Saving G29 Bed Level Correction Matrix to EEPROM

[Roxy]

Roxy,
I tried this and was not too successful. I will try again when I get more time.
Now, how do I revert the eeprom back to where it was. I don't seem to be able to remove the ABL matrix from it. I flashed the firmware ( with and without eeprom enabled, but both without the code above). I tried M502 then M500, I also tried M500.

You would need to have a version of the firmware that is not trying to save the Auto Bed Level Correction Matrix to the EEPROM. And then you could restore the default settings and save them. That would do it. Or you could change the version number of the EEPROM format at the top of ConfigurationStore.cpp so it doesn't trust the information in the EEPROM.

Alternatively... Add the M499 command to invalidate the EEPROM. It changes the EEPROM format version so the values won't be loaded until you do another save of the values to the EEPROM.

It can be found here: http://3dprintboard.com/showthread.p...4294#post34294

[Fri]

Thanks again Roxy,
I was able to get it back.

On another note. I have an issue on one machine since I changed the z offset on my viki lcd (this may not in anyway be related), but it is like that ever since.

I have the G28 deletion of plan bed leveling uncommented.

After g28 g29 the machine rests as it should in the far right corner. Now when I hit G28, it moves further back and right. Looks like it thinks it is home.
If I send a G x0y0 it goes to the home position, but it's not homeing (bouncing off the stops). If I lift z until the sensor goes out (I am using proximity sensors for z stop) it will home fine.
(It also homes fine if I don't do a G29 at all). Any ideas? I tried pretty much everything. It was working fine before.

[Roxy]

Thanks again Roxy,
After g28 g29 the machine rests as it should in the far right corner.

There isn't enough detail here. Is it going to the back far right corner? If you use a V option on the G29 such as G29 n 3 V it will not zig zag and the last point probed should be as far from the origin as possible. Can you redo it with a G29 n 3 v and report back?

Now when I hit G28, it moves further back and right. Looks like it thinks it is home.

It would be helpful to do a G28, a G29 and then a M114 to see where the printer thinks it is at. Then do a G28 and another M114 and see where it thinks it is at. That will give us some hints about what is going on.

If I send a G x0y0 it goes to the home position, but it's not homeing (bouncing off the stops).

That would make sense. You told it to go there. Not to home there. The homing code is special. It just starts heading there waiting for a limit switch to get pressed. Then it backs off and goes back towards the switch at a slower speed. But you are not giving it a home command. And my guess is if you look, the limit switches are pressed when you go to G1 X0 Y0

If I lift z until the sensor goes out (I am using proximity sensors for z stop) it will home fine.
(It also homes fine if I don't do a G29 at all). Any ideas? I tried pretty much everything. It was working fine before.

This is new information. What happens if you turn Z_SAFE_HOMING off? I don't know what code path it would follow (because I've never had an issue like this), but I can believe that the Z Probe firing could impact the X and Y homing.

[Fri]

There isn't enough detail here. Is it going to the back far right corner? If you use a V option on the G29 such as G29 n 3 V it will not zig zag and the last point probed should be as far from the origin as possible. Can you redo it with a G29 n 3 v and report back?

Yes, it is on the diagonal position opposite to HOME, I use 3 point grid N3V

It would be helpful to do a G28, a G29 and then a M114 to see where the printer thinks it is at. Then do a G28 and another M114 and see where it thinks it is at. That will give us some hints about what is going on.

G28, G29, M114 result is X190 Y175 Z1.65, when I hit G28 now, it goes straight into the wall. So I did a home X only, the x carriage moves approx 10mm away from the previous position (roughly 200 actual pos.) but the M114 reports X0



That would make sense. You told it to go there. Not to home there. The homing code is special. It just starts heading there waiting for a limit switch to get pressed. Then it backs off and goes back towards the switch at a slower speed. But you are not giving it a home command. And my guess is if you look, the limit switches are pressed when you go to G1 X0 Y0

You are right.

This is new information. What happens if you turn Z_SAFE_HOMING off? I don't know what code path it would follow (because I've never had an issue like this), but I can believe that the Z Probe firing could impact the X and Y homing.

True, but I cannot home without Z_Safe, the probe would be outside the bed. I did it anyway and and did a G29. Same result, it is in the right position 190,175,1.65, but when hitting home x it moves right or away from the last position.

[Roxy]

True, but I cannot home without Z_Safe, the probe would be outside the bed. I did it anyway and and did a G29. Same result, it is in the right position 190,175,1.65, but when hitting home x it moves right or away from the last position.

Oh! I started with my origin in the front left. That is where most people have it. But because the only place I could put a Z-Probe was on the front left, it was probing in negative space. What I did was move my origin to the back right. You do this by just changing this:

#define INVERT_X_DIR true //false // for prusa i3 set to false, x endstop should be on the left
#define INVERT_Y_DIR true //false // for prusa i3 set to false, y endstop should be in the back and opposite the y-motor
#define INVERT_Z_DIR false // for prusa i3 set to false, z endstop should be at the bottom

[Fri]

Okay, I did this on another machine too, but this has nothing to do with my problem or am I missing something? One thing that is really puzzling is that the same fw worked before flawlessly. I thought it was the board, swap it. Wiring...very unlikely. Eeprom reset.

[Fri]

I did this. No avail. I forgot to mention that where ever on the bed the stop switch is triggered, the x and y think they are home. So if I do a G29 then lift and do a G28, it will home fine, but if I hit G28 again, it will move from that x and y position. It's got to be something I am missing.
But again, it's the same fw that worked fine for month. Until I hit that z offset on the LCD.