Easy "Manual" ZProbe Calibration

[noddynomolos]

Great thread guys!
Art thanks for posting your spreadsheet. I have downloaded it and will be using it forthwith

[teoman]

....

And last night it was printing nicely until it decided to Picasso on me.
...

Hahaha ... it should be a verb, You've been Picassed

I bet that, if Picasso were alive, he would use a Rostock Max to create his prints

Indeed....

They actually look more like Dali's work. But still. Picasso sounds nicer.

So maybe Dali for bed adhesion problems where you end up with spaghetti and Picasso for calibration issues where everything is out of proportion.

[teoman]

Bumping this thread up so noobs with issues can see it.

[artexmg]

Bumping this thread up so noobs with issues can see it.

Great!

Actually, I had been using this method for a while with great success

[0110-m-p]

Was just reading through this an found an error.

In my "new" (or V2) I got 80, but in the "old" (or V1), the result is 106.6667.

This due to the pulleys are, in V1, 15 teeth, and in V2, 20 teeth. Everything else is the same.

Not all V1 machines use 15T pulleys. They changed to 20T pulleys sometime in the middle of last year. My machine ordered last October has 20T pulleys.

[teoman]

Sorry my bad. I have never seen a V1. Or any other product from SeeMeCnc except what i got in my box. It is probably the only one in my country.

[teoman]

In fact if you screw the screws all the way in, you can gain a couple of mm in build height!

[artexmg]

Was just reading through this an found an error.

In my "new" (or V2) I got 80, but in the "old" (or V1), the result is 106.6667.

This due to the pulleys are, in V1, 15 teeth, and in V2, 20 teeth. Everything else is the same.

Not all V1 machines use 15T pulleys. They changed to 20T pulleys sometime in the middle of last year. My machine ordered last October has 20T pulleys.

I think they even had other pulleys before 15T as well, not sure, but I recall when building mine (March 2013?) I got the wrong value and somebody told me that it was a previous configuration.

Any how, it is just a matter of counting the teeth in the pulley and adjust the number

[teoman]

Just want to thank you again for your post. I use it often and I think it is a very useful post.

[artexmg]

Just want to thank you again for your post. I use it often and I think it is a very useful post.

Your welcome!

Actually, I have kind of evolved the method using a excel spreadsheet to get all the feedback, and now is my only way to calibrate: no screws at all. And, can calibrate to very

I have been thinking on creating a (very) simple web app for this purpose (open source of curse!).... but I have had no time

[DGBK]

Thanks so much for this post. I was having quite a bit of trouble with the manual, but this got me through the calibration process very quickly.

[PeteD]

Agreed. Thanks for posting this. I had missed this thread until yesterday, but I was able to calibrate my printer much better using this method.

Any chance we can get this thread pinned?

[BenTheRighteous]

I found this on the repetier site, which seems to be exactly what you're describing, but easier? I haven't tried it, but it sounds like it accomplishes the same goal - use a stick to get all the motors to the exact same height, then the firmware figures out the offsets.

http://www.repetier.com/documentation/r ... z-probing/

Afterward then I guess it's still your job to figure out the horizontal radius, but maybe this helps speed up the whole process.

Delta printer need more calibration then normal cartesian printer. You need to calibrate the endstops first. This calibration changes the values you see in the eeprom configuration as “Tower X endstop offset [steps]” resp. Y and Z. The trick is simple. After homing to the max endstops, the firmware will move these steps down to position the extruder holder in the exact center. It is essential to define the exact center position or the nonlinear behavior of delta mechanics will result in wrong geometries. Measuring these values on your own is nearly impossible, so you need to use a simple trick. First home and remove any extruder offset using these commands:

G28
G131

Now position the 3 sliders, so that they are at the same height. To do this I use a stick and position all sliders, so that they press the stick lightly against the printer top, as I know it will be centered then. Depending on your printer it might be better to measure from the bottom (not the bed – that might be skewed in addition). The hard way is to use moves in x,y,z direction. The easy way is turn motors off with “M84″ and slide by hand. Be careful not to slide fast. Movements create a current in the motors flowing back to the electronics. Since some printers will move down when motors are disabled, newer versions have a command to only disable one motor for a time. So now the best way is to use M99 X0 to move the X motor for 10 seconds. You can change the time with S<time in seconds>. Use Y0 and Z0 for the other 2 axis. Once all sliders are at the exact same height, send:

G132 S1

That will measure the required offsets. S1 additionally stores the result in EEPROM, so you do not need to repeat it again.

