Those GW 405-31ZY were some of the very first motors we had located and peaked my excitement. Oh so long ago…obviously, they are not usable for the current driver board with stall currents so high, but at this point with Maslow being build with much bigger boards that can take the amperage, the potential field is very open.
dlang,
It is really cool that the development has taken many directions but remained tight in the main priorities of trying to gain accuracy and precision. Just wanted to personally thank you because your posts have helped me tremendously in being involved, even if not enough time to be involved as much as I would like.
It would be nice to have this in the wiki.
Maybe with a table of options.
One line stating the ideal Motor voltage / gear ratio / PPR / sprocket size for ‘stock maslow’
And then going up and down for more power / less power
Somehow this would need a multi dimensional table.
For those people who would like to build a bigger or smaller size machine this could help to get their heads around all the variables by visualising the differences somehow.
A downsized maslow with a compact router (500W) would need less powerful motors and if that machine would be used to cut 9mm plywood only then it could cut in just one pass at a higher feedrate perhaps?
I’m not sure if my train of toughts is correct, For me it then would make sense to start building a small lightweight and portable machine, With a slightly cheaper chain (bicycle chain perhaps?) etc.
Or a very large machine with a compact router?
Ultimately somekind of spreadsheet would be cool. Where one can input the total weight on the effector and the sheet then spits out a bunch of options that sow all motor options that have enough power for that kind of effector.
I realise this needs manpower to make such a sheet.
So at least my brainwave is out in the ‘free’ world
In a way it would be great if the motor manufacturers would offer such a spreadsheet, but i guess that china isn’t there yet. (hint to all the chinese motor manufacturers that are reading along)
Iceblu,
What I really would like to see is your board (driver) you are using. Eventually, I will build a higher amp version and like to see what people are using that can use the signal supplied by the maslow, but handle higher amperage.
Sort of funny: when waiting for my tiny motors to show up I messed around with two wheelchair motors. Could not think up a good use case for a 400 lb sled lol. Actually the gear ratio is not high enough to allow a powerless stop like in these motors we are using, so the sled would keep moving until it met the floor.
I have tons of high volt/amp controllers and lots with same ic setup as maslow board, but what makes a good middle ground driver? I know I could run parallel and increase the max amperage of the current maslow design, but never witnessed that working to well.
Funny to see these tiny motors (ones I ordered that matched pretty well the basic maslow but are literally 3 inches long-as I just wanted to get a working prototype and could size it based on motors). The crazy part is that I built a frame from old metal bedframe + wood that has a 4x4 working area with arms that you can hang on and they don’t flex (much). The sled is not small as originally intended because the router I needed to use is a Triton with a massive footprint. It has a lift built in so z axis is easy. All that and these 16 kg-cm motors have no problem reaching any of the work surface. They are slow though (13 rpm with load). Safe to say I was blown away that they even worked.
We know the motors that Bar picked for the stock maslow, we know how much power
they have and their speed
We don’t know what is actually required for the machine. We haven’t had anyone
experimenting with lots of different options (even the people building them from
scratch aren’t testing lots of options, just the option they are trying)
so we have zero comparisons of identical machines with different motors (we
don’t even have any with different gears yet)
we do know from simple math that how high the sled can go in the center is
limited by the amount of tension that the motors can put on the chain.
you can look at the http://lang.hm/maslow/v-plotter.py and experiment with
different dimensions and tension settings to see an approximation of what the
machine should be able to do.
These drivers are readily available all over the net. Half the circuity is just to make sure its L298 logic compliant. If Maslow spun their own PCB with an atmega2560 and three high current mosfet H-bridges it would be way cheaper. You could build a board for the same price as the Arduino Mega provided hands down.
My first Maslow attempt was before the original kickstarter was finished. I used some ford power window motors and wire rope. I didn’t have cash at the time to buy in but wanted to try out the idea.
I am going to see if the 24V 1:212 motors are going to be in stock soon. They are the higher torque version at 50kgcm and 29RPM under load.
I’m starting to get ever more curious, what motors seem to be the most ideal at this point in time? And why? If the price is in the same range and they don’t put too much load on the L298 I might want to give them a try. (If i can find a Shenzen retailer that feels right)
I’m wondering if i can find these motors in europe somewhere, If i can find a machine parts retailer who has some of these motors on stock i’m sure i can find a way ‘to try b4 buy’ different types.
My budget is kinda small so I will need to make educated choices. When ordering from china even more since it’s impossible to change them for others.
The critical thing is to use a gear on the motor that cannot be back-driven by the weight of the sled, if that can turn the motor, it won’t retain calibration when it’s turned off.
worm gears work this way, but there are other gear sets that have this property.
beyond that, you have speed, torque, and power requirements (which just have to match the driver board you are using)
The motors provided as part of the kit are small and (in quantity) cheap, they have a good amount of torque, and can spin up to 30 rpm (which is rather slow)
more torque doesn’t really help much (you can go a little higher, but that’s not much difference), although you can trade torque for speed with larger sprockets.
right now the maslow is artifically limited to about 30 ipm out of the 48 ipm it’s theoretically able to do. This is because we don’t consider acceleration in our movement. Raising this limit or going to faster movement (either through faster motors or larger sprockets) is going to run into problems as a result, but this is ‘just’ a software fix, so if you are selecting motors, faster is better
if you get motors that are weaker, you may need to slow the feed rate down, worst case you need to raise the motors to be able to cut at the top of your work area.
