another limit issue/homing issue
26 Apr 2014 22:26 - 27 Apr 2014 03:08 #46338
by rotccapt
another limit issue/homing issue was created by rotccapt
hi all i am at a loss with my machine. i have a cnc router that i had up and running with no issues except a lack of z axis travel due to machine design. so i decided to rebuilt the gantry and give myself more z axis travel.
so i am now having issues with my limit switches. particularly my z axis limit. as soon as i power on the machine through axis i get an immediate on limit error for joint 2 which is my z axis. i have looked at the hal meter and everything seams to be working as advertised. i have all limit/home switch loops on individual pins per axis with a combined ground all wired NC. while in hal meter all limits are false until i manually actuate a switch and then the appropriate pin goes true until i let go then it goes back to false. in the axis gui i see the arrow come on and go off when i actuate the switches.
things that i have changed,
-- all new switches (micro lever switches)- changed the x and y switches because i was having homing issues before the z switches are the same from the previous setup
-- all new wiring- since the machine is in a new place i had to extend the wiring but only by about a foot- as for wire i am using mil spec communication cord wire except my on machine limit wires, these are solid two conductor bell wire- the y and z wires run along the z axis motor cable when it goes from the main snake to the gantry slide
-- moved hobby cnc driver board and power supply (transformer,rectifier and cap) into same case with computer to conserve space
while the machine is powered on and sitting idle and axis is running but the machine is off according to axis i do randomly see the override limits button become active for a split second but it is not on long enough to do anything with. my next trouble shooting idea is to separate the limit wiring from the z axis motor wiring and see if that helps but im up for any ideas you all have. thank you for the help
so i am now having issues with my limit switches. particularly my z axis limit. as soon as i power on the machine through axis i get an immediate on limit error for joint 2 which is my z axis. i have looked at the hal meter and everything seams to be working as advertised. i have all limit/home switch loops on individual pins per axis with a combined ground all wired NC. while in hal meter all limits are false until i manually actuate a switch and then the appropriate pin goes true until i let go then it goes back to false. in the axis gui i see the arrow come on and go off when i actuate the switches.
things that i have changed,
-- all new switches (micro lever switches)- changed the x and y switches because i was having homing issues before the z switches are the same from the previous setup
-- all new wiring- since the machine is in a new place i had to extend the wiring but only by about a foot- as for wire i am using mil spec communication cord wire except my on machine limit wires, these are solid two conductor bell wire- the y and z wires run along the z axis motor cable when it goes from the main snake to the gantry slide
-- moved hobby cnc driver board and power supply (transformer,rectifier and cap) into same case with computer to conserve space
while the machine is powered on and sitting idle and axis is running but the machine is off according to axis i do randomly see the override limits button become active for a split second but it is not on long enough to do anything with. my next trouble shooting idea is to separate the limit wiring from the z axis motor wiring and see if that helps but im up for any ideas you all have. thank you for the help
Last edit: 27 Apr 2014 03:08 by rotccapt.
Please Log in or Create an account to join the conversation.
26 Apr 2014 23:14 #46339
by cncbasher
Replied by cncbasher on topic another limit issue
the chances are your picking up noise on the limit switch wiring , never have low voltage wiring shareing with motor wiring , it's asking for trouble
and always use shielded cable for switches .. , you don't mention what interface your using , so i presume it has pullup resistors on the inputs to dc
and always use shielded cable for switches .. , you don't mention what interface your using , so i presume it has pullup resistors on the inputs to dc
The following user(s) said Thank You: rotccapt
Please Log in or Create an account to join the conversation.
26 Apr 2014 23:23 - 26 Apr 2014 23:24 #46340
by rotccapt
Replied by rotccapt on topic another limit issue
the only other wire i have on hand is cat 5 network cable would that work for my limits? i just find it odd because my old machine configuration did not have any shielding on the limit wiring. ill look into it
the hobby cnc board has pull up resistors on the input pins
the hobby cnc board has pull up resistors on the input pins
Last edit: 26 Apr 2014 23:24 by rotccapt.
Please Log in or Create an account to join the conversation.
26 Apr 2014 23:45 #46342
by rotccapt
Replied by rotccapt on topic another limit issue
well you are right i must be getting interference from the bell wire, i separated the bell wire from the z axis motor wire bundle and the problem went away. so would using the network cable for y and z limits work better for the span that needs to run with the motor wires?
Please Log in or Create an account to join the conversation.
26 Apr 2014 23:56 - 27 Apr 2014 00:00 #46343
by cncbasher
Replied by cncbasher on topic another limit issue
yes cat 5 will do at a push , although it's not usually screened , just try and keep the signal wires away from the motor wiring as much as possible for now
i'd still recomend you change for screened at some stage , even if things seem to be working ok
i'd still recomend you change for screened at some stage , even if things seem to be working ok
Last edit: 27 Apr 2014 00:00 by cncbasher.
Please Log in or Create an account to join the conversation.
27 Apr 2014 00:02 - 27 Apr 2014 00:04 #46344
by rotccapt
Replied by rotccapt on topic another limit issue
thank you for the help. well in that case i will look into getting some shielded wire and just keep the setup the way it is for now, i dont need the machine right now so i can wait.
Last edit: 27 Apr 2014 00:04 by rotccapt.
Please Log in or Create an account to join the conversation.
