× Forum Header

stepConf and actual running precision issue

More
07 Sep 2016 20:44 #80175 by fannlow
fannlow replied the topic: stepConf and actual running precision issue
Hi Tommy,
thanks for helping out.

yes, on the X and U axis, its horizontal on each side, and i am using rack and pinion styled, so its about 82mm travel per 1 revolution.

and on the Y and V axis, its vertical on each side, I use a 5start ACME leadscrew for that.

the motors are direct drive, mounted on aluminum plates, then with a coupler to connect the two.

I have used this machine for over a year without any problems. and yes I did test with only 20mm/s, and also lower accel at about 50mm/s. while its slow, the difference in distance is still apparent.

what I don't really understand is that, how come only one axis of the horizontal and vertical is affected, while the other is working just fine with the same settings?

thanks
More
07 Sep 2016 20:47 #80176 by fannlow
fannlow replied the topic: stepConf and actual running precision issue
thanks for the quick help Andy

I will look into that and thanks for also pointing it out.

by the way, how do I see if its scale for degrees or distance (mm)?
More
07 Sep 2016 20:49 #80177 by fannlow
fannlow replied the topic: stepConf and actual running precision issue
Hi Andy did you meant

#V-AXIS
[AXIS_7]
TYPE = LINEAR
HOME = 0.0
MAX_VELOCITY = 95.0
MAX_ACCELERATION = 200.0
STEPGEN_MAXACCEL = 250.0
SCALE = 157.480314961
FERROR = 1
MIN_FERROR = .25
MIN_LIMIT = -1550.0
MAX_LIMIT = 0.001
HOME_OFFSET = 0.0

compared to Y-axis?

#Y-AXIS
[AXIS_1]
TYPE = LINEAR
HOME = 0.0
MAX_VELOCITY = 95.0
MAX_ACCELERATION = 200.0
STEPGEN_MAXACCEL = 250.0
SCALE = 157.480314961
FERROR = 1
MIN_FERROR = .25
MIN_LIMIT = -1550.0
MAX_LIMIT = 0.001
HOME_OFFSET = 0.0

can't tell what's different though?
mind pointing it out for me?
More
07 Sep 2016 21:13 #80180 by andypugh
andypugh replied the topic: stepConf and actual running precision issue

fannlow wrote: by the way, how do I see if its scale for degrees or distance (mm)?


LinuxCNC doesn't even "know" the units. You say move from X = 1 to X = 2 and it gives you SCALE pulses. The units could be degrees, inches, furlongs, angstroms... (and SCALE/10 pulses if you move from 100.2 to 100.3)

It is Stepconf that tries to be clever.
More
07 Sep 2016 21:15 #80181 by andypugh
andypugh replied the topic: stepConf and actual running precision issue

fannlow wrote: SCALE = 157.480314961
...
SCALE = 157.480314961
FERROR = 1


They are now the same. They weren't in the INI I looked at.

Are you sure that the drives are actually on the same micro step setting?
More
07 Sep 2016 21:20 #80183 by fannlow
fannlow replied the topic: stepConf and actual running precision issue
yes I did checked. the micro stepping are the same.

what i did notice is that while X and U are still 200 x 10 microstepping as i indicated. but now X moves further from intended distance while U is now accurate.

and previously they were same.

I does changing my steps cause any changes if my pitch doesn't change? I mean my actual distance traveled.

still using drive as 200 and just maybe change microstep from 10 to 5 or other values. of course the driver side must match as well.

thanks
More
07 Sep 2016 21:20 #80184 by tommylight
tommylight replied the topic: stepConf and actual running precision issue
That all looks OK, except the min and max limits. Is there any reason why you have that set to negative?
The other thing i find disturbing is having a pinion directly mounted on the motor, a few issues with that, but it is possible.
Issues: motor needs a lot of power or it will skip and miss steps, if it has enough power it will vibrate to much again causing it to miss steps mostly at lower speeds as it would resonate. Also asking a motor to move something 0.41mm per step is not efficient.
Again, that can work and it does but there are to many issues with it and getting it to work reliably is quite a challenge .
One more thing, if you have not already read about it, XYZ work normally regarding G0 and G1 feed speeds, ABC are not linear axis so you have avoided them< this is good, UVW axis do not comply to G1 feed speeds so they will tend to move at max speed all the time.
The last part about UVW, i am not sure about, but i remember reading about it lately.
The following user(s) said Thank You: fannlow
More
07 Sep 2016 21:27 - 07 Sep 2016 21:30 #80185 by fannlow
fannlow replied the topic: stepConf and actual running precision issue
hi Tommy you are right about the rack and pinion part. that's why I hate it when it goes slow... makes that weird sound resonance.

any recommendations for steps per rev for my spec? I think have been using 2000 steps for X and U (82mm/rev). and 800 steps for Y and V (12.7mm/rev).

also the negative travel is for vertical Y and V. as the top most is 0. so it travels down, and hence negative direction.

I have been using G1 for most parts, seems ok.

82mm รท 2000 steps = 0.041mm/step
correct me if I am wrong.

and please share more light about efficiency.

thanks
Last Edit: 07 Sep 2016 21:30 by fannlow.
More
07 Sep 2016 21:46 #80187 by tommylight
tommylight replied the topic: stepConf and actual running precision issue
Lets go back,
1.>>>What type of drives? < regarding this, can you please set the timing values at 15.000, 15.000, 20.000,20.000
2.>>>Set the base thread to 100.000
3.>>>You MUST know the drive micro stepping, no guessing there, so in case you can not find them you can follow this procedure, but only after the step No.1
---Set the micro stepping to 1 for all axis, leave the rest as you have it now, do a "test axis" and set velocity to 5 and accel to 30, at the "-+" choose "+" and set the movement to 50mm. Measure how much it moves? from that you can figure out the drive micro stepping as follows:
......if it moves 50mm you are good to move on
......if it moves 25mm you have 1/2 micro stepping so you need to set 2 at the drive micro stepping field in stepconf
......if it moves 12.5mm you have 1/4 so set it at 4..............and so on.
4.>>> Only when you get this right, you can move on to set the speeds and accelerations.
try this and report back.
thank you.
The following user(s) said Thank You: fannlow
More
07 Sep 2016 23:21 #80193 by BigJohnT
BigJohnT replied the topic: stepConf and actual running precision issue
Also when you get your scale figured out run a stepper test on each axis as shown in the manual to make sure you have not exceeded the maximum acceleration or something which may cause lost steps. Do your scale measurements at a slow speed.

JT
Time to create page: 0.085 seconds
Powered by Kunena Forum