Post by Stephen DubovskyThe positioning ability of an induction motor is no different to that
of a DC servo motor. When no torque is needed both go to zero
voltage/current to stop moving (though technically an induction motor
can maintain a DC current if it chooses to to provide braking -
similar to shorting out the DC motor leads.) When torque is needed to
maintain position excitation must be applied to each. Only difference
is one is DC and one is AC. There are plenty of AC induction servos.
They are typically one of the highest performance machines as the
rotor inertia can be made lower than DC or PMSM rotors. It does take
more advanced control software. Also, the wider torque vs speed
curves are typically unmatched.
Post by Dave ColeIs the positioning ability adequate?
Most squirrel cage motors are really bad at positioning under load
as they rely on slip to generate armature flux. No slip -> no flux
-> no torque But you are just trying to position the tool for the
changer, right?
The best I have seen for positioning with a load is basically a
coast/brake to a stop position and then a brake is applied at the
stop position.
Is there a reason why you did not use the hal component that Andy
pointed out? Or did you modify that?
Dave
Post by Marius Alksnysand in orientation / position mode I_out is constant and f_out is
controlled by PID.
I saw this again, and looked at the orient manpage, but for me it seems
inadequate.
It doesn't detail how the spindle control is switched from one pid to
the otherm but I'd assume a mux2 is involved.
Next, I see no output that could trigger a small solenoid to put some
pressure on the locking pin so that it would drop in as the locking hole
came by.
The solenoid however would be a bit bulky sticking out on the front of
the spindle brake/lock unless I were to rebuild it to become part of the
pin mount. Do-able but at how much time finding a bobbin and suitable
wire to wind the solenoid? I'd also have to find my missing round tuit.
My vision of that would be to have issued a slow rotation command of say
20 rpms, energizing the solenoid at the same time, and when the pin
dropped in, locking the spindle, the error would have went way up, so
use a comparator to issue a reverse at the same speed to assure the lock
pin was fully seated, and when the error went up again, shut the command
off but leave the solenoid enabled. Make that part of the manual tool
changer after Z has been run to the upper home switch for clearance.
When the tool has been changed, and the enter or click has been made,
release the solenoid, run z back to where it started, issueing the m3 a
second or so later.
In my case, making that very slow to lock position part of the m5 command
would be one way to stop it quickly as Jon's servo driver, with my
current PID settings, can take it from 2700 to 25 rpm in a very small
fraction of a second. All I hear is a click as I hit enter, and its
running at the new, slower speed. As Jon's pwm servo driver is a true 4
quadrant controller, that also dumps energy back into the filter banks
in the motor psu, putting the caps well above their surge voltage
ratings. Ditto the reversal in a G33.1, so I don't tap at more than
200-300 revs in deference to those caps. I profile that somewhat in a
rate limit module, but I need to slow that down to ease that pulse back
into the psu. I had put that into the ini file with a MAXACCEL = 1500
but just now added a decimal point so its 150.0 now. That may be too
boringly slow, we'll see. Since it also impinges on the depth of a
G33.1 in tap breaking overshoots, that will have to be looked at
carefully. I have some hal code in the lathe with its HEAVY chuck that
measures that overshoot that I should probably transfer to the mill if I
slow its velocity rate that much.
At least I'll know how much it is when cutting air. I run the code here
cutting air to see how much overshoot there is, and subtract 98% of that
from the requested depth, seems to work pretty close for smallish taps.
Above say 6mm, I have to figure out a better way to hold a tap in the
tool holder because either the tap slips in the chuck or the whole
darned rod the chuck is mounted on, clamped in a QC boring bar holder,
slips.
Game over. Power down and back it out by hand, which is a PIMA.
Fixing that will sure as hell break something else between the spindle
motor and the spindle. And I've looked at the 11x22VS lathe Grizzly and
others sells, and its spindle drive train isn't capable of withstanding
what a 107 volt DC supply and one of Jon's drivers can to to that 1hp
rated motor. Teeny little v-belt for starters, needs to be replaced
with a polygroove. If I thought it could be fixed at the one weak point,
I'd probably own one already. But I can easily see that fixing that will
break something else long before it has reached the cutting capacity
radius the chuck size and 11" swing implies. This so-called 7x12?, is
all tapped out at a 1.25" OD of 1065 in the chuck if you want to take a
cut that isn't so thin you need armored gloves or a magnet to clean up
the swarf. Carbide chips don't last long doing that, not enough mass in
the swarf to carry away the heat. Spit...
I gotta go see if my hand still fits a snow shovel...
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)
Genes Web page <http://geneslinuxbox.net:6309/gene>
