Skip Signal

[derekBPcnc]

Hi All,

I have tested the skip I/P with a simple micro switch and the G31 functions correctly with a simple program... so I'm now writing a macro for a tool setter.

Generally the macro design is within the power of my brain cell, just.... but I'd like to verify that the signal from the probe is going On - Off - On as expected as the tool touches retracts and touches again for the final "read".
If the probe is open / closed at the wrong time, then a #3006=1 (probe fail) would abort the setting routine and flag an alarm.

But,
I can't find in the Fanuc manuals (no surprise there ) how to read the status of the skip signal (or any of the general I/O).

Any help / points would be a great help.
Cheers
Derek.

[Al_The_Man]

The skip input cannot be made conditional so it does not show up in the ladder AFAIK.
When writing G31 routines or macro's it is customary to have a protected positioning move, this is a routine that would typically be used to move the probe to initial position in say rapid, IOW, any move of the probe should include a G31.
The standard G31 input is sent from PLC to CNC via the BMI, an optional High speed skip input is available through an add on board and option parameter that directs the high speed G31 skip input directly to the CNC side.
Al.

[derekBPcnc]

Hi Al,

I've been thumbing through the manual -
It says that the input interface signals 1 to 32 can be read by variables #1000 to #1031.
I'm hoping that the skip x4.7 is connected to one of these i/ps
I haven't got the electrical drawings with me, so I'll have to check when I get to work tomorrow.

I just want to check the probe closes on the back off move...just a bit of error checking realy, or that the tool hasn't mashed the tool setter ;-)

I wasn't sure if i could use a G31 with a G0 - I'll give that a try too.

ATB

[angelw]

Originally Posted by derekBPcnc Hi Al, I've been thumbing through the manual - It says that the input interface signals 1 to 32 can be read by variables #1000 to #1031. I'm hoping that the skip x4.7 is connected to one of these i/ps I haven't got the electrical drawings with me, so I'll have to check when I get to work tomorrow. I just want to check the probe closes on the back off move...just a bit of error checking realy, or that the tool hasn't mashed the tool setter ;-) I wasn't sure if i could use a G31 with a G0 - I'll give that a try too. ATB

I wasn't sure if i could use a G31 with a G0 - I'll give that a try too.
No you can't. G31 is a one shot command that is terminated by the slide reaching the end point or by the change of state of the switch, and functions similarly to G1. Accordingly, if you want rapid movement in the G31 block, include a fast feed rate.

I just want to check the probe closes on the back off move...just a bit of error checking realy, or that the tool hasn't mashed the tool setter ;-)
You have the logic correctly planned. The switch should be made when the tool/probe is in fresh air. A broken wire etc in the system will therefore fail safe.

Regards,

Bill

[Dan Fritz]

There are macro inputs and outputs that you can see with variables in the 1000s, but those are not related to the SKIP input. I suppose you could wire your probe output to BOTH the SKIP input and one of those 16 user macro inputs (Ui0-Ui15), but that may corrupt the latency of the SKIP input. These macro inputs have to be processed in the ladder, so if your ladder doesnt' have them, you may have to modify the ladder as well as wire up the extra input.

I suggest this trick instead: If you want to check if the probe is stuck on, move to a position where you know the probe be off, save the current absolute position in a local variable, then make a very short G31 move to another point where the probe should also be clear. If the SKIP axis position is not the same as the "target" position, then the probe is stuck, and you should generate an alarm. I used to do it this way when we were digitizing with Renishaw MP1 probes. They liked to stick in the "on" position when you move away after touching the part. I believe we even went to the extra effort to "bounce" the probe against the last known part surface, then back away again to center the probe, rather than throw an alarm. When you're touching 10,000 points in a digitizing cycle, you've got to have some way to recover from a stuck probe.

[derekBPcnc]

Hi Dan, Bill

Thanks for the replys.

The Fanuc manual says that #1000 to #1031 can be used to read the status of the external inputs.
trying to read anything over #1015 cause a variable out of range error, so ties up with what you say about Ui0 to Ui15.

