Re: [Emc-users] hands-on EDM experience

[Nicklas Karlsson]

On Sat, 23 Apr 2016 10:32:09 +0800
Thomas Powderly  wrote:

> ...
> in real work the tool is already at the discharge distance when the 
> initiation voltage is turned on
> so the initiation still occurs, followed by the discharge

I was thinking about this for the ionization voltage. Usually tool is already 
at the discharge distance but in case it is to far away ionization just a 
little bit above limit would prevent discharge from happen to far away.

> ...
> .75 amps never bothered me, that measure is of the supplied current,
> ( the current that would be delivered in a dead short)
>   which is much more than delivered current

It is the current delivered if I happen to touch tool that bother me, to high 
current and I wont be there to machine another day. There is a 
blue-green-yellow-red graph a bit down to the right on this page 
https://en.wikipedia.org/wiki/Electric_shock and I prefer to stay to the left 
of the yellow area.

> tomp

10mA seems to be enough for ionization although I will try higher later to 
check if it make a difference. Sparc distinguishing seems to work at if short 
circuited there is no current. I however have some problem to keep the process 
running manually.


Regards Nicklas Karlsson



> On 04/22/16 22:29, Nicklas Karlsson wrote:
> > .75 amps is very dangerous as soon as voltage is high enough. Low current 
> > and it should not be dangerous even for rather high voltage.
> >
> > Preferably I want to stay at low current at least until prototype leave 
> > desktop.
> >
> > Since dielectric breakdown strength only has volts and distance I guess 
> > volt should be adjusted so that ionization is not started at to long 
> > distance which would create a big crater.
> >
> > I however want to keep current low to stay safe but does the current change 
> > how large a channel is created or will it grow large enough anyway as soon 
> > as the real current start?
> >
> >
> > Regards Nicklas Karlsson
> >
> >
> >
> >
> > On Thu, 21 Apr 2016 22:03:32 +0800
> > Thomas Powderly  wrote:
> >
> >> initiation ONLY depends on voltage
> >> i use .75 amps and its overkill
> >> but it was easy to make the circuit
> >> the equation for dielectric breakdown strength only has volts and distance
> >> but
> >> higher voltage makes ionization easier
> >> typical value are 100 to 300VDC
> >> tho for some paarungs ( stahl bis stahl, copper to crystal ... ) upwards
> >> of 1000V= are used
> >> hth tomp
> >>
> >> On 04/21/16 20:13, Nicklas Karlsson wrote:
> >>> I certainly discovered the need to back off.
> >>>
> >>> As is now I use only 10mA for ionization and 10A for cuts. I want to keep 
> >>> ionization current down for safety reason, do you know if it is harder to 
> >>> get process running at low current?
> >>>
> >>> Regards Nicklas Karlsson
> >>>
> >>>
> >>>
> >>> On Thu, 17 Mar 2016 15:15:10 + (UTC)
> >>> Pete_Gruendeman  wrote:
> >>>
>  Hi Nicklas:
>    Did you set up and run the simple RC type EDM?  It's a pretty 
>  simple proposition, using a filtered DC power supply, a capacitor in 
>  parallel with the gap and the gap itself.  An electrolytic capacitor 
>  would work.  A motor-run capacitor, of 50-100 micro Farads would work 
>  even better.  Hands-on experience will answer many of you questions, 
>  especially in regard to the stability of the process.  No matter how 
>  stable your process is, backing up is part of this artform.
> 
>  Pete
>  
>  On Thu, 3/10/16, Nicklas Karlsson  wrote:
> 
> Subject: Re: hands-on EDM experience
> To: "Pete_Gruendeman" 
> Date: Thursday, March 10, 2016, 8:49 AM
> 
> > Hi Nicklas:
> > I wrote and sent this last
> night but it seems to have bounced instead of reaching
> you.  I want you to get some hands on experience with EDM,
> even if it's a simple RC type EDM that you clamp onto a
> milling machine and flush with a turkey baster you borrow
> from the kitchen.  Even though I had run commercial EDMs,
> it was eye opening to run the process manually.  The whole
> process can be complicated but it doesn't have to be.
> Read below.
> >
> > Hi
> tomp and Nicklas:
> >
>    Careful tomp as much of what you wrote is some
> pretty technical stuff.  It might be over Nicklas' head
> since he is new to EDM.  I barely recognize some of this
> but do recall that Agie offered auto-overcut compensation.
> I always turned that off as much of my work was something
> like a horizontally oriented cylinder and I typically
> orbited about the X-axis.
> >
> > Nicklas:  Have you done
> anything with EDM yet?, even with a simple resistive
>  

Re: [Emc-users] hands-on EDM experience

[Gene Heskett]

On Saturday 23 April 2016 02:17:58 Nicklas Karlsson wrote:

> On Sat, 23 Apr 2016 10:32:09 +0800
>
> Thomas Powderly  wrote:
> > ...
> > in real work the tool is already at the discharge distance when the
> > initiation voltage is turned on
> > so the initiation still occurs, followed by the discharge
>
> I was thinking about this for the ionization voltage. Usually tool is
> already at the discharge distance but in case it is to far away
> ionization just a little bit above limit would prevent discharge from
> happen to far away.
>
> > ...
> > .75 amps never bothered me, that measure is of the supplied current,
> > ( the current that would be delivered in a dead short)
> >   which is much more than delivered current
>
> It is the current delivered if I happen to touch tool that bother me,
> to high current and I wont be there to machine another day. There is a
> blue-green-yellow-red graph a bit down to the right on this page
> https://en.wikipedia.org/wiki/Electric_shock and I prefer to stay to
> the left of the yellow area.
>
> > tomp
>
> 10mA seems to be enough for ionization although I will try higher
> later to check if it make a difference. Sparc distinguishing seems to
> work at if short circuited there is no current. I however have some
> problem to keep the process running manually.

I should remind everyone that the lethal electric shock range is between 
20 micro-amps, and 20 milli-amps directly thru the heart muscle.  In 
this range fibrillation takes place and that is the killer because it 
continues after the shock has been removed.  Above 20 ma, the heart will 
freeze, and if the current is removed before you expire, it generally 
will start right back up when the current is removed.  You may have some 
2nd degree burns but they will heal.

The thing to watch out for as an aftereffect is the shingles as the 
body's immune system will take a huge hit while repairing other damages 
and may not be able to keep the chickenpox viri under control.

BTDT. Worst continuous, debilitating pain I have ever experienced, for 
about 2 months. I had no assistant at the time and finally at about 2 
weeks, had to get up and go to work and take care of the tv stations 
technical problems & broken equipment. I wasn't pleasant to be around as 
I had an extremely low tolerance for the idiocy the News Dept can 
perpetrate on equipment already, and it didn't get any better, but I did 
catch up eventually.

I would not wish the shingles on anyone, although I might make an 
exception for that pompous little twerp that runs North Korea. He will 
start the next war with his atomics.

