Hello Guest it is March 28, 2024, 12:48:11 PM

Author Topic: Plugin to load a file of points direct to DSPMC for High Speed Machining HSM  (Read 10531 times)

0 Members and 1 Guest are viewing this topic.

Offline piv

*
  •  91 91
    • View Profile
Ive been discussing trajectory planning with Art in the Tempest planner thread. The DSPMC offers the possibility of really high speed trajectories.  I am looking for a way of bypassing the Mach trajectory planner, probably by way of either a plugin or stand alone program that can load a file of points directly to the DSPMC.  Im looking for anyone else interested in HSM, fast contouring, pre planned trajectories, good chip load control and anyone with info or ability or advice to help me get this sorted out.

Offline piv

*
  •  91 91
    • View Profile
Rufi from VitalSystems DSPMC is going to help me with this.  I will let you know how it goes.
piv what happened with this? Thanks, Karl

Offline piv

*
  •  91 91
    • View Profile
Got the plugin and am in the process of writing a trajectory planner to test it out properly.
I just downloaded the Sept 29 version of Tempest from the other thread and tried it out on a 3D pocket. It was pretty much a no-go. Stopping and starting at what seemed like a million short segment curves. I had hoped for more but the benefit seems limited to 2D applications, unless I missed something.

I have a pretty bendy alu extrusion gantry router. It can fly through that pocket at 60ipm in regular Mach with MDF. I was hoping for a software fix that would allow me to let it rapid at high speed, but I get too many random jerks at the beginning and end of toolpaths to let it fly. So I override the rapid rate from 500 down to something like 100ipm, set the acceleration with jogs and then it seems to go OK.

Eventually I would like to build a steel servo-driven machine with proper slides etc, but this is a good learning machine. I am making Moth foils with it - see what you started back in 2000? Figured I could pay for a cheap router pretty easily by making a few molds.

Who knew mothing would drive me to CNC? =:-)

Anyway I have a welder and a flat granite work table so I might just draw up the next machine in Solid Works and start to weld up the major components, figuring to get it together over time. Perhaps by then you will have a plugin for the DSPMC worked out.

I have to say I like the idea of using machine physics to determine the feedrate at any point on the toolpath. But I fear I don't know enough about it to be of much help. I saw you were mentioning the 1ms interval and trying to put those points on the toolpath; obviously if it knows where it has to be 1ms in the future and knows how far it is to that point, it can accelerate the proper amount, but I need to learn more about G code modes to understand what you are up to more fully, and understand why 1ms is the desired interval.

Cool project. Thanks for sharing.

Offline piv

*
  •  91 91
    • View Profile
The 1ms interval is because that is what the DSPMC is expecting, although it could also work at 0.2ms but its not set up for that from Mach yet. Its called the "interpolation cycle".  Trajectory planning is not easy because because you have to work out the accelerations over an arbitrary path, for both speeding up and slowing down. But Ive got a way figured out that should work, its kind of half programmed now and it will also let me do RTCP for five axis machines, or any other kinematics required.

Where are you Karl?  Moths are a good challenge.

Mould making with short G code segments is a challenge. Even expensive controllers can and do have problems. At the end of the day, the things needed are a rigid machine with high natural frequencies and high stiffness and a control system that looks ahead and plans a decent trajectory that takes advantage of the machines acceleration.

I am now machining foil moulds from aluminium and steel on a machine that weighs 6 tons. I started carving wood centreboards using a table saw and hand plotted NACA foils when I was a kid 30 years ago. Apparently I am still a kid, just the foils are better.
Yes I am a late starter at 40. I am in LA. Better late than never as they say!

I just love the predicting acceleration bit. It seems like there would be some artificial intelligence way to do it; start off slow, put a six axis  MEMS sensor on it, run it around a few toolpaths, repeat. Maybe no easily accessible solution though. Might be worth a literature search.

Doing it from the toolpath seems like a lot of fun too, and more logical.

Someday I will have a "real" machine! Until then, I will confine myself to learning on something less expensive.

Offline piv

*
  •  91 91
    • View Profile
My interest is in high end high performance machining with high stiffness and high acceleration. Interestingly the trajectory planning requirements are similar for high end and low end machines but for different reasons. On high end machines the machining time and accuracy of complex surfaces is most influenced by the available acceleration. On low end machines the dynamics and stiffness are low and you need gentle acceleration and smooth motion so that you dont excite vibration in the structure, even at low feed rates.

I want to get to trajectory planning that will consider all of the limitation of the machine, not just the acceleration of the programmed axes. With five axis machines the acceleration and velocity of the physical axes (after transformation) is also important.

Rhino is a good starting point.  Have a look at MadCam.  Rhino has a good API, so you could get closely spaced trajectory points for both the tool tip and the physical axes (for machines with say 4 or 5 axes) directly from the surfaces and taking into account the local derivatives.  Its a good approach, but I need a solution that will work with other existing CAD-CAM packages.
I am actually using MadCAM. It works pretty well for what I do.

I have been doing a fair bit of reading scholarly articles on NURBS machining. Mainly this is driven by the trajectory planning issue - lots of papers are being written about it. Clearly defining the path as a NURBS curve lends itself more to good trajectory planning than defining the path as a bunch of connected segments.

I emailed Joakim and he is considering adding NURBS toolpath output to the next version. This would be a huge step forward, because none of the other low-end CAM programs can generate that code, which seems to be limiting the developers.

Of course there don't seem to be many controllers that will accept the NURBS gcode at the moment, but that is changing. EMC has a biarc interpolation method in 2D now apparently, and some other guys claim to have it working in 3D. But I haven't seen any video, other than a short one by Martin Duzi in 5 axis EMC, and he is using his own HAL interpolator to generate the toolpath from a regular line/arc toolpath.

In the interim, I think I need to mess around some more with the CV settings and the Tempest planner. When doing concentric surfacing paths, for instance, the jog from the end of one lap out to the start of the next one is too steep, which sets up resonance. There has to be a way to slow that movement down without slowing the rest of the toolpath.