Buying help

General discussion topics
Post Reply
Mothman
Posts: 2
Joined: Sun Jul 09, 2017 9:22 pm

Buying help

Post by Mothman » Mon Jul 10, 2017 12:45 am

Hey everyone,

I'm looking into buying a 3D printer and have been strongly considering the MakerGear M3. One of my largest concerns with the printer I'm purchasing is how quickly it is capable of printing. I know print speed is very dependent upon a lot of factors but I'm trying to establish a baseline across several systems to figure out how quickly I could build parts. To do this I've been slicing this car model (https://drive.google.com/open?id=0B0ccB ... np4QXM3TVU) at 1/10th size with a .2mm layer height. If someone could slice this and tell me approximately how long this would take to print it'd significantly help my decision making process.

Thanks,
Mothman

User avatar
ednisley
Posts: 1188
Joined: Fri Apr 11, 2014 5:34 pm
Location: Halfway up the Hudson
Contact:

Re: Buying help

Post by ednisley » Mon Jul 10, 2017 2:46 pm

Mothman wrote:slicing this car model
That model is severely non-manifold: slicers will refuse to process it or introduce weird defects while attempting to repair the broken geometry. There's no point in proceeding until you have a valid model.

Rather than attempting to build support into the model with rectangular bars, let the slicer handle that job. You won't like removing support, but it will work better: those side mirrors and wheel arches are un-printable without support.

At 1/10 scale, most of the fine detail will vanish, because all fused-filament printers have a minimum resolution around 0.5 mm in the XY plane.
how quickly it is capable of printing
All fused filament printers are limited by their nozzle diameter: gooey molten plastic doesn't flow well. In round numbers, a 0.35 mm nozzle can extrude 10 mm³/s, so divide the volume of your model by 10 to get the number of seconds. Round that up by a factor of 2 or 3 for a rough estimate.

Larger nozzles print faster by trading off XY resolution. If speed is the only thing you care about (it shouldn't be), then a larger nozzle can get the job done somewhat faster, limited by the hot end's ability to melt plastic, although the printed object will lack detail. For example, a 0.85 mm nozzle has an XY resolution of maybe 1.5 mm, about 1% of the reduced-size car model: a crappy printed part in maybe 1/3 the time.

wmgeorge
Posts: 200
Joined: Thu Jun 18, 2015 1:09 am

Re: Buying help

Post by wmgeorge » Mon Jul 10, 2017 3:36 pm

All 3D printers take some time to learn. 3D printing is not the answer for every project. This Forum is a great resource for the MakerGear. At one time I had another very well known brand, also made in USA. The build quality of the M2/M3 gives it a advantage over the other brands. It just works. The other one got sold.
Retired Master Electrician, Commercial HVAC/R,CNC Router

Mothman
Posts: 2
Joined: Sun Jul 09, 2017 9:22 pm

Re: Buying help

Post by Mothman » Mon Jul 10, 2017 10:06 pm

Thanks for the quick replies!

@ednisley So sorry I included the wrong file. Here is the one that I repaired for printing. (https://drive.google.com/file/d/0B0ccB6 ... sp=sharing) Unfortunately it lost a good deal of detail but it is what I used in my other tests. I left the support on the model so that the support added would be consistent between tests. I'm trying to avoid modding the machine too much so I'm trying to get an idea for out of the box speed.

Thanks again,
Moth

User avatar
ednisley
Posts: 1188
Joined: Fri Apr 11, 2014 5:34 pm
Location: Halfway up the Hudson
Contact:

Re: Buying help

Post by ednisley » Tue Jul 11, 2017 2:05 am

Mothman wrote:trying to get an idea for out of the box speed
The model still has non-manifold vertices and remains unprintable; the wheels now have detached parts and the model has far too much irrelevant detail. Despite that, I sliced it with my default settings, fed it to gcode.ws, and got back "not a number" time estimates, which suggests the G-Code is junk.

You must (learn how to) create geometrically correct models that work with 3D printers, optimize the models to eliminate irrelevant details, tune the model to match the slicing parameters, and iterate until the results converge. Do that for each printer and then you can get a valid comparison.

Comparisons between printers based on an invalid model using G-Code produced from different slicers and slicing parameters tells you nothing, because the numbers are meaningless.

In fact, for a given G-Code file, the final print speed has more to do with the axis accelerations set in the firmware, which are under your control if you're using a printer with GPL firmware. The actual print time can be slower than the estimates by anywhere from less than factor of two to more than a factor of ten, which generally outweighs the effect of the slicing parameters.

Bottom line: if speed is your only concern, have Shapeways print your (valid) models. The parts will come back in a few days and you can move on.

Post Reply