After the endstops are calibrated, you can run the bed leveling routine the same way as for cartesian printers.

[BenTheRighteous]

Welp I just tried the Repetier method that I posted. It's cool in theory but in practice it doesn't work. The M99 command mentioned doesn't do anything, and I couldn't get the stick method to produce acceptably consistent results. It would vary by around 20-30 steps each time which was pretty wild considering that I was manually adjusting by 3-5 steps at the end to get things really dialed in.

Long story short, I didn't find a shortcut, but the original method still kicks butt so thanks for sharing it.

[[email protected]]

Hi everyone.

After many unsuccessful hours trying to calibrate X, Y, and Z tower using the end-stop screws, I decided to try the calibration using only the EEPROM.
As teoman and some other users mentioned, it seems that I cannot use negative values for the Tower X, Y, Z endstop offset (it´s as if it takes negative values as zero, therefore it has no effect). So I had to increase the horizontal radius to make sure that all three towers would start at least above 0, and then start the adjustments always using positive values. I followed every step and was able to calibrate tower X, Y, and Z. This is how my EEPROM config looks like:

Baudrate 250000
Filament Printed [m] 0
Printer active 0
Max. inactive time [ms,0=off] 1800000
Stop stepper after inactivity [ms,0=off] 0
Steps per mm 80
Max. feedrate [mm/s] 300
Homing feedrate [mm/s] 120
Max. jerk [mm/s] 35
X home pos [mm] 0
Y home pos [mm] 0
Z home pos [mm] 0
X max length [mm] 250
Y max length [mm] 250
Z max length [mm] 396.622
Acceleration [mm/s^2] 1850
Travel acceleration [mm/s^2] 3000
Diagonal rod length [mm] 269
Horizontal radius [mm] 148
Segments/s for travel 80
Segments/s for printing 225
Tower X endstop offset [steps] 20
Tower Y endstop offset [steps] 449
Tower Z endstop offset [steps] 100
Alpha A (210) 210
Alpha B (330) 330
Alpha C (90) 90
Delta Radius A(0) 0
Delta Radius B(0) 0
Delta Radius C(0) 0
Z-probe height 39.91
Z-probe speed 2
Z-probe x-y-speed 150
Z-probe offset x 0
Z-probe offset y 0
Z-probe X1 100
Z-probe Y1 20
Z-probe X2 160
Z-probe Y2 170
Z-probe X3 20
Z-probe Y3 170

After doing this and checking the height at the center of the bed, I need to raise the nozzle by 2.20mm (to be clear, if I would send G28, G0 Z0, I would be 2.20mm below the bed…so I need to raise it by that much). I kind of got lost between steps 6, 7 and 8….What I understand is that since all towers are leveled, I need to increase the horizontal radius by X mm in order to raise the 0 at all towers and finally have the whole bed leveled, is this correct? I am not sure how to calculate that measure, the excel formula is not the same as the one explained in the original post.

Thanks in advance for your help.

[geneb]

Your horizontal radius value is WAY out of whack.

If you have the old style carriages, you should start with a HR of 130. If you have the new injection molded carriages, your starting HR should be 140.

Reset those tower offsets, put the screws in the carriages back to their starting points, grab a set of feeler gauges and head straight for Appendix B.

g.

[[email protected]]

Hi Gene, thanks for your answer.

I started following appendix B as you recommended (with a radius of 140), but I reached a point where I still needed to screw (or unscrew) at least one of the end stop screws in a particular tower and the screw is not long enough. If I make one more turn to the left, I would leave the screw loose…or if I have to tighten it, I would hurt the plastic. I feel that if those end stop screws were longer, I would have more distance to play with…not sure if it’s just me or what.

