On Wednesday 26 October 2016 00:42:02 Chris Albertson wrote:
> On Tue, Oct 25, 2016 at 9:11 PM, Gene Heskett
wrote:
> > On Tuesday 25 October 2016 23:07:55 bari wrote:
> >
> >
> > This is touted to be 100% compatible with R-Pi stuff. But has an
> > allwinner H3 brain, so
Does anyone here know of one verified case where a packet was dropped by a
switch? This just does not happen
Also, if you place a second Ethernet dongle on the Pi, the dongle and the
built-in port share a serial bus. This is not really an issue but it you
don't like the idea of a multiplexed
On Wednesday 26 October 2016 00:21:37 Chris Albertson wrote:
> I think SPI runs at 500 KHz. That is 50 times slower then Ethernet if
> you look at only the bit rate. But what matters, I think more is the
> time it takes to send one message. It might (maybe?) be faster on
> SPI because you
The Allwinner H3 is a quad core ARM processor... not intel. While the
orange Pi claims 100% compatibility with the raspberry Pi add-ons there
are reports of a few incompatibilities. Also note that it does have a
different graphics core and I think a different ethernet core as well
and not all of
On Tue, Oct 25, 2016 at 9:11 PM, Gene Heskett wrote:
> On Tuesday 25 October 2016 23:07:55 bari wrote:
>
>
> This is touted to be 100% compatible with R-Pi stuff. But has an
> allwinner H3 brain, so it runs on intel code, not arm. Its all ordered
> now Bari, so we'll find
> I think SPI runs at 500 KHz. That is 50 times slower then Ethernet if you
> look at only the bit rate. But what matters, I think more is the time it
> takes to send one message. It might (maybe?) be faster on SPI because you
> avoid the TCP/IP stack. With SPI your software is driving the
> ...
> Initiallly the hm2_eth driver provided no good recovery from dropped packets
> (it would hang for 100s of MS and cause a cascade of linuxcnc errors) but the
> errors occurred so seldomly that they were not a real problem (One test
> system
> had more than a year of uptime 24/7 at a 4
> The appeal of consumer products is that they can be much cheaper than
> industrial or custom products. Using a PC, and parallel port card with
> LinuxCNC can make for a very affordable machine controller, but I am
> always on the lookout for other options. Embedded processor cards are
>
I think SPI runs at 500 KHz. That is 50 times slower then Ethernet if you
look at only the bit rate. But what matters, I think more is the time it
takes to send one message. It might (maybe?) be faster on SPI because you
avoid the TCP/IP stack. With SPI your software is driving the bare
> >> The feedback comes from the rotary encoder that is attached to the back
> >> end of the servo shaft. So no backlash there.
> >
> > I was just thinking that looking at the command and feedback traces
> > might reveal any response issues which could be backlash, encoder
> > latency, or
I both succeeded and failed. I think inertia is high and friction low the axis
it failed.
On Tue, 25 Oct 2016 15:49:34 -0400 (EDT)
"Todd Zuercher" wrote:
> at_pid was a nice idea, but has anyone successfully made it work? (I'm
> pretty sure
On Tuesday 25 October 2016 23:07:55 bari wrote:
> @Gene
>
> Not sure how well SPI works on the Orange Pi. Since you've decided on
> the 7i90HD with SPI why don't use use the Rpiwith it?
>
> Orange Pi board, Ethernet > 7i92, hm2_ethdriver
>
>
> Rpi board, SPI >7i90, hm2_raspi driver
This
The surplus PC is only cheaper if it is indeed surplus. I just got two
for free.A free quad core I7 is the most compute power per dollar I'm
likely to find.
But for other applications I need a smaller physical size and low enough
power to run on batteries. Low cost is a nice plus if I can
On Tuesday 25 October 2016 21:52:08 Gene Heskett wrote:
> On Tuesday 25 October 2016 21:03:51 W. Martinjak wrote:
> > On 2016-10-25 23:10, Kirk Wallace wrote:
> > > So, I'm back to piecing together some sort of beagleduino
> > > thing if I want a pad class controller.
