background image background image background image
to DeskProto home page
Contact  
looking glass icon
to DeskProto Facebook page
to DeskProto Youtube page
to DeskProto Instagram page

Forum menu:

DeskProto user forum

Forum: communicate with other users


Forum home page main
Forum:
Thread:
Two sided cut off on y axis
You’re logged in as:guest
Online users:0
Online guests:8
 

page 1 of 2


SawdustandSmoke
2015-05-31
22:32 CEST
I have a problem that is very vexing, when I do a two sided cut(with rotary) the bottom half of the cut is off by about a tenth every time on the Y axis.

I have done everything I can think of to align my tailstock and chuck together,And make sure the 3 jaws are equal.

I though mach might have been losing position but it goes straight back to 0 when it is finished.

I have had no problems with cutting rotary parts at all.


Lex
2015-06-01
12:36 CEST

Is it about one tenth of a mm (very accurate job!) or of an inch ? Could that be the cutter radius, as that would indicate a problem in setting the zero point ?

For rotary machining an error in he workpiece zero point indeed is not be clearly visible. Still it should be visible as a slight deformation of the part.

What you can do is clamp a cilinder in your 3 jaw chuck,
and then let the side of the cutter touch the cilinder while coming from a negative Y (see image) and when coming from a positive Y.
The resulting Y values should be the same - except for a minus sign.

The Y=0 not being exactly above the rotation axis (so if the two resulting Y-values are not equal) is the most common cause for such misalignment.

Lex.

SawdustandSmoke
2015-06-07
20:32 CEST

Thanks for the reply, My cuts are off by 1/10 inch almost every single time using the automatic rotation mode, I have done about 25 cuts so far and am still unable to find the problem.

I have attached a picture that will hopefully explain it better.

I have zerod my Y off of almost everything I can think of, The last time I made a mark on the wood itself using a centering ruler and then used a v-bit to get it perfect but still no dice.

I have attempted to force using the rotary mode to make this cut but It is not possible without lots of air cutting.

Lex
2015-06-08
09:06 CEST
Unfortunately the photo does not help me much, as I have no clue to what I am seeing. The three jaw vise could be the rotation axis, however it is not on machine. And seeing a shift in the part (I assume that this shift is marked with pencil) does not explain what has happened.

What you can try is create toolpaths for a two-sided cut of a simple cube, using a flat cutter. For this cube the minimum and maximum Y-values in the NC file should be the same for both sides: you can open the NC file in a plain text editor to check.
If you can see the shift in the NC file the problem is caused by DeskProto, if in the NC file the values are equal then the cause of the problem must be in the Y=0 position.

Lex.


SawdustandSmoke
2015-06-13
16:19 CEST

Here is a better picture, The black line should go all the way across, The board lays flat as it is being cut so the Y is off by .2 or .3 inches.

I thought that perhaps the Y was being homed wrong somehow so two cuts were made with the Y home being of by 1/10th inch both plus and minus and it still has the exact same offset somehow.

The only other issue I have had is when doing a 4 sided cut with the same model the front of the model is on the back,
With rotational cuts there is no problem at all.

Lex
2015-06-15
09:20 CEST
This new picture is not yet really helpfull for me in understanding what happens. I cannot see how the block of wood was clamped in the rotary, and which are the X, Y and Z directions. A picture of the block on the machine would be better.
Also: both black lines seem on the same side of the block - or are these in fact two blocks ?
Have you tried what I suggested in my previous reply, and if yes: what was the result ?
Lex.

SawdustandSmoke
2015-06-16
22:44 CEST

Sorry I am not good with photos, The photos before were of two different blocks.

The piece of wood laying on top of the chuck is supposed to be a chair leg I did not expect it to come out right so did not bother getting the Z exact, The end you are looking at is what was facing the chuck.

The piece of oak with the screws in it is the exact height and width of the boards I am cutting, The picture does not show it but the carved area of the board that is facing upwards is off center as is the bottom in the opposite direction.

I have not been able to find a cube to try yet however I would be very very surprised if it was ok.

I am starting to think this is a mach3 problem, Before I purchased my rotary axis I was playing with a tangent cutter maybe one of the settings is off, Still it is strange that I can cut in roational mode without a problem though I would that that would me much less complex.-Thanks for the help BTW.

Lex
2015-06-17
10:17 CEST

Thanks: this is a clear picture, that nicely shows the Y-axis shift that you mean.
It does still seem to be an error in the Y=0 position, however perhaps indeed some setting in Mach3 may cause this as well. I do not know Mach3 very well so I cannot help you on that subject.

My idea about the cube was to check Y-values in the NC file (can be opened in a text editor), to see if it is a DeskProto problem or a machine problem. That can be done using this table leg as well. When the zero point is in the centre of the leg, the four indicated Y-values (at the lowest Z-value) should be equal (apart from the minus signs). When you can find the difference in the NC file coordinates it is a DeskProto problem, if not it is a machine problem.

And, I am not surprised that the error does not show when rotation axis machining. That is a far more forgiving method: even an error in the Y=0 will not clearly show (it will show as a small distortion of the shape, almost invisible).

Lex.


Lex
2015-06-17
16:32 CEST
One more idea:
on the picture above it seems that the remaining wood slab between the two machined halves is thicker on the right then it is on the left.
Perhaps a 180 degree rotation command for your A-axis on the machine does not result in exactly 180 degrees...
That would would be a valid explanation for the shift in Y that you see now.
Lex.

Lex
2015-07-07
09:01 CEST
Hi SawdustandSmoke,
Any news on this issue ?
When you have found the cause of the problem we'd be very interested to hear about this...
Lex.


page 1 of 2