Den 2022-05-21 kl. 14:02, skrev gene heskett:
On Saturday, 21 May 2022 04:52:28 EDT you wrote:
Den 2022-05-19 kl. 23:57, skrev gene heskett:
On Thursday, 19 May 2022 16:19:37 EDT Marius Alksnys wrote:
Gene,
how about setting min and max limits to zero (or other initial
value),
maybe also not enabling B drive?
This can be done in a bash script like:
halcmd unlinkp io-board-going-to-b-servo-drive.enable-pin
halcmd setp ini.4.min_limit 0
halcmd setp ini.4.max_limit 0
halcmd setp ini.b.min_limit 0
halcmd setp ini.b.max_limit 0
or in HAL with mux_generic
http://linuxcnc.org/docs/html/man/man9/mux_generic.9.html
First I've heard of this, sounds like it could be usefull. ATM I'm
carving up an e-stop coupling between linuxcnc and a couple of those
3
phase drives.
These 3 phase drives have STO Safe Torque Off you connect to a real
emergency stop button? You use a safety relay?
Not a relay in some cases, although on this machine there is a relay
because there's two power supplies. A 120 volt to run most stuff, and a
250 for the spindle motor.
The low voltage 127 feed and the 254 for the vfd are interlocked in hal
by being disabled when F2 is toggled off. In this machine by a relay, but
in most of my machines by BIG SSR's controlled by logic level signals.
The Fault outputs are simply linked into the E-stop to toggle F2 off, and
by cross linkage in the hal file, toggles F1 to the inactive state also.
A safety relay like this on connected to the emergency button. This one
seems to be German made, me myself have French pilz in my machine and
guess it work equally well.
It is common relay have an extra normally closed NC contact and plan is
to connect these in series with a magnetic unlock mechanism to the door.
If relays are correctly designed door will be locked until relays are
fully closed.
Nicklas Karlsson
Long answer, might have missed some details below.
So in the event of a Fault, the whole machine is powered down. This
powerdown is also the fault reset. Because they pulse the Fault while
powering up as you press F2 again, there is a oneshot triggered by F2 on
that disables the fault detection for about .8 seconds, otherwise it
faults on the power up. Forever.
The Sheldon lathe is all by SSR's as the 127 volt motor power is by one
leg of the 254. Control trigger straight from the 7i90.
The GO704 has an even more complex time delay startup to limit the inrush
as it spindle motor supply has too many microfarads that must be be
charged slowly else it clears a 30 amp breaker in the service if switched
straight on, so it gets a 50 ohm 200 watt resistor in series with its
line feed for about 4 seconds. So now it runs on a 20amp breaker that is
right at its trip point if the spindle motor stalls as the pico pwmservo
that drives it is set for an 18 amp max. There, if I had those motors on
it, but they are regular 2 phase steppers, I revert to resistor limits
for 1 second to give the motors time to get stopped, faster than just
dropping power. So there two SSR's involved there.
TLM is in the process of being converted to a full poweroff, but has
normal steppers. Basicly the machines are dead until F2 has been toggled
on, then pull the tool if its a lathe and rehome. The lathes get homed at
the X out stop first, then Z.
On the 6040, Z first, at the top to save broken tooling, then Y at the
front, X to the left, then B to whatever offset is zero degrees.
All that is subject to fine tuning yet as I need to recal Z for the new
motor too. The 6040 has had the spindle raised in its mount to gain some
vertical range at the same time, but I've not yet re-attached a shares
linear scale to re-calibrate it. Assuming the same u-step divisor in the
driver, I just multiplied its scale in the .ini file by 1.5, seems pretty
close, these motors are 1.2 degree motors, not 1.8 degree. Changing the Z
motor took the up speed max from about 20mm/sec to about 150 going up,
the OEM motor could just barely lift a decent spindle motor. This 1nm
motor lifts it with authority.
This is the third time I've rebuilt the electronics on this 6040, the OEM
supplied stuff was junk, incapable of moving the machine, so it went ino
the trash trailer about a week after I boght it. And it has a lot of
gingerbread I'm still tracing out, bringing to life in the .hal file,
mounting and wiring up. When done, the mist checkbox will only control
the liquid pump while the gui controls the volume, and the flood checkbox
will enable the air compressors to drive the mister with more gui
controlling the pressure, so I can clear sawdust with a dry air blow. I
intend to make some vise screws for my and others woodworking bench
vises, buttress threads on about a 1.9" diameter stick of hard maple, 2
start threads with a 12mm pitch. 3d printed in PETG half nuts, many times
stronger than PLA, are alike with 2 starts, and 2 starts for faster
clamping on a board. Current G-code is for a 12mm per turn pitch, fine
tuned to exactly match the nuts I'm designing in openscad that have
around a 9.5 to 10 thread engagement, about 2" of threads engaged.
Might not work, but if it does, it will beat anything on the market
today. You will see it advertised in Fine WoodWorking in that event.
Thats if I don't miss roll call first. At 87, I've made the reaper blink
first several times already. I went down to some of my wifes kin's shop
and had them make me 2 T-shirts "Built in the thirties, original and
unrestored, some parts still working". ;o)> So despite the aches and
pains of old age, I am having fun, pretending I still matter. ;)
> Nicklas Karlsson
Take care and stay well Nicklas.
Cheers, Gene Heskett.
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