I first tried following pages 250-260 on…where turning the end stop screws seemed to have no effect. Then I tried Appendix B, where I was able to modify the height at each tower, but I found this issue that the length of the screws doesn’t allow me to move the nozzle all the way I need to. I finally I tried the EEPROM alternative, that seemed to work fine (except I cannot use negative values and I still need to make that final change in the radius).

I guess I will reset everything and try again…but I am not sure if I am doing something wrong.

Thanks again.

[geneb]

Make sure your z-height is set correctly before you begin. You might want to look at the flowchart - you'd be surprised at how easy it is to forget a step. I've done it myself.

g.

[teoman]

What you are trying to achieve with this is to make all of the towers have the same length, or make them start at the same height if you prefer.

Examining the problem from this perspective, if you add say 100 to all of the tower offsets (each and every one), your Z height will be offset by 100 steps.


So... what you could do is add say 250-300 to each of the values there. Redo the Z height. And continue from there.

Once you have finished, everything should be good. If you really need the to use the very top mm of your build volume, then you can find the minimum of your 3 tower offsets and subtract that from all of the tower offsets (effectively making the smallest one 0). Then redo the Z height using the LCD. this will enable you to utilize the very top few mm of your build volume.

You should not fiddle with the horizontal radius besides it's intended purpose. That will make everything worse.

[teoman]

Example 1:

Lets say that you have followed the procedure through, but one of the towers has reached a negative number:

Tower offsets:
X:-80
Y:100
Z:120

Add 100 to all of the offsets
X:20
Y:200
Z:220

Redo the Z height and continue as usual.


Example 2:

Lets say that you have finished with your offsetting and everything is perfect
X:80
Y:100
Z:120

Our systems are configured to be 80 steps per mm. So looking at the offsets above (how many steps of distance a particular carriage should consider its 0 position relative to the limit switch) we see that the smallest is 80 steps. What this means is that we are wasting 80 steps = 1mm.

And if you really really want to use that last mm then you should subtract 80 from all of the offsets resulting in:
X:-0
Y:20
Z:40

When you do that you will notice that now your nozzle is sitting 1mm above the bed (because we subtracted 80 steps) and you have to readjust the z height.


I think that the reason why it does not accept negative values is that with the negative values you are driving the carriages towards the limit switches which could be problematic if you go too far negative.

[artexmg]

Hi everyone.

After many unsuccessful hours trying to calibrate X, Y, and Z tower using the end-stop screws, I decided to try the calibration using only the EEPROM.
As teoman and some other users mentioned, it seems that I cannot use negative values for the Tower X, Y, Z endstop offset (it´s as if it takes negative values as zero, therefore it has no effect). So I had to increase the horizontal radius to make sure that all three towers would start at least above 0, and then start the adjustments always using positive values. I followed every step and was able to calibrate tower X, Y, and Z. This is how my EEPROM config looks like:

Baudrate 250000
Filament Printed [m] 0
Printer active 0
Max. inactive time [ms,0=off] 1800000
Stop stepper after inactivity [ms,0=off] 0
Steps per mm 80
Max. feedrate [mm/s] 300
Homing feedrate [mm/s] 120
Max. jerk [mm/s] 35
X home pos [mm] 0
Y home pos [mm] 0
Z home pos [mm] 0
X max length [mm] 250
Y max length [mm] 250
Z max length [mm] 396.622
Acceleration [mm/s^2] 1850
Travel acceleration [mm/s^2] 3000
Diagonal rod length [mm] 269
Horizontal radius [mm] 148
Segments/s for travel 80
Segments/s for printing 225
Tower X endstop offset [steps] 20
Tower Y endstop offset [steps] 449
Tower Z endstop offset [steps] 100
Alpha A (210) 210
Alpha B (330) 330
Alpha C (90) 90
Delta Radius A(0) 0
Delta Radius B(0) 0
Delta Radius C(0) 0
Z-probe height 39.91
Z-probe speed 2
Z-probe x-y-speed 150
Z-probe offset x 0
Z-probe offset y 0
Z-probe X1 100
Z-probe Y1 20
Z-probe X2 160
Z-probe Y2 170
Z-probe X3 20
Z-probe Y3 170

