Advanced Search

It's Bridgeport CNC 1
My home switches locations are randomly placed but min/max are measured home to hard limit switch and then backed a little.

My X and Y axes home switches are cheap China proximity switches but Z axis switch are good mechanical. Thought maybe they are not accurate but DRO will show that machine have run over soft limit and next time soft limit is working.

Credit where credit is due: I built it but didn't design it, the credit goes to Mike Bax, who builds and sells the high voltage pulse generator used in his EDM machine design. Anyway, it's running and working fine; my questions are about using linuxCNC in general.

I'm having a heck of a time wrapping my head around how to properly set up a workpiece in linuxCNC so that it starts cutting at the expected point.  There are three markers in the Preview window, a large circle with crosshairs, and two smaller "X"s. I understand that the large marker is Machine Home but don't understand its usefulness in the display. The two small "X" figures move around when offsets are changed, but they don't always seem related to where the cut starts. My usual steps are:

Load an .nc job into linuxCNC, which appears in the Preview tab.

"Home All" is selected.

Bax supplies software with his power supply, including extensions to linuxCNC that provide utilities such as Edge Find. I run that, and the respective axis value changes to some value denoting the edge. So far, so good.

The first issue is that I don't know how the resulting value relates to the overall workpiece. The value changed somewhat during the Edge Find, so is offset by some amount. For this example when the cut starts right at the edge of the workpiece, I don't know if Touch Off is necessary, since the cutter is already immediately adjacent to the work piece. If a Touch Off value is required, I don't know what to enter.

The first photo shows a typical preview display with the large target and the two X's. My "guess" is that the two X's represent the start and end of the cutting path. Correct? The large target doesn't seem useful, but it's probably because I don't know what it represents.

The second photo really confused me. Here, the cut "started" at the free end of the red diagonal line, moving toward the upper left corner of the workpiece (note that this is inside the workpiece). Also note that neither of the X markers is at the start of the red line. While the red diagonal line was extending, the physical cutter was Not moving. The red line continued until it reached one of the X's and then began following the contour of the cut. The crazy thing is that the physical cutter Still wasn't moving and didn't until the red track came down even with its start. Only then did the physical cutter contact the work. As a result, the entire workpiece ended up missing everything above that point.

I'm fairly convinced that I'm not understanding and using Touch Off values correctly and am hoping that someone can explain it to me as someone would explain to a small child... Thanks.

Just what I needed. The rotary screw extruder is centered on the mounting plate and it can probably push filament faster than any NEMA17 motor could.

hackaday.com/2021/12/05/rolling-screw-extruder-goes-brushless/

Did you confirm the parport address is 0x378?

If not, post the output of
sudo lspci -v

The nozzle will be heated with induction heat. It will be interesting to find out whether the stainless steel tube 3mm od x 2mm id x 25mm long between the extruder and nozzle will have low enough heat conductivity to avoid overheating the extruder drive without additional cooling.

i responded to your git issue. the probe routines are correct, your issue is your machine setup. many new folks get things a bit twisted as they view motion from the perspective of the table when whats actually happening is motion relative to the spindle.

gantry machines (head moves around) DOES actually follow expect cartesian motion because the spindle is what is moving in all 3 axes. however when you have a machine where the table moves, everything is opposite to the motion of the table because the spindle is still in the x and y axis. so we are moving the work and not the spindle.

in a traditional mill where the table is what moves:

table moving rearward = Y minus
table moving forward = Y plus
table move right = X minus
table moving left = X plus

because the spindle motion relative to the work is what is moving.

Z axis is also tricky. on a knee mill it is also opposite, where the knee moves up it is Z minus

on a VMC, Head Moving up is Z plus, Head moving down is Z minu. this is because the tool is actually moving with expected cartesian coordinate system.

Hope this helps!

Here is a problem. The arms will catch the drive motor, no matter which way I rotate it. I probably have to lengthen the arms of the 'center piece' that holds the extruder. Mounting the extruder drive farther away from the nozzle is not an option. Unfortunately the filament path is not centered to the extruder.

The plan is to have multiple extruders living inside the upper part of the 3D printer, with some method of automatically swapping extruder heads. The nozzle is attached to the extruder head, the high frequency induction heater is attached to the 'center piece'.

There will be spring contacts between the motor and the 'center piece', to make the motor's and thermocouple's electrical connection.

I might use strong magnets to hold the extruder to the 'center piece'.

The linear carriage blocks will have the gearbelt sandwitched between the carriage block and the arm bracket, teeth out. The return path for the belt is between the arms. HDT5M x 20mm wide gearbelt not shown.

I bought a 4 axis ethercat stepper kit (reference) , now I need to find a use for it. Decided it will be a 3D printer that will (hopefully) do what mine already can't.

It will be enclosed and heated, build plate will also be heated. The three delta motors will be up on top outside the heated enclosure, driving the carriages up and down via gear belts.

The door will be something like polycarbonate. Two thirds of the sides will be sheet metal heavy enough so it doesn't sound like a tin can. One third of the sides will be the clear door spanning most of the distance between two corner posts.

The linear rails are 2000mm long, I might try 2pc 1000mm to make the full length.

The frame as shown is incomplete... The three corner posts are 2" steel square tube.

Yes, that makes sense. The “jump” after homing and the E87.1 you’re seeing are consistent with that temporary desync between the drive’s internal motion and LinuxCNC’s commanded position. During CiA-402 internal homing the drive is effectively moving on its own, so unless pos_cmd is kept in sync with pos_fb (or LinuxCNC is otherwise decoupled from the loop), you can get that kind of discontinuity when control is handed back. So your observation fits well with what rodw described about the need to resynchronize after homing.

On a side note (and not to derail the thread): I might eventually pick up a used EtherCAT drive/motor myself to learn and test this in practice. I saw your configuration in another post and got curious—what made you choose EtherCAT for your setup specifically? From the outside it looks powerful but also relatively costly and a bit of work to get everything “dialed in”, so I’d be interested in your reasoning.

While using probe basic I have found for outside corners probe selection.  I have to select all opposite corners except the center in the 3x3 (tic tac toe looking ) matrix.  Everything appears to be working correctly just have to invert.  I have checked all of my accesses and all seems to be fine except the probing function of probe basic. Anyone else experience this or have any thoughts?

Thank in advance,
D
 

OK that looks like exponential runaway which is an indication that the
feedback loop is backwards (positive rather than the desired negative feedback)

If the encoder feedback is in the correct direction, the fix is to change the sign
of the PWM scale value:

[JOINT_0]
TYPE = LINEAR
HOME = 0.0
FERROR = 10
MIN_FERROR = 5.0
MAX_VELOCITY = 200.0
MAX_ACCELERATION = 1250.0
P = 1
I = 0
D = 0
FF0 = 0
FF1 = 0
FF2 = 0

BIAS = -0.0105
############-0.0105 vorher Bias
DEADBAND = 0
MAX_OUTPUT = 10
ENCODER_SCALE = -4000
OUTPUT_SCALE = -10
OUTPUT_MIN_LIMIT = -10.0
OUTPUT_MAX_LIMIT = 10.0

Pictures of the machine and where the home switches are located?
Your min/max values seem random, any reason for that?

@Otis,
Your link to github is not working, try posting it on a new post as editing usually messes things.
Thank you.