> Regards Nicklas Karlsson
>
> > On 04/22/16 22:29, Nicklas Karlsson wrote:
> > > .75 amps is very dangerous as soon as voltage is high enough. Low
> > > current and it should not be dangerous even for rather high
> > > voltage.
> > >
> > > Preferably I want to stay at low current at least until prototype
> > > leave desktop.
> > >
> > > Since dielectric breakdown strength only has volts and distance I
> > > guess volt should be adjusted so that ionization is not started at
> > > to long distance which would create a big crater.
> > >
> > > I however want to keep current low to stay safe but does the
> > > current change how large a channel is created or will it grow
> > > large enough anyway as soon as the real current start?
> > >
> > >
> > > Regards Nicklas Karlsson
> > >
> > >
> > >
> > >
> > > On Thu, 21 Apr 2016 22:03:32 +0800
> > >
> > > Thomas Powderly  wrote:
> > >> initiation ONLY depends on voltage
> > >> i use .75 amps and its overkill
> > >> but it was easy to make the circuit
> > >> the equation for dielectric breakdown strength only has volts and
> > >> distance but
> > >> higher voltage makes ionization easier
> > >> typical value are 100 to 300VDC
> > >> tho for some paarungs ( stahl bis stahl, copper to crystal ... )
> > >> upwards of 1000V= are used
> > >> hth tomp
> > >>
> > >> On 04/21/16 20:13, Nicklas Karlsson wrote:
> > >>> I certainly discovered the need to back off.
> > >>>
> > >>> As is now I use only 10mA for ionization and 10A for cuts. I
> > >>> want to keep ionization current down for safety reason, do you
> > >>> know if it is harder to get process running at low current?
> > >>>
> > >>> Regards Nicklas Karlsson
> > >>>
> > >>>
> > >>>
> > >>> On Thu, 17 Mar 2016 15:15:10 + (UTC)
> > >>>
> > >>> Pete_Gruendeman  wrote:
> >  Hi Nicklas:
> >    Did you set up and run the simple RC type EDM?  It's a
> >  pretty simple proposition, using a filtered DC power supply, a
> >  capacitor in parallel with the gap and the gap itself.  An
> >  electrolytic capacitor would work.  A motor-run capacitor, of
> >  50-100 micro Farads would work even better.  Hands-on
> >  experience will answer many of you questions, especially in
> >  regard to the stability of the process.  No matter how stable
> >  your process is, 

Re: [Emc-users] Affordable, used lathes Re: Question on DM860 stepper drivers

[andy pugh]

On 23 April 2016 at 05:50, Gregg Eshelman  wrote:
> He needed a lathe long enough to handle the axle shafts.

This one is quite cheap at the moment.
https://www.apexauctions.co.uk/auction/itemDetails/197658

-- 
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

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Time estimates

[Danny Miller]

As I've said before, I'm used to Mach3.  In this case, Mach3 allowed me 
to click "Simulate Toolpath" and gave a 100% accurate figure for the 
file's runtime.  It actually simulated with the same path that would 
execute the code, just without waiting to step anything out, so it takes 
into account the max speed & accelerations of the machine.  It sometimes 
took a minute or 2 to evaluate large 3D files that took hours to run.

This is super-important for tuning.  I obsessed a lot about the best 
speeds & accelerations and it paid off.   Also kept me from getting into 
trouble with a job that would have taken 8 hrs.

I thought "Properties" would do that, but I was wrong.  That just takes 
distance, divides by feedrate, and adds them up.  This is not remotely 
valid for fine 3D carving, it requires so much acceleration that 
feedrate is almost irrelevant.  And I'd like to see what effect path 
tolerance has.  I'm not going to sit over the machine and run it with a 
stopwatch.  I tweak these parameters often.

Is there a way to do this?

Danny

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Affordable, used lathes

[Jeff Johnson]

I have found some incredible deals at HGR Surplus.

hgrinc.com 

Jeff Johnson
john...@superiorroll.com
Superior Roll & Turning
734-279-1831




--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] hands-on EDM experience (off topic)

[Gene Heskett]

On Saturday 23 April 2016 09:01:43 Nicklas Karlsson wrote:

> > > Thomas Powderly  wrote:
> > > > ...
> > > > in real work the tool is already at the discharge distance when
> > > > the initiation voltage is turned on
> > > > so the initiation still occurs, followed by the discharge
> > >
> > > I was thinking about this for the ionization voltage. Usually tool
> > > is already at the discharge distance but in case it is to far away
> > > ionization just a little bit above limit would prevent discharge
> > > from happen to far away.
> > >
> > > > ...
> > > > .75 amps never bothered me, that measure is of the supplied
> > > > current, ( the current that would be delivered in a dead short)
> > > >   which is much more than delivered current
> > >
> > > It is the current delivered if I happen to touch tool that bother
> > > me, to high current and I wont be there to machine another day.
> > > There is a blue-green-yellow-red graph a bit down to the right on
> > > this page https://en.wikipedia.org/wiki/Electric_shock and I
> > > prefer to stay to the left of the yellow area.
> > >
> > > > tomp
> > >
> > > 10mA seems to be enough for ionization although I will try higher
> > > later to check if it make a difference. Sparc distinguishing seems
> > > to work at if short circuited there is no current. I however have
> > > some problem to keep the process running manually.
> >
> > I should remind everyone that the lethal electric shock range is
> > between 20 micro-amps, and 20 milli-amps directly thru the heart
> > muscle.  In this range fibrillation takes place and that is the
> > killer because it continues after the shock has been removed.  Above
> > 20 ma, the heart will freeze, and if the current is removed before
> > you expire, it generally will start right back up when the current
> > is removed.  You may have some 2nd degree burns but they will heal.
> >
> > The thing to watch out for as an aftereffect is the shingles as the
> > body's immune system will take a huge hit while repairing other
> > damages and may not be able to keep the chickenpox viri under
> > control.
> >
> > BTDT. Worst continuous, debilitating pain I have ever experienced,
> > for about 2 months. I had no assistant at the time and finally at
> > about 2 weeks, had to get up and go to work and take care of the tv
> > stations technical problems & broken equipment. I wasn't pleasant to
> > be around as I had an extremely low tolerance for the idiocy the
> > News Dept can perpetrate on equipment already, and it didn't get any
> > better, but I did catch up eventually.
> >
> > I would not wish the shingles on anyone, although I might make an
> > exception for that pompous little twerp that runs North Korea. He
> > will start the next war with his atomics.
>
> So what is he supposed to do? Scrap the only weapon which could cause
> any significant damage on enemy and wait for the bombing and the
> following invasion?

He should quit the sabre rattling and use his starving but industrious 
people to develop his GDP and compete on the world market like SK does.  
Then he wouldn't have to worry about the bombs and invasions. He would 
have friendly relations with his neighbors.

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 

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Wildly OT: Strategic Musings [Was: hands-on EDM experience (off topic)]

[Erik Christiansen]

On 23.04.16 15:01, Nicklas Karlsson wrote:
> So what is he supposed to do? Scrap the only weapon which could cause
> any significant damage on enemy and wait for the bombing and the
> following invasion?

<2c>
His theatrics are almost exclusively for domestic consumption. IIRC,
he's already topped an uncle to maintain his position on the head of the
tiger, without slipping into the mouth. Where previously the carry-on
served to gain concessions such as oil and food, that doesn't seem to be
reported any more, and there doesn't seem to be much mileage there now.

If they did improve their unreliable launch vehicles, and make serious
radioactive trouble, then he'd be done in an hour, and glow for a long
time.

The only potential invader is the South, and the Giant Panda is
strategically reaching much further south at sea, complete with island
building. No serious invasion threat to the North exists, or can.

The only damage that would arise from ending the half-century-long
enduring state of war would be toppling of the regime once it's no
longer the protector against foreign bogeymen.


