Re: [Emc-users] triggering the mesa card encoder index pulse

[gene heskett]

On 11/3/22 20:33, andy pugh wrote:

On Thu, 3 Nov 2022 at 23:54, gene heskett  wrote:

That is what I am using but it is several years old, Is there a newer

one ddg can't find?


It's built in to LinuxCNC.

I just added this to my INI file [DISPLAY] section:

EMBED_TAB_NAME= cam_align
EMBED_TAB_COMMAND= qtvcp -d -c cam_align -x {XID} cam_align
And this is the same, a show stopper but the error msg is more specific, 
so It might be progress.

It now errors and quits with:

camera-postgui.hal:1: Pin 'mm_xacc.max' does not exist
Shutting down and cleaning up LinuxCNC...
Running HAL shutdown script

This is the 4th variation of the EMBED_TAB_COMMAND I've found tonight, 
none of which worked.
camera-postgui.hal has as line 1, the first of about 25 lines, all of 
which will be showstopper errors

because the 3rd argument of each "net" line in it, does not exist.

And all the missing mm_0nets there are sourced in check_constraints.hal, 
which is not included in any

postgui.hal reference. If its supposed to be, document it!

Andy, or whoever might be the expert, there ought to be ONE install.txt 
in that zip, describing once
only how to install this so it Just Works (with a compatible camera of 
course) but there's no such
complete help file that I've found. And I never did find a hit that 
would give me a fresh copy of this

5 or 6 year old cam_align.

And you can't just fix a typo and retry it, because that failure leaves 
all of qtvcp running
which ties up the video ports it needs until you call up htop, find it 
and kill it.


The camera works well, with cheese or v4l2. Why does it not work with 
linuxcnc?


And the LinuxCNC config running in a VM on my Mac started displaying the
webcam image from the built-in Mac camera. (which I am rather surprised
worked)




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, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] triggering the mesa card encoder index pulse

[andy pugh]

On Thu, 3 Nov 2022 at 23:54, gene heskett  wrote:

That is what I am using but it is several years old, Is there a newer
> one ddg can't find?


It's built in to LinuxCNC.

I just added this to my INI file [DISPLAY] section:

EMBED_TAB_NAME= cam_align
EMBED_TAB_COMMAND= qtvcp -d -c cam_align -x {XID} cam_align

And the LinuxCNC config running in a VM on my Mac started displaying the
webcam image from the built-in Mac camera. (which I am rather surprised
worked)

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is designed
for the especial use of mechanical geniuses, daredevils and lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] triggering the mesa card encoder index pulse

[gene heskett]

On 11/3/22 19:35, andy pugh wrote:

On Thu, 3 Nov 2022 at 14:14, gene heskett  wrote:


And fix my camview, again. Please...


Have you tried cam_align?

That is what I am using but it is several years old, Is there a newer 
one ddg can't find?



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, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] triggering the mesa card encoder index pulse

[andy pugh]

On Thu, 3 Nov 2022 at 14:14, gene heskett  wrote:

>
> And fix my camview, again. Please...


Have you tried cam_align?

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is designed
for the especial use of mechanical geniuses, daredevils and lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] triggering the mesa card encoder index pulse

[gene heskett]

On 11/3/22 14:50, andrew beck wrote:

Hey Andy your on to it.

I want to run a explicit homing cycle every tool change

Currently the tool change spindle zero point gets set the first time the
index gets set.  Which happens when I turn spindle on


So it's a good effort from linuxcnc really to work at all I guess.

As it keeps track in spindle orientation at 1rpm for hrs on end.

Just sometimes something happens and then it tries to spindle orientate 40
degrees out.  So I really need to home every tool change.

I'll post my gcode toolchange file and make a diagram of my set up.

I'm getting some funny behaviour with reading inputs with gcode to.

It's like they are not actually waiting for the input to read true.

Which leads me to suspect that your list of addf's in the .hal file might
be out of order.

You want an order that gets the hardware state first, then what you do to
it next and finally if something is waiting on this signal to do 
something else,
it is all in a 1,2,3,4 order, top to bottom  so the the final result fed 
to whatever depends on

it, is developed on the same 1 millisecond the chain reaction started with.

Strange things can and often will happen when a signal has to jump back
up this list to get to the next module. Everytime it has to jump backwards
up the list to get to the target adds another millisecond to the 
response time

because the correct state has to wait for the next pass thru the thread.

Do that 3 or so times thru a timing critical path will lead to fun and 
games,
for a millisecond or more before the final desired output is actually 
established.


Ghosts that bite you .1% can be exasperating to find, but printout your 
hal, follow
the signal path in hal, and put the position # beside that modules addf. 
You have it right

when a given signal ends up in the order its used.  Check 'em all.

In that case your tool change could run an explicit homing cycle prior to
toolchange.

M64 P04  ; set index-enable
G4 P0.1
M65 P04 ; toggle the tristate back
M3 S100 ; spin the spindle briefly
G4 P1 ; wait for at least one full rev
M19 P0 ; carry on.

On Fri, 4 Nov 2022, 01:33 andy pugh,  wrote:


On Thu, 3 Nov 2022 at 09:23, andrew beck  wrote:


looking for a way to set the index pulse high on the mesa card input

before

triggering a m19 for tool change as a safety check


The mesa index is a physical pin, so you would need to route out via a GPIO

But I don't understand your analysis of the problem, or the proposed fix.

If you just want to zero the encoder at a random point (which is what
toggling the index at a random point would do, if it did anything) then you
can do that via HAL.

Is it possible that the problem is that the spindle is simply not "homed"
sometimes? ie it has not yet passed through index with the index-enable
set, so the encoder counts are not aligned to the index mark?

In that case your tool change could run an explicit homing cycle prior to
toolchange.

M64 P04  ; set index-enable
G4 P0.1
M65 P04 ; toggle the tristate back
M3 S100 ; spin the spindle briefly
G4 P1 ; wait for at least one full rev
M19 P0 ; carry on.

As the index-enable is a bidirectional pin a bit of HAL is needed, using
the tristate-bit component.
That should be set up with the in-pin permanently high and the enable
connected to motion.digital-out-NN (NN = 04 in the example above)

--
atp
"A motorcycle is a bicycle with a pandemonium attachment and is designed
for the especial use of mechanical geniuses, daredevils and lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
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, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] triggering the mesa card encoder index pulse

[andrew beck]

That's a good idea

I'll try it

On Fri, 4 Nov 2022, 08:03 andy pugh,  wrote:

> On Thu, 3 Nov 2022 at 18:52, andrew beck  wrote:
>
> I'm getting some funny behaviour with reading inputs with gcode to.
> >
> > It's like they are not actually waiting for the input to read true.
>
>
> Try debouncing them in the HAL layer
>
> --
> atp
> "A motorcycle is a bicycle with a pandemonium attachment and is designed
> for the especial use of mechanical geniuses, daredevils and lunatics."
> — George Fitch, Atlanta Constitution Newspaper, 1912
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] triggering the mesa card encoder index pulse

[andy pugh]

On Thu, 3 Nov 2022 at 18:52, andrew beck  wrote:

I'm getting some funny behaviour with reading inputs with gcode to.
>
> It's like they are not actually waiting for the input to read true.


Try debouncing them in the HAL layer

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is designed
for the especial use of mechanical geniuses, daredevils and lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] triggering the mesa card encoder index pulse

[andrew beck]

Hey Andy your on to it.

I want to run a explicit homing cycle every tool change

Currently the tool change spindle zero point gets set the first time the
index gets set.  Which happens when I turn spindle on


So it's a good effort from linuxcnc really to work at all I guess.

As it keeps track in spindle orientation at 1rpm for hrs on end.

Just sometimes something happens and then it tries to spindle orientate 40
degrees out.  So I really need to home every tool change.

I'll post my gcode toolchange file and make a diagram of my set up.

I'm getting some funny behaviour with reading inputs with gcode to.

It's like they are not actually waiting for the input to read true.


In that case your tool change could run an explicit homing cycle prior to
toolchange.

M64 P04  ; set index-enable
G4 P0.1
M65 P04 ; toggle the tristate back
M3 S100 ; spin the spindle briefly
G4 P1 ; wait for at least one full rev
M19 P0 ; carry on.

On Fri, 4 Nov 2022, 01:33 andy pugh,  wrote:

> On Thu, 3 Nov 2022 at 09:23, andrew beck  wrote:
>
> >
> > looking for a way to set the index pulse high on the mesa card input
> before
> > triggering a m19 for tool change as a safety check
> >
>
> The mesa index is a physical pin, so you would need to route out via a GPIO
>
> But I don't understand your analysis of the problem, or the proposed fix.
>
> If you just want to zero the encoder at a random point (which is what
> toggling the index at a random point would do, if it did anything) then you
> can do that via HAL.
>
> Is it possible that the problem is that the spindle is simply not "homed"
> sometimes? ie it has not yet passed through index with the index-enable
> set, so the encoder counts are not aligned to the index mark?
>
> In that case your tool change could run an explicit homing cycle prior to
> toolchange.
>
> M64 P04  ; set index-enable
> G4 P0.1
> M65 P04 ; toggle the tristate back
> M3 S100 ; spin the spindle briefly
> G4 P1 ; wait for at least one full rev
> M19 P0 ; carry on.
>
> As the index-enable is a bidirectional pin a bit of HAL is needed, using
> the tristate-bit component.
> That should be set up with the in-pin permanently high and the enable
> connected to motion.digital-out-NN (NN = 04 in the example above)
>
> --
> atp
> "A motorcycle is a bicycle with a pandemonium attachment and is designed
> for the especial use of mechanical geniuses, daredevils and lunatics."
> — George Fitch, Atlanta Constitution Newspaper, 1912
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] triggering the mesa card encoder index pulse

[John Figie]

Gene said "
A 20 volt tool head should run on a 24 volt"

I'm pretty sure those 20V or V20 tools are just 18V nominal. At least that
is true for Dewalt and Craftsman.

John Figie

On Thu, Nov 3, 2022, 9:42 AM Sam Sokolik  wrote:

> I have written some simple hal components that set the index and then wait
> for the index to be seen before continuing.. (like shape turning/boring)
>
> probably not the greatest code  Seems to work though..
>
>
> Code snippet  (indexenable is hooked to index enable of encoder)
>
> // only set index enable once
> if (enable && !isindex) {
> isindex = true;
> indexenable = true;
> return;
> }
>
> // wait for spindle index before actually enabling
> if (enable && isindex && indexenable){
> return;
> }
>
>
>
>
> On Thu, Nov 3, 2022 at 9:14 AM gene heskett  wrote:
>
> > On 11/3/22 08:30, andy pugh wrote:
> > > On Thu, 3 Nov 2022 at 09:23, andrew beck 
> > wrote:
> > >
> > >> looking for a way to set the index pulse high on the mesa card input
> > before
> > >> triggering a m19 for tool change as a safety check
> > >>
> > > The mesa index is a physical pin, so you would need to route out via a
> > GPIO
> > >
> > > But I don't understand your analysis of the problem, or the proposed
> fix.
> > >
> > > If you just want to zero the encoder at a random point (which is what
> > > toggling the index at a random point would do, if it did anything) then
> > you
> > > can do that via HAL.
> > >
> > > Is it possible that the problem is that the spindle is simply not
> "homed"
> > > sometimes? ie it has not yet passed through index with the index-enable
> > > set, so the encoder counts are not aligned to the index mark?
> > >
> > > In that case your tool change could run an explicit homing cycle prior
> to
> > > toolchange.
> > >
> > > M64 P04  ; set index-enable
> > > G4 P0.1
> > > M65 P04 ; toggle the tristate back
> > > M3 S100 ; spin the spindle briefly
> > > G4 P1 ; wait for at least one full rev
> > > M19 P0 ; carry on.
> > >
> > > As the index-enable is a bidirectional pin a bit of HAL is needed,
> using
> > > the tristate-bit component.
> > > That should be set up with the in-pin permanently high and the enable
> > > connected to motion.digital-out-NN (NN = 04 in the example above)
> > >
> > An interesting thread Andy.
> >
> > I've considered doing something to aid in tool changing on my go704
> > which has an R8 spindle.
> > But the spindle locking consists of a home made clamp ring carrying a
> > spring loaded pin  I clamp
> > to the outside of the spindle housings bearing barrel when I want to do
> > a lot of it. It engages a
> > 6mm hole in the spindle nose to immobilize the spindle while cranking on
> > the drawbolt.
> > It also carries the camera I'd like to use for edge finding, but the fix
> > that made the camera
> > work a couple months back, apparently never made it into master, so the
> > camview tab is once again a
> > blank white screen. Insert frustrated Sigh here.
> >
> > Is there an unwritten rule that says it's never going to work in
> linuxcnc?
> >
> > Not mounted ATM because it can get in the way.
> >
> > And, using an an 8 point 10mm socket in an impact screw driver, the
> > impact is good enough to get an
> > R8 tight just by grabbing the spindle by hand. If I could find a pin
> > driver small enough to push the
> > locking pin in, and use something like the above to engage the pin, then
> > cause the spindle to move
> > until the pin drops in then turn it backwards to assure full penetration
> > of the pin, then I'd be tempted
> > make a motorized suspension to disengage the socket when the tool has
> > been changed, or to drop
> > the socket onto the drawbolt to be used in a semi-auto mode.
> >
> > Sawing the handle off the tool, and rigging a pair of  dc to dc SSR's in
> > place of the tools
> > trigger and direction switches, would allow me to actually change a tool
> > with only 2 hands.
> > Normally it needs at least 3...
> >
> > And by activating the spindle with a zero speed, would hold it
> > stationary enough for the
> > ratchet screwdriver to do the job. I do not do that now as a stopped
> > spindle is free.
> >
> > Sawing  up the tool itself is a minor problem. Picking a tool to saw up
> > presents a long term
> > problem because the batteries fail and replacement batteries can't be
> > found 5 years later.
> > Either that, or add yet another psu big enough to sub for the battery.
> >
> > A 20 volt tool head should run on a 24 volt supply scavenged from a 3d
> > printer, so those
> > are cheap enough. So are motors, belts, and linear bearings from failed
> > 3d printers, which
> > I have a house full of. So driving the tools vertical transport is only
> > a problem if I can't print it.
> > The Z drive from a dead printer could supply all of that. And that stuff
> > I have underfoot.
> >
> > Progress reports as it happens. ATM I'm rebuilding bigger 3d printers,
> > waiting on parts.
> >
> > And fix my 

Re: [Emc-users] triggering the mesa card encoder index pulse

[Sam Sokolik]

I have written some simple hal components that set the index and then wait
for the index to be seen before continuing.. (like shape turning/boring)

probably not the greatest code  Seems to work though..


Code snippet  (indexenable is hooked to index enable of encoder)

// only set index enable once
if (enable && !isindex) {
isindex = true;
indexenable = true;
return;
}

// wait for spindle index before actually enabling
if (enable && isindex && indexenable){
return;
}




On Thu, Nov 3, 2022 at 9:14 AM gene heskett  wrote:

> On 11/3/22 08:30, andy pugh wrote:
> > On Thu, 3 Nov 2022 at 09:23, andrew beck 
> wrote:
> >
> >> looking for a way to set the index pulse high on the mesa card input
> before
> >> triggering a m19 for tool change as a safety check
> >>
> > The mesa index is a physical pin, so you would need to route out via a
> GPIO
> >
> > But I don't understand your analysis of the problem, or the proposed fix.
> >
> > If you just want to zero the encoder at a random point (which is what
> > toggling the index at a random point would do, if it did anything) then
> you
> > can do that via HAL.
> >
> > Is it possible that the problem is that the spindle is simply not "homed"
> > sometimes? ie it has not yet passed through index with the index-enable
> > set, so the encoder counts are not aligned to the index mark?
> >
> > In that case your tool change could run an explicit homing cycle prior to
> > toolchange.
> >
> > M64 P04  ; set index-enable
> > G4 P0.1
> > M65 P04 ; toggle the tristate back
> > M3 S100 ; spin the spindle briefly
> > G4 P1 ; wait for at least one full rev
> > M19 P0 ; carry on.
> >
> > As the index-enable is a bidirectional pin a bit of HAL is needed, using
> > the tristate-bit component.
> > That should be set up with the in-pin permanently high and the enable
> > connected to motion.digital-out-NN (NN = 04 in the example above)
> >
> An interesting thread Andy.
>
> I've considered doing something to aid in tool changing on my go704
> which has an R8 spindle.
> But the spindle locking consists of a home made clamp ring carrying a
> spring loaded pin  I clamp
> to the outside of the spindle housings bearing barrel when I want to do
> a lot of it. It engages a
> 6mm hole in the spindle nose to immobilize the spindle while cranking on
> the drawbolt.
> It also carries the camera I'd like to use for edge finding, but the fix
> that made the camera
> work a couple months back, apparently never made it into master, so the
> camview tab is once again a
> blank white screen. Insert frustrated Sigh here.
>
> Is there an unwritten rule that says it's never going to work in linuxcnc?
>
> Not mounted ATM because it can get in the way.
>
> And, using an an 8 point 10mm socket in an impact screw driver, the
> impact is good enough to get an
> R8 tight just by grabbing the spindle by hand. If I could find a pin
> driver small enough to push the
> locking pin in, and use something like the above to engage the pin, then
> cause the spindle to move
> until the pin drops in then turn it backwards to assure full penetration
> of the pin, then I'd be tempted
> make a motorized suspension to disengage the socket when the tool has
> been changed, or to drop
> the socket onto the drawbolt to be used in a semi-auto mode.
>
> Sawing the handle off the tool, and rigging a pair of  dc to dc SSR's in
> place of the tools
> trigger and direction switches, would allow me to actually change a tool
> with only 2 hands.
> Normally it needs at least 3...
>
> And by activating the spindle with a zero speed, would hold it
> stationary enough for the
> ratchet screwdriver to do the job. I do not do that now as a stopped
> spindle is free.
>
> Sawing  up the tool itself is a minor problem. Picking a tool to saw up
> presents a long term
> problem because the batteries fail and replacement batteries can't be
> found 5 years later.
> Either that, or add yet another psu big enough to sub for the battery.
>
> A 20 volt tool head should run on a 24 volt supply scavenged from a 3d
> printer, so those
> are cheap enough. So are motors, belts, and linear bearings from failed
> 3d printers, which
> I have a house full of. So driving the tools vertical transport is only
> a problem if I can't print it.
> The Z drive from a dead printer could supply all of that. And that stuff
> I have underfoot.
>
> Progress reports as it happens. ATM I'm rebuilding bigger 3d printers,
> waiting on parts.
>
> And fix my camview, again. Please...
>
> 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, 1940)
> If we desire respect for the law, we must first make the law respectable.
>   - Louis D. Brandeis
> Genes Web page 
>
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> 

Re: [Emc-users] triggering the mesa card encoder index pulse

[gene heskett]

On 11/3/22 08:30, andy pugh wrote:

On Thu, 3 Nov 2022 at 09:23, andrew beck  wrote:


looking for a way to set the index pulse high on the mesa card input before
triggering a m19 for tool change as a safety check


The mesa index is a physical pin, so you would need to route out via a GPIO

But I don't understand your analysis of the problem, or the proposed fix.

If you just want to zero the encoder at a random point (which is what
toggling the index at a random point would do, if it did anything) then you
can do that via HAL.

Is it possible that the problem is that the spindle is simply not "homed"
sometimes? ie it has not yet passed through index with the index-enable
set, so the encoder counts are not aligned to the index mark?

In that case your tool change could run an explicit homing cycle prior to
toolchange.

M64 P04  ; set index-enable
G4 P0.1
M65 P04 ; toggle the tristate back
M3 S100 ; spin the spindle briefly
G4 P1 ; wait for at least one full rev
M19 P0 ; carry on.

As the index-enable is a bidirectional pin a bit of HAL is needed, using
the tristate-bit component.
That should be set up with the in-pin permanently high and the enable
connected to motion.digital-out-NN (NN = 04 in the example above)


An interesting thread Andy.

I've considered doing something to aid in tool changing on my go704 
which has an R8 spindle.
But the spindle locking consists of a home made clamp ring carrying a 
spring loaded pin  I clamp
to the outside of the spindle housings bearing barrel when I want to do 
a lot of it. It engages a
6mm hole in the spindle nose to immobilize the spindle while cranking on 
the drawbolt.
It also carries the camera I'd like to use for edge finding, but the fix 
that made the camera
work a couple months back, apparently never made it into master, so the 
camview tab is once again a

blank white screen. Insert frustrated Sigh here.

Is there an unwritten rule that says it's never going to work in linuxcnc?

Not mounted ATM because it can get in the way.

And, using an an 8 point 10mm socket in an impact screw driver, the 
impact is good enough to get an
R8 tight just by grabbing the spindle by hand. If I could find a pin 
driver small enough to push the
locking pin in, and use something like the above to engage the pin, then 
cause the spindle to move
until the pin drops in then turn it backwards to assure full penetration 
of the pin, then I'd be tempted
make a motorized suspension to disengage the socket when the tool has 
been changed, or to drop

the socket onto the drawbolt to be used in a semi-auto mode.

Sawing the handle off the tool, and rigging a pair of  dc to dc SSR's in 
place of the tools
trigger and direction switches, would allow me to actually change a tool 
with only 2 hands.

Normally it needs at least 3...

And by activating the spindle with a zero speed, would hold it 
stationary enough for the
ratchet screwdriver to do the job. I do not do that now as a stopped 
spindle is free.


Sawing  up the tool itself is a minor problem. Picking a tool to saw up 
presents a long term
problem because the batteries fail and replacement batteries can't be 
found 5 years later.

Either that, or add yet another psu big enough to sub for the battery.

A 20 volt tool head should run on a 24 volt supply scavenged from a 3d 
printer, so those
are cheap enough. So are motors, belts, and linear bearings from failed 
3d printers, which
I have a house full of. So driving the tools vertical transport is only 
a problem if I can't print it.
The Z drive from a dead printer could supply all of that. And that stuff 
I have underfoot.


Progress reports as it happens. ATM I'm rebuilding bigger 3d printers, 
waiting on parts.


And fix my camview, again. Please...

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, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] CAM software

[stjohn gold]

Hi Chris,
all very good points. For me the deal maker is the fact that posts are open
and can be modified. Community support is good. This I consider to be
really important. There are other cad/cam packages I like better but the
posts are often closed and much work to fine tune or edit...

cheers, St.john

On Wed, 2 Nov 2022 at 20:24, Chris Albertson 
wrote:

> You are correct.  The free version only does 3-axis milling. You need
> the paid version to get a 4th axis.
>
> The best you can do with the free version is  is "3.5 axis."  That means
> manual setup of the rotary table and multiple passes of 3-axis milling.
> You can not move the 4th axis while cutting with the free version.   You
> would need to design the parts with CAM limitations in mind.
>
> If this is a hobby, then of course cost matters a lot. But for business
> use, $545 per year is trivial.
>
> I'm seriously looking at Open Source CAM solutions.  But really, the free
> version of Fusion360 can do parts as complex as say a motorcycle transition
> housing. Or that shoulder bulkhead assembly I posted.
>
>
> Hi Chris,
> >
> > > -Original Message-
> > > From: Chris Albertson [mailto:albertson.ch...@gmail.com]
> > > Sent: November-02-22 9:55 AM
> > > To: Enhanced Machine Controller (EMC)
> > > Subject: Re: [Emc-users] CAM software
> > >
> > > Here is why hand-coded g-code files can't be used for "real work".
> > >
> > > I have a part (see link below) that is one of dozens of parts in a
> > > quadruped robot.
> > > https://a360.co/3sQkHm3
> > > The above is a simple web-based viewer that you get for "free" with
> > > Fusion360.
> >
> > > Also, I doubt any normal person could write the g-code for this part
> and
> > > get it correct.
> > >
> >
> > The problem is the moment you want that 4th Axis that can't be hand
> > coded.  At that point Fusion360 becomes way more expensive than say
> MecSoft
> > Visual CAD/CAM.
> >
> > John
> >
> >
> >
> >
> > ___
> > Emc-users mailing list
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
> >
>
>
> --
>
> Chris Albertson
> Redondo Beach, California
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] triggering the mesa card encoder index pulse

[andy pugh]

On Thu, 3 Nov 2022 at 09:23, andrew beck  wrote:

>
> looking for a way to set the index pulse high on the mesa card input before
> triggering a m19 for tool change as a safety check
>

The mesa index is a physical pin, so you would need to route out via a GPIO

But I don't understand your analysis of the problem, or the proposed fix.

If you just want to zero the encoder at a random point (which is what
toggling the index at a random point would do, if it did anything) then you
can do that via HAL.

Is it possible that the problem is that the spindle is simply not "homed"
sometimes? ie it has not yet passed through index with the index-enable
set, so the encoder counts are not aligned to the index mark?

In that case your tool change could run an explicit homing cycle prior to
toolchange.

M64 P04  ; set index-enable
G4 P0.1
M65 P04 ; toggle the tristate back
M3 S100 ; spin the spindle briefly
G4 P1 ; wait for at least one full rev
M19 P0 ; carry on.

As the index-enable is a bidirectional pin a bit of HAL is needed, using
the tristate-bit component.
That should be set up with the in-pin permanently high and the enable
connected to motion.digital-out-NN (NN = 04 in the example above)

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is designed
for the especial use of mechanical geniuses, daredevils and lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Protel99 SE Drill file to G-CODE for LinuxCNC or MACH3.

[Peter Homann]

Hi John,

If you have Sheetcam, I'm fairly sure that it can take in drill files 
and produce G code for Mach3 or LinuxCNC.


Cheers,

Peter


On 2/11/2022 8:19 am, John Dammeyer wrote:

I needed a small PC board.  I converted the short drill file .txt over to G-Code .nc 
or .tap.   Took a few tries to get it right for LinuxCNC which complained about a 
few things that MACH3 didn't.  Ended up using the CNC router with MACH3 because the 
router can do 20,000 RPM while my mill is limited to 3000.  Better to turn the 
0.035" drill bits at higher RPM I'm told.
  
Anyway.  Does anyone know of a Protel 99SE drill file conversion program to G-Code for CNC?  I've attached both.  Hopefully they will come through.
  
https://youtu.be/5zh-28CHdj4  
  
Here's a short (very short) video of a few of the holes being drilled.  It's been so long since I did this that of course I first drilled from the copper side rather than the component side.  Came out mirrored of course.  Then flipped the piece of PCB around and did it correctly.  The parts just barely fit but that's fine.  It's just a prototype.
  
John
  
"ELS! Nothing else works as well for your Lathe"

Automation Artisans Inc.
www dot autoartisans dot com
  



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users


--
-
Web : http://www.homanndesigns.com
email : gro...@homanndesigns.com
Phone : +61 421 601 665



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] triggering the mesa card encoder index pulse

[andrew beck]

hey guys

looking for a way to set the index pulse high on the mesa card input before
triggering a m19 for tool change as a safety check

i have a g code sub that i have remapped for m6 and it moves the spindle to
toolchange position and then activates m19 r0 which spins the spindle to
the correct angle.

that all works fine most of the time

every 2 weeks or so it doesn't work and i get a toolchange crash

and tonight i had enough lol time to fix it
my toolchanger just crashed and lost orientation

and broke a lovely endmill

sounds like its time to fix the spindle orientation finally

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users