On Thu, May 30, 2013 at 3:05 PM, MC Cason farmerboy1...@yahoo.com wrote:
On 05/30/2013 12:46 PM, andy pugh wrote:
On 30 May 2013 18:22, Andrew parallel.kinemat...@gmail.com wrote:
Last time I wish too often that I could do programming. Is it too late
to
learn in late 30s?
Probably
Gene;
Not no, but hell no Andrew. I saw my first micro cpu, an RCA 1802, in the
spring of '80 when I was 46. ...
You are a veritable spring chicken at 30. If you have the time to study,
the need to know how is a very strong incentive.
Mine was also the 1802, (maybe a year or two before
On Friday 31 May 2013 09:15:34 John Alexander Stewart did opine:
Gene;
Not no, but hell no Andrew. I saw my first micro cpu, an RCA 1802, in
the
spring of '80 when I was 46. ...
You are a veritable spring chicken at 30. If you have the time to
study, the need to know how is a
On Fri, 2013-05-31 at 07:19 -0400, John Alexander Stewart wrote:
Gene;
Not no, but hell no Andrew. I saw my first micro cpu, an RCA 1802, in the
spring of '80 when I was 46. ...
You are a veritable spring chicken at 30. If you have the time to study,
the need to know how is a very
2013/5/30 Jon Elson el...@pico-systems.com
5 ns is only one instruction. Do you think it can read, say, the 3
signals from 4 encoders
and keep count, all with only one instruction? Not possible. It would
probably take
20 -30 instructions for each encoder channel, and possibly quite a bit
Am 30.05.2013 um 09:07 schrieb Andrew parallel.kinemat...@gmail.com:
Talking about encoders, Beaglebone must be good with serial ones.
SPI should not be a problem at all, but what about BISS?
I got a few very nice motors with those encoders, and I'd like to test them
at least.
can you refer
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 5/29/2013 11:05 PM, Przemek Klosowski wrote:
Low power should not be a big deal---after all, LinuxCNC is likely
to drive electric motors rated in kiloWatts.
Low power is important when you put the computer in a sealed box to
keep it from getting
agreed and any increase in efficiency with no degrade in perfomance is
highly beneficial, also developing this will give us a platform for the
forseeable future. a friend of mine runs a computer recycling facility and
kills at least a hundred desktops a day , they will not be with us forever
,
On 30 May 2013 10:16, Michael Haberler mai...@mah.priv.at wrote:
can you refer to some reference implementation?
http://www.biss-interface.com/files/BiSS_BP1_profile_A4en.pdf
I think there is a Mesa module for them, (but no LinuxCNC driver)
I may even have one, I took a 17-bit multiturn
Jeremy,
Am 30.05.2013 um 12:36 schrieb jeremy youngs jcyoung...@gmail.com:
i
just wish the manufactureres would give us something with a little bit more
oomph than these bords have . *
there are stronger alternatives
second, and more important: where exactly should that 'oomph' go, and
On 5/30/2013 8:15 AM, andy pugh wrote:
On 30 May 2013 10:16, Michael Haberlermai...@mah.priv.at wrote:
can you refer to some reference implementation?
http://www.biss-interface.com/files/BiSS_BP1_profile_A4en.pdf
I think there is a Mesa module for them, (but no LinuxCNC driver)
I think this is off-topic,
but I was thinking in the following configuration for some projects:
- Raspberry PI computer
- A custom made SPI card with 4 DAC and 4 encoder input
DAC: http://ww1.microchip.com/downloads/en/devicedoc/21897a.pdf
Encoder:
2013/5/30 Michael Haberler mai...@mah.priv.at
can you refer to some reference implementation?
never heard of it, how widely is that used?
Pretty wide among modern absolute encoders, which are expensive and
therefore not so familiar for hobbyists.
