That is disappointing news. Definitely curious if you are talking about From 1 or Form 2. I’ll cling to hope that somehow you are the exception to the rule. In any case, I’ll be looking closely into precision and will report my findings…good or bad.
Thanks for the heads up.
EDIT: I see you’re using the Form 1. There is hope. The Form 1 wasn’t regarded so well in terms of precision. I believe the Form 2 is in fact more accurate. We’ll see.
I hope you are right that the 2 is significantly better than the 1±-I’d get one right away, if so–but I have not read anything that would lead me to believe it.
A perfect example of part failures are small interlocking features (like alignment pins & holes or slots & grooves). The Form 1+ requires the features in the model to be so significantly smaller than the actual molded part that the models must be edited with a Form configuration.
Another compromise you make with a Form are the supports; their size, quantity, and the impact on the dimensions and appearance of the part. For molded-part prototyping, you get to choose if you are making a sample for checking out the inside features, or if you are making one for external appearances. The labor to remove the blemishes is so heavy that it’s not worth my time.
In contrast, I got a “final” print from my outside shop, and I thought there was a scratch on it…I looked at it under the microscope, and it wasn’t a scratch, it was actually a support that was thinner than a sewing needle. They just missed one in their cleanup step, and a soft toothbrush will pull it off, and once it’s gone, there is no visible evidence it was there.
You may or may not be right regarding precision, CliffGriffin , but you need more than extrapolation to come to any final conclusion.
Just taking into consideration the improved peel mechanism and the addition of static heat regulation, makes the Form 2 a different animal from the Form 1+.
Though I am not an engineer, I have a weekend project coming up that requires more precision than the organic parts I have printed so far, so I will post my results.
Are there any engineers out there with a Form 2 that might share their professional opinion regarding this matter?
Hmmm… I printed 9x of these parts, dummy cylinders for a scale radial engine. I printed 6 of these in one run along with two other parts, and then 3 more in a separate run. I just measured the distance between the rocker boxes (the “ears” on top of the cylinders) and the front/back dimension of the rocker boxes. They varied by about +/-0.05mm over about a 32mm distance.
It seems like there’s some variability from machine to machine so YMMV.
This makes sense to me since the Galvanometers that move the mirrors are “open loop” control I think. That would mean they’re dependent on an external calibration, and so each machine is probably going to calibrate slightly differently… I suspect there’s a way to “fine tune” X and Y calibration, similar to the build platform Z height fine tuning calibration, but FL apparently doesn’t see the value in exposing those parameters via the user interface.
The Form 2 is a completely different machine than the Form 1/1+ * For accuracy, the new Formlabs designed galvos and digital galvo controller make a huge difference to accuracy. Temperature control of the resin and the peel mechanism with reduced peel forces also make a difference. Tolerances are highly dependent on geometry and material (just like with a milling machine) so it is difficult to boil down actual performance to one or two numbers. Check out some of the accuracy data that we posted for our surgical guide application http://formlabs.com/products/materials/dental-sg/
*Fun fact: there are only two non-fastener components in common between the Form 1+ and 2: the button and the bumpers that stop the cover in open position
HI I am a jeweller and I have created rings in a Cad program and then checked the sizes on a ring stick. and with vernier gauges Dimensionally they are perfectly circular and size wise spot on