Hello Guest it is March 28, 2024, 03:00:00 PM

Author Topic: Discombobulation  (Read 3955 times)

0 Members and 1 Guest are viewing this topic.

Discombobulation
« on: October 16, 2013, 09:57:37 AM »
A few months back, I started setting up my router and Mach 3 together and running through the set up of Mach 3. Since then I have cut a few parts, needed for completing the mechanical side of the table (brackets, Thein dust separator lid, etc). Running the machine to get more of an understanding of the programs (machine control and CAM). During this time I was using any piece big enough to act as a base for cutting on. Having obtained a "permanent" spoil-board,started to look at the table set-up again, and go into setting limits a permanent dedicated work co-ordinate zero position. Simultaneously setting up the parameters for surfacing the s/board, using 1 of the wizards to set the parameters. This brought up a whole rats nest of questions and discoveries. Discovery #1 was that the figures input into the wizard for distance travel for example was not what was generated in the code-box of the program run screen. My feeling for now is that this is a possible programming issue of the wizard. Discovery #2 was that if I input figures in the G54 fixtures of config. The axis moves to the work co-ordinate zero for X and Y. The machine co-ordinates for the axis read as follows (DRO's): X input 12.5 ~ DRO reads 12.4499 ...... Y input 20 ~ Dro reads 20.0040. The DRO readout at home in work co-ordinate read the same figures, but a (-). which is correct. The Question is why are the DRO's not reading the same as input figure. Is this a motor tuning issue and have to look at the steps per. Is it the kernel setting (M3 suggests 25 000 ~ Toshiba blue board recommends 35 000 ~ Mine set to 25 000. Should I look at the stepping go from 1/8 to 1/16 for instance. Where am I going wrong!
It would also be appreciated for further information of "debounce"  what it is and how to apply it.
I have also been having difficulty coming to terms with the setting and using of soft limits. This I will leave for now.
Guys your assistance is greatly appreciated.
Gary

Offline BR549

*
  •  6,965 6,965
    • View Profile
Re: Discombobulation
« Reply #1 on: October 16, 2013, 02:15:30 PM »
The problem you are seeing is your steps per setting do not have have the resolution to work at the .0001 level of Machine resolution.

To do the move of say X12.5 the machine can only get as close as 12.4999 due to step resolution. IF it took 1 more step it would go OVER the X12.5 and Mach3 does not allow that. It will stop short but never go over based on steps resolution.

Now in case you are wondering YES mach3 will make up the lost value in the next move it will add the lost .0001 to the next move IF it can. It never forgets the UNDER values for each move and will work it into the mix as it can.

(;-) TP

Offline BR549

*
  •  6,965 6,965
    • View Profile
Re: Discombobulation
« Reply #2 on: October 16, 2013, 02:19:44 PM »
Now for the Debounce , that is a form of signal filtering. The value you set is a time element in Msec. The signal MUST remain active for at least the time value you set in order for MAch3 to count it as ACTIVE.
Re: Discombobulation
« Reply #3 on: October 16, 2013, 04:02:24 PM »
TP
Thanks for the heads up, I was obviously looking at the posed question from the wrong angle; I' m telling this machine to move by X, it must by all means do! Between submitting my post and reading your reply, has given me time to mull this over in my noggin. I came to the conclusion is that I'm looking at less than a ball hair's difference, I should live with it as it's not actually that bad. Reading your post gives me better assurance and understanding of what is happening at the individual step level and piece of mind that M 3 picks up the under and overs and accommodates for them.
In the last 2 weeks or so I have some weird stuff happening to me, and every time I returned to a particular point or problem I was facing something new without getting back to the previous days problem. It eventually mean me stripping my Z axis off the Y, re-looking at all the mechanical's and rebuilding the machine.
During this time I had thought that some of my problems could be "noise" related and started noticing settings like debounce and ways to overcome dirty signal or signal loss. Given all this going on and also trying to make some headway in defining the work area of my table, soft limits, trying to get my spoilboard surfaced etc, started to put me into an information overload, at times it really felt as if my head would spin off.
I thank you for the peace of mind......... I still hav'nt reached a state of "nirvana", but geeting there hopefully.
Re: Discombobulation
« Reply #4 on: October 16, 2013, 07:36:17 PM »
Does the variances above relate to  an accuracy tolerance that the machine is capable of?
Over the last weeks, I'd written movement commands checking limit of travel and the  axes for any adverse movements in the motion of the motors and mechanics. Whilst cutting I had picked up 2 "hesitations" when the Y axis was travelling out to it's limit. Having then stripped, rebuilt and re-aligned the Y and Z, the movement commands were run again and so far nothing adverse detected. I can now get back to where I was and concentrate on what I was originally planing; getting my spoil-board setup and finish all the bits and pieces I need for completing my table and ready for "production".
While running the above commands, (I had also adjusted the "feed-rates"). I find that the maximum rate is around 1600 mm/minute. Should I be expect to get more (in free travel, no cutting), would I need more, or is this the limitation of the motors and drive system and or controller board. The table is rack and pinion driven with Nema 23 motors and using the dreaded Toshiba blue controller boards.
With the spoil-board, my aim is to have guides/stop-ends of aluminium angle (9 x 9 x 3 mm) running along the home side of the X and Y axes, the work zero set further in. The desired effect is to have a buffer zone between the work zero and home, having a rest for stock material to butt against, giving me a small area of "waste material".
I also wish to install an Auto Zero touch plate for the Z axis to reference it to the top of the material and set the limit of travel to the top of the spoil-board. It will be appreciated if I can get pointers to  installing an auto zero Z and installing the necessary macros, where to source and also whether someone has had experience in installing it with the Toshiba controller board and obtaining a pin to run it.

Offline RICH

*
  • *
  •  7,427 7,427
    • View Profile
Re: Discombobulation
« Reply #5 on: October 17, 2013, 06:15:16 AM »
Quote
Does the variances above relate to  an accuracy tolerance that the machine is capable of?

maximum rate is around 1600 mm/minute. Should I be expect to get more (in free travel, no cutting), would I need more, or is this the limitation of the motors and drive system and or controller board?

Gary,

Your machine will only be as accurate as the mechanical components of your system and also the electronic  / electrical part of the system.
You need to understand that it's a SYSTEM and that different parts of the system are interelated which can affect movement and accuracy.
The smallest step  ( movement ) you can expect to achieve is called resolution defined as 1/steps per unit. That's  theoretical and the actual will will be less. Each system component must be capable of providing for the accuracy you desire.

There are two movements, namely rapid and another at feedrate. With a stepper system, the motor torque decreases as the speed of the motor increases. So when you tune your system you are actualy interested is reliable / repeatable movement  for both a rapid movement  and when machining. Acceleration and velocity are interelated and through testing you will find what your machine is capable of. If you don't like what you have then you need to change something.

You have assembled a SYSTEM from parts so you will have what you have. You did not design the system for some set of parameters.

RICH
 


Offline BR549

*
  •  6,965 6,965
    • View Profile
Re: Discombobulation
« Reply #6 on: October 17, 2013, 11:47:45 AM »
I would FIRST get rid of the dreaded BLUE board and get a real high speed Breakoutboard.  ANY shortcomings there just ECHO throughout the entire system.

(;-) TP

Re: Discombobulation
« Reply #7 on: October 17, 2013, 03:55:07 PM »
Rich thanks for the clarification.
TP I hear you loud and clear, on my mind too! Will have to wait till funds are available.