Re: [Emc-users] Trouble with EMC spiral example

[Jeffrey Pease]

 > Jeffrey Pease wrote:
 >> I have no explanation for why this glitch is occurring in these  
spots,
 >> and no real idea how to fix it. Any ideas?
 >>
 >> My machine is using steppers and lead screws. Running EMC2 on Ubuntu
 >> 8.04.

Sebastian Kuzminsky wrote:

 > Which version of emc2?
EMC2 2.2.8
 > What hardware is driving your steppers?
I'm using the EZ Driver Board from HobbyCNC.com. Perhaps more  
usefully, this board uses SLA7078MR stepper drivers, datasheet here -> 
http://www.datasheetarchive.com/pdf-datasheets/Datasheets-29/DSA-575103.pdf
My board is configured for single stepping on all axes.
 > What config are you using?
 > What are your latency numbers like?
17901 ns latency, measured over almost an hour of messing around on  
the machine. I have both dirhold and dirstep for each motor set to  
37901 based on the tuning info found in Chapter 28 of the Integrator  
manual.

Hope that helps. Sorry, I know this is probably a tough issue to  
visualize as explained. Any suggestions would be appreciated.

Thanks,
Jeff


> Hi EMC,
>
> I'm tuning my machine, and I am having some weird problems with the
> EMC spiral example. This issue is entirely repeatable, and seems to
> happen without fail regardless of any feed rate, acceleration or speed
> settings that I have tried.
>
> What happens is on a particular place in every arc, the X and Y axis
> both seem to stutter in the same way - pausing for a second before
> continuing to finish the arc. This occurs in the area of the arc right
> before each axis changes direction to start drawing the other half of
> the arc - so the X axis stutters at 2:30, 3:30, 8:30, and the 9:30
> hour positions (the extreme left and right sides of each arc of the
> spiral) and the Y axis stutters at 12:30, 5:30, 6:30 and 11:30 (the
> extreme top and bottom of each arc). This strikes me as an unintuitive
> spot for this stuttering to occur for each axis, since it seems like
> the actual movement required at these points in each axis is
> relatively small - once everything speeds up, it seems fine again.
>
> I have lubricated all the guide rods and lead screws (thinking it
> might be some sort of binding issue), but that has not helped.
>
> The spiral appears to be perfectly smooth in these areas on the EMC
> preview, so I don't believe it's a programming issue.
>
> I have no explanation for why this glitch is occurring in these spots,
> and no real idea how to fix it. Any ideas?
>
> My machine is using steppers and lead screws. Running EMC2 on Ubuntu
> 8.04.
>
> Thanks,
> Jeff
>
> --

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

Re: [Emc-users] Trouble with EMC spiral example

[Leslie Newell]

Are you using backlash compensation? If so, what is your backlash 
compensation speed?

Les

Jeffrey Pease wrote:
> Hi EMC,
> 
> I'm tuning my machine, and I am having some weird problems with the  
> EMC spiral example. This issue is entirely repeatable, and seems to  
> happen without fail regardless of any feed rate, acceleration or speed  
> settings that I have tried.
> 
> What happens is on a particular place in every arc, the X and Y axis  
> both seem to stutter in the same way - pausing for a second before  
> continuing to finish the arc. This occurs in the area of the arc right  
> before each axis changes direction to start drawing the other half of  
> the arc - so the X axis stutters at 2:30, 3:30, 8:30, and the 9:30  
> hour positions (the extreme left and right sides of each arc of the  
> spiral) and the Y axis stutters at 12:30, 5:30, 6:30 and 11:30 (the  
> extreme top and bottom of each arc). This strikes me as an unintuitive  
> spot for this stuttering to occur for each axis, since it seems like  
> the actual movement required at these points in each axis is  
> relatively small - once everything speeds up, it seems fine again.
> 
> I have lubricated all the guide rods and lead screws (thinking it  
> might be some sort of binding issue), but that has not helped.
> 
> The spiral appears to be perfectly smooth in these areas on the EMC  
> preview, so I don't believe it's a programming issue.
> 
> I have no explanation for why this glitch is occurring in these spots,  
> and no real idea how to fix it. Any ideas?
> 
> My machine is using steppers and lead screws. Running EMC2 on Ubuntu  
> 8.04.
> 
> Thanks,
> Jeff
> 
> --
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users

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

Re: [Emc-users] Trouble with EMC spiral example

[Jeffrey Pease]

Les wrote:
> Are you using backlash compensation? If so, what is your backlash
> compensation speed?
>
> Les
>

