Homing works calibration doesn't.....
Posted by Sash
|
Homing works calibration doesn't..... August 17, 2015 12:07PM |
Hi again,
Well I have made some progress as homing works.
I cannot seem to calibrate my bed though enough to get printing well. I have managed a couple of descent prints. Generally though I can't print. I home and then use 'bed7' macro to calibrate the bed. It goes through the motions. On first pass it drags on areas of the build platform then second time I run bed7 it's fine no dragging. I run bed7 about 6 times before trying to print. But as you can see from the photo that the head still drags on the platform when trying to print. If I do get pass this stage and continue printing the bed will move sideways with the force of the head and the print will be misshapen. Any ideas would be welcome?
Cheers
Sash
Well I have made some progress as homing works.
I cannot seem to calibrate my bed though enough to get printing well. I have managed a couple of descent prints. Generally though I can't print. I home and then use 'bed7' macro to calibrate the bed. It goes through the motions. On first pass it drags on areas of the build platform then second time I run bed7 it's fine no dragging. I run bed7 about 6 times before trying to print. But as you can see from the photo that the head still drags on the platform when trying to print. If I do get pass this stage and continue printing the bed will move sideways with the force of the head and the print will be misshapen. Any ideas would be welcome?
Cheers
Sash
|
Re: Homing works calibration doesn't..... August 19, 2015 02:28PM |
Thanks again.
Well I proceeded to run bed.g several times here is the output:
I used a simple test print file. Here is the result (a little disappointing):
Well I proceeded to run bed.g several times here is the output:
>>>g28 SENDING:G28 >>>g32 SENDING:G32 Calibrated 6 factors using 7 points, deviation before 0.196 after 0.169 >>>g32 SENDING:G32 Calibrated 6 factors using 7 points, deviation before 0.164 after 0.164 >>>g32 SENDING:G32 Calibrated 6 factors using 7 points, deviation before 0.172 after 0.167 >>>g32 SENDING:G32 Calibrated 6 factors using 7 points, deviation before 0.177 after 0.164
I used a simple test print file. Here is the result (a little disappointing):
|
Re: Homing works calibration doesn't..... August 23, 2015 10:08AM |
Cool. Well if someone finds any of it useful then at least I'm not wasting my time. I did the stripdown & re-build yesterday/today. I've made notes of what I did and any changes with some relevant pics... I'll tidy it up and get here here when I get some free time.
In the mean time, here's a screenshot of the initial calibration and first print. I'm pretty happy with that, tbh. Yeh, it's stringy, but that's the material rather then the printer.
In the mean time, here's a screenshot of the initial calibration and first print. I'm pretty happy with that, tbh. Yeh, it's stringy, but that's the material rather then the printer.
|
Re: Homing works calibration doesn't..... August 24, 2015 10:34PM |
I've generated a model of a delta machine in Excel and have been running some scenarios with the 162mm arms and base scaled and the Z axis displacement correct and got some interesting results.
This chart compares a perfect 100mm circle on a perfectly calibrated machine (blue trace) with a machine with oversized arms (orange trace)
In order to better visualise the error on the orange trace I've multiplied it by 100, and found that it produces an interesting shape, its called a lobe its actually one of my favorite shapes because it has a very interesting property, that a casual inspection with calipers would convince you that its actually round...if you have a British 50p coin try it.... its also a shape that I used to anticipate with dread when working on machines, because it generally means your spindle bearings are knackered!
If you were to make both shapes and measure them with a pair of calipers the 100mm shape would measure 100mm (of course!), and the lobed shape would measure 100.21mm, which may be within a margin of error.
To detect a lobed shape you need to measure it with a three point measurement technique such as a placing it in a v-block and measuring the height as the shape is rotated.
It would be interesting to know if your printer with its oversized arms is actually printing lobes instead of cylinders.
Of course this doesn't mean this calibration technique is bad, if it enables parts to be produced within an acceptable margin of error then great, but it would be good to know if the actual consequence matches the prediction, if it doesn't then there is something else going on.
RepRapPro Mendel 3 Tricolour
RepRapPro Fisher
-Carbon Arms
-Easy adjust Carriage+effector
-axis stiffness mods
HE3D -600 delta
-Duet 0.8.5
-PanelDue
-DC42 Height probe
-RobotDigg metal components
Simplyfy3D
RS Design Spark CAD
This chart compares a perfect 100mm circle on a perfectly calibrated machine (blue trace) with a machine with oversized arms (orange trace)
In order to better visualise the error on the orange trace I've multiplied it by 100, and found that it produces an interesting shape, its called a lobe its actually one of my favorite shapes because it has a very interesting property, that a casual inspection with calipers would convince you that its actually round...if you have a British 50p coin try it.... its also a shape that I used to anticipate with dread when working on machines, because it generally means your spindle bearings are knackered!
If you were to make both shapes and measure them with a pair of calipers the 100mm shape would measure 100mm (of course!), and the lobed shape would measure 100.21mm, which may be within a margin of error.
To detect a lobed shape you need to measure it with a three point measurement technique such as a placing it in a v-block and measuring the height as the shape is rotated.
It would be interesting to know if your printer with its oversized arms is actually printing lobes instead of cylinders.
Of course this doesn't mean this calibration technique is bad, if it enables parts to be produced within an acceptable margin of error then great, but it would be good to know if the actual consequence matches the prediction, if it doesn't then there is something else going on.
RepRapPro Mendel 3 Tricolour
RepRapPro Fisher
-Carbon Arms
-Easy adjust Carriage+effector
-axis stiffness mods
HE3D -600 delta
-Duet 0.8.5
-PanelDue
-DC42 Height probe
-RobotDigg metal components
Simplyfy3D
RS Design Spark CAD
Sorry, only registered users may post in this forum.