Erik

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] hands-on EDM experience (off topic)

[Nicklas Karlsson]

> > Thomas Powderly  wrote:
> > > ...
> > > in real work the tool is already at the discharge distance when the
> > > initiation voltage is turned on
> > > so the initiation still occurs, followed by the discharge
> >
> > I was thinking about this for the ionization voltage. Usually tool is
> > already at the discharge distance but in case it is to far away
> > ionization just a little bit above limit would prevent discharge from
> > happen to far away.
> >
> > > ...
> > > .75 amps never bothered me, that measure is of the supplied current,
> > > ( the current that would be delivered in a dead short)
> > >   which is much more than delivered current
> >
> > It is the current delivered if I happen to touch tool that bother me,
> > to high current and I wont be there to machine another day. There is a
> > blue-green-yellow-red graph a bit down to the right on this page
> > https://en.wikipedia.org/wiki/Electric_shock and I prefer to stay to
> > the left of the yellow area.
> >
> > > tomp
> >
> > 10mA seems to be enough for ionization although I will try higher
> > later to check if it make a difference. Sparc distinguishing seems to
> > work at if short circuited there is no current. I however have some
> > problem to keep the process running manually.
> 
> I should remind everyone that the lethal electric shock range is between 
> 20 micro-amps, and 20 milli-amps directly thru the heart muscle.  In 
> this range fibrillation takes place and that is the killer because it 
> continues after the shock has been removed.  Above 20 ma, the heart will 
> freeze, and if the current is removed before you expire, it generally 
> will start right back up when the current is removed.  You may have some 
> 2nd degree burns but they will heal.
> 
> The thing to watch out for as an aftereffect is the shingles as the 
> body's immune system will take a huge hit while repairing other damages 
> and may not be able to keep the chickenpox viri under control.
> 
> BTDT. Worst continuous, debilitating pain I have ever experienced, for 
> about 2 months. I had no assistant at the time and finally at about 2 
> weeks, had to get up and go to work and take care of the tv stations 
> technical problems & broken equipment. I wasn't pleasant to be around as 
> I had an extremely low tolerance for the idiocy the News Dept can 
> perpetrate on equipment already, and it didn't get any better, but I did 
> catch up eventually.
> 
> I would not wish the shingles on anyone, although I might make an 
> exception for that pompous little twerp that runs North Korea. He will 
> start the next war with his atomics.

So what is he supposed to do? Scrap the only weapon which could cause any 
significant damage on enemy and wait for the bombing and the following invasion?

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Time estimates

[Nicklas Karlsson]

> I'd rather try to do it the other way round, i.e. instead of simulating
> a faster machine, better simulate the same machine with faster time. In
> fact, it should be possible to make the servo thread as fast as possible
> for the computer, while telling it that the time passes at 1ms/iteration
> (or whatever the nominal servo period is). However, I don't know if
> there are other parts in LinuxCNC having their own timebase. And the
> interpreter would have to run much faster too, which (as far as I know)
> does not run in the servo thread.
> 
> So it will still be a bit tricky, but should be possible.
> 
> Regards,
> Philipp

I done the similar before instead of simulating a faster machine in sort of 
simulate the same machine with faster time although it was not a machine that 
time.

Have no idea how it is implemented right now but said it before. It should be 
simpler if thread is implemented as function called regularly. Then it could be 
called from: real time operating system, inside interrupt routine, from a loop 
in a simulation script.


Nicklas Karlsson

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Time estimates

[Danny Miller]

I used Mach3 in 3D raster carving extensively.  I never noticed any 
error whatsoever in the "Simulate Toolpath" time estimate.

It should not be done in CAM software.  That's just not possible. Now in 
the case of 2D cutting some plywood, changing machine acceleration will 
likely make <1% difference.  Who cares, sure.  I'm not that much of a 
perfectionist.

BUT, in the case of 3D raster carving, the distance/feedrate calc might 
say 45 min yet the actual runtime is 2 hrs.  Because acceleration 
becomes the deciding factor.   There's no telling what it'll be like.  
Sometimes sharp surface textures which aren't even these deep, dramatic 
features really drag down the runtime.

Also, blending and path tolerance will have considerable effect. That 
implementation is very specific to the machine and I doubt CAM software 
could reliably reproduce its effects.

It's very important to have access to these figures.  For example, in 3D 
carving, I discovered the machine almost never went faster than 150 
ipm.  It would go much faster with smoother surfaces, but it rarely 
occurred- what having a 300 ipm speed DID do was create some 
inconsistency in surface finish in those few smooth areas, with only a 
petty improvement in runtime.  It wasn't worth it.  I decided to go with 
125 ipm and that only increased runtime by like 5% over 150 ipm, but 
with very consistent finish.

Similarly, I had to make a call- my drive system is VERY effective.  It 
is technically capable of fantastically high accelerations, but it will 
shake the machine all over the floor and could prematurely wear 
components.  I can see what toning down the acceleration is doing to the 
3D performance and make a decision.

So now I need to understand the effects of path blending tolerance (G64 
P0.001) in addition to accelerations.  This is a per-design tradeoff, 
IMHO.  If I start from "get this job runtime down to ~90 min", well, if 
increasing the path tolerance to 0.05 decreases runtime by 20%, I can 
decrease raster stepover by 20% which could result in a net INCREASE in 
quality.

But I got no idea what effect G64 will have on runtime in the real 
world.  And I have no way of finding out anything unless I actually run 
jobs over and over and time them.  Considering I'd really need to get 
hundreds of datapoints to fully understand G64 path tolerance vs 
acceleration vs raster detail, that's just not ever going to do it.  I 
could spend years of my life trying to understand the real-world effects 
if there's no accurate simulation of it.

In fact, in Vectric Aspire, there's a Tolerance figure there- I already 
discovered that one.   Similar to G64 path tolerance. It defaults to a 
very low value and increasing it can substantially decrease runtime.

Danny

On 4/23/2016 10:50 AM, Dave Cole wrote:
> That is more usually a function of the Cam software that I am used to using.
>
> I don't know if Mach3 is all that reliable when it comes to accurate
> numbers for run time.
> I recall that being a problem actually in some situations.
>
> There is some discussion of that on the web.
> https://www.machsupport.com/forum/index.php?topic=18557.0
> There are several other links available.
>
> Dave
>
>
> On 4/23/2016 4:19 AM, Danny Miller wrote:
>> As I've said before, I'm used to Mach3.  In this case, Mach3 allowed me
>> to click "Simulate Toolpath" and gave a 100% accurate figure for the
>> file's runtime.  It actually simulated with the same path that would
>> execute the code, just without waiting to step anything out, so it takes
>> into account the max speed & accelerations of the machine.  It sometimes
>> took a minute or 2 to evaluate large 3D files that took hours to run.
>>
>> This is super-important for tuning.  I obsessed a lot about the best
>> speeds & accelerations and it paid off.   Also kept me from getting into
>> trouble with a job that would have taken 8 hrs.
>>
>> I thought "Properties" would do that, but I was wrong.  That just takes
>> distance, divides by feedrate, and adds them up.  This is not remotely
>> valid for fine 3D carving, it requires so much acceleration that
>> feedrate is almost irrelevant.  And I'd like to see what effect path
>> tolerance has.  I'm not going to sit over the machine and run it with a
>> stopwatch.  I tweak these parameters often.
>>
>> Is there a way to do this?
>>
>> Danny
>>
>> --
>> Find and fix application performance issues faster with Applications Manager
>> Applications Manager provides deep performance insights into multiple tiers 
>> of
>> your business applications. It resolves application problems quickly and
>> reduces your MTTR. Get your free trial!
>> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
> 

Re: [Emc-users] Time estimates

[Dave Cole]

That is more usually a function of the Cam software that I am used to using.

