[Gregore]

any one know what might cause this? I get these random lines showing up in parts and I can not see it in the paths while in rhino , I just had a friend look at some Gcode and it is not visible in a Gcode simulator , the odd thing is if I mirror a part and run separate paths in madam both parts will have it in the same place.

I tend to get them quite a lot but in wax they are not a bother as it takes seconds to remove them by adding wax but in other materials I need to start over.

Attachment 275922

Attachment 275920

[svenakela]

What type of machine do you have, is it servo or stepper based?
I've experienced the same thing on a machine, couldn't figure out what the cause was until we had new workshop lights installed - it got worse... The machine was affected by the fluorescent light bulbs and when Z was raised too high it lost precision when the servo cable picked up noise and the feedback loop was disturbed. New shielded cables grounded in both ends and all problems gone.

This could be anything, as long as you don't see anything in Rhino nor in a simulator you have to investigate your machine.
This thread probably fits better in one of the machine forum.

[Gregore]

good idea to check that , but these are always in the same place on the parts no matter how many runs i do

[svenakela]

Can you give me the model and the G-code?

[victorofga]

just a remark here, on the second picture the marks are occured , where bit suddenly went deeper, or even poke trough material..
additionally the part already were loosened..


on the first picture marks are occured, where the first deep pass was...




my opinion would be, first make an outline where parts are laying.. then around drill holes and screw off the wax on a 3/4 plywood.. around the parts..

if you cut from two side, then within machining both side hold down with screws..

i dont think it either the gcode, nor the electric noise..

[Gregore]

i have the gcode for the top picture it is included here along with the model,

I had a question about the noise issue if it would be possible for very specific moves to create noise that upsets Z ?
does the noise throw the steppers off their count and if so I can not imagine with my machines set up that the steppers can refind the right indexing with out a homing sequence.

[Gregore]

[QUOTE=victorofga;1681524]just a remark here, on the second picture the marks are occured , where bit suddenly went deeper, or even poke trough material..
additionally the part already were loosened..

The model was not loose on the cut, the picture only shows a very small part of the model 1cm square area the rest had many tabs to hold it all solid also the cut lines shown happen on the first side before the flip and before anything is cut through. so it can not be slipping.

Almost always when I am using this 5mm thick wax I have aluminum bracing that holds the side of the wax all down the length .

[Gregore]

svenakela;

in case it makes a difference for pathing this is the cutter that was used

D 3.17
L30
F15.257
R.127
d.254
A 5.5

[Gregore]

here is a snap of the part from the Gcode sim

[victorofga]

greg

the gcode doesnt shows any erroneus movement

your machine dont loosing steps, that very much visible of the pictures

only issue is, when part getting close to the ready state it produces marks, on same spots..

there are no more option..

[Gregore]

only issue is, when part getting close to the ready state it produces marks, on same spots..

there are no more option..[/QUOTE]

I do not understand your last point, sorry

[Gregore]

If it is not in the Gcode and it is not missing steps then it has to be controller error , but how can a controller error be so consistent?

The lines are always in the same place in X and Y from 1 part run to the next but some times the depth in z is minutely more or less in Z .

[svenakela]

That's why I asked if your machine is servo or stepper based. A stepper will lose steps, a servo will re-position depending on the setup and type of noise.

If the errors are consistently in the same table coordinates, what happens if you move the setup or rotate the toolpath 90 degrees?

[Gregore]

I do not have the model with the original paths saved together , I could rerun them and see if it does it.

What would happen if in Gcode I changed all Y moves to X and all X moves to Y ? I will try this , seems easy enough.

[Gregore]

first test with X&Y swap seems that the lines are still on the model but they have shifted where they appear , so obviously they are not in the Gcode. any ideas where to look next
included pics of X run and Y run

[Gregore]

close ups / as you can see from the larger pics they are on the opposite side of the part or is that what would happen with an X/Y swap in Gcode

[Gregore]

when looked at this way they are mirror copies of each other and the marks are all the same the only changes are the depths of marks are not exactly matching

[svenakela]

Is your Z axle tight? Do you have a loose coupling somewhere? I am starting to suspect you have a screw, ballscrew, linear guide or joint somewhere that isn't tight enough.

[Gregore]

I will check that out

but I assume that if it slips it should stay slipped ,

[svenakela]

Not always if it's a loose mechanical part.