Machsupport Forum

Mach Discussion => General Mach Discussion => Topic started by: osman100 on April 13, 2009, 08:09:49 AM

Title: step error
Post by: osman100 on April 13, 2009, 08:09:49 AM
with mach 3 according to manual  i use the following program to check the step error
F1000 (i.e. faster than possible but Mach3 will limit speed)
G20 G90 (Inch and Absolute)
M98 P1234 L50 (run subroutine 50 times)
M30 (stop)
O1234
G1 X4
G1 X0 (do a feed rate move and move back)
M99 (return)
but after finishing it i found it not go back to the same start position
can u help me to solve this problem

Title: Re: step error
Post by: Hood on April 13, 2009, 09:57:53 AM
How much is it out? Do you have any backlash? If you have backlash comp switched on is it set correctly? Could you be missing steps by having your motors tuned to aggressively?

Hood
Title: Re: step error
Post by: osman100 on April 13, 2009, 10:38:19 AM
it's about 0.4%  and i haven't backlash 
how can i know if i missing steps

Title: Re: step error
Post by: Hood on April 13, 2009, 01:26:25 PM
You can usually hear if they are missing badly but if its just the occasional then you may not. Try increasing the pulse width to 5 and see if that helps, you could also try changing the active state  for Step and Dir pins.
Hood
Title: Re: step error
Post by: osman100 on April 13, 2009, 02:43:52 PM
i already increased the puls width to 5 for pulse width but not for direction puls
now i choose active low i'll try to use active high then i'll inform you
Title: Re: step error
Post by: osman100 on April 15, 2009, 04:24:41 PM
dear friends
i start today to test the machine but i'm very surpriues because when i change the feed rate i found the distance also change acoording to feed rate vatue can anyone help me
thanks
Title: Re: step error
Post by: Hood on April 15, 2009, 05:29:01 PM
Please attach your xml.
Hood
Title: Re: step error
Post by: osman100 on April 15, 2009, 05:38:07 PM
this xml
F1000 (i.e. faster than possible but Mach3 will limit speed)
G20 G90 (Inch and Absolute)
M98 P1234 L10 (run subroutine 50 times)
M30 (stop)
O1234
G4 P02
G1 Y40
G4 P02
G1 X40
G4 P02
G1 Y-40
G4 P02
G1 X0 (do a feed rate move and move back)
M99 (return)
%
Title: Re: step error
Post by: Hood on April 15, 2009, 05:43:13 PM
No the xml is your configuration file. If you are using the standard mill profile then go to the Mach3 folder and look for Mach3Mill.xml, copy it to your desktop and rename it osman.xml
On the reply page you will see additional options, use that to browse to your desktop and choose the renamed file.
Hood
Title: Re: step error
Post by: osman100 on April 16, 2009, 05:05:48 AM
dear hood this is the last xml ido yasterday
Title: Re: step error
Post by: Hood on April 16, 2009, 05:10:34 AM
Your motors are tuned too fast for the Kernel of 25KHz, either decrease the motor velocity or increase the Kernel speed to 35KHz.
Hood
Title: Re: step error
Post by: osman100 on April 16, 2009, 05:25:03 AM
i already did but the program give an error on of your motor axises is tunned to fast for the kernel speed so i increase kernel gradualy but i still have the same problem by the end i made kernel speed 1mhz i have all other xml files for yesterday i'll attach it
Title: Re: step error
Post by: osman100 on April 16, 2009, 05:31:10 AM
also that's other files
Title: Re: step error
Post by: Hood on April 16, 2009, 06:33:00 AM
Ok well nothing jumps out as being seriously wrong in your xml. Your Steps per unit seem a bit low but that will depend on your setup etc so could be correct.
 You definitely do not want to use a kernel speed that you do not require so dont go above 35KHz unless you need more due to the steps per/velocity you need.
 Have you run the DriverTest to see if you are getting a decent pulse?
 What kind of motors, drive, reduction, ballscrews etc do you have?
Hood
Title: Re: step error
Post by: osman100 on April 16, 2009, 07:20:08 AM
the motor is induction motor 1.1 kw with encoder 800 p/rev and the driver is omron f7 and the reduction is gearbox 1:40 then a reduction 2 gears 1:1.5 so that's mean the total reduction gear is  1:65  then i use a pinion and rack with pitch 6.2 mm
Title: Re: step error
Post by: Hood on April 16, 2009, 07:57:37 AM
With that setup the tuning will be extremely critical and even when perfect I very much doubt that you will have the accuracies required for a machine tool. I am at present downloading the manual but at first look it seems to be more use for prositional control where great accuracy is not required.
Hood
Title: Re: step error
Post by: osman100 on April 16, 2009, 09:15:46 AM
dear hood
with this inverter  (drive) i can use induction motor ac servo motor with high accuracy u can download manual from this link
http://industrial.omron.eu/en/products/catalogue/motion_and_drives/frequency_inverters/lifts_and_cranes/f7/default.html#productinfo13_1
Title: Re: step error
Post by: Hood on April 16, 2009, 09:28:43 AM
I have downloaded the manual and from first impressions it is suitable for such things as palletisers, lifts, cranes etc which although may require accuracy it is not the same accuracy that a machine tool requires. I will read further tonight but I think you are trying to use something that is not suitable for the application you have.
Hood
Title: Re: step error
Post by: osman100 on April 16, 2009, 10:16:11 AM
u r right if i use it as fvd without pg-b2 option card but with this card it work like a servo and with more power than stepper or servo
Title: Re: step error
Post by: Hood on April 16, 2009, 03:59:26 PM
That seems to me to be a speed control card so it will control the speed via closed loop but dont think it will control position as well.
Hood
Title: Re: step error
Post by: osman100 on April 17, 2009, 11:33:59 AM
dear hood this card can use a closed loop but i use it as v/f with pg and this make it work like servo to do that i set parameter a1-02 to 1 see pages 4-5,4-7,4-14,5-8 also parameter from f1-0 t0 f1-14 is to adjust pg page 5=36,5-37 and this also like mach 3 parameter
and the most important is puls train i/o parameters fro h6-0 to h6-7 page 5-51
page 6-30 it explain how puls train work 
so that by the end this inverter is controlling after this adjustment according puls train when we set the puls train reference for example 25khz equivalent to motor freq. 5o hz so if the freq. in from mach 3 is changed the speed will change because the freq. will  change the output freq. but the priod will be the same
all the best
Title: Re: step error
Post by: Hood on April 17, 2009, 12:07:28 PM
I have not read the manual fully and will look at the pages you have pointed out. However that board is a closed loop speed control and not a closed loop position control, there is to my mind a very big difference between these two things. I will reserve judgement until I have read fully read the manual but at the moment I would say you are wasting your time, then again I am usually wrong ;D