I don't know if Mach3 is all that reliable when it comes to accurate 
numbers for run time.
I recall that being a problem actually in some situations.

There is some discussion of that on the web.
https://www.machsupport.com/forum/index.php?topic=18557.0
There are several other links available.

Dave


On 4/23/2016 4:19 AM, Danny Miller wrote:
> As I've said before, I'm used to Mach3.  In this case, Mach3 allowed me
> to click "Simulate Toolpath" and gave a 100% accurate figure for the
> file's runtime.  It actually simulated with the same path that would
> execute the code, just without waiting to step anything out, so it takes
> into account the max speed & accelerations of the machine.  It sometimes
> took a minute or 2 to evaluate large 3D files that took hours to run.
>
> This is super-important for tuning.  I obsessed a lot about the best
> speeds & accelerations and it paid off.   Also kept me from getting into
> trouble with a job that would have taken 8 hrs.
>
> I thought "Properties" would do that, but I was wrong.  That just takes
> distance, divides by feedrate, and adds them up.  This is not remotely
> valid for fine 3D carving, it requires so much acceleration that
> feedrate is almost irrelevant.  And I'd like to see what effect path
> tolerance has.  I'm not going to sit over the machine and run it with a
> stopwatch.  I tweak these parameters often.
>
> Is there a way to do this?
>
> Danny
>
> --
> Find and fix application performance issues faster with Applications Manager
> Applications Manager provides deep performance insights into multiple tiers of
> your business applications. It resolves application problems quickly and
> reduces your MTTR. Get your free trial!
> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Time estimates

[Dave Cole]

Ironically I think that was one of the issues with Mach3's time 
estimates.   It ignored some of the effects of acceleration values.  My 
memory is coming back.   What really threw off Mach3
was short segments.  As in 3d printing, or cutting a lot of splined 
curves, or 3d contour cutting.   The time estimates would then be way 
off.  I did a bunch of Mach3 testing years ago.

It would be interesting to see if a Linuxcnc simulator setup could be 
done with increased velocities and accelerations to simulate a machining 
session running at 10x or 100x speed.

Dave

On 4/23/2016 12:02 PM, Philipp Burch wrote:
> Hi Dave!
>
> On 23.04.2016 17:50, Dave Cole wrote:
>> That is more usually a function of the Cam software that I am used to using.
> In my opinion, effects like limited acceleration should not be something
> the CAM has to care for. After all, we have G code as a (ideally)
> machine independent format, so it should not be necessary to tell the
> CAM software such limits. After all, LinuxCNC is already completely
> configured for the machine in question, so it could estimate the runtime
> almost exactly.
> I first thought about the simulation configurations, but these are most
> likely still linked to the wallclock. So it would be possible to
> simulate the tool path, but only in real time, which makes it less
> useful. But maybe such a feature is already present or in development?
>
> Regards,
> Philipp
>
>> I don't know if Mach3 is all that reliable when it comes to accurate
>> numbers for run time.
>> I recall that being a problem actually in some situations.
>>
>> There is some discussion of that on the web.
>> https://www.machsupport.com/forum/index.php?topic=18557.0
>> There are several other links available.
>>
>> Dave
>>
>>
>> On 4/23/2016 4:19 AM, Danny Miller wrote:
>>> As I've said before, I'm used to Mach3.  In this case, Mach3 allowed me
>>> to click "Simulate Toolpath" and gave a 100% accurate figure for the
>>> file's runtime.  It actually simulated with the same path that would
>>> execute the code, just without waiting to step anything out, so it takes
>>> into account the max speed & accelerations of the machine.  It sometimes
>>> took a minute or 2 to evaluate large 3D files that took hours to run.
>>>
>>> This is super-important for tuning.  I obsessed a lot about the best
>>> speeds & accelerations and it paid off.   Also kept me from getting into
>>> trouble with a job that would have taken 8 hrs.
>>>
>>> I thought "Properties" would do that, but I was wrong.  That just takes
>>> distance, divides by feedrate, and adds them up.  This is not remotely
>>> valid for fine 3D carving, it requires so much acceleration that
>>> feedrate is almost irrelevant.  And I'd like to see what effect path
>>> tolerance has.  I'm not going to sit over the machine and run it with a
>>> stopwatch.  I tweak these parameters often.
>>>
>>> Is there a way to do this?
>>>
>>> Danny
>>>
>>> --
>>> Find and fix application performance issues faster with Applications Manager
>>> Applications Manager provides deep performance insights into multiple tiers 
>>> of
>>> your business applications. It resolves application problems quickly and
>>> reduces your MTTR. Get your free trial!
>>> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
>>> ___
>>> Emc-users mailing list
>>> Emc-users@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/emc-users
>> --
>> Find and fix application performance issues faster with Applications Manager
>> Applications Manager provides deep performance insights into multiple tiers 
>> of
>> your business applications. It resolves application problems quickly and
>> reduces your MTTR. Get your free trial!
>> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
>
>
> --
> Find and fix application performance issues faster with Applications Manager
> Applications Manager provides deep performance insights into multiple tiers of
> your business applications. It resolves application problems quickly and
> reduces your MTTR. Get your free trial!
> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business 

Re: [Emc-users] Time estimates

[Gene Heskett]

On Saturday 23 April 2016 12:33:32 Philipp Burch wrote:

> Hi Dave!
>
> On 23.04.2016 18:21, Dave Cole wrote:
> > [...]
> >
> > It would be interesting to see if a Linuxcnc simulator setup could
> > be done with increased velocities and accelerations to simulate a
> > machining session running at 10x or 100x speed.
> >
> > Dave
>
> I'd rather try to do it the other way round, i.e. instead of
> simulating a faster machine, better simulate the same machine with
> faster time. In fact, it should be possible to make the servo thread
> as fast as possible for the computer, while telling it that the time
> passes at 1ms/iteration (or whatever the nominal servo period is).
> However, I don't know if there are other parts in LinuxCNC having
> their own timebase. And the interpreter would have to run much faster
> too, which (as far as I know) does not run in the servo thread.
>
> So it will still be a bit tricky, but should be possible.
>
> Regards,
> Philipp

I transferred the G0704's axis limits and homing of such to my sim on 
this machine, and thats been quite handy at locating bits of code that 
exceed the limits on the real machine.  Also to simulate a tool contact 
that I occasionally do to map something, or to measure TLO's is now 
simulated by a 1.5 second timer to satisfy the g38.2 function. Not at 
all accurate of course, but it lets me run the huge majority of my gcode 
concoctions in the sim for other error checking.

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 

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Wildly OT: Strategic Musings [Was: hands-on EDM experience (off topic)]

[Gene Heskett]

On Saturday 23 April 2016 09:38:37 Erik Christiansen wrote:

> On 23.04.16 15:01, Nicklas Karlsson wrote:
> > So what is he supposed to do? Scrap the only weapon which could
> > cause any significant damage on enemy and wait for the bombing and
> > the following invasion?
>
> <2c>
> His theatrics are almost exclusively for domestic consumption. IIRC,
> he's already topped an uncle to maintain his position on the head of
> the tiger, without slipping into the mouth. Where previously the
> carry-on served to gain concessions such as oil and food, that doesn't
> seem to be reported any more, and there doesn't seem to be much
> mileage there now.
>
> If they did improve their unreliable launch vehicles, and make serious
> radioactive trouble, then he'd be done in an hour, and glow for a long
> time.

