Another Jerk (and more!) controlled controller

[Mooselake]

Printrbot now has a TinyG2 (ARM based) printer controller board that replaces their AVR Printrboard.  They're shipping it in their $1000 entry level new Simple.  While I think they're pricing themselves out of the market, Moose have been wrong before, and it has the potential to become a widely distributed product.  Since it's "open source hardware" it might even make it into the far east manufacturing megalith and make higher order acceleration available from the usual sources in high volumes.

They aren't selling the controller separately yet so there's no pricing, but it could be interesting.  If it's close to the current $69 pricing I might even go for it.

Separately, Smoothieware has 7th order acceleration under development, and a pre-order Smoothie based controller has just arrived  in the Moosecave (a Coherent3D Mini).  It drops into the cheap K40 laser using the same connectors and pinouts as the factory shipped controllers.  It's waiting on a few parts from the aforementioned megalith sources, stepper drivers and an (optional) graphics LCD display.  Now I need to buy another laser (not to mention Mooseshop expansion) to compare Smoothie with Auggie; the no-snow work outside season is rapidly coming to a belated end with 3" planned for today, and there should be more time for the interesting stuff...

Kirk

[ArtF]

Kirk:

Cool, I was curious about those. Let me know how it works out. Hardware is fast becoming amazing..

Art

[Mooselake]

Cool, I was curious about those. Let me know how it works out. Hardware is fast becoming amazing..

Me, too :)  It's a purchased hand assembled pre-order, cost around a hundred bucks with cables and glcd adapter.  Iirc the target production price is around $70 (just like an AVR based Printrboard these days).

Hardware has been amazing since a 360/44 cost under a million dollars...

Just got an 8 port gigabyte ethernet smart switch for $25.  I can remember when a thousand dollars a port was cheap for ethernet, and I passed on pulling network wire in the mooselake manor after stripping the lath and plaster, since it was just too expensive to afford.  Think that was still in the coax days.  Whoops.

Kirk

[chicoxiba]

Hi:

  Interesting paper. Its basically Auggies planner again in almost every way, the single exception being the blend formulas,
in that I use a hermite curve rather than a bezier. The two share the most important qualities though. Im quite happy
with Auggies motion generally, though I have tweaked a few things recently in acceleration computations. I will be doing
more work on it this winter as I burn more with my laser and I'll see what improvements I can make. Auggie has few users for
motion, of course lasers via windows isnt a large field of people, and cnc is now a very different animal than when I wrote
Mach3, so I dont suspect Auggie's motion controller will ever grow to a large enough community to expand it too much. I
do have a galvo scanner Id like to add to it soon though so you never know what capabilities will show up in there. I add
anything I write to the release of course, so anything could show up. If you find motion bad in any particular way always let me
know so I keep it back of mind for when I open that code next.
 I know your troubles are usually surfboard related if I recall right, so speed at small segment is  the type of trouble you
hit.  Thats a very typical result on that type of work, I think a more bezier motion may help , some sort of nurbs capability to get rid
of all the small line segments.. or elliptical motion capability rather than arc.  Anyway, keep sending such papers,I do like to
keep up ono current thinking..

Thx
Art

Yes Art. I'm in the need for a bit more speed and smoothness. At the moment LCNC is leading the charge of PC based motion quality i think.... Tried many pc based.. most rubbish for 3D freeform G1.. Tried Mach3 tempest (the Sep29 last upload to the forum) too.. there something a little funny.. It is very interesting and almost there it just stutters in a smooth manner at decel and its enough to get things into a swingy vibration...  Its like a Wowowowowow just before X reversing direction.. It moves quick.. speed is not a real big issue with it.. If I set the Jerk high it moves its arse quite well. Bloody chinese pendants are good but are not as good Look and Feel as a PC so I don't want them. Any chance you want to develop Mach3 a bit more? I wouldn't mind using tempest in Mach3 but Need the ESS and Pause. LOL

regards,
F

[ArtF]

lol..never happen.

Art

[chicoxiba]

Thats ok :'( ;D

Thanks,
F

[chicoxiba]

Dear Art, would it be possible to set jerk limitation individual per axis within Auggie's Trajectory planner?

thanks,
F

[ArtF]

F:

Ill look into it, but Im not sure that doable mathematically. While it would be possible for something like Jogging,
Jerk when moving multiple axis is a measure of the accellerative energy of the entire motion. I know this can be hard to tune and will be working on it here to see what improvments I can do. I use a jerk that ranges from 50,000 to 500,000
while I know others use as low as 5000. If I load a vine for example, I can scream through it at a jerk of 500,000 but its slow going on point to point moves if its too much lowr. On a raster scan burnign 3d photo's I use lower at 50,000 to
more greatly smooth out the raster start and end points more as my table can vibrate.
  If I set it too high the autoslow for softlimits kicks in too early for me.. So there is work to do, but too few users of Auggie's motion control to dedicate too much time to that desire. I do continue to work on it though as I have one table completely Auggie driven where I use multiple types of speed strategies.

  As an experimental planner its doing far better than I had hoped, but I know its still a bit limited in some types of work unless jerk is pretty high. I will however, look into a per axis jerk to help when single axis are way off other axis in terms of the jerk they can handle.. Perhaps a dual system where the systemic jerk is respected as it is now, with a further planning per axis jerk for the initial feedrate planning. That would allow a Z for example to self limit itself when it
is involved in a motion..

Thx
Art

[ArtF]

F:

In thinking about this in the background, it occurs to me that in Auggie, the jerk is overridden by acceleration.
So my advice would be to set all accelerations very very high, tune jerk for a system that is overall smooth and fast
without regard to any axis that would need to be limited. Then, lower the accel only of the axis your worried about.
This will cause a jerk override on moves with that axis, while leaving the others at as high a jerk as is acceptable.
  Accel will limit jerk and jerk will limit accel, so it should be safe to set very high accel's while tuning jerk.

Art
 

[chicoxiba]

Hi Art

Thanks for ever considering all  these things.

I asked if it was possible after reading about a certain Siemens cnc controller where you have the ability to set jerk limit to each axis.. Then I thought it might bring something interesting to the table here.

Perhaps you could benefit from having this trajectory planner be used in Mach4? I would then be a very active tester as it would make my life a lot easier not having to deal with pokeys... I looked and looked at the  2 pokeys options and got unanimated.

Anyways.. just thoughts.

Thanks
F