Z axis calibration method?

General questions about the printer and printer software.

Z axis calibration method?

Postby brian.cook » Fri Aug 21, 2015 7:41 am

We have two Nano's. Prints from each are quite different in height. The rook for example on my printer is taller and slimmer than a control print made from a Makerbot. The other nano is producing prints that are shorter than the reference. Is there a procedure that can be used to calibrate the x-axis, to get both printers to print more alike?

Regarding the coupler that attaches the motor to the threaded rod, the nano that is printing shorter has a 1/4" of play in it at the motor, while the other one has virtually no movement when I gently pull up and down on the build plate assembly. What is correct?
brian.cook
 
Posts: 5
Joined: Fri Aug 21, 2015 7:20 am

Re: Z axis calibration method?

Postby ron.kundla » Fri Aug 21, 2015 10:31 am

If you think about it, any play in the z-axis will affect the accuracy of the print, or at least trying to zero the z-axis since the motor/screw assembly will move when the plate in embedded into the vat.

Which one gives you the best or closest Z accuracy? The one that has slack or the one that doesn't?

I haven't seen any calibration method for the Z-axis. Most of the debug log shows that the number of turns of the screw is hard-coded into their printing software. Maybe iBox can expose an adjustment where if you print a 10 mm cube and it isn't a full 10 mm high, you can input a ratio that will allow them to alter the number of steps required to move the screw 100 microns or whatever you are trying to do. Also, there could be accuracy issues based on the screw and stepper motor parameters where some layer heights are whole numbers of steps versus partial or fractional steps.
Super Early Beta Adopter
"I don't work for iBox, but I did stay at a Holiday Inn Express" ;)
ron.kundla
iBox Pro User
 
Posts: 164
Joined: Sun Jun 14, 2015 1:39 pm

Re: Z axis calibration method?

Postby brian.cook » Fri Aug 21, 2015 11:54 am

The one with the play in the threaded rod is actually closer to "reference", but in my opinion has less consistent layers. I'd like to know what's normal. Maybe the advanced user's guide will help explain some of the other software settings that might adjust this, but in theory I should be able to make two nano's print similar with the same settings, provided mechanically they are both doing the same thing. I believe the evidence suggests they are not. The order of the images I have attached is Nano2, Nano1, Makerbot (clear). The height of each Rook is as follows: Nano2: 30.71mm, Nano1: 33.07mm, Makerbot: 31.6mm. The difference in height between the 2 Nano's is 2.36mm (0.093") which seems like a lot.
Attachments
nano_top_view_rook.jpg
Top view
nano_top_view_rook.jpg (85.56 KiB) Viewed 2753 times
nano_height_comparison.jpg
Rook Height Comparison
nano_height_comparison.jpg (57.4 KiB) Viewed 2753 times
brian.cook
 
Posts: 5
Joined: Fri Aug 21, 2015 7:20 am

Re: Z axis calibration method?

Postby TieBox » Fri Aug 21, 2015 3:51 pm

Do nano1 and nano2 have the same settings for foundation layers? If nano1 is set to add 20 layers of model footprint foundation and nano2 is not this setting could account for the difference in the models. It appears to me that your nano2 and makerbot prints are more accurate to the 3D model of the rook.

The nano that has no z-play is ideal but there are easy ways for getting accurate prints even with some play.
1.) The easiest way would be to add model footprint layers. If you determine that your model is 2.3mm shorter than it should be than add 23 layers of model footprint foundation in your configuration settings. It will make the print take slightly longer but it will come out being the perfect size.

2.) You can add a teflon washer between the stepper motor and the drive screw to eliminate z-play. This is pretty hands on so I will make an instructional video on how to do this as soon as possible. The only down side of this method is that it can cause excess friction that leads to the z-axis sticking during prints. This z-stick goes away once you run the carriage all the way up and all the way down a few times.
Troy
Chemical Test Engineer at iBox Printers
Youtube Tutorial Guy- Please make suggestions for new tutorials, or ways to improve old ones
User avatar
TieBox
Chief Of Resin
 
Posts: 36
Joined: Fri Jun 05, 2015 12:11 pm

Re: Z axis calibration method?

Postby ron.kundla » Fri Aug 21, 2015 6:15 pm

brian.cook wrote:The height of each Rook is as follows: Nano2: 30.71mm, Nano1: 33.07mm, Makerbot: 31.6mm. The difference in height between the 2 Nano's is 2.36mm (0.093") which seems like a lot.


