On Saturday 21 March 2020 20:49:21 Leonardo Marsaglia wrote:
> I'm just guessing 200 rpm because of the speed I can develop on the
> Mazak now wich is 10 meter/min. I would love to have some more speed
> available but that will come with testing.
>
> What I intend to do with the grinder is something like this:
> https://www.youtube.com/watch?v=UGgR1GFOFhU
>
> This guy did it with a chinese control and used the spindle as a C
> axis. Is another way of doing it but I really preffer the spindle
> tracking method. I already have a cylindrical grinder that's waiting
> to be retrofitted with LCNC. In fact, right after the Mazak I plan to
> start working on that machine.
>
> I don't see it as a problem to compensate for the wheel radius because
> as you know HAL can handle almost anything. The only real problem I
> see is how to get a good measure of the grinding wheel as you dress
> it. The most primitive idea I have is to measure the wheel before
> placing it into the machine and then keep track of its diameter as it
> gets dressed. But sometimes we have to adjust the offset of the
> dressing tool because a diamond just detaches from the tool and then
> you need to correct for that difference. Although little variations in
> wheel radius are not that significant in such diameters (600 mm more
> or less) I would like to have a way to measure the wheel without
> stopping it. By the way these machines don't use the small CBN wheels
> because they are anything but new.
>
> The first thing I have in mind for measuring the wheel are two laser
> pointers one above the other so when the wheel "touches" the first
> beam you have the first point of the circumference and then the wheel
> continues moving until it "touches" the second beam and now you can
> calculate the lenght of the circular sector and hence the radius. The
> problem is, this is just theory and having such a system working with
> the accuracy needed escapes from my actual knowledge and experience in
> optics which is very poor. I attached a picture so it's more clear
> what I came up with in my head.
>
> So I guess the measure the wheel and keep track of the diameter will
> be my first method to compensate for the radius.
TL, read anyway.
You only need one beam. I would use the first beam interruption as a
second home switch of sorts, setting that with the home_offset when you
install a new wheel. The established home offset then becomes your new
wheel reference. This should then be considered a fixed reference and a
suitable distance from the work to prevent accidental contact during
setup.
Next write an xml file to draw a couple pyvcp buttons, something like
this perhaps:
!-- and stuff to control mist motor speed -->
<labelframe text= "mist motor speed">
<vbox>
<hbox>
<spinbox>
<width>("5")</width>
<halpin>"mist-on-adj"</halpin>
<min_>.000</min_>
<max_>.025</max_>
<initval>.005</initval>
<resolution>0.001</resolution>
<format>".3f"</format>
<font>("Helvitica",12)</font>
<param_pin>0</param_pin>
</spinbox>
<spinbox>
<width>("5")</width>
<halpin>"mist-off-adj"</halpin>
<min_>.000</min_>
<max_>.250</max_>
<initval>.100</initval>
<resolution>0.001</resolution>
<format>".3f"</format>
<font>("Helvitica",12)</font>
<param_pin>0</param_pin>
</spinbox>
</hbox>
</vbox>
Both of the spinboxes that makes have halpin outputs of their displays
but that mess, which I'm using to twiddle the on and offtimes of a timer
driving a peristaltic coolant pump to spray wet the tool in my 6040
mill, but here it could be reduced to one spinbutton by making the
second one a tach bar to display the outout of the laser's rx cell, and
the first one is used to twiddle a gcode var which becomes your wheel
wear amount, displayed by the spinbox, and ship it back across motion
with one of the m6x analogue channels. Subtract that from the radius of
the new wheel before doing the up/down calcs. From there on, your gcode
carves the cam with automatic wear comp. If wheel wear is so fast your
need to comp it mid-lobe, you can write a suitable pause and do it in
the middle of a lobe. You could even use a charge-pump output to drive
the adjustment in that mid-lobe correction making it an automatic null
and a lights out operation you wouldn't have to hover over. Don't forget
to turn the coolant off (and back on when done) since it would interfer
with the laser if running. And don't forget to refresh that variable in
your main code loop so if its been corrected in the middle of carving a
lobe, the new value is used by the rest of the calcs your code will do.
I'm obviously better at dreaming this stuff up than in writing a cogent
explanation but I hope this will help.
And obviously Leonardo, make a youtube video and tell us about it when
its working. :)
>
> El sáb., 21 mar. 2020 a las 20:35, Gene Heskett
> (<[email protected]>)
>
> escribió:
> > On Saturday 21 March 2020 16:11:44 andy pugh wrote:
> > > On Sat, 21 Mar 2020 at 17:44, Leonardo Marsaglia
> >
> > <[email protected]> wrote:
> > > > I intend to turn automotive camshafts, that is with a minimum of
> > > > 180º of base circle (sometimes called heel I think), and a
> > > > maximum lift of about 8 mm
> > >
> > > In that case the high resolution encoder might be good. I doubt
> > > that you will me machining camshafts at 20,000 rpm.
> > > (Especially not if the cutter needs to dive in and out)
> > >
> > > For grinding you need to consider how the contact point rolls
> > > above and below the centre line. I did do the maths, when I did
> > > the crank-grinding mock-up. But only for a circular but eccentric
> > > result.
> >
> > Another thought or 2 here.
> >
> > The weight of the grinder you'll have sitting on the crossfeed will
> > effect your usable spindle rpms by slowing down the maximum
> > attainable accel's. My nema 24, x on the sheldon, geared 2/1 has a
> > 3o ipm max limit, so that motor will need a serious boost in top
> > speed to even try doing that at 200 rpm.
> >
> > If room for its diameter, the use of a flat faced CBN wheel would
> > eliminate some of the math by eliminating the calculation of the
> > wheels up and down due to the curvature of the wheel face but I
> > think whats available is way too big for this. The use of CBN should
> > however lengthen the cutting life of the wheel compared to the alox
> > versions, particularly if running wet as I have seen it done in some
> > of the you tube videos.
> >
> > Cheers, Gene Heskett
> > --
> > "There are four boxes to be used in defense of liberty:
> > soap, ballot, jury, and ammo. Please use in that order."
> > -Ed Howdershelt (Author)
> > If we desire respect for the law, we must first make the law
> > respectable. - Louis D. Brandeis
> > Genes Web page <http://geneslinuxbox.net:6309/gene>
> >
> >
> > _______________________________________________
> > Emc-users mailing list
> > [email protected]
> > https://lists.sourceforge.net/lists/listinfo/emc-users
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
If we desire respect for the law, we must first make the law respectable.
- Louis D. Brandeis
Genes Web page <http://geneslinuxbox.net:6309/gene>
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users