> >
> > OK, then let me
@Gene
Not sure how well SPI works on the Orange Pi. Since you've decided on
the 7i90HD with SPI why don't use use the Rpiwith it?
Orange Pi board, Ethernet > 7i92, hm2_ethdriver
Rpi board, SPI >7i90, hm2_raspi driver
On 10/25/2016 08:52 PM, Gene Heskett wrote:
> On Tuesday 25
On 10/25/2016 06:04 PM, hubert wrote:
> Is there any way to move past home position.
>
> Each of my Axis, X, Y, and Z use three sensors to provide inputs for min
> and max limits and Homing. We have the Home switch set about 1 inch
> inside the max limit switch. However, We need to raise the Z
On 2016-10-25 23:10, Kirk Wallace wrote:
> So, I'm back to piecing together some sort of beagleduino
> thing if I want a pad class controller.
>
OK, then let me promote my hm2_raspi driver.
https://forum.linuxcnc.org/27-driver-boards/31753-raspberry-pi-and-mesa-7i90-spi-works-well#82070
You may
If you would include a link to your ini file it would be much easier to
diagnose and answer.
On Oct 25, 2016 6:11 PM, "hubert" wrote:
> Is there any way to move past home position.
>
> Each of my Axis, X, Y, and Z use three sensors to provide inputs for min
> and max limits and
There are multiple parameters:
1. You can declare the homing switch to be a coordinate other than 0.
e.g. "after we settle on the switch, this axis is offset so this is now
+1.25 inches"
2. You can set axis limits at anything. I probably would never set a
negative coord as a limit but
Just guessing, but you might have software limits set to 0. That'd be a
configuration issue, I'd guess the .ini file or the main *.hal file would
have those in it.
You should be able to travel past the home switch. On a side note, I have
no home switches and will 'manually' home in Axis by
On 10/25/2016 05:10 PM, Kirk Wallace wrote:
> the hardware inside is radio hardware, so the main board would be wasted on
> a machine controller
A lot of people like to listen to music or talk radio in the shop. Maybe
that radio hardware is not a waste after all. :-)
I need a dual core
Is there any way to move past home position.
Each of my Axis, X, Y, and Z use three sensors to provide inputs for min
and max limits and Homing. We have the Home switch set about 1 inch
inside the max limit switch. However, We need to raise the Z axis above
the home switch but inside the
The appeal of consumer products is that they can be much cheaper than
industrial or custom products. Using a PC, and parallel port card with
LinuxCNC can make for a very affordable machine controller, but I am
always on the lookout for other options. Embedded processor cards are
popular now,
On 10/25/2016 01:15 PM, Kirk Wallace wrote:
> On 10/25/2016 10:35 AM, Marius Liebenberg wrote:
>> The feedback comes from the rotary encoder that is attached to the back
>> end of the servo shaft. So no backlash there.
>
> I was just thinking that looking at the command and feedback traces
> might
On 10/25/2016 10:35 AM, Marius Liebenberg wrote:
> The feedback comes from the rotary encoder that is attached to the back
> end of the servo shaft. So no backlash there.
I was just thinking that looking at the command and feedback traces
might reveal any response issues which could be backlash,
at_pid was a nice idea, but has anyone successfully made it work? (I'm pretty
sure there is a good reason few if anybody is using it.)
- Original Message -
From: "Marius Liebenberg"
To: "Nicklas Karlsson" , "Enhanced Machine
On Tue, 25 Oct 2016, Stephen Dubovsky wrote:
> Date: Tue, 25 Oct 2016 14:40:46 -0400
> From: Stephen Dubovsky
> Reply-To: "Enhanced Machine Controller (EMC)"
>
> To: "Enhanced Machine Controller (EMC)"
On Tue, 25 Oct 2016 12:33:57 -0600
Charles Buckley wrote:
> I work in an environment at work where we do a lot of video broadcast over
> IP. That is all UDP multicast. Luckily, it is an environment where single
> dropped packet only corrupts that frame and section and is
All switches do at least store and forward today. Even the $20 cheap
ones. We use ethernet for real time control in some of our products.
Faster loop time than LinuxCNC. We use raw ethernet frames so that we
don't even need the ethernet stack (no DNS/etc needed either.) We also run
web
On Tue, Oct 25, 2016 at 11:09:04AM +0200, Alexander R??ssler wrote:
> Is there a way to change the rapid traverse rate (G0) ate run-time?
> Preferably without touching NML...
You can choose from the Rapid Override slider (percentage) and the
Max Velocity slider (velocity cap).
If neither of
I work in an environment at work where we do a lot of video broadcast over
IP. That is all UDP multicast. Luckily, it is an environment where single
dropped packet only corrupts that frame and section and is recovered the
next base frame (ie, anywhere from 1 to 16 frames).
The reason we use UDP
On 10/25/2016 01:46 PM, Chris Albertson wrote:
> I hate to say it but, a lot of old wives tales here. What is the
> latency added by a cut through switch? It's the packet header length
> divided by the Ethernet bit rate. If you are worried about this kind
> of stuff them you might as well
I hate to say it but, a lot of old wives tales here. What is the
latency added by a cut through switch? It's the packet header length
divided by the Ethernet bit rate. If you are worried about this kind
of stuff them you might as well start using shortened cables and wire
because of speed or
I take it that the PID values that where calculated will be seen in
halshow for that PID? One can then use those in the normal PID setup.
-- Original Message --
From: "Nicklas Karlsson"
To: "Marius Liebenberg" ; "Enhanced Machine
The feedback comes from the rotary encoder that is attached to the back
end of the servo shaft. So no backlash there.
-- Original Message --
From: "Kirk Wallace"
To: emc-users@lists.sourceforge.net
Sent: 2016-10-25 18:46:32
Subject: Re: [Emc-users] Problem
> ...
> A queue is not real time. Real time is *now* not when the queue decides
> to send. Losing packets in a critical environment is not a good thing.
Real time is usually within a certain time span, for example buffer half full
so it must be emptied before full.
> Let's suppose a limit
I have not. Dont know how I did not know about this. I will try it.
>Did you try the tune mode?
>
>It could be found in the hal configuratio and set via "setp" command.
>Manual http://linuxcnc.org/docs/html/man/man9/at_pid.9.html further
>down on page "pid.N.tune-"
>
>
>
>
>
>On Tue, 25 Oct
Because collisions do happen on modern switches, and using UDP with no
error correction in the protocol would cause those packets involved in a
collision to be dropped. Switches are another layer of complexity,
latency and expense when a dual port NIC is the cheapest and most
reliable way to
On 10/25/2016 12:52 PM, Chris Albertson wrote:
> Are you still certain you need two Ethernet ports? I can't see why
> you need two on a modern switched network.
>
> On Mon, Oct 24, 2016 at 7:05 PM, Mark Johnsen wrote:
>> On Mon, Oct 24, 2016 at 6:52 PM, Gene Heskett
My guess is that people are reading 20 year old information, using two
ports and
sure enough it works just fine. So they claim "I used two ports and it worked"
so the next person does the same and no one thinks about "why".
On Mon, Oct 24, 2016 at 7:42 PM, Stephen Dubovsky
On 10/25/2016 12:16 PM, Nicklas Karlsson wrote:
>> True dat. But why take the chance and not use a direct connection with
>> a cross-over cable? Typical UDP traffic in a switched LAN is fairly
>> fast but not necessarily totally reliable. And over the years I've seen
>> plenty of collisions on
Are you still certain you need two Ethernet ports? I can't see why
you need two on a modern switched network.
On Mon, Oct 24, 2016 at 7:05 PM, Mark Johnsen wrote:
> On Mon, Oct 24, 2016 at 6:52 PM, Gene Heskett wrote:
>
>> 1. someone said the 7i92H
On 10/24/2016 11:16 PM, Marius Liebenberg wrote:
> Hi All
> I have a problem with a machine using servos. It is a 4 axis machine
> with the 4th axis being rotary. The linear servos are tuned and working
> very well but the rotary axis is not behaving well. It has a severe
> oscillation and I
> True dat. But why take the chance and not use a direct connection with
> a cross-over cable? Typical UDP traffic in a switched LAN is fairly
> fast but not necessarily totally reliable. And over the years I've seen
> plenty of collisions on a switched network for both TCP and UDP. The
>
Did you try the tune mode?
It could be found in the hal configuratio and set via "setp" command. Manual
http://linuxcnc.org/docs/html/man/man9/at_pid.9.html further down on page
"pid.N.tune-"
On Tue, 25 Oct 2016 06:16:04 +
"Marius Liebenberg" wrote:
> Hi All
>
On 10/25/2016 11:26 AM, Przemek Klosowski wrote:
> On Mon, Oct 24, 2016 at 10:42 PM, Stephen Dubovsky
> wrote:
>> Why would UDP need resends on a shared ethernet port? There are no
>> collisions on a full-duplex port & switch (which is pretty much ALL of them
>>
On Mon, Oct 24, 2016 at 11:18 PM, Chris Albertson
wrote:
> Let's say we have two computers and each is sending UDP packets to two
> different Mesa cards. I can't see how those packets would ever be on
> the same physical cable, assuming a switched network. Each
On Mon, Oct 24, 2016 at 10:42 PM, Stephen Dubovsky wrote:
> Why would UDP need resends on a shared ethernet port? There are no
> collisions on a full-duplex port & switch (which is pretty much ALL of them
> now-a-days.) Passive hubs went the way of the dodo.
Of course,
On 09/10/2016 12:35 PM, Marius Liebenberg wrote:
> The cncprofi work very well and with no fuss. I used it many times.
I finally had time for my project again and I bought the cncprofi one. I
managed to connect it to the plasma and it seems to work (shows up and
down and it changes when I shout
You'd have to write VHDL code, compile into a bitfile, and then just use the
mesaflash command-line to upload to the card.
Danny
Les Newell wrote:
> I did wonder how expensive it is to write to the Mesa cards. I would
> like to have a go at converting one of
The ratio is 40:1. If I make the pulses per rev very high the system
becomes stable but the calibration is out off course. This supports your
idea I think if I understand you correctly.
-- Original Message --
From: "andy pugh"
To: "Marius Liebenberg"
On 10/25/2016 10:18 AM, Stephen Dubovsky wrote:
> Im fully aware UDP doesn't resend. But there are no collisions on modern
> ethernet LANs. No more reason a UDP packet would be corrupted going
> through a (shared) switch as a direct connection.
True dat. But why take the chance and not use a
Im fully aware UDP doesn't resend. But there are no collisions on modern
ethernet LANs. No more reason a UDP packet would be corrupted going
through a (shared) switch as a direct connection.
On Tue, Oct 25, 2016 at 8:23 AM, Mark wrote:
> On 10/24/2016 10:42 PM, Stephen
I did wonder how expensive it is to write to the Mesa cards. I would
like to have a go at converting one of my lasers one day. However I have
a mill and router that need retrofitting before I look at any laser stuff.
Les
On 25/10/2016 13:25, Todd Zuercher wrote:
> About 4-5kHz in a floating
On Tuesday 25 October 2016 08:25:14 Todd Zuercher wrote:
> About 4-5kHz in a floating point base-thread was about as fast as I
> was able to get the read/writes for my Mesa 5i25/7i77 combo to run at.
> There might be tricks to get it a little faster but I doubt there is
> a lot more to be gained
About 4-5kHz in a floating point base-thread was about as fast as I was able to
get the read/writes for my Mesa 5i25/7i77 combo to run at. There might be
tricks to get it a little faster but I doubt there is a lot more to be gained
with that combo. Smart serial communication seemed to be the
On 10/24/2016 10:42 PM, Stephen Dubovsky wrote:
> Why would UDP need resends on a shared ethernet port? There are no
> collisions on a full-duplex port & switch (which is pretty much ALL of them
> now-a-days.) Passive hubs went the way of the dodo.