No, no backlash compensation (at last not that I know of - I guess  
it's possible that stepconf snuck something in though). My lead screw  
nuts are of the anti-backlash variety, so I am not doing anything  
special for backlash in software.
Thanks,
Jeff

Jeffrey Pease wrote:
 > Hi EMC,
 >
 > I'm tuning my machine, and I am having some weird problems with the
 > EMC spiral example. This issue is entirely repeatable, and seems to
 > happen without fail regardless of any feed rate, acceleration or  
speed
 > settings that I have tried.
 >
 > What happens is on a particular place in every arc, the X and Y axis
 > both seem to stutter in the same way - pausing for a second before
 > continuing to finish the arc. This occurs in the area of the arc  
right
 > before each axis changes direction to start drawing the other half of
 > the arc - so the X axis stutters at 2:30, 3:30, 8:30, and the 9:30
 > hour positions (the extreme left and right sides of each arc of the
 > spiral) and the Y axis stutters at 12:30, 5:30, 6:30 and 11:30 (the
 > extreme top and bottom of each arc). This strikes me as an  
unintuitive
 > spot for this stuttering to occur for each axis, since it seems like
 > the actual movement required at these points in each axis is
 > relatively small - once everything speeds up, it seems fine again.
 >
 > I have lubricated all the guide rods and lead screws (thinking it
 > might be some sort of binding issue), but that has not helped.
 >
 > The spiral appears to be perfectly smooth in these areas on the EMC
 > preview, so I don't believe it's a programming issue.
 >
 > I have no explanation for why this glitch is occurring in these  
spots,
 > and no real idea how to fix it. Any ideas?
 >
 > My machine is using steppers and lead screws. Running EMC2 on Ubuntu
 > 8.04.
 >
 > Thanks,
 > Jeff
 >
 >  
--
 > ___
 > Emc-users mailing list
 > emc-us...@...
 > https://lists.sourceforge.net/lists/listinfo/emc-users

--

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

[Emc-users] Sunix Sun1888 parports broken in in-mode

[Haberler Michael]

after spending disordinate amounts of time I'd warn folks to use Sunix  
chipset parallel port cards

I have several Sunix 4008T parallel pci cards (distributed by Digitus)  
which show the same symptom: one cannot reliably set 'in' mode for  
pins 02-09. They seem to work ok in default (out) mode.

After swapping for a card with Netmos 9835 chipset (1xparallel, 
2xserial) all problems in in-mode are gone.

-Michael

emc2 version is 2.4.0 pre .

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

Re: [Emc-users] Trouble with EMC spiral example

[Dave Caroline]

steppers do go quiet on spiral as there is little to do in places

Dave Caroline

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

[Emc-users] how to debuild a changed .deb package

[Lindenmayr Bernd]

Hi EMC!

 

I use EMC2 for my final project. Therefore I did some changes in the souce
code and add some new hal - components to  EMC2 version emc2_2.4.0~pre. Now
I want to build the .deb packages, but I have troubles with gpg debsign.
Which steps I have to do for signing this new version? 

 

Thanks,

 

Bernd Lindenmayr

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

Re: [Emc-users] how to debuild a changed .deb package

[Alex Joni]

None, there is no need to sign the new built debs.
We only need to do that when we distribute the packages officially through 
www.linuxcnc.org.

Check one dir up, you should have the new deb's there, if no other error 
occured.

Regards,
Alex


- Original Message - 
From: "Lindenmayr Bernd" 
To: 
Sent: Wednesday, July 22, 2009 12:05
Subject: [Emc-users] how to debuild a changed .deb package


> Hi EMC!
>
>
>
> I use EMC2 for my final project. Therefore I did some changes in the souce
> code and add some new hal - components to  EMC2 version emc2_2.4.0~pre. 
> Now
> I want to build the .deb packages, but I have troubles with gpg debsign.
> Which steps I have to do for signing this new version?
>
>
>
> Thanks,
>
>
>
> Bernd Lindenmayr
>
> --
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
> 


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

Re: [Emc-users] Emc and Toolchange (Rack type toolchanger)

[Niels Jalling]

It doesn't work :-(

So someone with source insight please step in! How do I configure emc2 to
use a rack-tool-changer placed parallel to the Y-axis?

