I'm having problem with the rotary gear ratio, I jog the rotary axis (8th axis) E2 to 90 degree but it gives me around 52 degree, How do I correct this?
Rotary angle does not match
-
happytriger2000 -
December 9, 2013 at 11:48 AM -
Thread is marked as Resolved.
-
-
ask to designer for gear box ratio and write correct values to below line at $machine.dat
$RAT_MOT_AX[8]={N 0,D 1} ;UEBERSETZUNG MOTOR-ACHSE N = ZAEHLER, D = NENNER
-
Thanks.
I had a look at the Machine data and $RAT_MOT_AX[8] is:
$RAT_MOT_AX[8]={N 3119556,D 10000}.
Tweak around the N value, when set N = 185, then jog the rotary axis to 90, it gives 32.5 degree.
then tried $RAT_MOT_AX[8]={N 512.30,D 1}, gives 86 degree.
$RAT_MOT_AX[8]={N 5135.30,D 10}, 88degree~89degree....
Is this method accurate? -
Thanks.
I had a look at the Machine data and $RAT_MOT_AX[8] is:
$RAT_MOT_AX[8]={N 3119556,D 10000}.
Tweak around the N value, when set N = 185, then jog the rotary axis to 90, it gives 32.5 degree.
then tried $RAT_MOT_AX[8]={N 512.30,D 1}, gives 86 degree.
$RAT_MOT_AX[8]={N 5135.30,D 10}, 88degree~89degree....
Is this method accurate?What that actually references is the reduction ratio, meaning your external table has a 513.53:1 ratio. I think $RAT_MOT_AX is an integer so {N 51353, D 100} would be the correct way to express this.
As has been said check the gearbox ratio
-
Eusty,
It meant to be N5135, D10 not N 5135.30,D 10, .. , I'm aware of the integer.
Browsing through my photo library to see if I can find the rotary image with spec. sticker on it, this is the closest one:
I mentioned about this gear reducer before, it was taken from a KR125 A3 joint, need some info. on this Kuka gear reducer. -
We have used KR125's in the past, so I'll check an archive tomorrow...just got to remember!!
-
Thanks man, really need to get this done so I can test all 8 axes for milling.
-
My old backup for a KR125 says $RAT_MOT_AX[3]={N -185,D 1}. Which seems quite far off from what you're getting experimentally.
Working out $RAT_MOT_AX empirically is entirely legitimate. I've done it many times. The trick is that you have to have a means of measuring the actual physical rotation of the axis accurately. Your calibration will be only as good as your measurements, obviously.
Are you sure that's an A3? It sure doesn't look like an A3, more like A4 or A6. Are you using the same motor, or a different motor? The motors also have internal gearing, declared in $RAT_MOT_ENC. If you've swapped motors around, that would be an additional factor. KUKA motors generally have a RAT_MOT_ENC of 1/3 or 1/4.
-
Hi sky,
That gear reducer was purchased from a ex kuka Tw technician and he said it is a A3 of KR125 and the MADA for this axis is: $RAT_MOT_AX[8]={N185, D 1} before dis-assembled.
My measuring equipment is this inclinometer, I wouldn't say is an accurate measuring tool but good enough for rough measurement:The diameter is very close to A3 on KR2210, can't be A4 or A6. I measured it just now:
It is around 323mm DiameterThe servo motor of E2:
-
While I was tweaking the $RAT_MOT_AX[8] N value when save and exit return to navigator screen, highlight R1---> Program it loads very slow, even when highlighting other folder, then I use crtl + ESC to restart windows...REBOOTING.... after entering KSS on thr right screen where the window shows all axis value in X, Y, Z, E1, E2 all described Invalid value....the numbers appeared after I started jogging -Y, but only A4 rotates no matter I jog X Y Z, so what's going on here?
-
Undefined tool/base after boot up + Alpha 5 singularity and cartesian jogging?
-
-
Thanks Eusty.
$RAT_MOT_AX[3]={N -185,D 1}
$RAT_MOT_AX[3]={N 1904,D 10}
My current N value is $RAT_MOT_AX[8]={N 514,D 1} and it is very near to 90 degree when E2 is jogged to 90.004xx.
I have to jog E2 every time I change the N value to check the inclinometer, how do I run a program that turn E2 only to 90, 180, 270 degree? -
PTP {E2 90.0} etc
Sent using Tapatalk 4
-
A kr125/1 A 3 is
N 1904, D 10 if that sounds more like it
-
Thank you all, will try that tomorrow.
-
PTP {E2 90.0} etcSent using Tapatalk 4
However, make sure to always put the axis at 0 before you change $RAT_MOT_AX, otherwise that PTP {E2 90} command will not accurately reflect what the axis is actually doing. Basically, changing $RAT_MOT_AX when the axis is not physically at 0 will throw off your zero position until you re-master.
PTP_REL {E2 90} will generate a purely relative move starting from wherever the axis is when the command is executed, without needing to worry about the axis' absolute position. -
THe last one i did i rotated it to 360 then just adjusted the value till i got the figure i wanted in monitor robot position then it was all ok never had to remaster mmm
i might need to check it again !
but the external kinamatics worked fine as well
-
However, make sure to always put the axis at 0 before you change $RAT_MOT_AX, otherwise that PTP {E2 90} command will not accurately reflect what the axis is actually doing. Basically, changing $RAT_MOT_AX when the axis is not physically at 0 will throw off your zero position until you re-master.
PTP_REL {E2 90} will generate a purely relative move starting from wherever the axis is when the command is executed, without needing to worry about the axis' absolute position.I forgot to say when the E2 return to Home position is not actually @ 0 degree, it is 1 degree off, to correct this do I jog E2 to 0 degree then Setup > Master > Dial?
"The motors also have internal gearing, declared in $RAT_MOT_ENC. If you've swapped motors around, that would be an additional factor. KUKA motors generally have a RAT_MOT_ENC of 1/3 or 1/4."
I will try these $RAT_MOT_AX[3]={N -185,D 1} and $RAT_MOT_AX[3]={N 1904,D 10} and see what I get in the inclinometer, if it turns out to be somewhere between 50~70 degree, then does it mean I have to do something with RAT_MOT_ENC?.
Is it correct if I say the N value is equivalent to pulse?, like those in Stepper motor drive where stepper motor turn 1.8 degree per pulse and 360 degree = 200 pulse. -
Quote
"The motors also have internal gearing, declared in $RAT_MOT_ENC. If you've swapped motors around, that would be an additional factor. KUKA motors generally have a RAT_MOT_ENC of 1/3 or 1/4"there is no internal gearing (inside motor). this is motor/encoder position ratio which is number of poles of the motor (because for Kuka, motor and resolver must have same number of poles). in other words, on a 4 pole motor, 90deg motor shaft rotation produces full period sin/cos (2*pi) on resolver. kuka motors are actually built by Siemens. name plate of the motor contains both Siemens (1FK...) and Kuka part numbers (article number xx-xxx-xxx). you can download datasheet from Siemens website and confirm number of poles if you like.
one motor pole (one period of sin/cos on resolver) is 4096 counts (12-bit) on a KRC2. for comparison, on a KRC4 it is 16-bit (65536 counts).
so if we have a 3-pole motor, then one full motor revolution is 3x4096=12288 counts.
and if we have a 4-pole motor, then one full motor revolution is 4x4096=16384 counts. -