What do you really get for a $350 desk top engraver ?

All in all i was happy with what the machine could do. looking at a picture of a small tool and holding one in your hands is two different things  

one thing i did notice with this machine is that surface finish drops on the more complex cutting moves on say a angle cut or circular cut. i have been trying to pinpoint the cause of this but i am still looking, it's not horrible but it could be better. my first thought was machine resolution or belt tension, i tried a finer resolution which might have helped a tiny bit and did some tweaking on the X axis to get the belt tighter to try that today to see if that is the problem?
at first i thought it might have been a rigidity related issue but after playing around with finished passes quite a bit i ruled that out after noticing the straight cuts were really good. i would think a tandem axis is going to be more rigid with 2 bels and motors opposed to the X axis with a single belt and motor. i think tests have proved this for me. i can actually get a .06 stepover around ,5 deep with the side of the cutter on the Y axis but get into problems if i get that aggressive with a cut on the X axis.

one other observation is the closed loop steppers being a huge benefit to having something work well in the end. i would also say over sized motors are good way to go too. food for thought Tommy if you build one   

if i were to do it over again i would have doubled the size of the motor on the X axis in relationship to the size of the Y motors. i skimped on the power supplies and should have spent a little more for 48V rather than 36V too. wider belts might not have been a bad idea as well as long as sharp bends don't kill the life as a result.

this was a R & D type project for me and i had no idea what to expect but i'm happy to pass along my finings. the machine does well for what it is and i am just nit picking on little stuff. But regardless i tiny machine like this could be a time saver for you i think.

So i have been looking at a lot of belt drive idea's after doing this build. i have been playing around with a open ended reduction idea's and think i have something that would work and get a lot more from a smaller width belt. the down side to belts when you have some forces at play are what i believe to be minor flexing of the belt engagement into the drive pulley, belts just are not rigid like a ball screw in this respect so you either have to go to a wider belt or figure out a way to get more surface contact with the smaller belt.

Tommy i know how you do your reductions and i have looked at your designs quite a bit. The omega configuration on the open-ended side is a good way to get a few extra teeth engaged and is the basic configuration on my build minus the reduction. what i have built works mostly because the motors are oversized for the size of the machine. Keeping the design simple was a good base line to see what works and spot the tiny issues. for me it's surface finish between step downs
the idea i want to test is something like a serpentine belt reduction for open ended belts that would do away with the closed loop belt reduction on the drive shaft of the omega configuration that i see you using. my thoughts with a serpentine configuration is you end up with a lot more surface area in contact with the toothed pulley's which i would think would translate to a stiffer drive system.
I did some rough sketches to see what kind of footprint it would have and it seemed pretty compact  doing a 3 to 1 reduction with 3mm pitch belt. roughly 4.5 inched by 7.5 inches.

that kind of system solves the surface area contact so you get the most out of a belt, but it does not look to be a reduction from what i could tell. The idea i have could be configured to work the same way though.

when you get into the servo side of things the encoders are a lot better than what you find on these cheap closed loop steppers and the reduction is less of a big deal as long as the motor can do it[s job.

for what i built i have far more potential speed than space to ramp up to my top end speed so a 3 to 1 reduction would land me in the 4,000 micro step range with a .0002 resolution. one other thing would be the extra torque on the X axis which is powered by a single motor. the tandem axis has a lot more grunt with 2 motors.

i have seen some rotation reductions with ring gears and arc tooth belts that looked interesting. i was looking at size requirements and it is pretty compact and low profile but it's only around a 6 to 1 reduction. my thought would be plus 2 axis configurations or robotic arms having a use for something like this. The problem i see is on the programming side with CAM is the software being so expensive beyond the 3-axis hobby use Cam i use at home.

The commercial 5 axis CAM on average is probably somewhere in the ballpark of 10K per year which is out of reach for most on a hobby level.

machinedude wrote:

i have seen some rotation reductions with ring gears and arc tooth belts that looked interesting. i was looking at size requirements and it is pretty compact and low profile but it's only around a 6 to 1 reduction.

You could consider planetary gearboxes. These are not expensive out of China,
We wanted to use a 5:1 gearbox on a retrofit recently but we had to keep the original mounting geometry to a NEMA23 motor.
WUMA gears did a custom version with an 8mm dia shaft (usually 14mm) and a custoom shaft length. They charged an extra $10 for that but the total was still only USD $77 per gearbox. We bought 3,