I'm trying to use the patch from
http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?action=browse&diff=1&id=RackToolChanger

/Niels


> 2009/7/21 Niels Jalling :
>
>> Any hints how to get the rack to be parallel to the y-axis instead of the
>> x-axis?
>
> It looks like a fault in the parch file. It seems to read the
> increments into global variables, but doesn't use them.
>
> Bear in mind I am not a trained programmer, and have never programmed in
C, but I think the problem is here in the patch file:
>
> +generate_toolchange_move(
> +   FROM_EXT_LEN(TOOL_CHANGE_POSITION.tran.x) +
> slot*FROM_EXT_LEN(TOOL_SLOT_DELTA.tran.x),
> +   canonEndPoint.y, canonEndPoint.z, a, b, c, u, v, w);
>
> I think it should read contain references to TOOL_SLOT_DELTA.tran.y and
TOOL_SLOT_delta.tran.z too. Perhaps this?
>
> +generate_toolchange_move(
> +   FROM_EXT_LEN(TOOL_CHANGE_POSITION.tran.x) +
> slot*FROM_EXT_LEN(TOOL_SLOT_DELTA.tran.x),
> +   FROM_EXT_LEN(TOOL_CHANGE_POSITION.tran.y) +
> slot*FROM_EXT_LEN(TOOL_SLOT_DELTA.tran.y),
> +   FROM_EXT_LEN(TOOL_CHANGE_POSITION.tran.z) +
> slot*FROM_EXT_LEN(TOOL_SLOT_DELTA.tran.z),
> +   a, b, c, u, v, w);
>
> But I really, really, don't suggest you try this until someone with at
least a small clue concurs with me.
>
> --
> atp
>




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

Re: [Emc-users] how to debuild a changed .deb package

[Jeff Epler]

The key chosen for signing is the name listed in the most recent
debian/changelog entry.  You should write a changelog entry using the
'dch' command, which when properly configured will set that name to your
name.  This is a good idea, since you should also describe the changes
you've made, and give a version number that clearly identifies the
build as an unofficial one, such as 2.4.0~bernd0.

(configuring dch mostly consists of setting the environment variables
DEBEMAIL and DEBFULLNAME to appropriate values)

You can also choose not to sign the packages.  Far down in the debuild
manpage:
   The typical command line options to build only  the  binary
   package(s) without signing the .changes file (or the non-existent
   .dsc file):

  debuild -i -us -uc -b

Jeff

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

[Emc-users] PC/104 Install

[Pablo Pais]

Hello,
Someone knows about install linuxcnc on a PC/104 sistem.
Thanks
Pablo
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Trouble with EMC spiral example

[BRIAN GLACKIN]

What is your threading TPI on the lead screws.  In addition, have you tried
different stepping rates (1/2 or 1/4)?

I have the HCNC Pro Rev 1 board (basically the same as yours + Idle Current
Control).

I currently use 1/2 stepping with 10tpi acme threads and have not noticed
any similar issue.

As Dave noted, you are approaching the point on the spiral where the
movement on a given axis is at a minimum.  With the full step and a low
thread count, you might be encountering a detent on the motor perhaps?

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

Re: [Emc-users] PC/104 Install

[Peter C. Wallace]

On Wed, 22 Jul 2009, Pablo Pais wrote:

> Date: Wed, 22 Jul 2009 10:56:50 -0300
> From: Pablo Pais 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: emc-users@lists.sourceforge.net
> Subject: [Emc-users] PC/104 Install
> 
> Hello,
> Someone knows about install linuxcnc on a PC/104 sistem.
> Thanks
> Pablo

I think the trick there is to find a PC/104 CPU with decent latency
and a parallel port (if you are using parallel I/O to control the machine)

PC/104 CPUs are not cheap so selecting a CPU card by trial end error is 
liable to be an expensive project...


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

Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


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

[Emc-users] Machining question

[Rich Amaral]

Hey guys,

This is purely a machining question.

I'm currently retrofitting the lead screws on my home built cnc router  
with ACME thread screws. Yes, I run EMC2. The screws are about 3 feet  
long by 3/8 in and need to be turned a bit at either end to fit the  
stepper motor coupling and bearing (at the other end).

I have a Craftsman professional lathe (much like an Atlas) in decent  
shape. After turning both ends to size, I discovered the screw became  
warped (about 37 thousandths, a noticeable amount). I don't think the  
screw was warped before, although I'm not 100% sure. That leaves me  
and the way I held the screw in the lathe (I think).

I held the screw by wrapping some copper around the threads and  
clamping that portion in the chuck, leaving the remainder of the screw  
sticking out the left side of the lathe. The screw wobbled a bit but  
nothing that looked too bad, at least to me. Could this have warped  
the screw?

