@bar I think I remember you making a comment that the rollers now have the magnets glued in when you ship them. Is this correct?
I had a roller break the end off, so I printed a new end to put in it, but at some point the magnet fell off, so I need a new magnet, but if the rollers include the magnets, I’ll just buy a set of rollers.
The ones in the kits do, but the ones in the warehouse haven’t been restocked to include the magnets yet.
It’s been on my todo list, but there are some weird restrictions on shipping magnets so I was having a hard time getting them sent.
I do have a couple laying around here tho so I can send you one for sure. DM me the address you want them sent to and I’ll get them in the mail ASAP
What are the specs for the magnets? Are they north south across the diameter with just one pole on each side? Are they rated to a particular strength? Not ordering any now just wondering for documentation.
wouldchuck wrote:
What are the specs for the magnets? Are they north south across the diameter with just one pole on each side? Are they rated to a particular strength? Not ordering any now just wondering for documentation.
radial polorization, so across the diameter rather than up the axis. I don’t
think they are particularly sensitive to stength (the best thing there would be
to read the sensor datasheet
https://www.mouser.com/datasheet/2/588/AS5600_UG000254_2_00-1877354.pdf
since the bearing is 8mm, I believe these are 6mm diameter
one thing you need to know about the maslow, very little of it is “engineered”
as in there being a mathmatical model that then determines what each component
is. Instead it is designed and iterated “this seems like it will probably work,
build it, it does work” which doesn’t mean that some other variation may work as
well (or even better) but this set of components does work.
This is NOT a knock against Bar. He has built things that many experts would
have said aren’t going to work and produced thigns that work well enough for a
lot of people.
I am just setting expectaton that things like magnet strength are more ‘this is
a magnet that comes with the sensor’ or ‘this magnet is cross polorized and
fits’ followed by ‘try it, it works, stop thinking about it’ than any sort of
formal analysis of how strong the magnet needs to be
David Lang
They should be a match for the K&J magnet that AMS the encoder maker recommended. It’s been a long time since I looked at the datasheet but I’ll see if I can track that info down.
That being said my understanding is that magnets aren’t really like resistors or capacitors where they are ordered to an exact spec. It was more of talking with the factory to see what they had which was closest to the one in spec.
The most important thing is that they are magnetized so the north and south rotate. A lot of disk magnets are north and south up and down which obviously won’t work for us 
Thank you. That makes sense. I was asking so people in the future who are playing with these know where to start. I really enjoy how fun the Maslow is to play with from an engineering angle. The parts are simple, modular and their purpose is clear so that is really fun and a great way to design. I’ve been thinking about it and Almost all the parts could be made or rebuilt or sourced by hobbyists using their own resources. I guess the encoder boards and drive gears would be hard to source but even those could be ordered through a custom pcb service and the gears would not be hard to have machined (not cheap, but not hard) The sled and outside vertical pillars are self contained and swapable, the clamps and four support pillars as we see with the turtle clamp are user friendly. The z axis motors are a bit tricky but can be sourced or modded, The arms and belt guards are hard to print but workable.
My two questions would be:
Do people print their own spools? I haven’t tried and I don’t know how the gearing teeth would hold up to pressure.
Would it be feasable to mod a standard esp fluid ready board to replace the computer? Seems like it would work.
This wouldn’t be my personal focus for a project right now with Maslow, just thinking forward and looking to slowly document parts.
Rarely because the injection molded ones are cheap and stronger, but I have printed them for prototyping.
I found that only Prusament PC-CF filament was strong enough to hold up, but they did last a long time.
Maybe, but we use closed loop servo motors instead of stepper motors which is pretty different.
The board files are all available in the source repo so it would probably be cheaper and easier to just order the boards from JLCPCB or something like that even in small quantities