That was my first thought too, but we'd have a heck of a lot bigger mess 
radioactively than chernobyl and fukushima combined.  Serious damage to 
the environment because he has his facilities scattered all over that  
piece of dirt. And the only other nuclear power to do that bit of atomic 
surgery would have a front row seat to that and they hopefully know it.  
So those chances are between very slim and point triple-ought zip.  So 
figure on conventional weapons and boots on the ground.  And that will 
cost either of us, or the Russians even, more money and bodies than 
we'll expend unless he sends a nuke to one of his perceived enemies.

> The only potential invader is the South, and the Giant Panda is
> strategically reaching much further south at sea, complete with island
> building. No serious invasion threat to the North exists, or can.

As long as he's friendly to the Chinese, who have given every clue they 
can without officially saying it out loud, that they are equally tired 
of his insanity.
>
> The only damage that would arise from ending the half-century-long
> enduring state of war would be toppling of the regime once it's no
> longer the protector against foreign bogeymen.
> 
>
> Erik
>
> --
> Find and fix application performance issues faster with
> Applications Manager Applications Manager provides deep performance
> insights into multiple tiers of your business applications. It
> resolves application problems quickly and reduces your MTTR. Get your
> free trial!
> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
> ___
> 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 

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Can't get XHC-HB04 MPG to work

[Danny Miller]

FOUND it- well, most of the problem. It appears the Wiki itself has an 
error here!  What you need is:

loadusr -W xhc-hb04 -I xhc-hb04-layout2.cfg -H

That "xhc-hb04-layout2" is the CONFIG file, not the ini.  I believe you 
can specify it without the file type suffix and it hunts down the .cfg, 
but if you specify an .ini suffix, it does as ordered and grabs the 
.ini.  An ini file doesn't any info it needs and it'll quietly execute 
and exit without making any HAL connections.  But no errors.

http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Using_A_XHC-HB04_Wireless_MPG_Pendant

Says "In xhc-hb04.hal you can change your keyboard layout preference: 
loadusr -W xhc-hb04 -I xhc-hb04-layout1.ini -H"

Well that doesn't look right.  Like I say, it needs the .cfg, the .ini 
won't work.  I found another web page guide that clued me into the 
".cfg" requirement.

Danny

On 4/20/2016 5:29 AM, Gene Heskett wrote:
> On Wednesday 20 April 2016 02:29:18 Danny Miller wrote:
>
>> It's loading without error.   The LCD works, the handwheel does jog
>> the axes, and the selector knob does change the axes.  But none of the
>> buttons function, including the Step Size, so it's stuck moving by
>> 0.001".
>>
>> I have LinuxCNC 2.7.4  and the Linux Preempt-RTK.  I did what was
>> specified about adding USB permissions.
>>
>> So 7i92_general_XHC20.ini is the main ini file. I added a
>> [XHC_HB04_CONFIG] section which gives setup parameters. I added
>> HALFILE = xhc-hb20.hal to the [HAL] section.
>>
>> xhc-hb20.hal is a renamed copy of a modified xhc-hb04.hal. The homing
>> routines don't exist, nor does an A-axis, so that's removed. Loadusr
>> got changed to the local name "loadusr -W xhc-hb04 -I
>> xhc-hb04-layout20.ini -H"
>>
>> xhc-hb04-layout20.ini has layout2.inc added into it directly. For
>> whatever reason layout2.inc didn't seem to be loading correctly into
>> it so I just added it.
>>
>> The HAL viewer does show there's an XHC section.  However (I'm not in
>> front of the machine now) IIRC the halui.mdi-command-01 etc didn't
>> seem to exist under halui.  I found what should have been the STEP++
>> pin  but it didn't show TRUE when I pressed it.  But I'm not sure how
>> that works, if it's a momentary press and I just manually reload the
>> HAL status I would probably never catch it.
>>
>> Thanks,
>> Danny
> The halmeter is pretty close to real time, so if you can find that
> signals name in the show-hal, halmeter can tell you whats going on in
> millisecond response times.
>
> This might be a case for a rockhopper run, but that, by the time it is
> blown up enough to be readable, will take posterizer set for 6 to 12
> pages on a complex machine to show it all legibly.  That can be a quite
> informative troubleshooting tool.  Rockhopper has one gotcha, it does
> not record and draw, a pin with no target defined.
>
> And many times, a sudo dmesg -c, start linuxcnc, then a fresh dmesg might
> show you errors.  All are handy tools.
>
> Cheers, Gene Heskett


--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Wildly OT: Strategic Musings [Was: hands-on EDM experience (off topic)]

[Nicklas Karlsson]

> > The only potential invader is the South, and the Giant Panda is
> > strategically reaching much further south at sea, complete with island
> > building. No serious invasion threat to the North exists, or can.
> 
> As long as he's friendly to the Chinese, who have given every clue they 
> can without officially saying it out loud, that they are equally tired 
> of his insanity.

Nuclear weapons are really insane. They are effective against civilians but 
more effective against surrounding than military in tanks and bunkers.

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Affordable, used lathes Re: Question on DM860 stepper drivers

[Gene Heskett]

On Saturday 23 April 2016 11:46:43 Jon Elson wrote:

> On 04/22/2016 11:50 PM, Gregg Eshelman wrote:
> > So don't buy any LeBlond or Monarch with broken or missing parts.
> > Except perhaps a 10EE. There's plenty of those laying about in all
> > kinds of  sub-functional condition.
>
> This applies to ALL older machines!  There are a few
> machines where a huge supply of spares is constantly
> floating around.  These are such as the Atlas/Craftsman
> lathes and the Bridgeport vertical mills.  There are
> probably a few others.  There are still parts being produced
> for Atlas (Clausing Service Center) and Bridgeport (now
> Hardinge).
>
> I bought a Sheldon lathe (Really TOP of the line design,
> definitely LeBlond or Monarch level, possibly even better).
> But, going into it, I knew that parts would not be available
> at any reasonable price.  When I got it, I discovered a
> broken gear tooth in the QC box, and was amazed to find that
> a replacement gear actually COULD be obtained.  But,
> DeVlieg-Bullard II (the owner of the rights to Sheldon)
> wanted about $450 to make the gear for me.
>
> But, most of these big, old, production machines are very
> well made, and other than a severe crash, can be kept
> running for quite some time.
>
> Jon
>
I looked at a Sheldon on the HGR site, $1799 IIRC, but it was a couple 
feet longer and about a ton heavier. Weight concerns me as my garage 
floor is about 7 to 8" of poured 5000 lb concrete, but not a rebar in 
the mix. And its already full of hairline cracks.

It did look to be fairly complete, even had a QC tool holder mounted in 
the pix.  But to get it here would have needed riggers and a freight 
line van to get it to my driveway.  And it would be quite the fun to get 
more than a 20 foot straight box van in AND back out of here. That would 
likely add at least another thou to the $ total.

But I'll keep the site in mind in case something shows up that I can haul 
& handle with a 2000 lb rated cherry picker.  That looks to be about a 
days drive each way for me.

I discovered something else last night, hoss's (Daniel Kemp of 
hossmachine.xxx) locale is right here in WV, way up in the panhandle, 
perhaps 100 miles NW of Pittsburgh.  He has some cute stuff as I saw on 
one of his BF20/G0704's, what can only be a worm drive head rotation to 
tilt the head left or right.  But I didn't find a video on that on his 
site.  Looked to be maybe 5/8" thick and would be handier than bottled 
beer for my machine.  Hand cranked worm, but I might figure out how to 
motorize it if I could unlock it easily. Currently on mine, thats 4 big 
bolts to loosen and quite some contortions to set it dead level for 
horizontal rigid tapping (which worked great BTW).

