Back Lash compensation needed!!

[kinginnovator 6]

Hello everyone, first time here. Got my Ultimaker running perfectly just one minor bug to work out. My parts in the (y) direction are coming out too short. All round objects are slightly oval. Triple checked all belts, pulleys, motors, shafts, tensions, etc etc all is tight and sound. Is there a way to adjust backlash in the y?? I'm using skeinpypy and printrun...

Thank you!!

[Daid 306]

Skeinforge has an option for this. But it's a hardware problem, really. And even with backlash compensation you will not get perfect prints.

Did you check motor current setting?

[kinginnovator 6]

I didn't like skeinforge, I spent too much time trying to navigate and tweak settings to get good results, I also had crashes when printing didn't like how it jitterred around turns too. Where would I check motor current settings? Is this in Skeinpypy?

Thank you for responding by the way.

ST

[Daid 306]

The step motor current is a hardware setting:

http://wiki.ultimaker.com/Electronics_b ... or_drivers

Usually the defaults are good, but in some rare cases the driver current is wrong.

[kinginnovator 6]

Checked, tried to go up and down with the pot but no change. I need to adjust the motor speeds another way to compensate for the problem. Any other ideas? Do you recommend a better program with other settings? I can always try skeinforge again. Let me know if you have any other ideas. All I know is there is a direct relationship going on between the x and y. It doesn't matter if the part is large or small. The difference between the parts is averaging around .6mm difference. For example if I have a 12mm diameter ring to build and a 12mm wide square one measurement for either would be 12.2mm in x while y would be 11.6mm, an average difference between x and y 6mm. Very strange here. I've build 50 parts so far with a very precise average on this. Tried tensioning the builts in different ways but absolutely no change.

ST

[owen 19]

Checked, tried to go up and down with the pot but no change. I need to adjust the motor speeds another way to compensate for the problem. Any other ideas? Do you recommend a better program with other settings? I can always try skeinforge again. Let me know if you have any other ideas. All I know is there is a direct relationship going on between the x and y. It doesn't matter if the part is large or small. The difference between the parts is averaging around .6mm difference. For example if I have a 12mm diameter ring to build and a 12mm wide square one measurement for either would be 12.2mm in x while y would be 11.6mm, an average difference between x and y 6mm. Very strange here. I've build 50 parts so far with a very precise average on this. Tried tensioning the builts in different ways but absolutely no change.

ST

Gooday King

The trim pots adjust the maximum current that the stepper motor can draw. You need to change the Steps per mm for your X and Y in the frimware.

This is done by sending a M92 command either in your GCode header when you go to print.

M92 X78.74 Y78.74

or permanently set it in firmware by issuing

M500 stores current settings,

M501 retreives the data from EEPROM (also automatically done after a reset)

M502 reverts to the 'factory defaults' You'll need to send an M500 if you want to store this in EEPROM.

so from RepG or Printrun etc.

send M501 to view what they are currently

print something 100mm X 100mm

say your Y setting is currently 80 and your print is 90mm in Y direction

your new Y would need to be 80 * 100 / 90 = 88.889

So you would send

M92 Y88.889

M500

do the same for x and it should be all good

You can also use Daid's Build-me-Marlin to input your X and Y and update your firmware in one hit (It's the way I do it)

[kinginnovator 6]

Thank you so much! I'm going to give this a shot. I will let you know how this turns out. Thank you again so much.. Let me know if you need any cad done for free sometime... I also have urethanes ans silicones to prototype with. Been making things for over 10 years..

ST

[kinginnovator 6]

Worked like a charm! Just loaded new gcode. Machine is very fast, very precise, very happy...

King Innovator

[owen 19]

That's great. Thanks for the update. Enjoy your printing.

[almi 0]

Owen,

thanks for hints on calibrating the X,Y values.

I would wanna give it a try on my machine, too.

But of course need my actual values.

Your tip was "send M501 to view what they are currently".

I have sent M501 in printrun and it won't show me any information about the actual EEPROM values.

What went wrong? Any initial command to be sent for retrieving the data?

Thanks for any hint and cheers,

almi

[owen 19]

no that should do it. I'm not near my machine right now so I can't do it myself. I usually use repG myself. i can have a look tomorrow unless someone else can help

[almi 0]

I also tried to retrieve the EEPROM parameters in RepG (By sending an M501 command via GCODE dummy file) and got following:

"Couldn't show onboard paramters, driver doesn't implement onboard parameters"

Now I am using the Marlin Build 3 from Bernhard Kubicek.

What driver do you use?

sending: N4 T0 M105 *71

[22] received: ok t:51.27, raw:105 b:0.00

[22] sending: N5 M501 *2

[22] received: ok

[22] sending: N6 M108 S0 *79

[22] received: ok

[22] Scheduling machine state change event: READY

[22] Machine state changed to BUILDING

[22] Yes, I have a Realtime Control feature.

[22] Sending machine state change event: BUILDING

[22] Machine state changed to READY

[22] Supports RC

[22] Couldn't show onboard parameters: [driver doesn't implement onboard parameters]

[22] Yes, I have a Realtime Control feature.

[22] Supports RC

[22] Sending machine state change event: READY

[22] Done building.

[owen 19]

Hi almi

not sure how you're doing it in RepG but all you have to do is click in the white part

press Control+a

type M501

click on the build button

in the black window you should get something like this

[14] Machine stop called.

[14] Building...

[14] Estimating build time...

[14] Estimated build time is:

[14] Beginning build.

[14] echo:using default settings:

[14] echo:steps per unit:

[14] echo: m92 x79.06 y79.04 z533.33 e14.00

[14] echo:maximum feedrates (mm/s):

[14] echo: m203 x400.00 y400.00 z9.00 e150.00

[14] echo:maximum acceleration (mm/s2):

[14] echo: m201 x15000 y15000 z150 e200000

[14] Yes, I have a Realtime Control feature.

[14] Supports RC

[14] echo:acceleration: s=acceleration, t=retract acceleration

[14] echo: m204 s3000.00 t3000.00

[14] echo:advanced variables: s=min feedrate (mm/s), t=min travel feedrate (mm/s), b=minimum segment time (ms), x=maximum xy jerk (mm/s), z=maximum z jerk (mm/s)

[14] echo: m205 s0.00 t0.00 b20000 x20.00 z0.40 e5.00

[14] echo:pid settings:

[14] echo: m301 p22.20 i1.25 d99.00

[14] Yes, I have a Realtime Control feature.

[14] Supports RC

[14] Done building.

note: the return values are only to 2 decimal places of accuracy

[almi 0]

Thanks Owen.

I now solved the issue. The M501 command only works with the Marlin firmware.

In my case, for what reason ever I thought I had Marlin on, but actually it was a Sprinter version.

Now that I have Marlin on, I receive full EEPROm information after the M501 command.

So now I can try the X, Y Axes compensation.

Thanks,

almi

[owen 19]

That's great almi

good luck