On Fri, 18 May 2012 18:42:53 -0700, dave wrote:
>
> Matt and I were talking the other day about a next-gen linuxcnc
> processor and wandered into the area of a decent processor coupled to
> an fpga for all the stuff that a normal processor doesn't do well.
have you taken a look at Yishin Li's work
On Fri, 11 May 2012 12:10:27 -0400
"Kent A. Reed" wrote:
> Gentle persons:
>
> The recent email discussions about ARM-based cpus on the one hand and
> about progress with the PREEMPT_RT patch on the other got me to
> wondering about RTAI and where it is going.
>
> I see that two test versions
On 5/18/2012 5:43 PM, Eric Keller wrote:
> While it is behind
> a firewall, that doesn't stop the students from playing.
Oh, students... aka bright people with a great deal of curiosity and
boundless energy who seemingly can find as much time and as many
co-conspirators as it takes to pull off so
The advantage of long term support from Centos doesn't really help linuxCNC
too much, because they will be updating the kernel too. So even though it
may be an old kernel version, it will have security fixes in it that will
make any RTAI patches fail. The thing that's driving me nuts at work is
t
On 5/18/2012 4:28 PM, Eric Keller wrote:
> On Fri, May 11, 2012 at 12:10 PM, Kent A. Reedwrote:
>
>> Gentle persons:
>>
>> The recent email discussions about ARM-based cpus on the one hand and
>> about progress with the PREEMPT_RT patch on the other got me to
>> wondering about RTAI and where it is
On Fri, May 11, 2012 at 12:10 PM, Kent A. Reed wrote:
> Gentle persons:
>
> The recent email discussions about ARM-based cpus on the one hand and
> about progress with the PREEMPT_RT patch on the other got me to
> wondering about RTAI and where it is going.
>
Generally, the kernel and the distribu
Charles Steinkuehler writes:
>I'm really thinking more about the folks approaching LinuxCNC from the
>3D printing / Desktop CNC world, which seems to revolve around stepper
>motors, Polou microstep drivers, and Arduinos.
The RepRap folks are using the ATmega 644P, which has more I/O pins than
t
Depending on how the control software was written, a system that doesn't
use an operating system, or uses an operating system like MSDOS with no
multi-tasking or preemption can have really high performance on crummy
hardware. It's really frustrating to use if you want to go beyond very
simple act
David Armstrong wrote:
> would anyone concider this approach feasable
> http://code.google.com/p/miniemc2/ then attach a fpga sub board similar to
> the mesa 7i43 perhaps for a fully contained web browser style application
I see some problems. Only 64 BM of RAM, and only 400 MHz. The Beagle
Jan de Kruyf wrote:
> Jon:
> when I started working in the CNC field we had 20msec loops and they
> worked(cutting steel) from pure fanciness we would upgrade to 10msec, and
> with a well tuned servo drive you could feel the distinct steps of the DAC
> while accelerating.
Yes, with velocity servo a
On 18 May 2012 17:29, David Armstrong wrote:
> would anyone concider this approach feasable
> http://code.google.com/p/miniemc2/
I think it is feasible. I am not sure what it is for, though.
EMC2 on Xenomai / ARM is an interesting combination.
--
atp
If you can't fix it, you don't own it.
ht
would anyone concider this approach feasable
http://code.google.com/p/miniemc2/ then attach a fpga sub board similar to the
mesa 7i43 perhaps for a fully contained web browser style application Dave
---
Gentle persons:
I have been mulling over the need for some graphics to help
newbies---who knows, maybe even some veterans---understand the
architecture of a LinuxCNC system. If these already exist in the docs or
on the wiki please let me know because I don't recall them (annoying,
perhaps, bu
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 5/18/2012 8:11 AM, andy pugh wrote:
> At $80 for the 7i43 or the 5i25 I am not sure there is much
> "market value area" for an Arduino approach.
Well, the Arduinos are easier to purchase (click-and-buy on eBay), but
I'm really thinking more about t
On 5/18/2012 4:27 AM, Viesturs Lācis wrote:
> 2012/5/18 Jan de Kruyf:
>
>> In any case: from the grumbles in this thread we might perhaps deduct that
>> there is money in a good micro stepping board, that takes position or speed
>> input every msec or so; for the stepper people.
>>
> Yes,
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 5/18/2012 7:49 AM, Viesturs Lācis wrote:
> 2012/5/18 Charles Steinkuehler :
>>
>> My current plan is to interface the Arduino to the PC using the
>> EPP board, and off-load the step/dir generation similar to Mesa's
>> EPP FPGA 7I43.
>
> So do I un
On 18 May 2012 13:30, Charles Steinkuehler wrote:
> Mesa and other folks sell what looks like great hardware if you're at
> all serious, but I agree there should be a hobbyist path. The small
> stepper community around 3D printers has settled on the AVR/Arduino
At $80 for the 7i43 or the 5i25 I
2012/5/18 Charles Steinkuehler :
>
> My current plan is to interface the Arduino to the PC using the EPP
> board, and off-load the step/dir generation similar to Mesa's EPP FPGA
> 7I43.
So do I understand that it still means "game over" for software step generation?
--
Viesturs
If you can't fix
If you really want to warp your noodle. Take a look at "Styx on a
Brick"
http://www.vitanuova.com/inferno/rcx_paper.html
with complete source code, etc, but I am not sure of the current
status...
On Fri, 18 May 2012 07:30:51 -0500, Charles Steinkuehler wrote:
> -BEGIN PGP SIGNED MESSAGE--
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 5/18/2012 3:27 AM, Viesturs L?cis wrote:
> 2012/5/18 Jan de Kruyf :
>>
>> In any case: from the grumbles in this thread we might perhaps
>> deduct that there is money in a good micro stepping board, that
>> takes position or speed input every msec
velocity at 20msec intervals with a wel tuned analog drive (torque loop
inside a speed loop)
to do a real proper job you would adapt the torque (current!) loop to the
inductance of the motor, and the P and the I of the velocity loop to the
machine at hand.
There was some autotune magic in the posit
On 18 May 2012 09:05, Jan de Kruyf wrote:
> when I started working in the CNC field we had 20msec loops and they
> worked(cutting steel) from pure fanciness we would upgrade to 10msec
Was that a 20mS torque loop, or a 20mS velocity command loop?
I would imagine what you are describing there was
2012/5/18 Jan de Kruyf :
> Viesturs:
> Pray tell me which, I like to see what they did then to make the
> motherboard hardware perform better.
I am sorry, I do not understand Your question.
--
Viesturs
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto
-
Viesturs:
Pray tell me which, I like to see what they did then to make the
motherboard hardware perform better.
j.
On Fri, May 18, 2012 at 10:27 AM, Viesturs Lācis
wrote:
> 2012/5/18 Jan de Kruyf :
> >
> > In any case: from the grumbles in this thread we might perhaps deduct
> that
> > there is
2012/5/18 Jan de Kruyf :
>
> In any case: from the grumbles in this thread we might perhaps deduct that
> there is money in a good micro stepping board, that takes position or speed
> input every msec or so; for the stepper people.
Yes, and the problem is that for serious stepper machines it is ok
Jon:
when I started working in the CNC field we had 20msec loops and they
worked(cutting steel) from pure fanciness we would upgrade to 10msec, and
with a well tuned servo drive you could feel the distinct steps of the DAC
while accelerating. Afterwards I learned to tune a servo by holding the
tabl
26 matches
Mail list logo