After doing this and checking the height at the center of the bed, I need to raise the nozzle by 2.20mm (to be clear, if I would send G28, G0 Z0, I would be 2.20mm below the bed…so I need to raise it by that much). I kind of got lost between steps 6, 7 and 8….What I understand is that since all towers are leveled, I need to increase the horizontal radius by X mm in order to raise the 0 at all towers and finally have the whole bed leveled, is this correct? I am not sure how to calculate that measure, the excel formula is not the same as the one explained in the original post.

Thanks in advance for your help.

I hadn't seen the posts for a while, so I am just catching up.

I do not understand why you (and some, like Teoman) are not able to use negative values. I have now three Rostocks (one is a V1, cut in acrylic ) and all of them take the minus with not problem. I am using firmware 0.92

[artexmg]

Here is my most recent version of the excel I am using.

It would be easier to create it as a program, webpage, or even better, an app, that would ask only for the offset values of each tower ... but I guess I am too lazy for that (if you do it, please, let me know!)

By the way, I have been able to speed up the procedure by measuring the offset in the THREE TOWERS before doing any modification in the EEPROM, then I enter the three offsets values at once.

It saves some steps. Also, most of the time when recalibrating, the offset in the three towers are very close. If so, then I enter the average value as the offset for the RADIUS in the excel, then enter the three tower AND the radius in the EEPROM at once.

Any way, here is the spreadsheet I am using; it has the historical data I had entered in the last year calibrating one of my printers (so called, the whispering death!)

I hope it helps.

[artexmg]

I found this on the repetier site, which seems to be exactly what you're describing, but easier? I haven't tried it, but it sounds like it accomplishes the same goal - use a stick to get all the motors to the exact same height, then the firmware figures out the offsets.

http://www.repetier.com/documentation/r ... z-probing/

Afterward then I guess it's still your job to figure out the horizontal radius, but maybe this helps speed up the whole process.

Delta printer need more calibration then normal cartesian printer. You need to calibrate the endstops first. This calibration changes the values you see in the eeprom configuration as “Tower X endstop offset [steps]” resp. Y and Z. The trick is simple. After homing to the max endstops, the firmware will move these steps down to position the extruder holder in the exact center. It is essential to define the exact center position or the nonlinear behavior of delta mechanics will result in wrong geometries. Measuring these values on your own is nearly impossible, so you need to use a simple trick. First home and remove any extruder offset using these commands:

G28
G131

Now position the 3 sliders, so that they are at the same height. To do this I use a stick and position all sliders, so that they press the stick lightly against the printer top, as I know it will be centered then. Depending on your printer it might be better to measure from the bottom (not the bed – that might be skewed in addition). The hard way is to use moves in x,y,z direction. The easy way is turn motors off with “M84″ and slide by hand. Be careful not to slide fast. Movements create a current in the motors flowing back to the electronics. Since some printers will move down when motors are disabled, newer versions have a command to only disable one motor for a time. So now the best way is to use M99 X0 to move the X motor for 10 seconds. You can change the time with S<time in seconds>. Use Y0 and Z0 for the other 2 axis. Once all sliders are at the exact same height, send:

G132 S1

That will measure the required offsets. S1 additionally stores the result in EEPROM, so you do not need to repeat it again.

After the endstops are calibrated, you can run the bed leveling routine the same way as for cartesian printers.

Yeah, I am pretty sure I am not the only one who has figured out. Actually, I am pretty sure there SHOULD be a better, easier, painless way to do it. Young people call it "an app"

[artexmg]

Great thread guys!
Art thanks for posting your spreadsheet. I have downloaded it and will be using it forthwith

Happy that it has been used over time!

[artexmg]

Thanks so much for this post. I was having quite a bit of trouble with the manual, but this got me through the calibration process very quickly.

Glad it helped, really!