More importantly, how should the screw be held in the lathe to prevent  
warping or other damage during machining? Thanks!


-Rich





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

Re: [Emc-users] Machining question

[Dave Caroline]

centrifugal force on the unsupported length can bend it easily if the
RPM is high enough

Dave Caroline

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

Re: [Emc-users] Machining question

[John Kasunich]

Rich Amaral wrote:
> Hey guys,
> 
> This is purely a machining question.
> 
> I'm currently retrofitting the lead screws on my home built cnc router  
> with ACME thread screws. Yes, I run EMC2. The screws are about 3 feet  
> long by 3/8 in and need to be turned a bit at either end to fit the  
> stepper motor coupling and bearing (at the other end).
> 
> I have a Craftsman professional lathe (much like an Atlas) in decent  
> shape. After turning both ends to size, I discovered the screw became  
> warped (about 37 thousandths, a noticeable amount). I don't think the  
> screw was warped before, although I'm not 100% sure. That leaves me  
> and the way I held the screw in the lathe (I think).
> 
> I held the screw by wrapping some copper around the threads and  
> clamping that portion in the chuck, leaving the remainder of the screw  
> sticking out the left side of the lathe. The screw wobbled a bit but  
> nothing that looked too bad, at least to me. Could this have warped  
> the screw?

Hard to say - depends on how fast it was turning, among other things.

As an aside - long thin things sticking out the left end of the lathe 
can be VERY BAD.  Above a critical speed (which isn't necessarily that 
fast), centrifugal force can make the part abruptly turn into a giant L, 
and the end whipping around can smash the hell out of anything nearby, 
including you.

> More importantly, how should the screw be held in the lathe to prevent  
> warping or other damage during machining? Thanks!

One step to take would be to turn and bore a bushing that fits nicely 
into the left end of the spindle, and holds the screw centered at that end.

Regards,

John Kasunich

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

Re: [Emc-users] Trouble with EMC spiral example

[Jeffrey Pease]

Hi Brian,

> As Dave noted, you are approaching the point on the spiral where the
> movement on a given axis is at a minimum.  With the full step and a  
> low
> thread count, you might be encountering a detent on the motor perhaps?

Bingo, I think that nailed it. I was running 5 tpi lead screws at full  
step with 200 step/rev motors. I changed each of these to 1/4 step it  
it seems to run MUCH smoother on the spiral. It seems sort of obvious  
in hindsight, but it didn't occur to me that any loss of precision  
from running full step would actually make that perceptible of a  
difference in the stuff I do with my lowly hobbyist setup.

I'm glad I picked the spiral as my test exercise as otherwise I might  
not have caught this and would have been pretty frustrated when making  
actual projects requiring any precision, I think. This is my first  
machine so I'm still getting a feel for the tradeoffs between low and  
high tpi leadscrews.

I'm still running it through its paces, but so far it's a night and  
day difference. Thanks a lot!

Jeff

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

Re: [Emc-users] Trouble with EMC spiral example

[BRIAN GLACKIN]

>
>  Given this is a hobby machine, the 5TPI while sexy could lead to this
> type of thing.  I use 10TPI on a 2'X4' work surface and can get 120IPM but
> stick to 90ipm due to slop in my machine.  If your carving out templates
> from large sheets of 1/4" MDF or the like, then 5tpi is the way to go,
> otherwise, its an accuracy and speed versus actual needs.
>

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

Re: [Emc-users] Machining question

[cmgfam]

>
A fabricated bushing as mentioned, split if need be.  Indicate the work if
a jaw chuck is used,

 and

A bushing at the left end of the headstock to hold the rod true is advised.

That said,  a three foot long piece of 3/8th rod is not likely to be
straight.
EVER.

CalG



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

[Emc-users] Stepper conversion

[Terry]

Hi

I have a Bridgeport Boss 5 retrofitted with an Ah-Ah system (stepper)
and I am going to keep the drivers and change out the software with
EMC (still using steppers).
I am a servo person so be kind when answering the following two
questions.

