Dual bipod modified for plotter

I got the dual bipod hotwire working correctly.
Now I want to turn it into sort of a pen plotter configuration...
here is what I have done;
re-routed 4 wires to all be in the horizontal plane
they are all joined together holding a pen in center of machine frame.

I thought it would be a simple matter of inverting the right pins in .ini file to change motor rotation.
But no matter what combination of invert/not... I always get at least one of XYUV moves going wrong direction.
can not get all 4 wires in sync.

My idea is to have pen supported only by wires, not needing any mechanical slides etc.
If I can get this small proof of concept to work, then it should scale up easily to a VERY large "plotter"
Imagine a large framework rolled onto football field with spray paint can supported by wires painting large graphics

Any ideas?
Thanks
~*Brian

The rotation direction of individual motors are changed in the hal file, like this:But to change the direction of UV or XY coordinates is different. I assume the only way is a modified kinematic.
Kep in mind, the ini file parameters in the AXIS_N sections are referring to joints in reality, not to axis.

If I understand your geometry correctly, you have to invert the direction of the V axis.
The same effect you can achieve at g-code. Just move V opposit to Y.

Could be also what you want is better to implement with a tripod. I think there was an example somewhere.

Gerd

Gerd,
you are right, the V axis needs to be inverted, and probably must be done with modified kinematics...that is beyond me unless someone would be willing to guide my editing of kinematics?

modified g-code would be rather tedious to do on every program.

The reason I am trying 4 motors instead of 3 (tripod) is it should have greater working area, yes?
I think 4 motors would be more accurate and 'balanced' (faster) than 3 ? ... I imagine motors pulling against each other creating a balanced condition that would go faster and change directions easier?

One of my main goals of this machine configuration is the ability to "raster" engrave in any direction just as easily as straight X or Y moves; angles, circles, squares, spirals,etc. ( no mechanical constraints of traditional slides & lead screws)
Acceleration-Deceleration-stop-change direction of the weight of tool head and linear rails seems to limit raster engraving speed and smoothness.

~*Brian

bvandiepenbos wrote:

Gerd,
you are right, the V axis needs to be inverted, and probably must be done with modified kinematics...that is beyond me unless someone would be willing to guide my editing of kinematics?

modified g-code would be rather tedious to do on every program.

hi Brian,
the kinematic code is not very well documented, but inverting one axis is not a very big job. I published recently a new document here: code.google.com/p/emc2hotwinch/downloads/list, maybe it can help to understand the math (in german language :( )

I would consider modified g-code too (at least for a try). How do you create your g-code ? Maybe you have to do some postprocessing anyway ?

Gerd

Hello Gerd, thanks for the info. I will study your docs and see if I can understand how to change kinematics.

I have not really decided on what to use for g-code generation. Have tried a few options, the most promising one is plugin for inkscape.
I use Corel Draw at my 'real' day job and know it really really well, inkscape is very much like Corel.

I have Bobcad for my CNC Bridgeport mill at home... but I hate it, very crappy program I think. Besides it is PC based.
Would prefer everything in Linux.

Post processing might be a temporary work around. good idea.

~*Brian