Thanks Jon.  If I do buy a Sheldon, I may pester you for hints & 
tricks. :)

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 

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Time estimates

[Philipp Burch]

Hi Dave!

On 23.04.2016 17:50, Dave Cole wrote:
> That is more usually a function of the Cam software that I am used to using.

In my opinion, effects like limited acceleration should not be something
the CAM has to care for. After all, we have G code as a (ideally)
machine independent format, so it should not be necessary to tell the
CAM software such limits. After all, LinuxCNC is already completely
configured for the machine in question, so it could estimate the runtime
almost exactly.
I first thought about the simulation configurations, but these are most
likely still linked to the wallclock. So it would be possible to
simulate the tool path, but only in real time, which makes it less
useful. But maybe such a feature is already present or in development?

Regards,
Philipp

> 
> I don't know if Mach3 is all that reliable when it comes to accurate 
> numbers for run time.
> I recall that being a problem actually in some situations.
> 
> There is some discussion of that on the web.
> https://www.machsupport.com/forum/index.php?topic=18557.0
> There are several other links available.
> 
> Dave
> 
> 
> On 4/23/2016 4:19 AM, Danny Miller wrote:
>> As I've said before, I'm used to Mach3.  In this case, Mach3 allowed me
>> to click "Simulate Toolpath" and gave a 100% accurate figure for the
>> file's runtime.  It actually simulated with the same path that would
>> execute the code, just without waiting to step anything out, so it takes
>> into account the max speed & accelerations of the machine.  It sometimes
>> took a minute or 2 to evaluate large 3D files that took hours to run.
>>
>> This is super-important for tuning.  I obsessed a lot about the best
>> speeds & accelerations and it paid off.   Also kept me from getting into
>> trouble with a job that would have taken 8 hrs.
>>
>> I thought "Properties" would do that, but I was wrong.  That just takes
>> distance, divides by feedrate, and adds them up.  This is not remotely
>> valid for fine 3D carving, it requires so much acceleration that
>> feedrate is almost irrelevant.  And I'd like to see what effect path
>> tolerance has.  I'm not going to sit over the machine and run it with a
>> stopwatch.  I tweak these parameters often.
>>
>> Is there a way to do this?
>>
>> Danny
>>
>> --
>> Find and fix application performance issues faster with Applications Manager
>> Applications Manager provides deep performance insights into multiple tiers 
>> of
>> your business applications. It resolves application problems quickly and
>> reduces your MTTR. Get your free trial!
>> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
> 
> --
> Find and fix application performance issues faster with Applications Manager
> Applications Manager provides deep performance insights into multiple tiers of
> your business applications. It resolves application problems quickly and
> reduces your MTTR. Get your free trial!
> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
> 



signature.asc
Description: OpenPGP digital signature
--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Affordable, used lathes Re: Question on DM860 stepper drivers

[Jon Elson]

On 04/22/2016 11:50 PM, Gregg Eshelman wrote:
> So don't buy any LeBlond or Monarch with broken or missing parts. Except 
> perhaps a 10EE. There's plenty of those laying about in all kinds of  
> sub-functional condition.
>   
>
This applies to ALL older machines!  There are a few 
machines where a huge supply of spares is constantly 
floating around.  These are such as the Atlas/Craftsman 
lathes and the Bridgeport vertical mills.  There are 
probably a few others.  There are still parts being produced 
for Atlas (Clausing Service Center) and Bridgeport (now 
Hardinge).

I bought a Sheldon lathe (Really TOP of the line design, 
definitely LeBlond or Monarch level, possibly even better).
But, going into it, I knew that parts would not be available 
at any reasonable price.  When I got it, I discovered a 
broken gear tooth in the QC box, and was amazed to find that 
a replacement gear actually COULD be obtained.  But, 
DeVlieg-Bullard II (the owner of the rights to Sheldon) 
wanted about $450 to make the gear for me.

But, most of these big, old, production machines are very 
well made, and other than a severe crash, can be kept 
running for quite some time.

Jon

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Time estimates

[Philipp Burch]

Hi Dave!

On 23.04.2016 18:21, Dave Cole wrote:
> [...]
> 
> It would be interesting to see if a Linuxcnc simulator setup could be 
> done with increased velocities and accelerations to simulate a machining 
> session running at 10x or 100x speed.
> 
> Dave
> 

I'd rather try to do it the other way round, i.e. instead of simulating
a faster machine, better simulate the same machine with faster time. In
fact, it should be possible to make the servo thread as fast as possible
for the computer, while telling it that the time passes at 1ms/iteration
(or whatever the nominal servo period is). However, I don't know if
there are other parts in LinuxCNC having their own timebase. And the
interpreter would have to run much faster too, which (as far as I know)
does not run in the servo thread.

So it will still be a bit tricky, but should be possible.

Regards,
Philipp



signature.asc
Description: OpenPGP digital signature
--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Semi OT: Heidenhain incremental encoders

[Nicklas Karlsson]

> Anyway my only doubt now is how much better are differential encoders
> compared to normal ones. I always used normal encoders with no signs of
> noise but since this is for a lathe with 0.01 mm increments I would like to
> go with the safest option.
> 
> As always Andy, thanks you!

The common ground is a problem but it also depends on cable length. Basic 
problem is then there is a flank on one of the signals ground bounce in the 
receiving end. With differential signals ground current stay there it is and it 
will not affect the other signal.

You also need a differential receiver. It is important current at transitions 
are between differential signals and not the ground.


Nicklas Karlsson

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Wildly OT: Strategic Musings [Was: hands-on EDM experience (off topic)]

[Leonardo Marsaglia]

All of this reminded me about the great Carl Sagan pale blue dot monologue.
Here's the link in case you want to take a look.

https://www.youtube.com/watch?v=wupToqz1e2g

2016-04-23 15:27 GMT-03:00 Nicklas Karlsson :

> > > The only potential invader is the South, and the Giant Panda is
> > > strategically reaching much further south at sea, complete with island
> > > building. No serious invasion threat to the North exists, or can.
> >
> > As long as he's friendly to the Chinese, who have given every clue they
> > can without officially saying it out loud, that they are equally tired
> > of his insanity.
>
> Nuclear weapons are really insane. They are effective against civilians
> but more effective against surrounding than military in tanks and bunkers.
>
>
> --
> Find and fix application performance issues faster with Applications
> Manager
> Applications Manager provides deep performance insights into multiple
> tiers of
> your business applications. It resolves application problems quickly and
> reduces your MTTR. Get your free trial!
> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



-- 
*Leonardo Marsaglia*.
--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Semi OT: Heidenhain incremental encoders

[andy pugh]

On 23 April 2016 at 22:29, Leonardo Marsaglia
 wrote:
> That's what I was not sure about them. And I can't find the manuals of
> these series so I can't check wheter I can wire them directly as TTL or if
> I need that interpolator box to make them work.


The description says that they are normal TTL output, and that he has
the pinout available.

-- 
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

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Can't get XHC-HB04 MPG to work

[andy pugh]

On 23 April 2016 at 19:25, Danny Miller  wrote:
> FOUND it- well, most of the problem. It appears the Wiki itself has an
> error here!

As a general observation: A lot of the Wiki is simply wrong. It is a
collection of pages written by users. Some are massively out of date,
some are just plain wrong.

If you find something wrong in the Wiki, please correct it.

-- 
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

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Excedrin headache of a near infinite number

[John Thornton]

I'm confused by your home offset, I think normally you home the X axis 
toward you and call that X0 with maybe enough home offset to get off the 
switch. Soft limits seem trivial after homing move to the limits with no 
offsets and record the numbers. Adjust your ini file and test.