1. Is there an advantage to using a board from mesa vs parallel port?

2. What do these numbers mean in the EMC world?:

6 us minimum step pulse width
15 us off time
capable of 45,000 steps per second

(this is all the infro I have on these drivers)

Thanks in advance


Terry
 Msg sent via CableONE.net MyMail - http://www.cableone.net
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Machining question

[Ian W. Wright]

   <>>>

That would almost certainly cause the screw to bend 
What you should do is use a fixed steady on the lathe bed 
and have maybe 2/3 of the screw sticking out of the chuck 
with its free end being restrained by the steady. I can't 
remember what the tailstock is like on a Craftsman lathe 
but, if you can swap the positions of the tailstock and 
saddle, you can use the tailstock as a steady passing the 
work through temporary bushes at each end of the tailstock 
barrel. What you are trying to achieve is like a Swiss Auto 
lathe where the part of the work you are turning is held 
securely in a close-fitting bush near to the tool. The part 
of the screw still in the headstock spindle can be packed 
with a bit of rag to stop it flying about.

Ian
_
Ian W. Wright
Sheffield  UK

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

Re: [Emc-users] PC/104 Install

[Lars Andersson]

Hi,
I have several Advantech PCM-9375 PC104 boards (project leftovers). These
run plain Ubuntu 8.04 (LAMP) very well but freezes near the end of an
attempted install of EMC2 from an USB stick. I could not find any diagnostic
feedback, mostly because I don't know where to look in a noncomplete
install. This was a while ago, and I have sort of given up for now, but it
would be a nice package if it worked. 
I did not get as far as being able to check on latency.

FWIW,
Lars


> -Original Message-
> From: Peter C. Wallace [mailto:p...@mesanet.com]
> Sent: den 22 juli 2009 17:58
> To: Enhanced Machine Controller (EMC)
> Subject: Re: [Emc-users] PC/104 Install
> 
> On Wed, 22 Jul 2009, Pablo Pais wrote:
> 
> > Date: Wed, 22 Jul 2009 10:56:50 -0300
> > From: Pablo Pais 
> > Reply-To: "Enhanced Machine Controller (EMC)"
> > 
> > To: emc-users@lists.sourceforge.net
> > Subject: [Emc-users] PC/104 Install
> >
> > Hello,
> > Someone knows about install linuxcnc on a PC/104 sistem.
> > Thanks
> > Pablo
> 
> I think the trick there is to find a PC/104 CPU with decent latency
> and a parallel port (if you are using parallel I/O to control the
> machine)
> 
> PC/104 CPUs are not cheap so selecting a CPU card by trial end error is
> liable to be an expensive project...
> 
> 
> >
> ---
> ---
> > ___
> > Emc-users mailing list
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
> >
> 
> Peter Wallace
> Mesa Electronics
> 
> (\__/)
> (='.'=) This is Bunny. Copy and paste bunny into your
> (")_(") signature to help him gain world domination.
> 
> 
> ---
> ---
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


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

Re: [Emc-users] Stepper conversion

[Stephen Wille Padnos]

Terry wrote:

>Hi
>
>I have a Bridgeport Boss 5 retrofitted with an Ah-Ah system (stepper)
>and I am going to keep the drivers and change out the software with
>EMC (still using steppers).
>I am a servo person so be kind when answering the following two
>questions.
>
>1. Is there an advantage to using a board from mesa vs parallel port?
>  
>
Software stepping through the parallel port can come close to the 
maximum the drivers can handle.  A Mesa card will definitely be able to 
reach the maximum the Ah-ha drivers can do, and much more if you decide 
to upgrade later (to faster stepper drives or to servos).

Out of curiosity, do you have the Ah-ha PC plug-in card?

>2. What do these numbers mean in the EMC world?:
>
>6 us minimum step pulse width
>  
>
This means that steplen needs to be 6000 or higher (that's 6 
microseconds, expressed in nanoseconds)

>15 us off time
>  
>
This probably means that stepspace needs to be 15000 or higher.

>capable of 45,000 steps per second
>  
>
You will need to set your velocity limit such that 45000 steps/sec is 
not exceeded.  So if you have 1 steps/inch, you will need to set the 
velocity limit to 4.5 IPS.  This will need to be lower if you need to 
use backlash compensation, since the backlash accel is added to the 
trajectory planner accel.

These specs don't say what the "dirsetup" and "dirhold" times should be, 
so I'd set them to the same as stepspace, 15000, to be safe.  These are 
the times to wait after changing the direction output and before issuing 
a step, and after issuing a step before changing direction.

Hope this helps
- Steve


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

Re: [Emc-users] Machining question

[Andy Pugh]

2009/7/22 Rich Amaral :

> I have a Craftsman professional lathe (much like an Atlas) in decent
> shape. After turning both ends to size, I discovered the screw became
> warped (about 37 thousandths, a noticeable amount). I don't think the
> screw was warped before, although I'm not 100% sure.

