On Tue, 1 Apr 2008 23:42:59 +0300
"Alex Joni" <[EMAIL PROTECTED]> wrote:


> looks like an interesting design. If it would be to work semi-independently, 
> then you'll have to name them : X, Y&V, Z & W.

 ack.
 
> >  The X axis might in fact have two motors. I did a research in the
> > archive and it seems this gantry axis setup is supported.
> 
> gantry axis setup can be done, there is a bit of code missing in order to be 
> able to home it without certain hacks.
> If you want we can go into further details/specifics, but I don't see a big 
> issue in this direction..

 I'd appreciate it. I'll ask you more details on this if this project
 will go ahead.


> >  Y1 and Y2 will share the same rail but have independent motors
> > and limit switches.
> 
> call them Y and V, different joints, etc. Again .. there might be parts of 
> emc2 that need some adjustment, but nothing majour (mostly because there 
> aren't that many machines in use with this configuration).

 ok.

 
> steppers and encoders have been proven as functional with emc2, as long as 
> you don't expect to correct position based on the encoder. (in other words: 
> the encoders are mainly used to see when stepping errors appear, and 
> everything is halted).

 mmm.. I believe it's expected to be able to correct, but i'll ask.
 what's the reason behind this?

 
> >  I should be able to influence the development a bit, so if
> > there's anything that would make it easier to use it
> > with EMC2 I would love to hear it.
> 
> hmm, the things so far are a bit vague to be able to tell..

 Please ask any detail that I've skipped overy in my description,
 I'll be more than happy to ask what I don't know and report back.


> I think there are quite a number of people working with emc2 and 5i20, that 
> doesn't mean other cards aren't equally well supported. What is not working 
> at the moment is step generators embedded in the 5i20 FPGA, although that 
> might probably come one of these days.

 I need to choice the best card for this project so I will evaluate each
 driver. The problem lies in the driver or in the fpga? I've a bit of
 experience on both sides and might be able to help

 
> >  I'm also planning to write the code to use an USB jog/shuttle
> > device for manual axis control, if it's not already supported.
> 
> That's not necessarely needed, quite a few USB (and non-USB) devices are 
> supported by hal_input, and thus are beeing used to jog emc2 machines.

 I've checked both the sources of hal_input, the linux hid code and a
 program that I know uses that controller.. it seems that it is
 already supported. 


> >  I would also have to do most of the development without the real
> > machine, would the "axis" software be able to simulate it?
> 
> Yes and no. AXIS by itself is just a GUI. You would have to set up a machine 
> simulation, and that can be done by using HAL (lots of simulation 
> components) and vismach (a special tool to visualise/simulate a machine). 
> Look at the sample configs for scara and puma to see how that is done for 
> robots. I think a simulation like this exists for some gantry type machines.

 ok, I'll do my homework. Will start with a simpler machine and then
 progress toward the final one.
 
 thanks a lot for your help.

-- 

 Best regards,

 Alessandro Zummo,
  Tower Technologies - Torino, Italy

  http://www.towertech.it


-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to