I thought that the input registers would be mapped to the #1000 variables - obviously not the case (and outputs to # 1100 - #1131 I think).

SO - I'll take your advice and use some logic comparing the absolute position before and after skip to make sure the expected move has been made and not 'halted' by a stuck probe (tool setter stylus).
As a side note - I see andthe X4.7 Skip in the PMC maintenance status screen and it toggles when the probe is tripped- I guess that the X4.7 signal is not being passed to the CNC...but it must be for it to process the G31 command ...that's the confusing part.

The other way I was going to try to extract the status of skip signal was to read the whole of the X4 register and process that Byte to extract BIT7 - I think #1032 to #1035 reads the whole register??? IF that can be done in Fanuc (I'm thinking microprocessor register techniques, so might be complete tosh when it comes to CNC

I haven't seen the Renishaw macros, so I dont know how Renishaw error check probe open / stuck to flag an error.

Fun learning if nothing else :-)

ATB
Derek

[Al_The_Man]

Originally Posted by derekBPcnc As a side note - I see andthe X4.7 Skip in the PMC maintenance status screen and it toggles when the probe is tripped- I guess that the X4.7 signal is not being passed to the CNC...but it must be for it to process the G31 command ...that's the confusing part. ATB Derek

As I mentioned, the PMC skip is passed over to the CNC side via the BMI. Basic Machine Interface, this is the bus that the CNC and PMC talk to each other, normally the PMC and the CNC communicate via G and F registers, it is a while since I implemented a Fanuc ladder, but I believe there is no G register bit available to enabling conditioning the Skip input, the High Speed skip bypasses the BMI/PMC and inputs direct to the CNC.
Al.

[sinha_nsit]

Originally Posted by derekBPcnc ... The Fanuc manual says that #1000 to #1031 can be used to read the status of the external inputs. trying to read anything over #1015 cause a variable out of range error, so ties up with what you say about Ui0 to Ui15. I thought that the input registers would be mapped to the #1000 variables - obviously not the case (and outputs to # 1100 - #1131 I think). ...

#1000 - #1015 and #1032 are defined.
#1016 - #1031 may not be defined.
The same applies to output signals also.
See the attachment for more details.

[JMS4287]

Originally Posted by Dan Fritz I suggest this trick instead: If you want to check if the probe is stuck on, move to a position where you know the probe be off, save the current absolute position in a local variable, then make a very short G31 move to another point where the probe should also be clear. If the SKIP axis position is not the same as the "target" position, then the probe is stuck, and you should generate an alarm.

Here's a quick sample prg of that same concept.

O9032(REN FLASH ON)
G98

#2=#5042-#5082
#3=[#2-[.002]]
#4=0
N2
IF[#4GT4]GOTO4
G31Z#3F20.
G4U.1
#6=#5042-#5082
G4U.1
IF[#6LT#2]GOTO5

G0Z#2

M55
G4U.7
M54(M CODE TO TURN PROBE ON)
G4U.7
M55(RESET M CODE)

#4=#4+1
GOTO2
N4
#3000=101(PROBE START UP FAILURE-CHECK BATTERY)
N5
G0Z#2
M99

[derekBPcnc]

Originally Posted by sinha_nsit #1000 - #1015 and #1032 are defined. #1016 - #1031 may not be defined. The same applies to output signals also. See the attachment for more details.

Hello Sinha,

Thanks for the info - I'll 'digest' in conjunction with the manual tomorrow when I'm back in the office, as I understand now, that X4.7 skip is not mapped to a user variable.:-(
Another quick question - of the variables in the range #1000 to #1015 - how do I work out the mapping of a specific input (say X2.1) to a particular variable?
-------------------
Hi Dan,
Thanks for the section of code- looks like a plan is forming
I got the MI-8 interface delivered today -so let the real fun commence

ATVB
Derek

[sinha_nsit]

Originally Posted by derekBPcnc ...Another quick question - of the variables in the range #1000 to #1015 - how do I work out the mapping of a specific input (say X2.1) to a particular variable?

Read the attachment which has more detailed information.