As others have said, it might have whipped during turning. You might
even have had a lucky escape.
If it had bent through whipping then you should be able to find out
where the bend it (probably just behind where the chuck jaws held it)
and by holding it the other way round at the same point you might well
be able to bend it straight again. My feeling would be that there
won't be any significant loss of pitch accuracy doing this. You may
have different definitions of "significant"

-- 
atp

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

Re: [Emc-users] Machining question

[Gene Heskett]

On Wednesday 22 July 2009, cmg...@sover.net wrote:
>A fabricated bushing as mentioned, split if need be.  Indicate the work if
>a jaw chuck is used,
>
> and
>
>A bushing at the left end of the headstock to hold the rod true is advised.
>
>That said,  a three foot long piece of 3/8th rod is not likely to be
>straight.
>EVER.
>
>CalG
>
Even supposedly good acme stock isn't straight. I bought a 3 foot piece of 
1/2" 10tpi from McCarr that had a 3/16" bow in the middle right out of the 
packing, which showed no signs of miss-handling.

So I rigged up a length of plank with risers on each end, put a V in the 
middle of the risers, cut a similar v in the end of a handy hunk of cherry, 
and whaled away at it with a 4 pounder until I couldn't see any noticeable 
wobble when I rolled it in the v's.

Then I cut it to a swag length and turned one end to take my homemade 
couplings.  That was when it was still on the back of the post on my 
micromill.  Even with a 425 motor on it, I could only get about 5 pounds of 
downforce due to racking of the head on the post binding it up.  So I used the 
rest of it to make a new, longer screw, built a new overhead to hold the motor 
and the belt drive to the nut(s), that bolted to the sides of the post, turned 
the gear head 90 degrees to expose some of the heads casting in front of the 
post, and planted the end of the screw in a hole in the casting about 2.25" in 
front of the post.  The motor now drives the nut(s) and the screw rises and 
falls with the head.

I can now push the down button on the keyboard with bathroom scales under the 
spindle & make 155 pounds when the 425 cogs the first time. I'd say a 31x 
increase in the available force was worth it, as now I can program to drill 
holes and if the steel isn't too fancy, just drill the hole(s) at a very good 
feed rate.

Fancy steel, like in a 10" Avanti (Freud) table saw blade a inch & a half from 
the spindle hole, I still have to use EDM for.  Boy are those holes neat & 
clean. :)  But darn is it noisy, the last EDM power supply I built is about 80 
volts, 4 to 5 amps and 10 uf.  With a playdough dam around the hole, and 1/4" 
of kero in it, my ears were ringing noticeably by halfway through the first 
one, so I went after my gun muffs for the rest of that job.  I'm already deaf 
enough from wearing out rifle barrels in past years.

I probably should find a smaller capacitor which would be quieter & raise the 
frequency too & now that the old analog GE transmitter has been shut down 
forever, I may go see if I can dig around in the spares. :)

Darn, get me started and I don't know when to shut up. :)

-- 
Cheers, Gene
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
The NRA is offering FREE Associate memberships to anyone who wants them.


Certainly there are things in life that money can't buy,
But it's very funny -- did you ever try buying them without money?
-- Ogden Nash


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

Re: [Emc-users] Axis Question

[Christopher Purcell]

I am testing a rotary axis (added to my 3 axis router), to which the  
part is to be mounted. The rotary axis is mounted on the X axis, which  
in turn is mounted on the Y axis. The Z axis is independent and  
carries the cutter. Following the advice in Jeff Epler's post quoted  
below, AXIS shows a tool path that reflects the effects of all 4 axes,  
but I think from the part's frame of reference, which makes it hard  
for me to follow the action.

Can AXIS be configured to show the tool path from the operator's frame  
of reference
(i.e., the global coordinate system)? I would like to see the part  
rotate about A, and translate along X and Y, and the cutter move along  
Z.
Does this make sense, or have I inhaled too much varnish tonight?

> From: Jeff Epler 
> Subject: Re: Axis question
> Newsgroups: gmane.linux.distributions.emc.user
> Date: 2009-04-19 19:32:59 GMT (13 weeks, 3 days, 2 hours and 53  
> minutes ago)
> Yes, this is one of the new features in emc 2.3.0. I modified the  
> sample axis_9axis config file to specify [DISPLAY] GEOMETRY=XAYZ The  
> meaning of [DISPLAY]GEOMETRY is described in the documentation: 
> http://linuxcnc.org/docs/2.3/html/config_ini_config.html#sub 
> :AXIS-Interface besides A rotating around X, the other assumption it  
> makes when plotting is that the center of "A" rotation is machine  
> Y0Z0. Here is an example of the preview plot and backplot it  
> produces: 
> http://emergent.unpy.net/index.cgi-files/sandbox/axis2.3-rotary-moves.png 
>  Jeff
>
christopherpurc...@mac.com




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

Re: [Emc-users] Axis Question

[Chris Radek]

On Wed, Jul 22, 2009 at 08:58:33PM -0300, Christopher Purcell wrote:
> I would like to see the part  
> rotate about A, and translate along X and Y, and the cutter move along  
> Z.

