Re: [Emc-users] Parallel port in the future - query

[Jon Elson]

alan wrote:
> On Thu, 2008-10-30 at 20:30 -0500, Jim Coleman wrote:
>   
>> alan, have you considered replacing the fan?  this seems like the cheapest
>> solution to me :-D  and you could probably find a complete heatsink with fan
>> for your laptop on ebay, people part out laptops alot on there.
>> 
> I took it to a local computer store (not PC world) and they tried to get
> a replacement fan with no success.
>   
If it is a CPU fan, just replace the whole heatsink assembly.  They just 
clip onto the CPU, and come with the fan packaged as an assembly.  Other 
than high-grade commercial systems, the clip-on heatsink fans are awful 
junk where the plastic starts to crumble within months.  These should be 
around $12 new.  I wouldn't spend even $1 on any of the home-grade stuff 
used, it will be already dying.

Power supply and case fans should be pretty standard sizes, look in the 
Digi-Key catalog.

Jon

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Jon Elson]

tomp wrote:
>
> i dont think firewire card is rare or expensive
> siig syba & other mfctrs
> tiger direct 20$
> mirco center 20$
> newegg 12.99$
> so not rare
> and not expensive
> more important, it already works realtime linux rtai for motion control
>
>   
OK, they have REALLY come down, I haven't looked at FireWire in a long 
time.  It was actually declared "dead" several years ago, but the users 
wouldn't allow it.  But, there are these REALLY cheap USB- and 
Ethernet-enabled CPU chips and CPU-less USB target device chips that 
could be used (if suitable) to control some other motion hardware.  
Using one of the USB FIFO chips to send out step pulses sounds like it 
might actually be pretty easy.  (Note the "might"!)

Jon

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Jon Elson]

Kirk Wallace wrote:
> Wasn't there some talk, last year, about real-time Ethernet? Ethernet,
> to my uneducated brain, seems much more appealing than USB.
>   
Jan Kiszka did this for RT Linux (its called rtnet), and it is 
apparently a complete project, and is being maintained.
BUT, the slave side is NOT done.  That's the problem.  If you wanted to 
have some slave motion control devices that were commanded by the EMC 
system using rtnet, you'd have to do some tricky stuff, at the least, to 
make them perform the way rtnet requires.  Basically, time slots are 
allocated based on the number of devices and their requirements for data 
transfers.  Then, the master sends out sync packets every interval, and 
all the slaves have a time relative to the sync that they may respond 
in, and at no other time.  it seems like many of the really nice, cheap 
and flexible Ethernet-enabled chip products, such as the many Arm-based 
micros with USB and net ports, may not be flexible enough to bend their 
Ethernet ports to work with rtnet.  Too much of the protocol is handled 
in hardware to adapt to rtnet, or at least that is the impression I got 
while diging through a number of data sheets.  Maybe there is a way to 
do it, but the off the shelf protocol stacks that have been written for 
these micros certainly won't work without massive changes, if at all.
> Another thought, is that as long as there are desktop PC's, there will
> be slots (maybe?) and for real-time, a slot card controller would be the
> most efficient. I guess this doesn't address the bare-bones systems were
> cost and simplicity are the major issues. So far, there have always been
> parallel port cards to fit the latest slots, so maybe it's not yet worth
> worrying about?
>
>   
I haven't tried a PCIe parallel port card yet, as I don't have a PC with 
PCIe slots, yet.  But, that definitely SHOULD work.  If a PCI parallel 
port works, a PCIe should, too.
> I wonder if there will be a time when EMC2 will need to branch, to cater
> to different system types? How long will those, that want to do the most
> with the least, and those that want to do the most with the most, be
> able to use the same EMC?
>
>   
EMC is already a very flexible system, that can do both servo and 
software-generated steps, and even mix them in the same system!
Robots and hexapods, too.  Why branch?

Jon

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[tomp]

Kirk Wallace wrote:
> On Fri, 2008-10-31 at 17:57 +, Dave Caroline wrote:
>   
>> I have a feeling if we learn enough about USB2 a fixed bi directional
>> micro frame protocol could be deterministic for EMC
>> I need a second and third read of the spec
>> Dave Caroline
>> 
>
> Wasn't there some talk, last year, about real-time Ethernet? Ethernet,
> to my uneducated brain, seems much more appealing than USB.
>
> Another thought, is that as long as there are desktop PC's, there will
> be slots (maybe?) and for real-time, a slot card controller would be the
> most efficient. I guess this doesn't address the bare-bones systems were
> cost and simplicity are the major issues. So far, there have always been
> parallel port cards to fit the latest slots, so maybe it's not yet worth
> worrying about?
>
> I wonder if there will be a time when EMC2 will need to branch, to cater
> to different system types? How long will those, that want to do the most
> with the least, and those that want to do the most with the most, be
> able to use the same EMC?
>
> Who made the coffee, this morning?
>
> Kirk
>
>
> -
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>   

rt-ethernet is enabled by the rt-firewire used with rtai's xenomai

http://www.rtfirewire.org/

"Via the implementation of real-time Ethernet-over-FireWire, an 
application-layer module emulating Ethernet interface over FireWire 
hardware, RT-FireWire enables RTnet , a hard 
real-time communication framework over Ethernet to work on Firewire."

reghards
tomp

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Peter C. Wallace]

On Fri, 31 Oct 2008, Matt Shaver wrote:

> Date: Fri, 31 Oct 2008 14:39:47 -0400
> From: Matt Shaver <[EMAIL PROTECTED]>
> To: "Enhanced Machine Controller (EMC)" 
> Subject: Re: [Emc-users] Parallel port in the future - query
> 
> On Thu, 2008-10-30 at 08:00 -0500, Jeff Epler wrote:
>> It can go to pci and post-pci internal busses using intelligent cards
>> like the mesa 5i20. (this is the path I personally favor)
>
> I'm kind of hoping for pci express over cable:
>
> http://www.onestopsystems.com/pcie_over_cable.html
>
> http://www.innovative-dsp.com/products.php?product=PCI%20Express%20X1%
> 20Cable%20Adapter
> This board is $175 today, but they will likely become common and
> inexpensive in the near future (1-3 years I would guess). A 1 meter
> cable is $28, 3 meters is $45, this will go down as well.
>
> At the other end of the cable would be a MegaMesa2000 board with FPGAs
> and 50 pin Opto22 bus connectors (like a 5i22 but PCIe-Cable instead of
> regular PCI), and maybe a daisy chain connector to add additional
> boards.
>
> Here's hoping,
> Matt
>

We will have the 6I71 soon (PCIe slot to 1 lane PCIe cable+equallizer) its $69 
in singles. Around March we should have the 3X20 available (External PCIe FPGA 
card) 2M gate, 144 I/O with smaller and cheaper varients to follow...


Peter Wallace
Mesa Electronics

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


-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Matt Shaver]

On Thu, 2008-10-30 at 08:00 -0500, Jeff Epler wrote:
> It can go to pci and post-pci internal busses using intelligent cards
> like the mesa 5i20. (this is the path I personally favor)

I'm kind of hoping for pci express over cable:

http://www.onestopsystems.com/pcie_over_cable.html

http://www.innovative-dsp.com/products.php?product=PCI%20Express%20X1%
20Cable%20Adapter
This board is $175 today, but they will likely become common and
inexpensive in the near future (1-3 years I would guess). A 1 meter
cable is $28, 3 meters is $45, this will go down as well.

At the other end of the cable would be a MegaMesa2000 board with FPGAs
and 50 pin Opto22 bus connectors (like a 5i22 but PCIe-Cable instead of
regular PCI), and maybe a daisy chain connector to add additional
boards.

Here's hoping,
Matt



-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[alan]

On Thu, 2008-10-30 at 20:30 -0500, Jim Coleman wrote:
> alan, have you considered replacing the fan?  this seems like the cheapest
> solution to me :-D  and you could probably find a complete heatsink with fan
> for your laptop on ebay, people part out laptops alot on there.
I took it to a local computer store (not PC world) and they tried to get
a replacement fan with no success.

Alan


-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[tomp]

Jon Elson wrote:
> tomp wrote:
>   
>> maybe this is of interest
>> RT-Firewire on Linux, using Xenomai (RTAI )
>> used for robotics, for realtime control of positioning systems
>> it also enables RTnet over firewire
>> several nice papers at http://www.rtfirewire.org
>>   
>> 
> Practically all PCs have USB and Ethernet.  The hardware is CHEAP!  
> After some searching, the Real time Linux USB projects do NOT seem to be 
> very active, unfortunately.  Anybody know anything about what is going 
> on with them?  There is a USB4RT and a USB 2.0 for RT project.  The 
> former was last updated in 2005, the latter in 2006.  That seems to 
> indicate they are dead.  With all the new hardware out, it would seem 
> there would need to be some maintenance work, at least.
>
> Firewire is a rare thing, and not so cheap.
>
>
> Jon
>
> -
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>   
i dont think firewire card is rare or expensive
siig syba & other mfctrs
tiger direct 20$
mirco center 20$
newegg 12.99$
so not rare
and not expensive
more important, it already works realtime linux rtai for motion control

regards
tomp

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Kirk Wallace]

On Fri, 2008-10-31 at 17:57 +, Dave Caroline wrote:
> I have a feeling if we learn enough about USB2 a fixed bi directional
> micro frame protocol could be deterministic for EMC
> I need a second and third read of the spec
> Dave Caroline

Wasn't there some talk, last year, about real-time Ethernet? Ethernet,
to my uneducated brain, seems much more appealing than USB.

Another thought, is that as long as there are desktop PC's, there will
be slots (maybe?) and for real-time, a slot card controller would be the
most efficient. I guess this doesn't address the bare-bones systems were
cost and simplicity are the major issues. So far, there have always been
parallel port cards to fit the latest slots, so maybe it's not yet worth
worrying about?

I wonder if there will be a time when EMC2 will need to branch, to cater
to different system types? How long will those, that want to do the most
with the least, and those that want to do the most with the most, be
able to use the same EMC?

Who made the coffee, this morning?

Kirk


-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Jack Coats]

I was just reading SERVO, and a guy is using one of the PicoITX boards.
Low power consumption, small, and could do well, but I would like a board
with 3 or 4 parallel ports available, even using a 'plug in board'.  
That would give
all the expansion I could ever think of.  But building a parallel port 
board can't
be that big a deal.

I don't know what happened to the real time projects.  I suspect that 
many of them
migrated to doing embedded systems, or virtuilization.  Unless someone 
has a real
need for hard real time (like machine control, etc) using an attached 
embedded
system may be easier and cheaper (given the time cost in the software).  
But that
is just my guess.

As computers get faster, and if lightly run response time shrinks, near 
real time
might be good enough for non-supercritical projects.

Jon Elson wrote:
> tomp wrote:
>   
>> maybe this is of interest
>> RT-Firewire on Linux, using Xenomai (RTAI )
>> used for robotics, for realtime control of positioning systems
>> it also enables RTnet over firewire
>> several nice papers at http://www.rtfirewire.org
>>   
>> 
> Practically all PCs have USB and Ethernet.  The hardware is CHEAP!  
> After some searching, the Real time Linux USB projects do NOT seem to be 
> very active, unfortunately.  Anybody know anything about what is going 
> on with them?  There is a USB4RT and a USB 2.0 for RT project.  The 
> former was last updated in 2005, the latter in 2006.  That seems to 
> indicate they are dead.  With all the new hardware out, it would seem 
> there would need to be some maintenance work, at least.
>
> Firewire is a rare thing, and not so cheap.
>
>
> Jon
>
> -
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>   

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Dave Caroline]

I think some of us need to pick up and even help or fork those dead
projects and see where we can go

Dave Caroline

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Jon Elson]

tomp wrote:
>
>
> maybe this is of interest
> RT-Firewire on Linux, using Xenomai (RTAI )
> used for robotics, for realtime control of positioning systems
> it also enables RTnet over firewire
> several nice papers at http://www.rtfirewire.org
>   
Practically all PCs have USB and Ethernet.  The hardware is CHEAP!  
After some searching, the Real time Linux USB projects do NOT seem to be 
very active, unfortunately.  Anybody know anything about what is going 
on with them?  There is a USB4RT and a USB 2.0 for RT project.  The 
former was last updated in 2005, the latter in 2006.  That seems to 
indicate they are dead.  With all the new hardware out, it would seem 
there would need to be some maintenance work, at least.

Firewire is a rare thing, and not so cheap.


Jon

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Dave Caroline]

I have a feeling if we learn enough about USB2 a fixed bi directional
micro frame protocol could be deterministic for EMC
I need a second and third read of the spec
Dave Caroline

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Jon Elson]

Jon Elson wrote:
> Dave Caroline wrote:
>   
>> The amount of data one could put in a USB microframe, may allow for
>> more loops per frame so make it easier
>>   
>> 
> I can see some ways that one could export the software step generation 
> to a USB parallel port.
Umm, I obviously didn't say what I meant, here.  What I was thinking is 
that these USB parallel FIFO modules could be pretty easily set up to 
pump out a steady flow of 8-bit bytes from the FIFO.  I'm not too clear 
on how seamless the transition between packets (USB frames) would be, 
that is where the research comes in.  But, if the step generation task 
could buffer up a bunch of output words and then send them in one 
packet, the USB target could spoon them out steadily over a millisecond.

Jon

Jon

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[John Harris]

Hi All,
You might want to take a look at a Saxo board sold through the 
www.knjn.com site. It combines the Cypress CY7C68013, with an Altera FPGA 
and an ARM7 microcontroller. both the FPGA and the ARM7 can be reprogrammed 
through the USB port, which simplifies code development. Even if you do not 
use this board, the fact that you can update embedded code through the USB 
port is worth knowing.

Separate subject: In my 3-axis microscope stage controller, I have a means 
of mixing  imperative commands like Stop into the the positioning command 
stream. At the upstream end of the USB connections, the imperatives are 
inserted ahead of buffered positioning commands, and at the downstream end 
these commands are acted on without going through the FIFO buffer used for 
the positioning commands. Each positioning command, that is like a G Code 
block in binary is given a 1/100 second timestamp at the upstream end and 
this timestamp tells the embedded control system when the command is to be 
executed.

John Harris

- Original Message - 
From: "Jon Elson" <[EMAIL PROTECTED]>
To: "Enhanced Machine Controller (EMC)" 
Sent: Thursday, October 30, 2008 10:05 PM
Subject: Re: [Emc-users] Parallel port in the future - query


> John Kasunich wrote:
>
>> This subject seems to come up every month or two.  The answer is and
>> will probably always be the same.
>>
>> Yes, it is possible to make a box that lives outside the PC and can
>> queue up enough motion "in advance" that it doesn't require realtime
>> performance from the PC.
>>
>>
> I agree that taking the PC completely out of the real-time game is a
> mistake in two senses.  First, we've already done the "heavy lifting",
> and it works.  Second is, if you are NOT real-time, then there is no
> upper bound to latency, and one day, the system won't get the data there
> in time, and the buffer will run empty.  The machine will crash to a
> stop, possibly with disasterous consequences.
>
> But, there might be another way.  Keep EMC real-time, but have a USB
> device that can pump out parallel bytes from a FIFO.  This requires EMC
> to sync to the USB clock instead of the system timer.  Every tick of
> that clock, a whole batch of step pulses, just like they are spoon-fed
> to the parallel port now, would be buffered and sent to the USB device
> to be de-buffered at a constant rate such that as the last byte was
> sent, the next buffer would be ready.  I think the Cypress CY7C68013 can
> do all this in hardware, once configured.
>> You can do that if you want, but you lose one of the big benefits of PC
>> based control.  On a PC based control, it is easy to find problems, add
>> features, and update the software.  If you embed everything in some
>> external box with no keyboard, no screen, no development tools, etc, you
>> are rather locked down.  It's one thing to embed simple stuff that can
>> be tested and proven robust, and that is unlikely to change.  It is much
>> tougher to embed complex code that will need to be debugged and changed
>> to meet evolving needs.
>>
> Well, using the Cypress chip, or some other USB-FIFO device, is an
> intermediate step.  If it only has a one ms buffer, the user would never
> know the difference.  And, it would not be moving all the motion control
> of EMC into some external device, it is just a FIFO and an interface 
> device.
> Of course, it needs real-time determinism on the USB.  I have no idea
> what state that is in, but I think there has been some work done in that
> area.
> A quick Google search appears to show that EVERY document containing the
> text "USB" also contains "real time", using it to means something
> happening  within a couple of seconds!  UGH!  But, it looks like Jan
> Kiszka, who did the rt-net package also has a USB stack for rtai,
> originally started by Joerge Langenberg.  I should point out that I am
> NOT volunteering for this project, although after I get some hardware
> and software expertise with this Cypress chip, I might be willing to
> contribute to such a project.
>
> I have a bigger interest in possibly using this chip to connect my other
> boards to a PC without parallel ports.  It might also increase the
> performance, as the CPU having to process each byte laboriously through
> the parallel port is becoming a bottleneck.  But, I don't know if the
> chip, or the USB model, is really conducive to the existing boards'
> model of communication.  The fact that it can do one-way FIFO transfers
> without intervention of the slow 8051 CPU is tantalizing, though.  With
> some additional FPGA logic to format blocks of data to be read from the
> FPGA to the CP

Re: [Emc-users] Parallel port in the future - query

[tomp]

Jon,

Jon Elson wrote:
> John Kasunich wrote:
>
>   
>> This subject seems to come up every month or two.  The answer is and 
>> will probably always be the same.
>>
>> Yes, it is possible to make a box that lives outside the PC and can 
>> queue up enough motion "in advance" that it doesn't require realtime 
>> performance from the PC.
>>
>>   
>> 
> I agree that taking the PC completely out of the real-time game is a 
> mistake in two senses.  First, we've already done the "heavy lifting", 
> and it works.  Second is, if you are NOT real-time, then there is no 
> upper bound to latency, and one day, the system won't get the data there 
> in time, and the buffer will run empty.  The machine will crash to a 
> stop, possibly with disasterous consequences.
>
> But, there might be another way.  Keep EMC real-time, but have a USB 
> device that can pump out parallel bytes from a FIFO.  This requires EMC 
> to sync to the USB clock instead of the system timer.  Every tick of 
> that clock, a whole batch of step pulses, just like they are spoon-fed 
> to the parallel port now, would be buffered and sent to the USB device 
> to be de-buffered at a constant rate such that as the last byte was 
> sent, the next buffer would be ready.  I think the Cypress CY7C68013 can 
> do all this in hardware, once configured.
>   
>> You can do that if you want, but you lose one of the big benefits of PC 
>> based control.  On a PC based control, it is easy to find problems, add 
>> features, and update the software.  If you embed everything in some 
>> external box with no keyboard, no screen, no development tools, etc, you 
>> are rather locked down.  It's one thing to embed simple stuff that can 
>> be tested and proven robust, and that is unlikely to change.  It is much 
>> tougher to embed complex code that will need to be debugged and changed 
>> to meet evolving needs.
>>   
>> 
> Well, using the Cypress chip, or some other USB-FIFO device, is an 
> intermediate step.  If it only has a one ms buffer, the user would never 
> know the difference.  And, it would not be moving all the motion control 
> of EMC into some external device, it is just a FIFO and an interface device.
> Of course, it needs real-time determinism on the USB.  I have no idea 
> what state that is in, but I think there has been some work done in that 
> area.
> A quick Google search appears to show that EVERY document containing the 
> text "USB" also contains "real time", using it to means something 
> happening  within a couple of seconds!  UGH!  But, it looks like Jan 
> Kiszka, who did the rt-net package also has a USB stack for rtai, 
> originally started by Joerge Langenberg.  I should point out that I am 
> NOT volunteering for this project, although after I get some hardware 
> and software expertise with this Cypress chip, I might be willing to 
> contribute to such a project.
>
> I have a bigger interest in possibly using this chip to connect my other 
> boards to a PC without parallel ports.  It might also increase the 
> performance, as the CPU having to process each byte laboriously through 
> the parallel port is becoming a bottleneck.  But, I don't know if the 
> chip, or the USB model, is really conducive to the existing boards' 
> model of communication.  The fact that it can do one-way FIFO transfers 
> without intervention of the slow 8051 CPU is tantalizing, though.  With 
> some additional FPGA logic to format blocks of data to be read from the 
> FPGA to the CPU, though, I think it MIGHT work.  I need to learn more 
> about how many balls this chip can keep in the air at once.  But, the 
> idea is : Every micro-frame, the FPGA sends encoder position and digital 
> inputs to the CPU, and every micro-frame, EMC sends new velocity and 
> digital output info the the FPGA, based on the last data it processed.  
> It would always be one microframe out of sync, as the commands it is 
> sending to the FPGA are based on readings taken 2 microframes ago.  But, 
> as long as it is deterministic, it should work.
>
> Jon
>
> -
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>   

maybe this is of interest
RT-Firewire on Linux, using Xenomai (RTAI )
used for robotics, for realtime control of positioning systems
it also enables RTnet over firewire
several nice papers at http://www.rtfirewire.org
regards
TomP

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge

Re: [Emc-users] Parallel port in the future - query

[Jon Elson]

John Kasunich wrote:

> This subject seems to come up every month or two.  The answer is and 
> will probably always be the same.
>
> Yes, it is possible to make a box that lives outside the PC and can 
> queue up enough motion "in advance" that it doesn't require realtime 
> performance from the PC.
>
>   
I agree that taking the PC completely out of the real-time game is a 
mistake in two senses.  First, we've already done the "heavy lifting", 
and it works.  Second is, if you are NOT real-time, then there is no 
upper bound to latency, and one day, the system won't get the data there 
in time, and the buffer will run empty.  The machine will crash to a 
stop, possibly with disasterous consequences.

But, there might be another way.  Keep EMC real-time, but have a USB 
device that can pump out parallel bytes from a FIFO.  This requires EMC 
to sync to the USB clock instead of the system timer.  Every tick of 
that clock, a whole batch of step pulses, just like they are spoon-fed 
to the parallel port now, would be buffered and sent to the USB device 
to be de-buffered at a constant rate such that as the last byte was 
sent, the next buffer would be ready.  I think the Cypress CY7C68013 can 
do all this in hardware, once configured.
> You can do that if you want, but you lose one of the big benefits of PC 
> based control.  On a PC based control, it is easy to find problems, add 
> features, and update the software.  If you embed everything in some 
> external box with no keyboard, no screen, no development tools, etc, you 
> are rather locked down.  It's one thing to embed simple stuff that can 
> be tested and proven robust, and that is unlikely to change.  It is much 
> tougher to embed complex code that will need to be debugged and changed 
> to meet evolving needs.
>   
Well, using the Cypress chip, or some other USB-FIFO device, is an 
intermediate step.  If it only has a one ms buffer, the user would never 
know the difference.  And, it would not be moving all the motion control 
of EMC into some external device, it is just a FIFO and an interface device.
Of course, it needs real-time determinism on the USB.  I have no idea 
what state that is in, but I think there has been some work done in that 
area.
A quick Google search appears to show that EVERY document containing the 
text "USB" also contains "real time", using it to means something 
happening  within a couple of seconds!  UGH!  But, it looks like Jan 
Kiszka, who did the rt-net package also has a USB stack for rtai, 
originally started by Joerge Langenberg.  I should point out that I am 
NOT volunteering for this project, although after I get some hardware 
and software expertise with this Cypress chip, I might be willing to 
contribute to such a project.

I have a bigger interest in possibly using this chip to connect my other 
boards to a PC without parallel ports.  It might also increase the 
performance, as the CPU having to process each byte laboriously through 
the parallel port is becoming a bottleneck.  But, I don't know if the 
chip, or the USB model, is really conducive to the existing boards' 
model of communication.  The fact that it can do one-way FIFO transfers 
without intervention of the slow 8051 CPU is tantalizing, though.  With 
some additional FPGA logic to format blocks of data to be read from the 
FPGA to the CPU, though, I think it MIGHT work.  I need to learn more 
about how many balls this chip can keep in the air at once.  But, the 
idea is : Every micro-frame, the FPGA sends encoder position and digital 
inputs to the CPU, and every micro-frame, EMC sends new velocity and 
digital output info the the FPGA, based on the last data it processed.  
It would always be one microframe out of sync, as the commands it is 
sending to the FPGA are based on readings taken 2 microframes ago.  But, 
as long as it is deterministic, it should work.

Jon

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Jon Elson]

Dave Caroline wrote:
> The amount of data one could put in a USB microframe, may allow for
> more loops per frame so make it easier
>   
I can see some ways that one could export the software step generation 
to a USB parallel port.  Both the FTDI2232 and the Cypress 7C68013
(EZ-USB FX2 chip) can take a USB frame and crank it out byte by byte to 
an 8-bit output.  I think the FTDI would need an external "pacer", the 
Cypress chip could be made to pump out bytes at a programmable rate.  If 
EMC2 could be set up to schedule the base thread at some rate to keep in 
sync with the frames sent to the USB peripheral, it would pretty much 
solve the disappearing parallel port problem.  If we could come up with 
a scheme to emulate several hundred dispatches of the base thread to 
send one big packet of parallel port data writes to the USB device, it 
would probably reduce CPU overhead and even allow increased step pulse 
rates, while solving the par port crisis.

The FTDI chip is a pretty fixed-purpose device, but it can pump data 
to/from a FIFO at high rates.  The Cypress chip has an 8051 on it, but 
that is used primarily to set up the hundreds of control registers on 
it.  It can autonomously pump out data as 8- or 16-bit words, and can do 
a lot of flow-control, handshaking, or just fixed-rate throttling of the 
data.

I am going to be doing some work at "work" with the Cypress chip, I just 
got in a little board with it.  A complete set of open-source tools is 
available for it.  I will be learning how to use it under NON-RT Linux, 
and I don't know how much trouble there'd be in communicating with it 
from the real-time side.  Obviously, sharing the system's USB controller 
between RT and non-RT is non-trivial.

Jon

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Jim Coleman]

alan, have you considered replacing the fan?  this seems like the cheapest
solution to me :-D  and you could probably find a complete heatsink with fan
for your laptop on ebay, people part out laptops alot on there.
-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[tomp]

Jim Combs wrote:
> BUT, USB is a shared resource.  Up to 127 devices can be on the bus and
> there is no guarantee of timed delivery.
>
> You could identify the device a as a video interface and the USB bus would
> allocate specific bandwidth to an interface.
>
> Making a USB based IO EMC2 hardware controller would be the way to go.  I
> know enough to do the hardware side, but
> not enough to do the Ubuntu side.
>
> Then the Parallel port could go away and RIP.  It has served it's time VERY
> WELL!
>
>
>
>
>  "Dave Caroline"   
>[EMAIL PROTECTED]>   
> To 
>"Enhanced Machine Controller (EMC)" 
>  10/30/2008 12:50 
>  PM cc 
>
>Subject 
>  Please respond to Re: [Emc-users] Parallel port in
>  "Enhanced Machine the future - query  
> Controller 
>  \(EMC\)"  
>  <[EMAIL PROTECTED] 
>  sourceforge.net>  
>
>
>
>
>
> The amount of data one could put in a USB microframe, may allow for
> more loops per frame so make it easier
> Dave Caroline
>
> -
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>
> -
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>   

Jim,
you may be interested in 'determinism'
this is a characteristic of communications needed for realtime operation
firewire is determinisitic
usb is not

(in general, usb is not considered suitable for realtime by 
professionals in the industry)
you have a good idea, but maybe usb is not the route
and maybe firewire is not the route
but big companies have spents lots to find this out
it may be expensive for you to find the same answer.
or
you may find what they missed

best o luck
tomp

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[John Kasunich]

John Harris wrote:
> Hi Jim,
> I'm very interested in your proposed project, of moving the 
> 'Real Time' out of the box and into a USB replacement for the parallel port 
> breakout board.
> 
> I design motorized microscope XY stages and focus knob drives for Z, with a 
> 3-axis controller running from a USB port. I also have two mills that are 
> potential targets for a EMC2 based CNC upgrade.
> 
> When Microsoft (can I use that word?) upgraded to Win3.x and became 
> multi-threaded with the added feature of latency, I moved all my real-time 
> stuff out of the box. A FIFO command buffer takes care of the latency and a 
> bunch of FPGA code allows complete parallel control of the three axis.
> 
> I have been thinking some time, that a similar solution could be made for 
> EMC2, to switch to USB when there are no more parallel ports
> 
> So where would you divide EMC2 to split off the latency susceptible 
> real-time stuff, and what would be the software to software interface 
> through the USB port?
> 
> I have not done box software other than LabVIEW since Win95, so I'm a little 
> reluctant to dive into the EMC2 source code on my own.
> 
> John Harris
> 

This subject seems to come up every month or two.  The answer is and 
will probably always be the same.

Yes, it is possible to make a box that lives outside the PC and can 
queue up enough motion "in advance" that it doesn't require realtime 
performance from the PC.

The problems start when you start asking yourself how much is "enough"?
A hundredth of a second?  A tenth of a second?  A second?  Five seconds?

If all you are doing is following a path that doesn't depend on anything 
external, then sure, queue up 10 seconds or even 10 minutes of motion.

But what if the user hits stop?  Do you stop now, or after you've 
executed the 10 seconds of queued motion?  What if they turn up the feed 
override (or turn it down)?  What if they are doing a probe move, where 
the distance traveled isn't fixed, but depends on when the probe hits 
something?  What about threading, where the movement needs to slow down 
or speed up if the spindle speed changes, and needs to start just as a 
specific spot on the spindle goes by?

EMC is designed doing all of these things in the PC, because the PC 
(when running Linux with a realtime kernel) is capable of doing realtime 
control.  If you try to take "realtime" out of the PC, it isn't enough 
to just move step generation, or simple motion, out to some dedicated 
hardware - you have to move EVERYTHING that EMC currently does in 
realtime out of the PC.

You can do that if you want, but you lose one of the big benefits of PC 
based control.  On a PC based control, it is easy to find problems, add 
features, and update the software.  If you embed everything in some 
external box with no keyboard, no screen, no development tools, etc, you 
are rather locked down.  It's one thing to embed simple stuff that can 
be tested and proven robust, and that is unlikely to change.  It is much 
tougher to embed complex code that will need to be debugged and changed 
to meet evolving needs.

Regards,

John Kasunich





-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[John Harris]

Hi Jim,
I'm very interested in your proposed project, of moving the 
'Real Time' out of the box and into a USB replacement for the parallel port 
breakout board.

I design motorized microscope XY stages and focus knob drives for Z, with a 
3-axis controller running from a USB port. I also have two mills that are 
potential targets for a EMC2 based CNC upgrade.

When Microsoft (can I use that word?) upgraded to Win3.x and became 
multi-threaded with the added feature of latency, I moved all my real-time 
stuff out of the box. A FIFO command buffer takes care of the latency and a 
bunch of FPGA code allows complete parallel control of the three axis.

I have been thinking some time, that a similar solution could be made for 
EMC2, to switch to USB when there are no more parallel ports

So where would you divide EMC2 to split off the latency susceptible 
real-time stuff, and what would be the software to software interface 
through the USB port?

I have not done box software other than LabVIEW since Win95, so I'm a little 
reluctant to dive into the EMC2 source code on my own.

John Harris

- Original Message - 
From: "Jim Combs" <[EMAIL PROTECTED]>
To: "Enhanced Machine Controller (EMC)" 
Sent: Thursday, October 30, 2008 11:04 AM
Subject: Re: [Emc-users] Parallel port in the future - query


> Guess it would do well for me to sign my message!
>
> BUT, USB is a shared resource.  Up to 127 devices can be on the bus and
> there is no guarantee of timed delivery.
>
> You could identify the device a as a video interface and the USB bus would
> allocate specific bandwidth to an interface.
>
> Making a USB based IO EMC2 hardware controller would be the way to go.  I
> know enough to do the hardware side, but
> not enough to do the Ubuntu side.
>
> Then the Parallel port could go away and RIP.  It has served it's time 
> VERY
> WELL!
>
> Jim Combs (Lexington, Ky)
>
>
> -
> This SF.Net email is sponsored by the Moblin Your Move Developer's 
> challenge
> Build the coolest Linux based applications with Moblin SDK & win great 
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the 
> world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
> 


-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Jim Combs]

Guess it would do well for me to sign my message!

BUT, USB is a shared resource.  Up to 127 devices can be on the bus and
there is no guarantee of timed delivery.

You could identify the device a as a video interface and the USB bus would
allocate specific bandwidth to an interface.

Making a USB based IO EMC2 hardware controller would be the way to go.  I
know enough to do the hardware side, but
not enough to do the Ubuntu side.

Then the Parallel port could go away and RIP.  It has served it's time VERY
WELL!

Jim Combs (Lexington, Ky)


-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Jim Combs]

BUT, USB is a shared resource.  Up to 127 devices can be on the bus and
there is no guarantee of timed delivery.

You could identify the device a as a video interface and the USB bus would
allocate specific bandwidth to an interface.

Making a USB based IO EMC2 hardware controller would be the way to go.  I
know enough to do the hardware side, but
not enough to do the Ubuntu side.

Then the Parallel port could go away and RIP.  It has served it's time VERY
WELL!



   
 "Dave Caroline"   

To 
   "Enhanced Machine Controller (EMC)" 
 10/30/2008 12:50 
 PM cc 
   
   Subject 
 Please respond to Re: [Emc-users] Parallel port in
 "Enhanced Machine the future - query  
Controller 
 \(EMC\)"  
 <[EMAIL PROTECTED] 
 sourceforge.net>  
   
   



The amount of data one could put in a USB microframe, may allow for
more loops per frame so make it easier
Dave Caroline

-
This SF.Net email is sponsored by the Moblin Your Move Developer's
challenge
Build the coolest Linux based applications with Moblin SDK & win great
prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users


-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Dave Caroline]

The amount of data one could put in a USB microframe, may allow for
more loops per frame so make it easier
Dave Caroline

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Jon Elson]

alan wrote:
> So any suggestions for a cheap replacement computer. What is the
> smallest, cheapest desktop or laptop with a parallel port? I thought
> about using a pc/104 card but again here in the UK they seem to be quite
> expensive? Has anyone got experience using pc/104 ?
>   
Stuart Stevenson put me onto the Gateway Profile 3, which is basically a 
laptop MoBo and screen in a pedestal-mount case, with separate keyboard. 
it has one parallel port, USB and network.  it has a cardbus slot, but 
so far I have not been able to get that to work for a second parallel 
port.  These go fairly cheap on eBay.  1 GHz CPU, 256 MB on many of 
them.  Perfectly adequate for EMC2.  So, if a single parport is good 
enough, then it is one option.

Dell has some boxes that are the size of a brick, with hard drive, mobo, 
parallel port, etc. in them.  I don't remember the model number.  You 
would need a keyboard, mouse and video screen for these.
> So really my query is also about the future. Where will emc go if
> parallel ports become a thing of the past?
>
>   
Yeah..  At least with larger desktops, you can still plug a parallel 
port into the PCI or PCIe slots.  I still think there may be a hope for 
USB.  I have gotten a USB target board the size of a large postage stamp 
for some digital control retrofits at work.  This is not real time 
stuff, however.  But, if the USB clock were to be made to trigger the 
periodic interrupts that schedule EMC, then it may be possible to 
control servo systems over USB.  High speed mode has "micro-frames", 
allowing up to 8 transactions per millisecond.  That might allow 2 servo 
updates per ms, allowing for compute time.  (Maybe even 8 if you allow 
read/compute/write to lag one microframe.)

Another possible scheme would be to use any machine that had 2 network 
ports.  One for machine control, one for regular network.
(You could even do it with just one net port, but you'd have to keep the 
local network driver from messing with the port when EMC was using it.)
100 mb Ethernet should provide VERY good servo update times.

Jon

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Andrew Ayre]

Chris - what is a typical number of pulses per second that EMC2 can 
generate on one of these with 256Mb of RAM? I'd like to compare it to 
what I am currently using.

Andy

Chris Radek wrote:
> On Thu, Oct 30, 2008 at 02:10:59PM +, alan wrote:
>> I have built an ornamental turning lathe with stepper control, on a
>> moveable stand and I am really looking for something that sits on a
>> narrow shelf under the lathe next to my stepper controller box. So I am
>> looking for a eorocard or smaller sized pc with sufficient memory size
>> (whatever that is ). Anyone got a best buy suggestion?
> 
> 
> http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=300268637293
> 
> I have used these successfully.  They have fine realtime performance.
> They have parallel and serial ports.  They take a full size IDE hard
> disk.  There are lots of them around and they are cheap.
> 
> If you buy one beware it will only handle 256MB of RAM and also be
> sure you get the power pack with it.  The RAM is the only real
> limitation.  I have not tried the later (P4) models.


-- 
Andy
PGP Key ID: 0xDC1B5864

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Sebastian Kuzminsky]

Jeff Epler wrote:
> On Thu, Oct 30, 2008 at 12:34:36PM +, alan wrote:
>> So really my query is also about the future. Where will emc go if
>> parallel ports become a thing of the past?
> 

> It can go into embedded systems that are as powerful as today's desktop
> PCs, but with appropriate I/O (e.g., PC/104)

It should be possible to fit a nice EMC2 computer in about a liter 
volume, including power supply, CPU, storage, and I/O.

A PC-104/Plus motherboard like this (1.4 GHz Pentium M, 1 GB RAM, 
Ethernet, USB, parport, etc): 


A PC-104/Plus IO board like the 4i65 or 4i68 (very similar to the 5i20 
and 5i23, respectively): , 
about 2/3 of the way down the page

All in an extruded enclosure like this: 



Cheap?  No.

Neat?  Yes.


-- 
Sebastian Kuzminsky
what i like best about 12648430 is the coffee

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[tomp]

fan alan wrote:
> The fan on my laptop quit the other day so I started looking around for
> a replacement computer. Looking on the Ebuyer website here in the UK it
> seems that most new laptops (especially the smaller ones) now only have
> usb outputs. There is a wonderful little acer computer just released
> very small, onboard radio network and cheap, but it just has usb ports
> so is not suitable.
>
> I use a laptop because it is portable between my home and workshop
> (rather too grand a title , its a shed). I don't want to dedicate a
> large pc box to run my modest cnc needs (maybe I will have to) and I
> don't want to have to struggle carrying a large pc box between my home
> office and shed.
>
> So any suggestions for a cheap replacement computer. What is the
> smallest, cheapest desktop or laptop with a parallel port? I thought
> about using a pc/104 card but again here in the UK they seem to be quite
> expensive? Has anyone got experience using pc/104 ?
>
> So really my query is also about the future. Where will emc go if
> parallel ports become a thing of the past?
>
> Alan
>
>   
Alan
fan died on my HP ze1250 laptop
I lay it on top of a $1.98 110V fan
still working 2 yrs later
maybe worth a try
signed
cheap bastard

;)
tomp


-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[John Kasunich]

Jack Coats wrote:
> If you get a little brave, design a smaller case that is more 
> appropriate.  Wood is good!
> Put a good fan or two in it with filters, and keep a 'positive pressure' 
> in the case.  This will
> keep much of the dust/dirt/chips out.
> 

Personally I wouldn't go for wood - metal provides better grounding and 
electrical noise immunity.  But Jack's idea in general is a good one. 
PC cases aren't some mystical black art.  Just hack away at it till you 
have something you like.

A couple examples - for my own machine, I chopped up a large tower case 
into something much smaller:
http://jmkasunich.dyndns.org/cgi-bin/blosxom/shoptask/pc-cased-and-mounted-12-14-07.html

On the Mazak that we retrofitted at the CNC workshop, all the computer 
parts are simply mounted to a flat plate, which we attached to the 
inside of the cabinet door as you can see in this photo, to the left of 
Dave Engvall's head: http://fenn.freeshell.org/retrofest/dcp_0366.jpg
The strange blue glow in the upper left corner is from the tacky blue 
LED fan on the power supply.

Regards,

John Kasunich

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Jack Coats]

If you get a little brave, design a smaller case that is more 
appropriate.  Wood is good!
Put a good fan or two in it with filters, and keep a 'positive pressure' 
in the case.  This will
keep much of the dust/dirt/chips out.


In the OLD days before the IBM PC was a dream in Boca Raton FL, a guy 
George Morrow
was selling computers.  He had such a problem keeping folks in his 
company from shipping
the computer he was using for his desk top, he took a mother board out 
of the case and used
screws to mount it to the side of his wood desk.  Now he was able to 
keep a computer for him
to use. 
... You could use the same technique, nothing says a computer MUST be in 
a UL certified metal box!
... Make it just the right size for you, and screw it to the wall, so 
you have more table space!

... Idea payment required: post a picture of what you do! :) And let us 
know about it!

alan wrote:
> On Thu, 2008-10-30 at 08:00 -0500, Jeff Epler wrote:
>   
>> Cast-off Pentium-III or AMD Athlon desktops are probably the best
>> systems to seek out to run emc today.
>>
>> 
> I have come to appreciate that this is the case. I even have an Athlon
> that I can use (if I can persuade my wife to give it back) but I don't
> want to use it because the case is too big and I really haven't got the
> space for it in my shed. 
>
> I have built an ornamental turning lathe with stepper control, on a
> moveable stand and I am really looking for something that sits on a
> narrow shelf under the lathe next to my stepper controller box. So I am
> looking for a eorocard or smaller sized pc with sufficient memory size
> (whatever that is ). Anyone got a best buy suggestion?
>
> Alan
>
>
> -
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>   

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Chris Radek]

On Thu, Oct 30, 2008 at 02:10:59PM +, alan wrote:
> 
> I have built an ornamental turning lathe with stepper control, on a
> moveable stand and I am really looking for something that sits on a
> narrow shelf under the lathe next to my stepper controller box. So I am
> looking for a eorocard or smaller sized pc with sufficient memory size
> (whatever that is ). Anyone got a best buy suggestion?


http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=300268637293

I have used these successfully.  They have fine realtime performance.
They have parallel and serial ports.  They take a full size IDE hard
disk.  There are lots of them around and they are cheap.

If you buy one beware it will only handle 256MB of RAM and also be
sure you get the power pack with it.  The RAM is the only real
limitation.  I have not tried the later (P4) models.


-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[alan]

On Thu, 2008-10-30 at 08:00 -0500, Jeff Epler wrote:
> Cast-off Pentium-III or AMD Athlon desktops are probably the best
> systems to seek out to run emc today.
> 
I have come to appreciate that this is the case. I even have an Athlon
that I can use (if I can persuade my wife to give it back) but I don't
want to use it because the case is too big and I really haven't got the
space for it in my shed. 

I have built an ornamental turning lathe with stepper control, on a
moveable stand and I am really looking for something that sits on a
narrow shelf under the lathe next to my stepper controller box. So I am
looking for a eorocard or smaller sized pc with sufficient memory size
(whatever that is ). Anyone got a best buy suggestion?

Alan


-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Jack Coats]

As others have said, laptops have issues.  Dumpster diving and using 
discards from offices is
probably the cheapest way to support this habit :)

Eventually, I see EMC2 going to a 'client server' model, taking the real 
time portion and putting
it in an external controller, like a pc/104 or even a beefy PIC system.  
Communicating back to
its 'display system' for AXIS kind of display and to get the NC files 
from.  Communication could
be USB, but it could as well be Bluetooth (shudder), ethernet, or 
fiber.  Even a serial connection
(USB is just a fast serial port with more protocols running on it, for 
our purposes anyway) that is
'sufficiently fast' would work then.  By then the 'controller' could be 
the size of a business card if
we can attach the motor controllers and sensor wires to it!

