer 3600 or some large multiple of 360 and see if it ends in the same spot. This will magnify errors so it will be obvious if you have an incorrect setting.

[bill.young]

Bob,

Support is your best bet for sure, but I have a somewhat similar setup on my Max and do a lot of turning with it. I don't use the MTC posts though, and have a 3" indexer.

If you want to email me your Aspire and sbp files I'll be glad to run them and see if I get the same results. It sounds like it fails in that section without any load, so testing with the smaller indexer might be OK?

billyoung@shopbottools.com

Bill

[bobmann]

So, I was mounting another piece of wood for more testing, and as I was tightening the chuck I noticed that the chuck will rotate.
Power is on, I can rotate the indexer manually.
This does not seem right to me, I would think that with the power on to the chuck that it would not rotate easily by hand.

I do not want to take the thing apart for fear of voiding the warranty.

I will send this to Shopbot to see what they say.


https://www.youtube.com/watch?v=ioy6oDs_pTY

[bobmann]

So, I was mounting another piece of wood for more testing, and as I was tightening the chuck I noticed that the chuck will rotate.
Power is on, I can rotate the indexer manually.
This does not seem right to me, I would think that with the power on to the chuck that it would not rotate easily by hand.

I do not want to take the thing apart for fear of voiding the warranty.

I will send this to Shopbot to see what they say.


https://www.youtube.com/watch?v=ioy6oDs_pTY

SB support directed me to the location of a split coupler screw that connects the motor to the gearbox. That was indeed loose, and tightening the screw fixed this issue.
I am just before making another couple cuts to see if this had an effect on the other problems I am having.

[bobmann]

SUCCESS!

So I just completed a test piece that used both the 3D 2 rail cut, a VCarve, and a pocket.
All cuts did exactly what they were supposed to do.

I theorize that the loose collar only showed up when I did the 3D cut and not the othercuts was because the rotation of the indexer was around one degree per step for the 3D, and the shock of the motor trying to turn the gearhead caused minute slippage with each step of the motor.
That is why rotating manually to verify that the 360 degree rotation was accurate did not reveal the problem, as the slippage was not so severe that the slower and less jerking rotation allowed the motor to turn the gearbox without slipping.
Regardless, it looks like it is now fixed, and I can keep what little hair I have left to pull out another time.
I will create some more parts soon, and hopefully no further issues will surface.

I want to thank everyone who took the time to respond with suggestions.

Bob

[bill.young]

Excellent!

[srwtlc]

Great to hear! :cool: