Re: [Emc-users] Bringing the saw beveler back to life

[Mark]

On 03/05/2018 10:33 AM, John Kasunich wrote:


On Mon, Mar 5, 2018, at 10:25 AM, Mark wrote:


At any rate, what the question above is asking, does the correction
happen while the X axis is moving to smoothly blend the adjustment, or
does it try to jump to the correction without blending the two axis's moves?

The "lin" in lincurve stands for linear interpolation.  So yes, it blends.


I was hoping that was the case.  ;-)




Yes, there are groups of X axis stations that have the same offset. What
I'm trying to figure out is how to map those X axis coordinates that are
in that group.  If the X axis stations between say 15 and 25 are all the
same, but the next group of X axis stations which are the same between
26 and 31 are different than the previous group, how would I map that
between the x,y coordinates in the lincurve statements?

Example:

The error between X=0 and X=4 is zero, then it ramps up to 0.002 at X=5,
hits 0.003 at X=6, then remains at 0.003 for the next 8", until X=13.  Then
it ramps back down to zero at X=14, and hits -0.001 at X=15, where it stays
until X=21.

So you would use the following X,Y pairs:

0, 0.000
4, 0.000
5, -0.002
6, -0.003
13, -0.003
14, 0.000
15, 0.001
21, 0.001

You can use up to 16 pairs.  Best approach is to plot your measured errors
on a piece of paper or in a spreadsheet, then find the set of 16 (or fewer)
straight line segments that best matches your measured error plot.



Ah ha!  The lightbulb is beginning to illuminate.  I was planning on 
doing exactly what you mentioned in the last paragraph, and from the 
measurement groupings I have it should take a lot less than 16 pairs.


Thanks for the clarification John.  I might nick this beast yet. ;-)

Mark

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Bringing the saw beveler back to life

[John Kasunich]

On Mon, Mar 5, 2018, at 10:25 AM, Mark wrote:

> At any rate, what the question above is asking, does the correction 
> happen while the X axis is moving to smoothly blend the adjustment, or 
> does it try to jump to the correction without blending the two axis's moves?

The "lin" in lincurve stands for linear interpolation.  So yes, it blends.

> Yes, there are groups of X axis stations that have the same offset. What 
> I'm trying to figure out is how to map those X axis coordinates that are 
> in that group.  If the X axis stations between say 15 and 25 are all the 
> same, but the next group of X axis stations which are the same between 
> 26 and 31 are different than the previous group, how would I map that 
> between the x,y coordinates in the lincurve statements?

Example:

The error between X=0 and X=4 is zero, then it ramps up to 0.002 at X=5, 
hits 0.003 at X=6, then remains at 0.003 for the next 8", until X=13.  Then
it ramps back down to zero at X=14, and hits -0.001 at X=15, where it stays
until X=21.

So you would use the following X,Y pairs:

0, 0.000
4, 0.000
5, -0.002
6, -0.003
13, -0.003
14, 0.000
15, 0.001
21, 0.001

You can use up to 16 pairs.  Best approach is to plot your measured errors 
on a piece of paper or in a spreadsheet, then find the set of 16 (or fewer)
straight line segments that best matches your measured error plot.


-- 
  John Kasunich
  jmkasun...@fastmail.fm

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Bringing the saw beveler back to life

[Gene Heskett]

On Monday 05 March 2018 09:17:00 Mark wrote:

> On 03/05/2018 08:56 AM, andy pugh wrote:
> > On 5 March 2018 at 13:07, Mark  wrote:
> >> I'm leaning towards lincurve.  In order to utilize that, I've got
> >> to upgrade linuxcnc to a bit new version.
> >
> > That's not strictly necessary, the file is a standalone module, you
> > could just download the comp file and
> > sudo comp --install lincurve.comp
> > https://github.com/LinuxCNC/linuxcnc/blob/master/src/hal/components/
> >lincurve.comp Right-Click the "Raw" button and "Download file as..."
>
> Probably better to upgrade to the current version anyway.  v2.5.5 is
> getting pretty long in the tooth, and the improvements added to the
> newer version might well be welcome.
>
> > But: To use the external offsets you need to run the experimental
> > external-offsets branch, and it isn't a standard LinuxCNC version.
> > You might be able to get it from the buildbot. But you would need
> > someone else who knows their way around the buildbot to find the
> > .deb if it exists (I just build from source)
>
> When I built the machine and configured linuxcnc we sorta had to jump
> through hoops to get the configuration set up the way it was needed
> for the operations it did.  Changing to something like that might mean
> a whole re-write of the hal code that I got a lot of help on from some
> very kind folks on the IRC.
>
> > You can use the "offset" HAL component though, so all is not lost.
> > You can even use both with your 2.5.5 install
> > http://linuxcnc.org/docs/2.7/html/man/man9/offset.9.html
>
> I'll look into that one too.  Thanks for the info.
>
> >> I couldn't find any examples of lincurve in use and the man page is
> >> a bit thin on how to actually implement it.  Also, the man page
> >> mentions x,y coordinates.
> >
> > Those refer to the axes of the imaginary graph paper that lincurve
> > is drawn on. You can use any input at all, and do anything at all
> > with the output. This is HAL.
> > You set the curve up (rather clunkily) in HAL. If X = 0 is nominal
> > and X = 3" is 5 thou high:
> > setp lincurve.0.x-val-00 0
> > setp lincurve.0.y-val-00 0
> > setp lincurve.0.x-val-01 3
> > setp lincurve.0.y-val-01 -0.005
> >
> > and so-on.
>
> Okay, so how do I translate the 'y-val' to the Z axis position?  I may
> have mentioned I'm a tad weak on hal coding since I haven't done it in
> ages.  Looking at those lines above, I'm guessing that the Z axis
> would "curve" between station 0 to station 1 to accommodate the -0.005
> difference?
>
> If that is true, how would I map say something like this:  stations 0
> - 5 are the median height.  Stations 6 through 15 are .003" high.
> Stations 16 through 25 are .002" low and so on.  The values of
> deviance from the mean table height are rather nicely grouped with
> maybe one or two .001" deviations in the mix, but I'm not going to
> worry about that.
>
> > You might want to keep the curve definition in a separate HAL file
> > for convenience.
>
> You lost me on this one.  What is the curve definition?  Speaking of
> HAL, is the placement order in the main HAL file for something like
> this going to bite me if it's put in too early or late in the
> declarations? I remember Gene having issues with some of his HAL code
> by coding something in too early or too late in the file.
>
That was in the addf order Mark, the only place in a halfile where order 
is important. There, for most uses, you want the order to be such that 
even in a 10 module chain, the calculations "fall thru" so that the 
results are ready at the end of the servo threads individual cycle. So 
you wind up with a halfile that starts with the loadrt or loaduser 
statements, then a list of addf's, followed by the sets's, setp's, 
net's, and other rules that tie all the logic together. Order is not 
important, so I tend to group that stuff into functions as much as 
possible so its easier to read and follow what its doing. One can also 
instantiate a slower thread, which I've done with the jog wheel stuff in 
the pi driving the Sheldon lathe.

There are of course creative exceptions to that rule.

And I had forgotten about the offset module, which replaces the sum2's 
I mentioned with a single module to do the same thing.

> Thanks again for your help Andy.
>
> Mark
>
> --
> Check out the vibrant tech community on one of the world's
> most engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> ___
> 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)
Genes Web page 

Re: [Emc-users] Bringing the saw beveler back to life

[andy pugh]

On 5 March 2018 at 14:17, Mark  wrote:

>> You set the curve up (rather clunkily) in HAL. If X = 0 is nominal and
>> X = 3" is 5 thou high:
>> setp lincurve.0.x-val-00 0
>> setp lincurve.0.y-val-00 0
>> setp lincurve.0.x-val-01 3
>> setp lincurve.0.y-val-01 -0.005
>>
>> and so-on.
>
>
> Okay, so how do I translate the 'y-val' to the Z axis position?  I may have
> mentioned I'm a tad weak on hal coding since I haven't done it in ages.
> Looking at those lines above, I'm guessing that the Z axis would "curve"
> between station 0 to station 1 to accommodate the -0.005 difference?

As I wrote the lincurve component its operation is so obvious to me
that I don't even understand the question.

lincurve just maps an input value to an output value, using an
imaginary graph of points linked by straight lines.
So for an X value between two X points it linearly interpolates
between the two corresponding Y values.

> If that is true, how would I map say something like this:  stations 0 - 5
> are the median height.  Stations 6 through 15 are .003" high. Stations 16
> through 25 are .002" low and so on.

If you have 25 points to map then it is a bit tricky as lincurve only
has 16 points.
You can skip some, though, if they all have the same offset.

>> You might want to keep the curve definition in a separate HAL file for
>> convenience.

> You lost me on this one.  What is the curve definition?

The set of "setp" commands that define the (x,y) pairs.

>  Speaking of HAL, is
> the placement order in the main HAL file for something like this going to
> bite me

If the curve definition HAL file is separate then it needs to be
loaded after the file that loads the lincurve component, or there
won't be any parameters to setp.

-- 
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, 1916

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Bringing the saw beveler back to life

[Gene Heskett]

On Monday 05 March 2018 08:01:58 Mark wrote:

> On 03/04/2018 12:53 PM, Gene Heskett wrote:
> > That is something that Andy wrote, called lincurve, see man 9
> > lincurve for a basic explanation. What isn't mentioned is that in a
> > motion scenario, it will probably feed a pair of sum2's, one summing
> > the command from motion to the machine in order to sum the command
> > from motion and the correction into the stepgen controlling the x
> > (height) axis, and a 2nd sum2 with its input from the lincurve set
> > for a gain of -1, and its feedback from the stepgen summed so that
> > the effect of the first "offset" is subtracted from the feedback
> > back to motion, effectively hiding any corrective motions from the
> > motion controller which would generate either corrections, or
> > following errors.
>
> I did a bit of research yesterday.  lincurve is not in the elder
> version of linuxcnc I have, v2.5.5.  I'm willing to upgrade to get the
> utility of lincurve, but I'd like to understand the utility a bit
> better. Downloaded the latest linuxcnc and burned the DVD yesterday
> ever.
>
> > The confusing part for me would be that the x input would be derived
> > from the saws z (length along the trackage) position, and the
> > corresponding y output controls the error corrections applied to the
> > height motor drive.
>
> I saw it was an x,y thing.  Maybe Andy can help figure how to apply
> this to an X and a Z axis.
>
> > As the man page says, it is a maximum of 16 points long, and I've no
> > clue if it could be cascaded to more points, however from your
> > description of the errors, 16 points seems like more than enough to
> > do a considerable error reduction. The calibration, done with a
> > series of setp's in the hal file, can be easily corrected for wear
> > error over time.
>
> 16 points would probably be overkill so no worries.  I've already got
> the cal numbers for each inch along the X axis.  We're talking at most
> about a few thou of deviation from the mean at any given point on the
> table.
>
> > Your ability to measure the error will likely determine how accurate
> > the correction can be.
>
> Got that nailed down pretty good with repeatable numbers.
>
> > Beyond this, Andy can probably explain it better than I can since he
> > wrote it. My only experience with it was in trying to linearize the
> > spindle speed control, a miserable failure trying to correct the
> > errors caused by some bob optoisolators so slow as to need stakes
> > set and a surveyer called.
>
> 
>
> > But I have some half built furniture to finish first. I've
> > had another blood clot reminder that I am on the downhill side of
> > life, and I do want to get them done before I miss roll call. So my
> > warfarin dose is back up...
> >
> > Alone ATM, the missus is in the shop with a broken leg, possible
> > cast removal in May. Osteoporosis etc, to go with the COPD. She did
> > that the night before I was scheduled to have a sore leg looked at.
> > The ultrasound tech called the ER, and they kept me for 6 days,
> > pouring in the heperin. Its better, but I've been instructed to get
> > the hell out of this chair every 30 minutes.
>
> Hope you and the Missus are mending.  Best wishes.
>
>
> Cheers,
> Mark
>
Thanks Mark. Now I need to go make a lap around the house. And its too 
darned quiet.

-- 
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)
Genes Web page 

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] email failures.

[Gene Heskett]

On Monday 05 March 2018 07:17:54 Charles Steinkuehler wrote:

> On 3/5/2018 4:19 AM, andy pugh wrote:
> > On 5 March 2018 at 02:20, Kurt Jacobson  
wrote:
> >> I have noticed lately that several old SourceForge and
> >> Yahoo list are moving to groups.io, so I guess others are either
> >> having trouble with SF or think it is not going to be around for
> >> much longer
> >
> > It might be simpler to run mailman on a server we control.
>
> I do this for a low-traffic neighborhood list and it's somewhat of a
> pain in today's email world.  Microsoft, Yahoo, and some others use
> very strict anti-spoofing settings which break pretty much any mailing
> list if you don't mangle or wrap the original message, both of which
> have drawbacks:
>
> https://wiki.list.org/DEV/DMARC
>
> I have also found it necessary to play with outgoing mail routing
> rules as the various MX machines I have access to (my personal server,
> my ISP's MX, and my company's MX) get added or removed from various
> blacklists, causing delivery failures to various user domains.
>
> Personally, I'd suggest Google groups, but I do still admin a Mailman
> list, so I can't recommend against it too much.  :)

google groups has been such a source of spam that I've been routing it 
to /dev/null for at least 2 years. Maybe longer.

-- 
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)
Genes Web page 

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Bringing the saw beveler back to life

[Mark]

On 03/05/2018 08:56 AM, andy pugh wrote:

On 5 March 2018 at 13:07, Mark  wrote:


I'm leaning towards lincurve.  In order to utilize that, I've got to upgrade
linuxcnc to a bit new version.

That's not strictly necessary, the file is a standalone module, you
could just download the comp file and
sudo comp --install lincurve.comp
https://github.com/LinuxCNC/linuxcnc/blob/master/src/hal/components/lincurve.comp
Right-Click the "Raw" button and "Download file as..."


Probably better to upgrade to the current version anyway.  v2.5.5 is 
getting pretty long in the tooth, and the improvements added to the 
newer version might well be welcome.





But: To use the external offsets you need to run the experimental
external-offsets branch, and it isn't a standard LinuxCNC version.
You might be able to get it from the buildbot. But you would need
someone else who knows their way around the buildbot to find the .deb
if it exists (I just build from source)


When I built the machine and configured linuxcnc we sorta had to jump 
through hoops to get the configuration set up the way it was needed for 
the operations it did.  Changing to something like that might mean a 
whole re-write of the hal code that I got a lot of help on from some 
very kind folks on the IRC.




You can use the "offset" HAL component though, so all is not lost. You
can even use both with your 2.5.5 install
http://linuxcnc.org/docs/2.7/html/man/man9/offset.9.html


I'll look into that one too.  Thanks for the info.




I couldn't find any examples of lincurve in use and the man page is a bit
thin on how to actually implement it.  Also, the man page mentions x,y
coordinates.

Those refer to the axes of the imaginary graph paper that lincurve is
drawn on. You can use any input at all, and do anything at all with
the output. This is HAL.
You set the curve up (rather clunkily) in HAL. If X = 0 is nominal and
X = 3" is 5 thou high:
setp lincurve.0.x-val-00 0
setp lincurve.0.y-val-00 0
setp lincurve.0.x-val-01 3
setp lincurve.0.y-val-01 -0.005

and so-on.


Okay, so how do I translate the 'y-val' to the Z axis position?  I may 
have mentioned I'm a tad weak on hal coding since I haven't done it in 
ages.  Looking at those lines above, I'm guessing that the Z axis would 
"curve" between station 0 to station 1 to accommodate the -0.005 difference?


If that is true, how would I map say something like this:  stations 0 - 
5 are the median height.  Stations 6 through 15 are .003" high. Stations 
16 through 25 are .002" low and so on.  The values of deviance from the 
mean table height are rather nicely grouped with maybe one or two .001" 
deviations in the mix, but I'm not going to worry about that.





You might want to keep the curve definition in a separate HAL file for
convenience.


You lost me on this one.  What is the curve definition?  Speaking of 
HAL, is the placement order in the main HAL file for something like this 
going to bite me if it's put in too early or late in the declarations?  
I remember Gene having issues with some of his HAL code by coding 
something in too early or too late in the file.


Thanks again for your help Andy.

Mark

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Bringing the saw beveler back to life

[andy pugh]

On 5 March 2018 at 13:07, Mark  wrote:

> I'm leaning towards lincurve.  In order to utilize that, I've got to upgrade
> linuxcnc to a bit new version.

That's not strictly necessary, the file is a standalone module, you
could just download the comp file and
sudo comp --install lincurve.comp
https://github.com/LinuxCNC/linuxcnc/blob/master/src/hal/components/lincurve.comp
Right-Click the "Raw" button and "Download file as..."

But: To use the external offsets you need to run the experimental
external-offsets branch, and it isn't a standard LinuxCNC version.
You might be able to get it from the buildbot. But you would need
someone else who knows their way around the buildbot to find the .deb
if it exists (I just build from source)

You can use the "offset" HAL component though, so all is not lost. You
can even use both with your 2.5.5 install
http://linuxcnc.org/docs/2.7/html/man/man9/offset.9.html

> I couldn't find any examples of lincurve in use and the man page is a bit
> thin on how to actually implement it.  Also, the man page mentions x,y
> coordinates.

Those refer to the axes of the imaginary graph paper that lincurve is
drawn on. You can use any input at all, and do anything at all with
the output. This is HAL.
You set the curve up (rather clunkily) in HAL. If X = 0 is nominal and
X = 3" is 5 thou high:
setp lincurve.0.x-val-00 0
setp lincurve.0.y-val-00 0
setp lincurve.0.x-val-01 3
setp lincurve.0.y-val-01 -0.005

and so-on.

You might want to keep the curve definition in a separate HAL file for
convenience.

-- 
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, 1916

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Bringing the saw beveler back to life

[Mark]

On 03/04/2018 01:17 PM, andy pugh wrote:

On 4 March 2018 at 12:53, Mark  wrote:


There was something Andy talked about that he wrote, but I don't remember
what that was. It was something in hal code if I remember correctly.

Sounds like a job for the external_offsets branch, and either a
distance sensor of some sort or the "lincurve" HAL component to apply
a fixed offset depending on axis position.


Andy,


I'm leaning towards lincurve.  In order to utilize that, I've got to 
upgrade linuxcnc to a bit new version.  I'm running v2.5.5 right now, 
but lincurve isn't in that distro.  I'll be doing the upgrade over the 
next day or soif lincurve will do what I need.



I couldn't find any examples of lincurve in use and the man page is a 
bit thin on how to actually implement it.  Also, the man page mentions 
x,y coordinates.  Will I be able to mapthis to and X axis position and a 
Z axis height?  Also where are the mapping values stored?  A lookup 
table in a separate file, a listing in the hal file, or something else?  
I'm a bit weak on hal coding not having to write new hal stuff for a 
number of years.  Notreally sure how I need to implement this utility. 
Do I need to use any other hal utilities to make something like this work?



Thanks for your help,

Mark


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Bringing the saw beveler back to life

[Mark]

On 03/04/2018 12:53 PM, Gene Heskett wrote:

That is something that Andy wrote, called lincurve, see man 9 lincurve
for a basic explanation. What isn't mentioned is that in a motion
scenario, it will probably feed a pair of sum2's, one summing the
command from motion to the machine in order to sum the command from
motion and the correction into the stepgen controlling the x (height)
axis, and a 2nd sum2 with its input from the lincurve set for a gain
of -1, and its feedback from the stepgen summed so that the effect of
the first "offset" is subtracted from the feedback back to motion,
effectively hiding any corrective motions from the motion controller
which would generate either corrections, or following errors.


I did a bit of research yesterday.  lincurve is not in the elder version 
of linuxcnc I have, v2.5.5.  I'm willing to upgrade to get the utility 
of lincurve, but I'd like to understand the utility a bit better.  
Downloaded the latest linuxcnc and burned the DVD yesterday ever.