The problem is that AXIS doesn't show the part, so it cannot rotate.

It shows the tool and the tool path only.  Can you be more specific
about what you want?

> Does this make sense, or have I inhaled too much varnish tonight?

I'm not sure yet :-)

Chris

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

Re: [Emc-users] Axis Question

[Christopher Purcell]

You're right -  I was taking the tool path to be a representation of  
the "part".
So with that image in mind, when I want to do some additional work on  
the part, for example, with MDI commands, I would like to see the  
"part" moving as the operator would see it.


On Wed, Jul 22, 2009 at 08:58:33PM -0300, Christopher Purcell  
wrote:  I would like to see the part rotate about A, and translate  
along X and Y, and the cutter move along Z.


The problem is that AXIS doesn't show the part, so it cannot rotate.  
It shows the tool and the tool path only. Can you be more specific  
about what you want?



 Does this make sense, or have I inhaled too much varnish tonight?


I'm not sure yet 
<>
 Chris

christopherpurc...@mac.com



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

Re: [Emc-users] Stepper conversion

[Terry]

Sorry I sent this only to Mr.Padnos by mistake
So here...

On Wed Jul 22 18:03 , Terry  sent:

>
>Stephen,
>
>>These specs don't say what the "dirsetup" and "dirhold" times should be, 
>
>
>I looked a little deeper and found this:
>
>10 microseconds direction to step delay?
>
>
>>Out of curiosity, do you have the Ah-ha PC plug-in card?
>
>Yes,it is an If-16 ?
>Artisian software,12amp drivers.
>Would you like to poke and prod around at the card and see
>what makes it tick?You can have at it.
>
>It was one of the last kits they put out,was new in 2005
>I have no problems with the setup but the GUI needs to be more
>user friendly.I am selling this machine to a non-cnc programmer
>friend and I dont want to answering his questions all the time.
>(Well not GUI related questions anyway)
>
>Thanks
>
>Terry
>
>
>
>On Wed Jul 22 15:27 , Stephen Wille Padnos spad...@sover.net> sent:
>
>>Terry wrote:
>>
>>>Hi
>>>
>>>I have a Bridgeport Boss 5 retrofitted with an Ah-Ah system (stepper)
>>>and I am going to keep the drivers and change out the software with
>>>EMC (still using steppers).
>>>I am a servo person so be kind when answering the following two
>>>questions.
>>>
>>>1. Is there an advantage to using a board from mesa vs parallel port?
>>>  
>>>
>>Software stepping through the parallel port can come close to the 
>>maximum the drivers can handle.  A Mesa card will definitely be able to 
>>reach the maximum the Ah-ha drivers can do, and much more if you decide 
>>to upgrade later (to faster stepper drives or to servos).
>>
>>Out of curiosity, do you have the Ah-ha PC plug-in card?
>>
>>>2. What do these numbers mean in the EMC world?:
>>>
>>>6 us minimum step pulse width
>>>  
>>>
>>This means that steplen needs to be 6000 or higher (that's 6 
>>microseconds, expressed in nanoseconds)
>>
>>>15 us off time
>>>  
>>>
>>This probably means that stepspace needs to be 15000 or higher.
>>
>>>capable of 45,000 steps per second
>>>  
>>>
>>You will need to set your velocity limit such that 45000 steps/sec is 
>>not exceeded.  So if you have 1 steps/inch, you will need to set the 
>>velocity limit to 4.5 IPS.  This will need to be lower if you need to 
>>use backlash compensation, since the backlash accel is added to the 
>>trajectory planner accel.
>>
>>These specs don't say what the "dirsetup" and "dirhold" times should be, 
>>so I'd set them to the same as stepspace, 15000, to be safe.  These are 
>>the times to wait after changing the direction output and before issuing 
>>a step, and after issuing a step before changing direction.
>>
>>Hope this helps
>>- Steve
>>
>
>
> Msg sent via CableONE.net MyMail - http://www.cableone.net


