Welcome! Log In Create A New Profile

Advanced

Ammissible deflection in 3d printing

Posted by Yaro 
Ammissible deflection in 3d printing
October 31, 2014 10:47AM
Hi all,

I want to ask how much is an ammissible deflection of mechanical parts in a 3d printer. Since we can't build a no deflection structure, how much is maximum deflection in axis or other components of commons user 3d printers? How much is in a professional 3d printer? and how much it should be so we won't see printing errors with an human eye.

Thank you.
Re: Ammissible deflection in 3d printing
November 25, 2014 08:18AM
Very tiny. Everything that causes the extruder to deviate from the gcode path in the XY plane will have a visible effect in the print. Even vertical errors will be visible because the plastic line being laid down will vary in width.

The prints we make can still be functional and aesthetically pleasing even with the errors- that is why there are so many machines using laser cut plywood in the structure of the machine- it is good enough for most people's requirements.
Re: Ammissible deflection in 3d printing
December 27, 2014 11:15PM
Get yourself a cheap dial indicator and measure the actual deflection at various places with a small amount of force. I think you'll be shocked at far it actually deflects, and you'll likely find that certain parts deflect way more than others. You fix those big deflections first and you'll probably see a big improvement, and not bother with the others. I found this to be true when I was building packaging machinery that reciprocated a large mass at fairly high speeds.

I used to say: You can't fix what you can't measure!
Re: Ammissible deflection in 3d printing
December 30, 2014 05:56PM
The human eye can perceive minute irregularities in a consistent pattern.

I'd personally treat xy deflection as a percentage of extrusion width, and z deflection as a percentage of layer height. More than about 5% will probably be visible.

Note that as long as the printer deflects the same amount on each layer there will be no visible finish problems (just dimensional inaccuracies). That situation will probably only occur on very simple prismatic shapes (where each layer is similar) though.


Stratasys publications suggest their machines can repeat to within 0.002" or better, which I believe. The outer surface is done at lower speed (And possibly acceleration) which limits deflection.

To estimate your own deflection under load you can put a dial indicator against the axis and apply a force equal to the acceleration times the mass of the extruder carriage.

Edited 2 time(s). Last edit at 12/30/2014 05:59PM by 691175002.
Sorry, only registered users may post in this forum.

Click here to login