Hood
Title: Re: step error
Post by: Hood on April 17, 2009, 02:26:15 PM
Afraid all I can see from reading the manual is you have precise control over the speed but I fear you have no real control over position, at least to the degree of accuracy that is required for a machine tooll. Your out of position tests which vary with each speed  seem to verify this.
Hood
Title: Re: step error
Post by: osman100 on April 17, 2009, 03:10:12 PM
even if this according to mach3 manual i can do through speed control chapter 4 page 4-12 pwm control as follow
3. PWM motor control
As an alternative to Step and Direction control, Mach3 will output a pulse width modulated
signal whose duty cycle is the percentage of full speed that you require. You could, for
example, convert the duty cycle of the signal to a voltage ( PWM signal on for 0% of time
gives 0 volts 50% gives 5 volts and 100% gives 10 volts) and use this to control an
induction motor with a variable frequency inverter drive
. Alternatively the PWM signal
could be used to trigger a triac in a simple DC speed controller.
Figures 4.12 and 4.13
show the pulse width at
approximately 20% of
the cycle and 50% of the
cycle.
In order for the PWM
spindle speed signal to be
turned into direct current
(actually a direct voltage
is generally used as the input to variable speed drives, but you know what we mean) the
pulse signal it must transformed. In essence a circuit is used to find the average of the pulse
width modulated signal.
The circuit can be a
simple capacitor and
resistor or be much more
complex depending (a)
on how linear you want
the relationship between
the width and the final
output voltage and (b) on
the speed of response
you need to the changing pulse width.
Title: Re: step error
Post by: Hood on April 17, 2009, 03:13:13 PM
Yes that is all relevant BUT for a spindle not for axis control.
Hood
Title: Re: step error
Post by: osman100 on May 12, 2009, 12:49:49 PM
dear all
thanks for your help and I'm sorry for long time to contact you but rely I'm busy to adjust my machine with this type of inverter . it isn't easy but by the the end and after long fitting today i just finish to adjust x axis and I'm rely succeed 100 % INSHAA' ALLAH
as i told u before now it's work as servo but to reach this that's very important point
1- make motor tuning to adjust motor with encoder and inverter and this tuning should be running tuning
2-adjust pg parmeter
3- adjuster ASR parmeter to adjust feedback from encoder
4-use mode v/f with pg feedback
5- adjust v/f curve and parmeter
thanks for all
Title: Re: step error
Post by: Hood on May 12, 2009, 02:24:36 PM
Glad to hear you got it working, what is the accuracy for positioning?
Hood
Title: Re: step error
Post by: osman100 on May 12, 2009, 03:36:04 PM
actually now it's less than 1 mm  but the acuirte value I'll cheek tomorrow and tell u because today i chick the value with caliber 1 mm only
Title: Re: step error
Post by: Hood on May 12, 2009, 03:38:17 PM
Ok thanks.

Hood
Title: Re: step error
Post by: osman100 on May 13, 2009, 12:01:10 PM
dear hood
today i check and i found it 0.2 mm but I'm in need some time to adjust back lash and some parameter on inverter i think after this fine tuning the accuracy will be 0.02 mm but let me try and tell you the result
Title: Re: step error
Post by: Hood on May 13, 2009, 01:27:20 PM
That is better than I would have expected but in reality no where near enough for a machine tool such as a mill or a lathe even if you do manage to get it as good as 0.02mm. I think I remember you saying it was a granite cutter you have, that kind of accuracy is more than likely much better than you need :)
Hood
Title: Re: step error
Post by: osman100 on May 17, 2009, 09:25:31 AM
thanks a lot
can u help me for the accurate and best way for home switch r it limit or proximity switch or photocell or other
Title: Re: step error
Post by: Hood on May 17, 2009, 09:31:08 AM
probably an optical switch would be the most accurate but it would depend on the environment whether it would be the best. If there is a lot of dirt and dust then and optical would need to be enclosed and the flag operated by a plunger which exits from the side of the encosure so that dust isnt a problem.
 You may be better with a prox switch if it is no magnetic material you will be cutting, they can be accurate as well.
Hood