JT

On 4/23/2016 6:28 PM, Gene Heskett wrote:
> Greetings all;
>
> I have never been able to make sense out of the homing vs soft limits on
> this darned lathe.
>
> Running it in mm mode (g21), and considering that with my tool holder
> shortage, a QC tool post I often have to rotate oddly to get tool access
> to the work, my homing sequence assigns some relative lavues that work,
> and leave the carriage setting about 3/8 in from the outer switch.
>
> It screwed up, running the wrong direction when I hit ctl+home today, and
> mangled, for the umptieth time, the brass cover over the rear extension
> of the x ball screw.  So now I am bound to make the soft limits work
> because my x screw sounds like some of Orville's Redenbachers popcorn
> now from all the dirt that has migrated into that area behind the
> carriage, caused by previous damage to this cover.  My X motor is on the
> rear of the carriage so I have a carriage extension just thick enough so
> the motor mount clears, and the area in front of it covered by a glued
> up contraption of brass sheet, old felt hat for dirt seals, and light
> alu angle for the upper corners, all hopefully glued in place with Goop.
>
> Playing the 10,000 monkeys, I finally did get an x min limit that works,
> but the number bears minus zero resemblance to what I see in the DRO.
>
> But setting the max limit to 72, where the home offset is 74 and its
> final parking place is 70, actually gets me after it has homed x, a
> message that axis 0 is beyond the software limits.
>
> Considering that when I run inward after homing, it stops with the dro
> showing -55mm diameter or -27.5 radius, while what I think is the figure
> in the ini that stops it says -2.5mm.
>
> At one point I homed it, which left it sitting 4mm (assuming that is
> radius diff and that we are indeed measuring millimeters) from the home
> switch, then did a g53 g1F150 x0, and it ran in and crushed the brass
> cover again, after having spent at least half an hour hammering it out
> from the last such incident, can someone please explain to me a bullet
> proof procedure to set this stuff?
>
> Here is the current axis 0 section in the .ini file
> ===
> #
> # Axis X
> #
> [AXIS_0]
> TYPE = LINEAR
> HOME = 60.0
> FERROR = 1.0
> MIN_FERROR = 0.1
> MAX_VELOCITY = 25.0
> MAX_ACCELERATION = 350.0
> # The values below should be 25% larger than MAX_VELOCITY and
> MAX_ACCELERATION
> # If using BACKLASH compensation STEPGEN_MAXACCEL should be 100% larger.
> STEPGEN_MAXVEL = 50.0
> STEPGEN_MAXACCEL = 700.0
> P = 50.0
> I = 0.0
> D = 0.0
> FF0 = 0.0
> FF1 = 1.0
> FF2 = 0.0
> BIAS = 0.0
> DEADBAND = 0.0
> MAX_OUTPUT = 40.0
> # these are in nanoseconds
> DIRSETUP   = 6000
> DIRHOLD= 6000
> STEPLEN= 2500
> STEPSPACE  = 2500
> STEP_SCALE = 315.150
> BACKLASH = 0.005
> MIN_LIMIT = -2.5
> MAX_LIMIT = 72.62
> HOME_OFFSET = 70.0
> HOME_SEARCH_VEL = -10.00
> HOME_LATCH_VEL = -2.17
> HOME_FINAL_VEL = 18.33
> HOME_USE_INDEX = NO
> HOME_SEQUENCE = 0
> =
>
> Thanks all.
>
> Cheers, Gene Heskett


--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Semi OT: Heidenhain incremental encoders

[Leonardo Marsaglia]

2016-04-22 14:05 GMT-03:00 andy pugh :

>
> Heidenhain mainly do sin/cos encoders and one uses an interpolation
> box to convert to TTL.
> I suspect that these encoders have a built-in 25x interpolator to give
> the 5000ppr.
>
> The price looks great, especially including the fancy mounts.


That's what I was not sure about them. And I can't find the manuals of
these series so I can't check wheter I can wire them directly as TTL or if
I need that interpolator box to make them work.

There's another link with a gorgeous Allen Bradley encoder but they only
have one.

http://www.ebay.com/itm/ALLEN-BRADLEY-845H-SJDZ14DRY2C-Optical-Incremental-Encoder-5000-PPR-Type-4-13-/281274676932?hash=item417d46f6c4:g:w0gAAOxy0bRTDnbv

Anyway my only doubt now is how much better are differential encoders
compared to normal ones. I always used normal encoders with no signs of
noise but since this is for a lathe with 0.01 mm increments I would like to
go with the safest option.

As always Andy, thanks you!



-- 
*Leonardo Marsaglia*.
--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Affordable, used lathes Re: Question on DM860 stepper drivers

[Gene Heskett]

On Saturday 23 April 2016 19:23:52 Jon Elson wrote:

> On 04/23/2016 01:38 PM, Gene Heskett wrote:
> > I looked at a Sheldon on the HGR site, $1799 IIRC, but it was a
> > couple feet longer and about a ton heavier. Weight concerns me as my
> > garage floor is about 7 to 8" of poured 5000 lb concrete, but not a
> > rebar in the mix. And its already full of hairline cracks.
>
> Sheldon made a range of machines.  The only one that is a
> "toolroom precision" grade is the R-series.
> I have an R-15.  80 threading feeds, feed by driveshaft
> separate from leadscrew.  It has clutches for carriage and
> crossfeed that can be adjusted to slip, so you don't break
> things if you have a crash.
> Spindle is D1-6, with a 2.25" through hole.  The standard
> model will do 40 - 1250 RPM, the high speed model (which I
> don't have) will go up to 2500 RPM!  Yikes!
>
> > Thanks Jon.  If I do buy a Sheldon, I may pester you for hints &
> > tricks. :)
>
> There is a newsgroup, and one of the family members (John
> Knox) is there, and he has access to a wealth of info.  If
> you need a standard part that was in there, he can get the
> maker and their part number.  I needed to replace some seals
> in the apron, and with his info, I was able to order them
> locally very easily.
>
> I only know the R-series Sheldons.
>
> Jon

Thanks Jon, I'll mentally limit my Sheldon search for models starting 
with an R.

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 

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Semi OT: Heidenhain incremental encoders

[Leonardo Marsaglia]

2016-04-23 18:44 GMT-03:00 Nicklas Karlsson :

> The common ground is a problem but it also depends on cable length. Basic
> problem is then there is a flank on one of the signals ground bounce in the
> receiving end. With differential signals ground current stay there it is
> and it will not affect the other signal.
>
>
Well the longest cable would be approximately 2 meters long. I guess to be
sure I will go with differential.


> You also need a differential receiver. It is important current at
> transitions are between differential signals and not the ground.
>

I'm planning to use the Mesa 7i33 that I have bought but never used so I
have the option of driving normal encoders or differential ones.

Thank you Nick!


