Velocity change by HAL possible?

Hello...
I try to do laser scanner and like to have all program in one file. I just got an idea to somehow tweak the acceleration in X scanning. (laser firing with M62/M63)

For example sequence like this:
G0X0
M62P1 G1X40
M63P1 G1X80
M62P1 G1X120
M63P1 G1X121 (this move too short and traj. planer will decelerate)
M62P1 G1X160

Will slow down at X121 move as segment is shorter than acceleration of machine.
But the machine doesn't need to accelerata/decelerate at all as it's straight move.
So I got the idea to do something like this:

G0X0
G1X5 (the actual acceleration as motor was reversing)
M62P3 (override to fast acceleration)
G1X10 (long enought move to avoid decceleration
Rastering program here
...
...
M63P3 (kill fast acceleration)
G1 X(lastX+5) (decelerate ramp)

So actualy when machine got speed I want to set acceleration for this axix to very big value and further M62/63 will make ramp short enought to stay on speed. Just before line end I will restore correct acceleration and make one more move long enought to safely stop the axis.

The only problem I have that I didn't find the hal pin to change the acceleration in live.

Any idea?

Hi

Have a read through this thread, we established why you cannot set acceleration on the fly, when emcPT wanted to do something similar, for different reasons

www.linuxcnc.org/index.php/english/forum...-functionality#47283

regards

I do check with ini.* and stepgen.* and setp the new walue. Parameter are changed, but real thing on machine doesn't. So values are queued somewhere.

the lookahead in only one line now. But I read somewhere than new TP in developed to have wider lookahead..

...but there must be some nice solution....

You need to read the thread, they are stored somewhere and are only re-read when reloaded, so changing them makes no difference.

If you have a lot of short segments, the new TP might be better, you will have to try it.

regards

ArcEye wrote:

You need to read the thread, they are stored somewhere and are only re-read when reloaded, so changing them makes no difference.
regards

I do read... And now I know why my test fail (I do that before I post this thread)

ArcEye wrote:

If you have a lot of short segments, the new TP might be better, you will have to try it.
regards

Actulay there are short segments. It may be lot of them of near nill. Raster scanning make segment every time I need to chanhe power of laser.

I didn't try the new TP yet.