Machsupport Forum
Mach Discussion => VB and the development of wizards => Topic started by: nso on December 20, 2013, 01:45:51 PM
-
is it possible check( encoder dro )and( machine coordinate dro)
have some deviation during just mach3 do E-stop ?
i need to know can do this with macro pump or Brian.
if any one can advise to me make me happy...
thanks
-
this one may be worth a try.
see pict.
-
If you are just looking to see the deviation error, check OEMDRO's 127, 128, 129.
Error between Encoder 1 and abs X position DRO 127
Error between Encoder 2 and abs Y position DRO 128
Error between Encoder 3 and abs Z position DRO 129
-
first of all i say thanks guys for quick reply... :)
cncalex ;i tried today if work or not say here .
Bodini : it just dro for Error between Encoder ,so i need this:
if have a deviation error on each axis then active output ,or do oem button reset. thanks again
-
Is this closer to what you want ?
alex
-
i tried right now and yes that is working great thanks.
yesterday i tried something like this with brain . it is working like your brain, but in your brain better....
now i have a new problem :
in this case i use servo motor encoder on my machine and encoder pulse = 2500 per revoltion
i guess mach3 can not reading pulse maybe encoder pulse is too fast or big ?!!! ???
when axis moving slow dro encoder pulse is true but during axis moving like G0 then encoder dro false
how can solved this problem?
sorry for my bad english
-
If after a rapid move the axis position and encoder position are the same then it looks like yor problem is the follow error of the servo.
in this case you can try tu tune the servo more stiff. Or increase the allowable lag in the Brain.
If after a rapid move the axis position and encoder position are completely different then I think the pulses came too fast.
sorry for my bad english
no prob., my is not even better ;D
alex
-
One of the problems you run into doing this is it takes a LOT of CPU resources to read all the encoders at speed. Mach3 is not really up to do this reliably. The OEM (Rogers Machine) that has done this USED a seperate high speed counting board to COUNT the encoders and report to Mach3 via a plugin. This takes the load off of mach3 and it works fine. You can EVEN have mach3 auto correct the position if it gets off track.
Just a thought, (;-)TP
-
dear cncalex , i am sure my servo motor and drive is tuning correct ...so perhaps encoder pulse too fast because 2500 per r
if use (1000 per r or less) encoder then possible solved this problem?
BR549 hi,thanks for replay i see rogers website and i know how to work...i can t buy that (from my country) :-\
so i tried find result and i know maybe cant do that but just tried...
-
Hi,
AFTER a long rapid move is there a difference between the axis dro and encoder dro?
alex
-
yes exactly true .... for example : when move axis rapid ,230 mm then dro show true but encoder show 237,234,236 in fact random ...
-
Ok, in this case I would say yes. You said you get nice results at half rapid speed then reducing the encoder resolution ( PPR) should work.
But as TP said your system will have lot of work to read the inputs.
For that I would choose the lowest encoder resolution where you can live with lets say 400 PPR or lower.
Alex
-
thanks dear friend ,you kind of to me ,so your comment is my result ... after see your last reply, i searching at pdf of servo drive manual about
decrease encoder pulse, and i find a parameter can change ratio pulse output i guess may can help me but i don't try yet... ;D
-
Elaborating on what TP said, I will give you a few more clues. Rogers encoder board uses a plug-in. Brains or VB/macropump just arent fast enough to monitor closed loop with much success. The plug-in compares steps sent by Mach to counts sent back from encoders. It doesnt use DROs at all, but rather the raw data behind the scenes in Mach. The encoder board is the only way Mach can read and compare it all fast enough.
However, if you are having any sort of luck at all, I encourage you to keep pushing and see what you can get it to do! :-)
-
Bodini ,thanks for your support...
However, if you are having any sort of luck at all, I encourage you to keep pushing and see what you can get it to do! :-)
-
I assume you want it for a kind of safty, if something fails to e stop the system.
a kind of closed loop isn't possible with a brain ( or macro ) not even close to.
the brain runs fix at 10HZ so 100ms for one run.
that means you have a lag of 1mm at 600mm feedrate or 10mm at 6000mm feedrate and so on.
plus the lag time of the incoming pulses what is small compared to the brain.
Alex
-
Alex ,you right... i tried with 0.1 mm lag error at 2000 mm feed rate the 9 mm !! so that is too big and i looking for less like( 0.2 or 0.3 mm) maximum in this case i take my answer with your advise ,in fact can t do this just because brain & macro too slow...
anyway i want say thank you and other :)