Personally, I would get a 'dumpster diver' machine to leave in the 
shed.  Put your stuff on a USB
key and carry it back and forth.  If you have an ethernet network strung 
(wired or wireless) you
can monitor your rig from the comfort of your house while your machine 
is grinding way in the
workshop!  But you do want the real machine control to be 'local to your 
machine' in the shop.

If you can get a couple (or more) of similar machines free/cheap, stash 
the 'spare' away in the
shop for when you eventually need parts!

Does that help?

alan wrote:
> The fan on my laptop quit the other day so I started looking around for
> a replacement computer. Looking on the Ebuyer website here in the UK it
> seems that most new laptops (especially the smaller ones) now only have
> usb outputs. There is a wonderful little acer computer just released
> very small, onboard radio network and cheap, but it just has usb ports
> so is not suitable.
>
> I use a laptop because it is portable between my home and workshop
> (rather too grand a title , its a shed). I don't want to dedicate a
> large pc box to run my modest cnc needs (maybe I will have to) and I
> don't want to have to struggle carrying a large pc box between my home
> office and shed.
>
> So any suggestions for a cheap replacement computer. What is the
> smallest, cheapest desktop or laptop with a parallel port? I thought
> about using a pc/104 card but again here in the UK they seem to be quite
> expensive? Has anyone got experience using pc/104 ?
>
> So really my query is also about the future. Where will emc go if
> parallel ports become a thing of the past?
>
> Alan
>
>
> -
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>   

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Emory Smith]

On Thu, Oct 30, 2008 at 08:00, Jeff Epler <[EMAIL PROTECTED]> wrote:
> On Thu, Oct 30, 2008 at 12:34:36PM +, alan wrote:
>> So really my query is also about the future. Where will emc go if
>> parallel ports become a thing of the past?
>
> It can go to systems that still have parallel ports, even after they
> become unusual in new consumer PCs.
>
> It can go to systems where parallel ports are available on pci and
> post-pci add-on cards.
>
> It can go to systems where non-parport dumb I/O cards are available.
>
> It can go to pci and post-pci internal busses using intelligent cards
> like the mesa 5i20. (this is the path I personally favor)
>
> It can go into embedded systems that are as powerful as today's desktop
> PCs, but with appropriate I/O (e.g., PC/104)
>
> It can go in ways I haven't even considered in writing this e-mail.
> Whatever someone develops and contributes to emc, that's where emc will
> go.
>
> All that said, for the time being it is simply not a good idea to buy a
> machine without a parport or PCI slot.  I would 100% stay away from
> laptops.
>
> It is not a good idea to buy a laptop for emc (even one with a parallel
> port), because typically you have to choose between leaving SMI
> interrupts enabled (which will introduce huge, stall-inducing delays
> between steps) or disabling them and (according to Intel CPU and chipset
> documentation) burning up the system because fans and other thermal
> management techniques like CPU throttling may not work correctly
> without SMI interrupts.
>
> Cast-off Pentium-III or AMD Athlon desktops are probably the best
> systems to seek out to run emc today.
>
> Jeff
>
> -
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

As Jeff said, I'd stay away from laptops. Some have severe latency problems.
Our 2nd CNC box was an IBM PIII/800 from someone's trash. A 2Gig HD got
it running. Our current machine (#3) came from an office where they were getting
rid of old equipment. It was a P4/1.6. I dropped a P4/2.66G in it, bumped the
memory to 512 and it runs EMC fine. I took its 40G HD for myself and put a 10G
in it.  There are good finds in dumpster diving, curb shopping and being lucky.
Another office cleanout resulted in 12 PIII/1G, 256MB machines with 40G HDs.
A few of them had PCI parport cards for additional I/O.
Check out the local thrift shops too. They will dump machines that have simple
problems like dead HDs, etc.
It's amazing what people throw away.

-- 
First they ignore you. Then they laugh at you. Then they fight you.
Then you win.
Emory

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Parallel port in the future - query

[Jeff Epler]

On Thu, Oct 30, 2008 at 12:34:36PM +, alan wrote:
> So really my query is also about the future. Where will emc go if
> parallel ports become a thing of the past?

It can go to systems that still have parallel ports, even after they
become unusual in new consumer PCs.

It can go to systems where parallel ports are available on pci and
post-pci add-on cards.

It can go to systems where non-parport dumb I/O cards are available.

It can go to pci and post-pci internal busses using intelligent cards
like the mesa 5i20. (this is the path I personally favor)

It can go into embedded systems that are as powerful as today's desktop
PCs, but with appropriate I/O (e.g., PC/104)

It can go in ways I haven't even considered in writing this e-mail.
Whatever someone develops and contributes to emc, that's where emc will
go.

All that said, for the time being it is simply not a good idea to buy a
machine without a parport or PCI slot.  I would 100% stay away from
laptops.

It is not a good idea to buy a laptop for emc (even one with a parallel
port), because typically you have to choose between leaving SMI
interrupts enabled (which will introduce huge, stall-inducing delays
between steps) or disabling them and (according to Intel CPU and chipset
documentation) burning up the system because fans and other thermal
management techniques like CPU throttling may not work correctly
without SMI interrupts.

Cast-off Pentium-III or AMD Athlon desktops are probably the best
systems to seek out to run emc today.

Jeff

-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Parallel port in the future - query

[alan]

The fan on my laptop quit the other day so I started looking around for
a replacement computer. Looking on the Ebuyer website here in the UK it
seems that most new laptops (especially the smaller ones) now only have
usb outputs. There is a wonderful little acer computer just released
very small, onboard radio network and cheap, but it just has usb ports
so is not suitable.

I use a laptop because it is portable between my home and workshop
(rather too grand a title , its a shed). I don't want to dedicate a
large pc box to run my modest cnc needs (maybe I will have to) and I
don't want to have to struggle carrying a large pc box between my home
office and shed.

So any suggestions for a cheap replacement computer. What is the
smallest, cheapest desktop or laptop with a parallel port? I thought
about using a pc/104 card but again here in the UK they seem to be quite
expensive? Has anyone got experience using pc/104 ?

So really my query is also about the future. Where will emc go if
parallel ports become a thing of the past?

Alan


-
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users