The confusing part for me would be that the x input would be derived from
the saws z (length along the trackage) position, and the corresponding y
output controls the error corrections applied to the height motor drive.


I saw it was an x,y thing.  Maybe Andy can help figure how to apply this 
to an X and a Z axis.




As the man page says, it is a maximum of 16 points long, and I've no clue
if it could be cascaded to more points, however from your description of
the errors, 16 points seems like more than enough to do a considerable
error reduction. The calibration, done with a series of setp's in the
hal file, can be easily corrected for wear error over time.


16 points would probably be overkill so no worries.  I've already got 
the cal numbers for each inch along the X axis.  We're talking at most 
about a few thou of deviation from the mean at any given point on the table.




Your ability to measure the error will likely determine how accurate the
correction can be.


Got that nailed down pretty good with repeatable numbers.



Beyond this, Andy can probably explain it better than I can since he
wrote it. My only experience with it was in trying to linearize the
spindle speed control, a miserable failure trying to correct the errors
caused by some bob optoisolators so slow as to need stakes set and a
surveyer called.




But I have some half built furniture to finish first. I've
had another blood clot reminder that I am on the downhill side of life,
and I do want to get them done before I miss roll call. So my warfarin
dose is back up...

Alone ATM, the missus is in the shop with a broken leg, possible cast
removal in May. Osteoporosis etc, to go with the COPD. She did that the
night before I was scheduled to have a sore leg looked at. The
ultrasound tech called the ER, and they kept me for 6 days, pouring in
the heperin. Its better, but I've been instructed to get the hell out of
this chair every 30 minutes.


Hope you and the Missus are mending.  Best wishes.


Cheers,
Mark

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Mark]

On 03/04/2018 08:22 PM, Chris Albertson wrote:

On Sun, Feb 18, 2018 at 2:10 PM, Gene Heskett  wrote:

Seem like you machine runs into some form of resonance at 7 to 20 ipm.
   I'd like to see a controllers that can be reconfigured in real time.
for low speeds you might use 1/8 steps them move to /14/ and full
steps at higher RPM.  It could be built.   But simply going to a
digital drive might be enough.


Gecko drives are designed to do just that.



Mark



--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] email failures.

[Charles Steinkuehler]

On 3/5/2018 4:19 AM, andy pugh wrote:
> On 5 March 2018 at 02:20, Kurt Jacobson  wrote:
>> I have noticed lately that several old SourceForge and
>> Yahoo list are moving to groups.io, so I guess others are either having
>> trouble with SF or think it is not going to be around for much longer
> 
> It might be simpler to run mailman on a server we control.

I do this for a low-traffic neighborhood list and it's somewhat of a
pain in today's email world.  Microsoft, Yahoo, and some others use
very strict anti-spoofing settings which break pretty much any mailing
list if you don't mangle or wrap the original message, both of which
have drawbacks:

https://wiki.list.org/DEV/DMARC

I have also found it necessary to play with outgoing mail routing
rules as the various MX machines I have access to (my personal server,
my ISP's MX, and my company's MX) get added or removed from various
blacklists, causing delivery failures to various user domains.

Personally, I'd suggest Google groups, but I do still admin a Mailman
list, so I can't recommend against it too much.  :)

-- 
Charles Steinkuehler
char...@steinkuehler.net

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[andy pugh]

On 5 March 2018 at 01:23, Chris Albertson  wrote:
> I'm not contributing to EMC development so I have no say in how to do
> it.  But I'd suggest a move to Github.

The code moved to Github some time ago. ( https://github.com/LinuxCNC )

Moving a mailing list is more tricky as people will have the existing
address in their address lists, and will have the list whitelisted
etc.

-- 
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, 1916

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] email failures.

[andy pugh]

On 5 March 2018 at 02:20, Kurt Jacobson  wrote:
> I have noticed lately that several old SourceForge and
> Yahoo list are moving to groups.io, so I guess others are either having
> trouble with SF or think it is not going to be around for much longer

It might be simpler to run mailman on a server we control.

-- 
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, 1916

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users