27 Apr 2014 03:21 #46348
by rotccapt
Replied by rotccapt on topic another limit issue/homing issue
ok i decided to just continue on my current tread with my new issue
so also during the rebuild i decided to turn the entire table 180 so that the stepper for x is closer to the computer. so i now have issues when i go to home the machine. z homes first which works great, x is next and after some fiddling i can get the machine to jog in the right direction (home is in the lower left corner) in the dro screen when i jog right the indicator and the gantry move right, when i jog left same thing everything moves left. so when i hit "home all" at first the machine starts homing to the right so in my ini file i changed the sign to - on the home search, now the machine homes in the right direction until it hits the home switch then it backs off like it is supposes to but then tries to offset to the left into the hard limit of the machine where did i go wrong? also after hitting the hard limit it will not allow me to jog right even if i override the limit.
here is the x portion of my ini file
[AXIS_0]
TYPE = LINEAR
HOME = 0.0
MAX_VELOCITY = 4.5
MAX_ACCELERATION = 55.0
STEPGEN_MAXACCEL = 68.75
SCALE = 4000.0
FERROR = 0.05
MIN_FERROR = 0.01
MIN_LIMIT = -1.0
MAX_LIMIT = 25.0
HOME_OFFSET = 0.125
HOME_SEARCH_VEL = -0.050000
HOME_LATCH_VEL = 0.050000
HOME_IGNORE_LIMITS = YES
HOME_SEQUENCE = 1
so also during the rebuild i decided to turn the entire table 180 so that the stepper for x is closer to the computer. so i now have issues when i go to home the machine. z homes first which works great, x is next and after some fiddling i can get the machine to jog in the right direction (home is in the lower left corner) in the dro screen when i jog right the indicator and the gantry move right, when i jog left same thing everything moves left. so when i hit "home all" at first the machine starts homing to the right so in my ini file i changed the sign to - on the home search, now the machine homes in the right direction until it hits the home switch then it backs off like it is supposes to but then tries to offset to the left into the hard limit of the machine where did i go wrong? also after hitting the hard limit it will not allow me to jog right even if i override the limit.
here is the x portion of my ini file
[AXIS_0]
TYPE = LINEAR
HOME = 0.0
MAX_VELOCITY = 4.5
MAX_ACCELERATION = 55.0
STEPGEN_MAXACCEL = 68.75
SCALE = 4000.0
FERROR = 0.05
MIN_FERROR = 0.01
MIN_LIMIT = -1.0
MAX_LIMIT = 25.0
HOME_OFFSET = 0.125
HOME_SEARCH_VEL = -0.050000
HOME_LATCH_VEL = 0.050000
HOME_IGNORE_LIMITS = YES
HOME_SEQUENCE = 1
Please Log in or Create an account to join the conversation.
27 Apr 2014 03:51 #46351
by pi.r
Replied by pi.r on topic another limit issue
Using a cable with a tightly twisted pair (like in CAT5 cables) could be more important than using a screened cable.
High currents in the motor cables will generate "magnetic field" interferences, and screening don't give any protection for that kind of interferences. But twisted wires do... Screening protects against "electrical field" interferences from high voltages, that also could be present, so using both is always the best...
When using a CAT5 cable (or any other twisted pair cable with more than one pair) it's essential to use the 2 wires form one twisted pair to connect one switch.
I'd also recommend avoiding cables with solid wires in any application where the cables move. Always use cables with strained wires for that. So if you want to use CAT5 cables, that is very good and cheap, look for "patch cable" and avoid "installation cable". Screened CAT5 cable is usually marked "FTP" (or "SFTP" for individually screened pairs = overkill), and unscreened "UTP". Complete patch cables with plugs, is available in suitable lengths at low cost.
High currents in the motor cables will generate "magnetic field" interferences, and screening don't give any protection for that kind of interferences. But twisted wires do... Screening protects against "electrical field" interferences from high voltages, that also could be present, so using both is always the best...
When using a CAT5 cable (or any other twisted pair cable with more than one pair) it's essential to use the 2 wires form one twisted pair to connect one switch.
I'd also recommend avoiding cables with solid wires in any application where the cables move. Always use cables with strained wires for that. So if you want to use CAT5 cables, that is very good and cheap, look for "patch cable" and avoid "installation cable". Screened CAT5 cable is usually marked "FTP" (or "SFTP" for individually screened pairs = overkill), and unscreened "UTP". Complete patch cables with plugs, is available in suitable lengths at low cost.
Please Log in or Create an account to join the conversation.
27 Apr 2014 04:44 #46355
by rotccapt
Replied by rotccapt on topic another limit issue/homing issue
thank you all for your help i finally got it all figured out it took a lot of trial and error but i found out that i had to invert the x axis step and direction and then change the latch velocity to - as well. now everything works as advertised
Please Log in or Create an account to join the conversation.
28 Apr 2014 23:31 #46435
by andypugh
Considering that homing is slow, and a small delay (of a few mS) in detecting the switch is not super-critical, and that false-triggering of the limit switches can ruin parts and/or be annoying, it is probably not a bad idea to add debounce to your limit switches (Mine have it)
www.linuxcnc.org/docs/html/man/man9/debounce.9.html
Replied by andypugh on topic another limit issue/homing issue
while the machine is powered on and sitting idle and axis is running but the machine is off according to axis i do randomly see the override limits button become active for a split second but it is not on long enough to do anything with.
Considering that homing is slow, and a small delay (of a few mS) in detecting the switch is not super-critical, and that false-triggering of the limit switches can ruin parts and/or be annoying, it is probably not a bad idea to add debounce to your limit switches (Mine have it)
www.linuxcnc.org/docs/html/man/man9/debounce.9.html
Please Log in or Create an account to join the conversation.
Time to create page: 0.096 seconds