Hi there,
I would like to use the Maslow4.1 for cutting ply sheets with a dimension of 3.1 x 1.53 meter. How big would the frame need to be, in order to cut along all edges of the sheet?
I want to cut frames and plates with finger puzzle joints for a boat building project.
Many thanks for your reply and best regards, Nicolas
take a look at http://lang.hm/maslow/maslow4_frame.html and play around with
numbers
the green area is not an absoltute limit, but the further you go outside of it
the more likely you are to have problems.
David Lang
Hi David, thanks for your reply.
What is to be expected in the white areas? Does the accuracy start to lack and if, are there already experience about how much?
So, depending on the space which will be available - which dont know exactly, yet - I could maybe span an area of 5 x 3.4 meters or so for the anchor points.
But, this would require a belt length of more than 4.7 meters. I guess this is more than the spools are holding, right?
I have alread read from belt extensions. How long would such extensions need to be?
Best regards,
Nicolas
Triops wrote:
What is to be expected in the white areas? Does the accuracy start to lack and if, are there already experience about how much?
In both the white and the red area, the arms hit the uprights, and so there is
an angle between the arm and the belt that the machine doesn’t know about. As a
result the effective length of the belt is shorter than the machine thinks it
is.
The further into the red/white areas, the more of an angle there is, and the
bigger the effect. This is a trig problem, and if you remember the cosine curve,
it starts off fairly flat, but develops more slope as you move. This is what
happens with the error, initially there is almost no difference, but it gets
bigger in a non-linear way.
So, depending on the space which will be available - which dont know exactly, yet - I could maybe span an area of 5 x 3.4 meters or so for the anchor points.
But, this would require a belt length of more than 4.7 meters. I guess this is more than the spools are holding, right?
the belts that are provided are about 4.4m long (14.5 ft, but you don’t want to
run them fully out)
I have alread read from belt extensions. How long would such extensions need to be?
as long as needed to reach your anchors 
but they aren’t supported yet.
David Lang
one more thing, you notice that the page has options for angles (from bar, from
cad, etc) those there because we are not completely sure what the angles are
that can work.
Bar designed this to be 20 degrees from the post, but cad indicates that we may
be able to get just a smidge closer.
as for how big the error is
if the belt + arm is supposed to be 1m long
angle effective length
1 999.98
2 999.921
3 999.822
4 999.684
5 999.507
6 999.29
7 999.034
8 998.738
9 998.403
10 998.029
look at the angles that are there from bar and cad, then select custom and
change the angles and you can see how much further you can get by allowing a
little error. As you see above, going a degree or two out of the green area
(into either white or red) is probably not going to be noticable, but going a
lot out will be much more likely to be noticed. But it all depends on what you
are doing. If you are doing big shapes for a theater stage or holiday yard
figures, you will not notice a few mm of error. But if you are making cabinets,
you will notice if one side is a few mm different from the other side.
We also don’t know how much error we are getting from the belts stretching
(depends on the manufacturer stretch factor, the belt length, and the tension
applied, and we only really know the belt length)
David Lang
@dlang do you have the source for that posted somewhere? I’d like to play around with adding more variables
Bar wrote:
@dlang do you have the source for that posted somewhere? I’d like to play around with adding more variables
I just fired up onshape and made a sketch with a triangle (153mm, 1000-153mm,
measure the other side) and changed the angle between the two fixed sides and
typed in the length of the remaining side
It was faster than putting together the right math in a spreadsheet
what other variables do you want to add? I can think of stretch and frame flex
off the top of my head
David Lang
Hi David, thanks for the explanation.
My intention is to use the machine to cut parts for a 8 meter plywood sailboat from 8-12 mm thick sheets. So, the error should be kept as small as possible.
Is it already planned to support belt extensions in the software? I already read about a similar discussion in another thread.
With regards to the stretching of the belts. I assume that the motor current is already measured? Thus, the motor torque and with that the tension in the belts could be approximated?
We are just in the beginning of the design phase and the actual start with building is schedulled for end of next year. So we still have some time, waiting for good updates.
However, we plan to pre-order the M4.1 already to do some testings and to get familiar with the device.
Best regards,
Nicolad
I’d like to make everything gradients so we can tell the difference between areas where the angle won’t really matter and areas where it will. I’d also like to do belt tension and angle
Here’s my take on a simulator for the anchor points. It will let you turn on and off different priorities to see how they will impact your effective work area.
You can play around with it here:
Triops wrote:
Is it already planned to support belt extensions in the software? I already read about a similar discussion in another thread.
it is planned, I wrote a series of 14 patches that did a lot, including that,
but something in the series broke, and it’s so invastive that I’m going to have
to re-do it and push for them to get merged quickly (looking for testers to try
the code)
With regards to the stretching of the belts. I assume that the motor current is already measured? Thus, the motor torque and with that the tension in the belts could be approximated?
except that you don’t know how much friction there is between the spool and the
arm, or how much extra force is needed because the idler and motor are too close
together.
you can say that on this arm, more current means more tension, but cannot say
that more current on arm A is higher or lower tension than less current on arm B
David Lang
Bar wrote:
I’d like to make everything gradients so we can tell the difference between areas where the angle won’t really matter and areas where it will. I’d also like to do belt tension and angle
the problem is knowing where it does and doesn’t matter
I was thinking of adding lines outside the green area and inside the red area
with each new line being 1 degree more angle allowed. Any suggestions for better
colors? I like green for the ‘it is safe’ but the red is too strong. a couple
different yellowish colors? and what color for these new lines? (green? other??)
David Lang
I changed the graph at http://lang.hm/maslow/maslow4_frame.html
instead of red/white it’s now gold in the corners and yellow top and sides
I also added a green line for each degree of bend and added the table above
so you can almost always go one band out of the green, probably at least two,
but I think that by about 5 (~ 0.5mm of error on a 1m desired belt length) you
have probably gone too far (on the chain maslow, 1mm or chain error could be as
much as 2mm of position error, on the maslow 4 it gets MUCH more complicated)
David Lang