 Msg sent via CableONE.net MyMail - http://www.cableone.net
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Machining question

[Jon Elson]

Rich Amaral wrote:
> Hey guys,
>
> This is purely a machining question.
>
> I'm currently retrofitting the lead screws on my home built cnc router  
> with ACME thread screws. Yes, I run EMC2. The screws are about 3 feet  
> long by 3/8 in and need to be turned a bit at either end to fit the  
> stepper motor coupling and bearing (at the other end).
>
> I have a Craftsman professional lathe (much like an Atlas) in decent  
> shape. After turning both ends to size, I discovered the screw became  
> warped (about 37 thousandths, a noticeable amount). I don't think the  
> screw was warped before, although I'm not 100% sure. That leaves me  
> and the way I held the screw in the lathe (I think).
>   
How fast did you spin the screw in the lathe?  A 3 foot bar 3/8" in 
diameter will easily develop "whip" when spun.
You need to support it every 6" to a foot, or turn VERY slowly, like 100 
RPM if you leave the back end unsupported.

Jon

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

Re: [Emc-users] OpenGL on EMC2? + 7i43

[Richard Acosta]

El 20/07/2009 11:41 a.m., Chris Radek escribio':
> On Mon, Jul 20, 2009 at 11:24:29AM -0300, Richard Acosta wrote:
>   
>> I really don't want to keep this going and going
>> 
>
> This does not appear to be true.
>
> Chris
>   
Not while there are people writing emails to me directly to talk 
nonsense and insulting, people who never intervened to offer help but 
just intrude with irrelevant personal stuff.

> --
> Enter the BlackBerry Developer Challenge  
> This is your chance to win up to $100,000 in prizes! For a limited time, 
> vendors submitting new applications to BlackBerry App World(TM) will have
> the opportunity to enter the BlackBerry Developer Challenge. See full prize  
> details at: http://p.sf.net/sfu/Challenge
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>   

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

Re: [Emc-users] Axis Question

[Jeff Epler]

On Wed, Jul 22, 2009 at 09:28:14PM -0300, Christopher Purcell wrote:
> You're right -  I was taking the tool path to be a representation of the 
> "part".
> So with that image in mind, when I want to do some additional work on  
> the part, for example, with MDI commands, I would like to see the "part" 
> moving as the operator would see it.

That's not currently possible.

Jeff

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

[Emc-users] Try to figure out/remember what I have

[mike langeweg]

   I started a cnc knee style retrofit to emc about 6 years ago.  I haven't 
done anything with it since then.  Marriage and kids took a priority.  I am 
starting to have some more free time and would like to get back into it.
Some of what I have is: 

3.0 ghz pentium 4 dual boot with Ubuntu/emc2 ready to install, but i'm not sure 
abou the ini file installed off of live cd. and xp on other hard 
drive-disconnected when installed ubuntu, I just select through bios.
ppmc purchased from pico systems:
   -motherboard
   -dac card
   -dio card
   -encoder card
2 copley control servo amps model 422 (what I could afford at the time need 
more servo amps but would like to see progress first)
I looked at an encoder on the end of my brushed dc servo. This is what I seen
pin#1 vdc
  2 common
  4 A output
  5 B output
  6 M output
  7 A comp
  8 B comp
  9 M comp
from motion control devices I would assume era early 1970's
Model # 80-dm-1000-5-C0-C-T
they originally only used pins 1,2,4,8,9

I would guess this to be a differential encoder with a index (M?)

If possible I would like to prove this at home(basement) with power supply, 
servo amps etc(not the mill for now, I have more time at home than in the shop)

I am not sure if my encoder board will do differntial or just single, I think 
it is just single because I don't see any jumpers.
I breifly started to look over emc2 users manual, and I didn't see/overlooked a 
lot of servo information.  I believe it said something about a different manual?

I would apreciate if I could get some information on a good way to get started 
and some help or direction with the ini file for use with servos

Thank you
 Mike Langeweg

_
Windows Live™ Hotmail®: Celebrate the moment with your favorite sports pics. 
Check it out.
http://www.windowslive.com/Online/Hotmail/Campaign/QuickAdd?ocid=TXT_TAGLM_WL_QA_HM_sports_photos_072009&cat=sports
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] EMC2 Developer Manual - broken pdf

[Haberler Michael]

it seems the PDF linked from the Wiki 
http://www.linuxcnc.org/docs/devel/EMC2_Developer_Manual.pdf 
  ist broken

-Michael

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