Just to make sure, you have the same profile on each Nano, right? - haha already asked by TieBox :-P

I've seen what appeared to be layer compression, or my zero point was off initially and the first 'n' layers were not right. I wanted to run a test with one of those 5 mm pyramid calibration cubes so you can look at each layer and verify the height. It would prove whether I was seeing problems with the base layers, or it was an issue across the entire build height. Never got around to it...
Super Early Beta Adopter
"I don't work for iBox, but I did stay at a Holiday Inn Express" ;)
ron.kundla
iBox Pro User
 
Posts: 164
Joined: Sun Jun 14, 2015 1:39 pm

Re: Z axis calibration method?

Postby brian.cook » Fri Aug 21, 2015 8:57 pm

Yes, the profiles were verified to be the same, initially they were not, that's why I have been curious about whether z axis calibration is possible. Thanks for the suggestion about the pyramid, that sounds like it might be a good way to check what's happening especially with nano2. What can I do about nano1 where the rook is too tall? This is the nano that has almost no play in the z axis, removing foundation layers will change the layer height, but not the overall scaling. Thanks.
brian.cook
 
Posts: 5
Joined: Fri Aug 21, 2015 7:20 am

Re: Z axis calibration method?

Postby Xios1306 » Fri Aug 21, 2015 10:15 pm

I would suggest adding a foundation set to model foot print so the nano spends all the play it has in printing the foundation and when it starts printing the model all the play is now gone.
Harsha B C
Intern - Computer Engineer
harsha@iBoxPrinters.com
            iBox Printers
┌───────────────────┐
│██   ██   ██   ██   ██   ██  \
│  ██   ██   ██   ██   ██   ██  \
│   ▀▀   ▀▀   ▀▀   ▀▀   ▀▀   ▀▀ │
│   ▄▄   ▄▄   ▄▄   ▄▄   ▄▄   ▄▄ │
│  ██   ██   ██   ██   ██   ██  /
│██   ██   ██   ██   ██   ██  /
└───────────────────┘
             you make it
User avatar
Xios1306
iBox Pro User
 
Posts: 174
Joined: Wed Jun 03, 2015 12:15 pm

Re: Z axis calibration method?

Postby ron.kundla » Sat Aug 22, 2015 12:58 am

I think the problem with that is really knowing how much play there is on the assembly and if it is consistent between Z-axis zeroing.

Using a support base, assuming the print can survive the peel forces, would help get you past any initial compression.

For the too tall nano, print the 5mm pyramid (it was on Thingiverse). Maybe what you have there is that the stepper motor is giving off too many steps based on that calculation? It is like a measurement to Z stepper motor scaling issue.
Super Early Beta Adopter
"I don't work for iBox, but I did stay at a Holiday Inn Express" ;)
ron.kundla
iBox Pro User
 
Posts: 164
Joined: Sun Jun 14, 2015 1:39 pm

Re: Z axis calibration method?

Postby brian.cook » Sat Aug 22, 2015 1:03 pm

Some of the difference is definitely related to the zeroing of the build plate to the vat. I made a few changes to my approach and the setup of the nano. I switched to Qsil for the vat coating. The entire build plate assembly was not level to the bottom of the vat, so I used the 2mm nano model and turned off foundation layers. I then shimmed the build plate until the model measured an even thickness across the plate (1.2mm in my case). Next, instead of allowing the motor to drive all the way down to the vat, I placed a small amount of ibox nano resin in the vat and lowered the build plate until it squeezed out from around the plate, then I bumped the plate down 1 step at a time until it was in full contact with the vat. I loaded the chichen_itza10mm model and printed it with the foundation layers turned off. I used a micrometer to measure the height of the finished model and it measured 10.35mm. The model is 104 layers, so that is within 50 microns of what it should be, not what I was expecting since I was prepared to use this measurement to adjust the number of steps per 100 microns. This was for nano1 that has no play in the z axis. I will try the same approach on the other one, but it might not be until a week from now due to travel. I am printing another Rook to see how that goes, but I am not getting the pop that I was getting on release with the FEP tape so far.

I would still like to get the procedure for how to add a teflon washer to minimize the play so I can do a before and after. Thanks for the assistance.
brian.cook
 
Posts: 5
Joined: Fri Aug 21, 2015 7:20 am

Re: Z axis calibration method?

Postby ron.kundla » Sat Aug 22, 2015 7:56 pm

The Qsil flexes so you won't get the pops anymore when the print separates from the plate. This is the reason why I have not gone back to FEP on my vat.

Another positive side effect is that you can actually do away with the foundation layers using Qsil since the peel is less violent.

Nice testing. Please keep us up to date with your progress!
Super Early Beta Adopter
"I don't work for iBox, but I did stay at a Holiday Inn Express" ;)
ron.kundla
iBox Pro User
 
Posts: 164
Joined: Sun Jun 14, 2015 1:39 pm

Next

Return to General

Who is online

Users browsing this forum: No registered users and 1 guest

cron