On 25 June 2010 10:18, Alexey Starikovskiy <[email protected]> wrote:

Firstly, many thanks for looking at my module.

> You use C from previous call in case of dx=0 and dy=0.
> It could be anything and could move you anywhere.

That is intentional. "C" represents the direction that the cutting
head is moving in, and should hold its value if the axis stops.
I might add code at a later point that limits the speed that C can
swing at and enforces a cw/ccw  direction, but for the moment it seems
to work.

Mathematically the module works exactly as expected and desired in a
"sim" configuration, it is only when running on a "live" RTAI
installation that problems occur, and the problems seem to be separate
from the kinematics mathematics. No matter how wrong my calculations
are, I don't understand why stepgen.0.position-fb increases every time
the forwards kinematics function runs, even though there is no signal
connected to its input. Then, when I remove line 53 (which refers to a
C-axis which is not connected to any stepgen at all in the HAL) the
problem goes away.
I am almost tempted to add another stepgen and just not use stepgen.0,
but that would only solve one of the problems.

Something in my module appears to be writing "out of bounds" into the
memory occupied by the stepgen. Also, though I have not managed to
track down where yet, valid numerical values from my module are giving
NaN values in downstream calculations

Something I haven't tried yet is changing the order in which functions
are declared and added to threads in the HAL file (and, in fact, I
didn't write the HAL file)

-- 
atp

------------------------------------------------------------------------------
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo
_______________________________________________
Emc-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to