-- 
*Leonardo Marsaglia*.
--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Affordable, used lathes Re: Question on DM860 stepper drivers

[Jon Elson]

On 04/23/2016 01:38 PM, Gene Heskett wrote:
>
> I looked at a Sheldon on the HGR site, $1799 IIRC, but it was a couple
> feet longer and about a ton heavier. Weight concerns me as my garage
> floor is about 7 to 8" of poured 5000 lb concrete, but not a rebar in
> the mix. And its already full of hairline cracks.
Sheldon made a range of machines.  The only one that is a 
"toolroom precision" grade is the R-series.
I have an R-15.  80 threading feeds, feed by driveshaft 
separate from leadscrew.  It has clutches for carriage and 
crossfeed that can be adjusted to slip, so you don't break 
things if you have a crash.
Spindle is D1-6, with a 2.25" through hole.  The standard 
model will do 40 - 1250 RPM, the high speed model (which I 
don't have) will go up to 2500 RPM!  Yikes!
>
>
> Thanks Jon.  If I do buy a Sheldon, I may pester you for hints &
> tricks. :)
>
There is a newsgroup, and one of the family members (John 
Knox) is there, and he has access to a wealth of info.  If 
you need a standard part that was in there, he can get the 
maker and their part number.  I needed to replace some seals 
in the apron, and with his info, I was able to order them 
locally very easily.

I only know the R-series Sheldons.

Jon

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Excedrin headache of a near infinite number

[Gene Heskett]

Greetings all;

I have never been able to make sense out of the homing vs soft limits on 
this darned lathe.

Running it in mm mode (g21), and considering that with my tool holder 
shortage, a QC tool post I often have to rotate oddly to get tool access 
to the work, my homing sequence assigns some relative lavues that work, 
and leave the carriage setting about 3/8 in from the outer switch.

It screwed up, running the wrong direction when I hit ctl+home today, and  
mangled, for the umptieth time, the brass cover over the rear extension 
of the x ball screw.  So now I am bound to make the soft limits work 
because my x screw sounds like some of Orville's Redenbachers popcorn 
now from all the dirt that has migrated into that area behind the 
carriage, caused by previous damage to this cover.  My X motor is on the 
rear of the carriage so I have a carriage extension just thick enough so 
the motor mount clears, and the area in front of it covered by a glued 
up contraption of brass sheet, old felt hat for dirt seals, and light 
alu angle for the upper corners, all hopefully glued in place with Goop.

Playing the 10,000 monkeys, I finally did get an x min limit that works, 
but the number bears minus zero resemblance to what I see in the DRO.

But setting the max limit to 72, where the home offset is 74 and its 
final parking place is 70, actually gets me after it has homed x, a 
message that axis 0 is beyond the software limits.

Considering that when I run inward after homing, it stops with the dro 
showing -55mm diameter or -27.5 radius, while what I think is the figure 
in the ini that stops it says -2.5mm.

At one point I homed it, which left it sitting 4mm (assuming that is  
radius diff and that we are indeed measuring millimeters) from the home 
switch, then did a g53 g1F150 x0, and it ran in and crushed the brass 
cover again, after having spent at least half an hour hammering it out 
from the last such incident, can someone please explain to me a bullet 
proof procedure to set this stuff?

Here is the current axis 0 section in the .ini file
===
#
# Axis X
#
[AXIS_0]
TYPE = LINEAR
HOME = 60.0
FERROR = 1.0
MIN_FERROR = 0.1
MAX_VELOCITY = 25.0
MAX_ACCELERATION = 350.0
# The values below should be 25% larger than MAX_VELOCITY and 
MAX_ACCELERATION
# If using BACKLASH compensation STEPGEN_MAXACCEL should be 100% larger.
STEPGEN_MAXVEL = 50.0
STEPGEN_MAXACCEL = 700.0
P = 50.0
I = 0.0
D = 0.0
FF0 = 0.0
FF1 = 1.0
FF2 = 0.0
BIAS = 0.0
DEADBAND = 0.0
MAX_OUTPUT = 40.0
# these are in nanoseconds
DIRSETUP   = 6000
DIRHOLD= 6000
STEPLEN= 2500
STEPSPACE  = 2500
STEP_SCALE = 315.150
BACKLASH = 0.005
MIN_LIMIT = -2.5
MAX_LIMIT = 72.62
HOME_OFFSET = 70.0
HOME_SEARCH_VEL = -10.00
HOME_LATCH_VEL = -2.17
HOME_FINAL_VEL = 18.33
HOME_USE_INDEX = NO
HOME_SEQUENCE = 0
=

Thanks all.

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 

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Affordable, used lathes Re: Question on DM860 stepper drivers

[andy pugh]

On 24 April 2016 at 00:23, Jon Elson  wrote:
> Sheldon made a range of machines.  The only one that is a
> "toolroom precision" grade is the R-series.

Intersting, I can't find them at http://www.lathes.co.uk/sheldon/index.html

-- 
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

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Affordable, used lathes Re: Question on DM860 stepper drivers

[Jon Elson]

On 04/23/2016 06:33 PM, andy pugh wrote:
> On 24 April 2016 at 00:23, Jon Elson  wrote:
>> Sheldon made a range of machines.  The only one that is a
>> "toolroom precision" grade is the R-series.
> Intersting, I can't find them at http://www.lathes.co.uk/sheldon/index.html
>
The Sheldon R-series was their high-end lathe, to compete 
with LeBlond and other toolroom lathes.
The early R-series was fairly advanced, then they re-did it 
in about 1968.  After that, they had an 80-speed QC gearbox, 
and added a feed shaft parallel to the threading leadscrew.  
Feeds are via clutches, so that an overload won't break gears.

The main gearbox is in the pedestal, so the only gears in 
the headstock are the backgear.

The headstock casting is about the size of a smallblock V8, 
and similar weight.  Even the tailstock is too heavy for me 
to pick up without separating the upper and lower half.  The 
cross feed dial is about 4" diameter.
There are a bunch of other construction details that show 
how hard they thought about every single bit of the lathe.

See http://pico-systems.com/sheldon.html  for the saga of 
getting the lathe moved inside, and some pics of it.

Jon

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Excedrin headache of a near infinite number

[andy pugh]

On 24 April 2016 at 00:28, Gene Heskett  wrote:
> but the number bears minus zero resemblance to what I see in the DRO.

That's normal, unless you are displaying machine coordinates.

-- 
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

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Excedrin headache of a near infinite number

[Gene Heskett]

On Saturday 23 April 2016 19:49:52 andy pugh wrote:

> On 24 April 2016 at 00:28, Gene Heskett  wrote:
> > but the number bears minus zero resemblance to what I see in the
> > DRO.
>
> That's normal, unless you are displaying machine coordinates.

How do I make that the default display?  It wants to display relative 
until I change it.  Is that something in the gcode preamble I have 
screwed up?

That worked a treat by the way Andy, thanks.

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 

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Excedrin headache of a near infinite number

[Jon Elson]

On 04/23/2016 10:44 PM, Gene Heskett wrote:
> On Saturday 23 April 2016 19:49:52 andy pugh wrote:
>
>> On 24 April 2016 at 00:28, Gene Heskett  wrote:
>>> but the number bears minus zero resemblance to what I see in the
>>> DRO.
>> That's normal, unless you are displaying machine coordinates.
> How do I make that the default display?  It wants to display relative
> until I change it.  Is that something in the gcode preamble I have
> screwed up?
>
>
I think you can switch between machine and part systems with 
either @ or #.  (I rarely use these, so I have forgotten.)  
You generally never use machine coordinates except when 
setting up limit and home switches.
You set the relative, or part, coordinates with the 
touch-off button in Axis after measuring/aligning to the 
workpiece.

Jon

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Time estimates

[Dave Cole]

On 4/23/2016 2:44 PM, Danny Miller wrote:
> It should not be done in CAM software.  That's just not possible.

http://support.vectric.com/aspire-questions/item/how-to-get-estimate-of-machining-time

I can't think of a cam system that doesn't have a time estimate function.
How accurate they are is a different issue.

Dave

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users