2013/5/30 Dave e...@dc9.tzo.com
BISS and
On Thu, May 30, 2013 at 10:46 AM, Andrew parallel.kinemat...@gmail.com wrote:
LinuxCNC definitely lacks support for modern hardware like SSI and BiSS
absolute encoders, CANopen, EtherCAT drives etc. In fact, step/dir and
+-10V remain the only supported control interfaces for almost 20 years
Am 30.05.2013 um 16:46 schrieb Andrew parallel.kinemat...@gmail.com:
Absolutely!
LinuxCNC definitely lacks support for modern hardware like SSI and BiSS
absolute encoders, CANopen, EtherCAT drives etc. In fact, step/dir and
I agree violently! now go prime your editor and send those
] Olinuxino/Beagleboard/bone, Xenomai, SPI?
2013/5/30 Michael Haberler mai...@mah.priv.at
can you refer to some reference implementation?
never heard of it, how widely is that used?
Pretty wide among modern absolute encoders, which are expensive and
therefore not so familiar for hobbyists
Charles Steinkuehler wrote:
Low power is important when you put the computer in a sealed box to
keep it from getting sprayed with metal chips and coolant, and you
have no fans to avoid one of the most common parts to fail.
You can also get completely fan-less systems with the Intel D525MW
My thoughts Michael ,
and this is not a strong point of mine so forgive technical ignorance
please.
I think a dual core to handle realtime on one and all other processes would
be wonderful 1.5 or better ghz and iIthink ram is lacking in the bone and
rpi . something like 4 gig would give us a great
2013/5/30 Eric Keller
eekel...@psu.eduhttps://mail.google.com/mail/u/0/?view=cmfs=1tf=1to=eekel...@psu.edu
Nothing stopping anyone from making any of this stuff work that I can
see. There isn't a big group of developers sitting around waiting for
jobs to do, that should be obvious.
On Thu, May 30, 2013 at 11:34 AM, Michael Haberler mai...@mah.priv.at wrote:
as for EtherCAT - read this first, its a license minefield:
http://www.xenomai.org/pipermail/xenomai/2013-May/028597.html
We went through this agonizing process with open source licenses a
while back. Since that was
We went through this agonizing process with open source licenses a
while back.
yes we did I seem to recall a few weeks of banter about it . where does one
find information to guide himself in thes directions . I wish to cure my
ignorance of these things so as to be able to be a more productive
On 30 May 2013 18:22, Andrew parallel.kinemat...@gmail.com wrote:
Last time I wish too often that I could do programming. Is it too late to
learn in late 30s?
Probably not. You might even enjoy the process.
Then I better start MESAing with 5i25 + 7i77 or 7i76 when I get it. I guess
7i43 is
2013/5/30 andy pugh
bodge...@gmail.comhttps://mail.google.com/mail/u/0/?view=cmfs=1tf=1to=bodge...@gmail.com
On 30 May 2013 18:22, Andrew
parallel.kinemat...@gmail.comhttps://mail.google.com/mail/u/0/?view=cmfs=1tf=1to=parallel.kinemat...@gmail.com
wrote:
Last time I wish too often that I
On 30 May 2013 19:19, Andrew parallel.kinemat...@gmail.com wrote:
No reason it wouldn't work just as well on the 7i43 as on any other Mesa
card.
OK, then I guess I should start from compiling MESA firmware.
That's probably quite a hard place to start, and I guess that Pete
might already have
Am Donnerstag, 30. Mai 2013 schrieb Eric Keller:
On Thu, May 30, 2013 at 11:34 AM, Michael Haberler mai...@mah.priv.at
wrote:
as for EtherCAT - read this first, its a license minefield:
http://www.xenomai.org/pipermail/xenomai/2013-May/028597.html
We went through this agonizing process
Subject: Re: [Emc-users] Olinuxino/Beagleboard/bone, Xenomai, SPI?
On 30 May 2013 19:19, Andrew parallel.kinemat...@gmail.com wrote:
No reason it wouldn't work just as well on the 7i43 as on any other Mesa
card.
OK, then I guess I should start from compiling MESA firmware.
That's
Am 30.05.2013 um 19:22 schrieb Andrew parallel.kinemat...@gmail.com:
I agree violently! now go prime your editor and send those patches ;)
I meant no demands, of course. Just saying from not-so-advanced end user
side.
Last time I wish too often that I could do programming. Is it too late
On Thu, May 30, 2013 at 5:02 PM, Michael Haberler mai...@mah.priv.at wrote:
(a likely followup by certain inmates will be and it shows ;-)
or as a long-gone EMC contributor told me, I've seen your code
--
Introducing
On Thursday 30 May 2013 20:32:49 Andrew did opine:
2013/5/30 Eric Keller
eekel...@psu.eduhttps://mail.google.com/mail/u/0/?view=cmfs=1tf=1to
=eekel...@psu.edu
Nothing stopping anyone from making any of this stuff work that I can
see. There isn't a big group of developers sitting around
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 5/28/2013 11:11 PM, Michael Haberler wrote:
Eric,
Am 29.05.2013 um 00:37 schrieb Eric Keller eekel...@psu.edu:
note that we're not restricted to x86 and ARM; Xenomai builds on
Blackfins and ppc as well and then some, and RT-PREEMPT on
2013/5/29 Charles Steinkuehler
char...@steinkuehler.nethttps://mail.google.com/mail/u/0/?view=cmfs=1tf=1to=char...@steinkuehler.net
I would also like to point out we are not limited to low-end hardware
for experiments. I am targeting the BeagleBone primarily because of
my focus on 3D
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 5/29/2013 5:59 AM, Andrew wrote:
2013/5/29 Charles Steinkuehler
char...@steinkuehler.nethttps://mail.google.com/mail/u/0/?view=cmfs=1tf=1to=char...@steinkuehler.net
I would also like to point out we are not limited to low-end
hardware for
On Wed, May 29, 2013 at 7:22 AM, Charles Steinkuehler
char...@steinkuehler.net wrote:
The PRU can do a great job with step/dir generation. Anything that
has encoders should really be running with FPGA hardware and something
like the Mesa boards.
Why do you say so? PRU is running at 200MHz
Hi Przemek,
Am 30.05.2013 um 06:05 schrieb Przemek Klosowski przemek.klosow...@gmail.com:
On Wed, May 29, 2013 at 7:22 AM, Charles Steinkuehler
char...@steinkuehler.net wrote:
The PRU can do a great job with step/dir generation. Anything that
has encoders should really be running with
On Thu, May 30, 2013 at 12:24 AM, Michael Haberler mai...@mah.priv.atwrote:
The 5ns is 'on paper'; since there's more involved than a CPU cycle - the
bit takes a bit of time to work through to/from the pin , the actual
figures are slower
Charles has done the mother of all homework on this,
Przemek Klosowski wrote:
On Wed, May 29, 2013 at 7:22 AM, Charles Steinkuehler
char...@steinkuehler.net wrote:
The PRU can do a great job with step/dir generation. Anything that
has encoders should really be running with FPGA hardware and something
like the Mesa boards.
Why do
Hi Anders,
on the Beaglebone status:
I had prepared an SD image with a 3.2.21 xenomai kernel; literally hours after
it was out it became apparent there is a fatal bug in the 3.2.21 xenomai patch
and it will be retracted
by now I do have a working xenomai kernel based on 3.8.13; however the
On 5/28/2013 3:50 AM, Michael Haberler wrote:
well it runs off an SD card root filesystem to start with, and Ethernet is in
place so yes
Since Anders mentioned datalogging...
I think I would use the USB port for the purpose because the Beaglebones
employ microSD cards, not SD cards. Maybe
Am 28.05.2013 um 16:35 schrieb Kent A. Reed kentallanr...@gmail.com:
On 5/28/2013 3:50 AM, Michael Haberler wrote:
well it runs off an SD card root filesystem to start with, and Ethernet is
in place so yes
Since Anders mentioned datalogging...
I think I would use the USB port for the
In spite of myself
http://www.youtube.com/watch?v=ona-RhLfRfc
;)
On 2013-05-28 16:35, Kent A. Reed wrote:
On 5/28/2013 3:50 AM, Michael Haberler wrote:
well it runs off an SD card root filesystem to start with, and Ethernet is
in place so yes
Since Anders mentioned datalogging...
I
On 5/28/13 09:43 , Michael Haberler wrote:
Am 28.05.2013 um 16:35 schrieb Kent A. Reed kentallanr...@gmail.com:
As an aside, in its current state, the O/S + LinuxCNC are too big to fit
into the Beaglebone Black's internal storage so at least some of the
microSD card capacity is needed for
--- On Tue, 5/28/13, Michael Haberler mai...@mah.priv.at wrote:
what would make sense is to export a _production_
environment onto the builtin flash once done with the
development builds; that surely can be done within the
confines of 2GB
With everything configured for a specific machine,
On Tue, May 28, 2013 at 12:05 PM, Sebastian Kuzminsky s...@highlab.com wrote:
I'd be happy to help with this. Any thoughts on what/when you want
merge with the mainline of LinuxCNC?
I would also like to help. I need to test the old toolkit to see what
is happening with that.
I keep hoping
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 5/28/2013 5:37 PM, Eric Keller wrote:
I keep hoping to hear of some break out board replacement for the
BBB because it would probably be useful to have a target to aim
for. Eric
What sort of break-out board are you wanting to see?
I am looking
On Tue, May 28, 2013 at 6:45 PM, Charles Steinkuehler
char...@steinkuehler.net wrote:
What sort of break-out board are you wanting to see?
Charles,
Basically I just want a printer port output. I have thought of just
getting one of the proto board capes, but it seems like a BOB
interface would
design.
-- Ralph
From: Eric Keller [eekel...@psu.edu]
Sent: Tuesday, May 28, 2013 4:01 PM
To: Enhanced Machine Controller (EMC)
Subject: Re: [Emc-users] Olinuxino/Beagleboard/bone, Xenomai, SPI?
On Tue, May 28, 2013 at 6:45 PM, Charles Steinkuehler
char
On 28 May 2013 23:45, Charles Steinkuehler char...@steinkuehler.net wrote:
I've been considering whipping up something (schematic or circuit
board) if it would be helpful, but I'm not sure who has interest in
using the 'Bone for something that _isn't_ a 3D printer.
Not quite the same thing,
On 5/28/2013 11:43 AM, Michael Haberler wrote:
Am 28.05.2013 um 16:35 schrieb Kent A. Reed kentallanr...@gmail.com:
On 5/28/2013 3:50 AM, Michael Haberler wrote:
well it runs off an SD card root filesystem to start with, and Ethernet is
in place so yes
Since Anders mentioned datalogging...
Am 29.05.2013 um 02:30 schrieb Kent A. Reed kentallanr...@gmail.com:
I have tried, but getting the image below 2GB requires IMO a bit too many
functional compromises, at least using wheezy as a base, and under the
assumption that a full development system and deep git clone is retained
Eric,
Am 29.05.2013 um 00:37 schrieb Eric Keller eekel...@psu.edu:
On Tue, May 28, 2013 at 12:05 PM, Sebastian Kuzminsky s...@highlab.com
wrote:
I'd be happy to help with this. Any thoughts on what/when you want
merge with the mainline of LinuxCNC?
I would also like to help. I need
Michael Haberler wrote:
hm, that'd be interesting. I'm somewhere around 2.1GB and am already pressed
to choose between nice-to-have-on-board packages. Maybe I've overlooked some
whoppers.
The biggest win right now would be replacing the git repos by shallow clones
with just a bit of
51 matches
Mail list logo
