On Tue, 10 Dec 2019, Gene Heskett wrote:
Date: Tue, 10 Dec 2019 13:52:47 -0500
From: Gene Heskett <ghesk...@shentel.net>
Reply-To: "Enhanced Machine Controller (EMC)"
<emc-users@lists.sourceforge.net>
To: emc-users@lists.sourceforge.net
Subject: Re: [Emc-users] Need a power signal independent of linuxcnc.
On Tuesday 10 December 2019 11:52:06 Peter C. Wallace wrote:
On Tue, 10 Dec 2019, Gene Heskett wrote:
Date: Tue, 10 Dec 2019 11:18:11 -0500
From: Gene Heskett <ghesk...@shentel.net>
Reply-To: "Enhanced Machine Controller (EMC)"
<emc-users@lists.sourceforge.net>
To: emc-users@lists.sourceforge.net
Subject: Re: [Emc-users] Need a power signal independent of
linuxcnc.
On Tuesday 10 December 2019 10:16:16 Peter C. Wallace wrote:
On Tue, 10 Dec 2019, Gene Heskett wrote:
Date: Tue, 10 Dec 2019 06:46:16 -0500
From: Gene Heskett <ghesk...@shentel.net>
Reply-To: "Enhanced Machine Controller (EMC)"
<emc-users@lists.sourceforge.net>
To: emc-users@lists.sourceforge.net
Subject: Re: [Emc-users] Need a power signal independent of
linuxcnc.
On Monday 09 December 2019 22:45:26 Gene Heskett wrote:
Greetings PCW and all;
I find I need a way to do a powerdown reset on the interface box
of one of my machines. And it needs to be separate from linuxcnc.
That, or some way to tickle the 7i76D to reset it before I start
LinuxCNC the second time because hal can't find spinout.
I am finding that little power bumps, not enough to do a full
powerdown reset, are killing the 5i25 to 7i76's serial com.
Short of rigging a 3rd hand switch on the shelf to toggle the
interface power, or getting out a stepladder to access the one on
the rear of the shelf, once the 7i76 has been bumped, the seriel
port is dead until a full, several seconds dark powerdown, there
seems to be no other recovery method that works.
Ideas anyone??? Or do I go get a switch plus socket and position
it where I can operate it from the front of the machine?
Thanks all.
Shucks, I was hoping there was a secret handshake I could put in a
rapper for linuxcnc that would tickle the right 5i25 bit(s) to do
a full reset on the 7i76D just before linuxcnc was started.
I would verify that your field an 5V power are OK at the 7I76 end.
I dont see any way a short power failure/glitch should get the
system into a unrecoverable state.
The supplies are separate x volt by several amps little boxes, and
voltages are +- .1 volts of 5 and 12 volts(field power)
This has occurred several times now and has always needed a
powerdown of some seconds duration to recover from. About a week
ago during a storm 20 miles north I experienced about 15 minutes
worth of flickering lights, but no bumps to total darkness of the
overhead lights.
The next evening the mobo in this box went up in flames with the
fire starting at where one of the usb2 breakouts was plugged into
the mobo.
I moved the boot drive to an old dell 745 I pulled out of the midden
heap, but it didn't have enough memory to do much more than email.
Took me a few days to order enough stuff to put a new mobo of
considerably higher performance in this box, and when I had it up
and running, by moving that same boot drive back into the drive cage
in this boxes tower, linuxcnc refuses to run over an "ssh -Y shop"
link because the 5i25 isn't talking to the 7i76D over the serial
link. Because I am switching power to other things around this
interface with this interface, the interface itself is powered
continuously, with no means to reset it other than taking a short
ladder to the back of the shelf where power comes to the shelf, and
toggling the switch on one of those long 6 outlets gismos.
At the moment, that ladder is blocking my access and has all the
chucks, tailstocks and goodie plates for a BS-1 table on it, so just
to move the ladder, I first have to move about 80 lbs of stuff
someplace else. And I'm going to have to invent a skyhook just to
get the BS-1 onto the milling machines table as its 56kg all by
itself. And my back is no longer up for that sort of weight at 85
yo.
IMO the watchdog timer should initiate a reset after making a gui
entry that its bitten. Clearing the gui should initiate the reset.
Then restart linuxcnc and its all working again at the restart.
I still suspect something else is going on here, as we have never seen
an issue like this. I'll try and duplicate by cycling power slowly to
a 7I76 multiple times, but Like I say I have never seen this behavior
before. The PIC code has brownout detection and a hardware watchdog so
should never be able to get into a state that requires power cycling
to recover.
But it does Peter, its hit the other machine with a 7i76D (my GO704) in a
nearly identical interface, but I've a manual switch for stepper power
there that also switches the interface power now. So as long as thats
turned on before starting linuxcnc its fine. On this machine, lcnc
controls motor power but that requires the interface to be powered full
time. If there have been any bumps during that idle time, then I get:
Are you sure that the 5I25 is jumpered for external 5V power on the
daughtercard? This can cause odd power related issues
---------------------------
Starting LinuxCNC...
.
Found file(REL): ./6040-5i25-7i76.hal
./6040-5i25-7i76.hal:66: Pin 'hm2_5i25.0.7i76.0.0.spinout' does not exist
Shutting down and cleaning up LinuxCNC...
---------------------------
And I have to go do a powerdown reset on the interface box, long enough
for both supplies to die which takes a good 10 seconds, to recover. Its
fine after the next powerup. The only diff between the two interface
boxes is the wiring and the brand of BoB on the 5i25's p2, its a
modified SainSmart on the GO704's version, and a rev-4 C1G on the 6040.
The SainSmart was used on the GO704 because its dead easy to bypass the
way too slow opto's on the SainSmart. I love the cnc4pc C1G, but the
rev-4, the newest one is both lots bigger because it has all those bells
and whistles, and out of the catalogs for about a decade at cnc4pc.com.
Priced way too high for a BoB, but what a BoB.
Earlier revisions are a different critter entirely, with optos all over
the place, slow ones too, worthless for anything moving faster than
software stepping. The rev-4 has no opto's, heavier current outputs and
only about 10 ns delay everywhere. And leds for status everywhere
Everything a BoB should be but it was also $80+ bucks.
So I guess I'll have to stop at Lowes on my way home from the
horsepistol tomorrow, and get one of those combo switch and outlet
gizmo's and a handibox.
What I did do was buy a small, refurbed 650W cyberpower from tigerdirect,
for $35.30 with taxes and free shipping today only. Even with only one
battery in it, TD has to be losing their shirt on shipping. At this
light a load, holdup s/b half an hour or more. :) Worth it to me if it
fixes this.
Can I take any measurements for you before I reset it?
Thank you Peter.
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
If we desire respect for the law, we must first make the law respectable.
- Louis D. Brandeis
Genes Web page <http://geneslinuxbox.net:6309/gene>
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
Peter Wallace
Mesa Electronics
(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