UDP won't, and can't do resends. Unlike TCP
On 2016-10-25 12:34, andy pugh wrote:
> On 24 October 2016 at 23:01, Sebastian Kuzminsky wrote:
>>> I don't _think_ that Machinekit has NML any more. I think it has all
>>> been replaced by something else (0mq possibly)
>> I believe Machinekit still uses NML in exactly the same
On 25 October 2016 at 07:16, Marius Liebenberg wrote:
> The 4th axis used to be a stepper and that worked well but we changed the
> stepper for a servo to get more speed through a reduction planetary gearbox.
> I attached the HAl and INI files.
What is the ratio of the
On 25 October 2016 at 10:09, Alexander Rössler wrote:
> Is there a way to change the rapid traverse rate (G0) ate run-time?
> Preferably without touching NML...
Not in Auto mode, I don't think, but there are
ini.joint.N.max-velocity and ini.axis.L.max-velocity pins.
But
Hi Rene
I am using the standard analog servo channel on the 7i37 with an encoder
feedback that comes from the servo drive.
-- Original Message --
From: "Rene Hopf"
To: "Marius Liebenberg" ; "Enhanced Machine
Controller (EMC)"
I have no idea who or why machinekit was forked. Usually if or then I
contribute something to open source it is something I want for myself and I
guess it is similar for most, discussions to make it fit with other parts
or solutions is usually part of this.
2016-10-25 12:34 GMT+02:00 andy pugh
On 25 October 2016 at 04:09, Danny Miller wrote:
>
> Second, I ended up with the NPN proximity sensors shown here on the top
> left:
>
> https://oceancontrols.com.au/IBS-0051.html
>
> Those are insidiously awkward to interface,
Why? You feed them with 24V for power, and
On 24 October 2016 at 23:01, Sebastian Kuzminsky wrote:
>> I don't _think_ that Machinekit has NML any more. I think it has all
>> been replaced by something else (0mq possibly)
>
> I believe Machinekit still uses NML in exactly the same way that
> LinuxCNC does.
I suppose I
On 25 October 2016 at 09:48, Les Newell wrote:
>
> Is there any reason why the realtime part of the raster component can't
> run in a faster thread, say 10kHz? As long as you have hardware PWM (e.g
> Mesa FPGA) you should be able to get pretty good resolution.
Hi,
First thing I notice, you dont have any ff term. I guess you have ho external
current or velocity regulator, just a position pid outputting voltage via pwm?
That is not really the best setup, but still should be possible to tune…
Rene
> On 25 Oct 2016, at 08:16, Marius Liebenberg
Is there a way to change the rapid traverse rate (G0) ate run-time?
Preferably without touching NML...
Thanks,
Alexander
--
The Command Line: Reinvented for Modern Developers
Did the resurgence of CLI tooling catch you
On 24/10/2016 22:25, dan...@austin.rr.com wrote:
> So there is another mode I didn't mention. When accelerating or
> decelerating, you're not at the specified cut speed, and the laser power must
> be reduced proportional to the speed.
>
A better way to solve this problem when raster engraving
On Monday 24 October 2016 22:42:31 Gregg Eshelman wrote:
> How about some pics of that modification?
You wouldn't see much, Gregg, that $9 (ebay) bob used individual 4 pin
opto's, so I had axis start the spindle, used a scope to locate which
one was carrying the pulse, removed it and put a
Hi All
I have a problem with a machine using servos. It is a 4 axis machine
with the 4th axis being rotary. The linear servos are tuned and working
very well but the rotary axis is not behaving well. It has a severe
oscillation and I cannot seem to get the PID trimmed to stabilize the
servo.
70 matches
Mail list logo