John Cross wrote:
Is there a way to set soft limits on the Z axis? Or change the step increment
automatically when you change the units. Or convert the US to the metric
system. Whatever is easiest
yes, $20=1 turns on soft limits (only works after homing/aka retract/extend/take
slack is complete)
$130-132 set the soft limits
$132=50 would set the soft limit to 50mm
I think it works in mm internally even when you are in inches in gcode.
David Lang
Yeah, I think you’re right. I was thinking it would be less than that, then I remembered that the teeth of the belt aren’t engaging the spool because they face the other way.
Thanks David, that helps.
Also, thanks for your frame sizer website. It’s an impressive bit of programming that really helped me decide to make a 11’W X 9’T vertical frame.
Regarding the tension on the connection between the belt and its anchor into the spool, I did a little research* and found that the reduction of tension when wrapping around a cylinder is governed by the Euler-Eytelwein equation:
T2 = T1 e^(μθ)
where T2 is the tension on the belt-spool connection, T1 is the tension on the belt approaching the spool, μ is the coefficient of friction between the belt and the spool, and θ is the angle of the wrap. To put this in simpler terms, the tension is reduced exponentially the further you wrap. Note that this equation was really intended for a rope wrapped in a spiral around a pole, so θ can be more than 360 degrees. For the spool, it becomes more complicated once the wrap starts overlapping itself.
*By research I mean I searched “how wrapping a rope around a pole increases friction” and clicked on the first link I saw, which was this article
I received my maslow earlier this week and finally had a chance to try calibrating it yesterday. I didn’t have any luck getting the calibration pass. I was getting the frame size dialed in a little more accurately I think, but my belt snapped about 6 ft from the end on the last attempt. I didn’t notice any damage on the belt during setup. My frame size was 3600 x 2451. I will email Anna with hopes of getting a new belt, but thought I would share my experience.
You’re not the first one that has happened to. It’s not common, but it has happened. Could be a defect in the belt, could be it got overstressed. When you get your replacement, pay close attention to how everything goes together; make sure it all moves freely. Examine your encoder board for anything that looks funky. We are still looking for the causes of problems, any clues could be useful.
Good luck!
Finally had time to get everything back together and updated. Had the calibration fail a few times due to a fluorescent light that was too close knocking out the WiFi. I appreciate all the help and look forward to start using it.
Really think the next iteration needs to be belt replacement friendly. I have a feeling that over time with repetitive use the belts will weaken and fail.
Tim Seidlitz wrote:
Really think the next iteration needs to be belt replacement friendly. I have
a feeling that over time with repetitive use the belts will weaken and fail.
possibly, but they are being flexed far less than in 3d printers
David Lang
True but I suspect the tension is higher on the Maslow.
And they are exposed for spans a that are far larger than the entirety of most 3d printers, will definitely be touched/bumped more often, etc. Not to mention the risks of hitting one with a workpiece or something else while mounting them (not everyone will take the Maslow off the frame every time to mount a workpiece), or the strain that would potentially be put on them by people leaning on the sled while changing bits.
The spools not being designed for easier replacement just seems like an unforced error, all that in mind. It’s definitely the part that I would hate to replace the most while also feeling like it is the most likely I will have to.
For now, I would be happy (if it were available) to buy a completely assembled individual arm to have as a spare – all included, motor, encoder, belt everything.
Totally agree with this. However, I would go for a basic option of a fully assembled arm with the encoder, but not the motor.
In fact, I would think that this should be the standard way that the arms are shipped in a regular kit. Then separately there could be the various individual parts for people who wanted to repair an arm for themselves.
Same here - belt snapped during calibration. Although it also looked like the machine was giving too much slack on the belts when moving leading to the belt being wrapped around the drum and I guess the belt snapped when the motor tried to retract. Is there a way to reduce the force that pulls on these belts? Also - and that might be a user error - I’m starting with a smaller 2x4 ft frame but reading through the forums people used smaller frames so not sure if the angles are a problem during calibration with smaller frames
it’s more likely just a defective belt, tontact @bar and @anna to get a
replacement.
you don’t want to lower the tension on the belts.
David Lang
By 2x4 ft frame do you mean:
I suspect you would not get good results with option 1 above.
Rectangular work area for 2ft by 4ft frame would be under 1ft by 2 ft.
yep I was trying to start small 
But I just think the different and larger angles from the rotors to the anchor points make it too difficult. Once I have the replacement I start with a bigger footprint - probably directly on the floor.
Is there a pinned link the frame calculator somewhere? I see you entered a frame size in inches and the auto calculation worked. I suspect that I am using an old link (because the auto calculate is not working and I see no way to select metric vs english units).Thx!