Yes, raising the motors to access more of the work surface is the primary reason these little motors I have are working so well. Like I have said, I must build in 10-15 min at a time, which is frustrating, but not enough to get a divorce lol. But, because I had to hack the thing together, I wound up doing a lot of design by trial, which is why I wouud be embarrassed to put up photos. I like to take pride in my project builds but this one looks like a hack lol.
Once I clear out the space in the back shop I will be able to set it up permanently and really play with it. Currently, even though modular, I have to set up and calibrate and then very little time to experiment.
That 24v 1:212 sounds great, especially because your driver can handle the stall current.
Dlang,
Good point about not knowing the ideal motor for the maslow in its current form.
I do need to work out some attachments to run triangular setup but my router is huge, diameter wise. It is a Triton that has a lift built in sort of. Made it easy to attach a motor for the z.
I have been trying to catch up on the sled with triangular attachment thread but not really done with it yet. Anybody know offhand what a practical max leghth the bars can be? I need to really see how big the diameter in router plus stand.
But, because I had to hack the thing together, I wound up doing
a lot of design by trial, which is why I wouud be embarrassed to put up
photos. I like to take pride in my project builds but this one looks like a
hack lol.
post pictures, people learn from hacks and mistakes as well as successes.
Dcoumenting what you tried and had to re-do may help someone else avoid the
problem.
Dlang,
Good point about not knowing the ideal motor for the maslow in its current form.
I do need to work out some attachments to run triangular setup but my router
is huge, diameter wise. It is a Triton that has a lift built in sort of. Made
it easy to attach a motor for the z.
I have been trying to catch up on the sled with triangular attachment thread
but not really done with it yet. Anybody know offhand what a practical max
leghth the bars can be? I need to really see how big the diameter in router
plus stand.
max length of the bars is when they start hitting things in the top corners
how wide is your router? I’m making a larger top set for Gero, I can make
multiple copies of it.
Dear Sir / Madam,
Please give me more details about the encoder.
I need to know the ->Pulse per rotation (PPR)<- of the encoder.
Is the encoder the same on all motors?
The answer is:
Dear
The encoder has 16 pulses
Different motor installation of the encoder is different, so the pulse is different
Do you have other questions ?
Best regards!
Miss Merry
Hey David,
Sorry took so long to post as I have been out of town. I may take you up on that really nice offer. I will have a chance to check exact measurements this weekend and will send to get your input. Plus, once I move it inside I will take some pics. I live in Florida so it has been safe to keep outside (I remove all the hardware -in less than five mins) if I get my area of “stuff” organized today or tomorrow I can move it in and get serious about playing with it.
Thanks,
Todd
We started our build as part of our middle/high school robotics program and tested lots of motor options that we had and could order. Also spent lots of time calibrating some work gear boxes we built. The time to invest was limited when compared to other projects. It has been a couple months since we have tried to work on the project. Woke up from a dream and remembered back in the day that FIRST supplied window motors that had a built in worm gear that you couldn’t back drive and amazing torque. Looked to see what the latest was and appears an option exists for car seat motors and has a built in hall effect sensor that does 172 pulses per shaft rotation. Found one of the older motors for window and rotation speed was a little fast but seems reasonable.
Provided a spec sheet for motors that are donated by bosch to FIRST robotics teams. Wanted to open up a different path to the group to see if someone can find an ideal window or seat motor with built in encoder.
Those motors have less gearing than what the kit provides, so they draw more
power, but as long as your motor controller has enough capacity, that shouldn’t
be a problem
They are at least in the ballpark of the stock motors.
172 pulses per output shaft rotation won’t work, you will need to add an encoder
(we’ve posted a few options here in the past), something around 1000p/r (which
translates into 4000 steps/rev) should work for you.
Looks like this seller states the PPR of the motor shaft. 16 pulses per one rotation of the motor shaft
I also have seen sellers who state PPR of the output shaft. Stuff like that make orient express shopping a little more exiting.
So the PPR of the output shaft will be 16 x (whatever gear ratio you pick).
The on-board encoder in that Bosch unit doesn’t provide quadrature signals, so you can’t decode direction. The resolution is stated at one full cycle per armature rotation, that would be the motor shaft.
the maslow needs to have a quadrangle signal, two sensors 90 degrees out of phase with each other.
This motor has a single pulse per revolution, and only one sensor That means you can’t tell which way things are turning, which means that when you are moving slowly near the edge of a pulse, you can get what look like many pulses and don’t know if the thing is spinning wildly or bouncing back and forth across a boundry.
but that motor with a separate encoder (and supports for the output shaft) should work well.
The motor that Gero found has a quadrature signal, but the connector is wired differently from the one in the Maslow kit - the Vcc (3) and GND (4) on the Maslow are GND (3) and Vcc (4) on the one in the link. Not a show stopper, but should be noted.
That’s easy enough to modify those connectors, a toothpick pushed against the conductor ‘barb’ will allow the wires to be pulled out and switched around.
(hint to all the chinese motor manufacturers that are reading along)
