NH4000 Strange behaviour

[Leigh @ Kodiak]

This just started occuring last Friday on programs that have been proven running for months.

 

When calling a program from the data server  (m198) we started getting "009 P/S ALARM (ILLEGAL ADDRESS INPUT)"

 

 To get rid of this error, all I had to do was remove all the comments at the beginning of the file. This is baffling as these comments have always been there without issue until last Friday. Has anyone ever seen this behaviour before, or have a clue as to what may be causing it?

 

Here is an example of what I removed:

 

(NH4000 - PROGRAM IS TO BE RAN FROM DATASERVER)

(MACHINE GROUP-1)

(MCX FILE  - W:\MACHINING FILES AND CUSTOMER FILES\SUPERCROSS\SX MINI CRANK\CAM\SX MINI CRANK 150.MCX-9)

(SX MINI CRANK 150 - ROUGHING)

(PROGRAM   - O0150.NC)

(DATE      - SEP-18-2015)

(TIME      - 1:57 PM)

(T26  - 1" 2FL MITSUBISHI INDEXABLE EM, 2.65SO, .5FL, .08"CR - H26  - D26  - D1.0000" - R0.0800")

(T98  - 5/8" 3FL ROUGH EM 1.625FL, 2.0SO - H98  - D98  - D0.6250" - R0.0300")

(T62  - 2" MITSUBISHI FACEMILL, .08CR - H62  - D62  - D2.0000" - R0.0800")

(T15  - 3/16", 3FL ENDMILL, 3/4" RELIEF - H15  - D15  - D0.1875")

(T76  - 1/2" SC, 2FL BN, 1.75FL,SO - H76  - D76  - D0.5000" - R0.2500")

(T8   - 3/8 2FL SC EM 1.0 LOC HIGH HELIX - H8   - D8   - D0.3750")

(T100 - 5/8" VP EM, 2.0FL, 2.4SO - H100 - D100 - D0.6250")

(T16  - 2 DEG TAPERED ENDMILL, 1" FLUTE LENGTH, 5/64 TIP DIA, - H16  - D16  - D0.0781")

(T92  - 1/2" ENDMILL W/ .125 CR - H92  - D92  - D0.5000" - R0.1250")

(T119 - 15 DEG TAPERED ENDMILL W/1" FLUTE LENGTH, .1875 DIA TIP - H119 - D119 - D0.1875")

(T42  - 3/8 SC CHAMFERMILL, 1.4 SO - H42  - D42  - D0.3750")

(T63  - 1/4" X 90 DEG CARBIDE SPOT DRILL - H63  - D63  - D0.2500")

(T95  - 7/32 DRILL           - H95  - D95  - D0.2180")

(T10  - 4.6MM NACHI SG DRILL - H10  - D10  - D0.1811")

(T12  - M5 X .8 FORM TAP     - H12  - D12  - D0.1969")

(T13  - 12-1 CARBIDE THREADMILL - H13  - D13  - D0.4000")

(T14  - TM450-20A CARBIDE THREADMILL - H14  - D14  - D0.4500")

(T4   - 1/4 2FL SC EM, 1.15SO, .95FL - H4   - D4   - D0.2500")

(OVERALL MAX - Z12.)

(OVERALL MIN - Z-.5284)

[gcode]

Is it possible there is a blank line at the top of the program

I've seen this cause all sorts of trouble with some machines

[Leigh @ Kodiak]

Is it possible there is a blank line at the top of the program

I've seen this cause all sorts of trouble with some machines

 

Perhaps, but it wouldn't explain why it worked on Thursday, and not on Friday.

[?Mark]

I've seen similar things happening when using Mori's user area memory but reset always cleared it. Never a problem when using their data server though.

I have a feeling that someone "fat fingered" something in there

[Leigh @ Kodiak]

Nobody has access  to the dataserver to "fat finger" anything. This one seriously has me stumped.

[cincy k]

Check for a () inside of another ().  Or a ( inside a () or ) inside a (). Think we had this happen today.

[Colin Gilchrist]

Could it be some type of Parameter change on the machine?

[Leigh @ Kodiak]

Did a bit of trial and error investigating on this one. It seems that for some reason the control no longer likes the "\" in the file path. Interestingly, if you try to input this character on the control (shift +W) the character ¥ shows up on the screen instead. if you try to input that character into the code, there is a message along the lines of that character not being registered for use in nc programs, which is absolutely correct. No I need to figure out why when triying to input "\", I actually get "¥"...anybody have a clue on this one??

[Craig-B]

I've had a similar issue and It was that someone turned on the TV Check parameter (000). 

On 000  bit 0 should be 0 and it was set to a 1. Check your machine and see what it is set to.

[Leigh @ Kodiak]

You got my hopes up there, but bit 0 is already set to 0

[Guest]

Never had any trouble with "/" or "\" on a 30i/31i/32i control. What control is it?

[Leigh @ Kodiak]

It's a Mori badged MSX 501 III, which I believe is an 18 underneath. The problem is definitely not the \, but that when the machine sees "\". it thinks it is seeing "¥"

[Matthew Hajicek - Singularity]

 

 

Perhaps, but it wouldn't explain why it worked on Thursday, and not on Friday.

 

It wanted to go home early.

[Leigh @ Kodiak]

I also found out this morning that the control no longer will take programs with an "_" underscore in the comments, but this has not been a problem before.

[Rstewart]

Time to call DMG/MORI I would think, that's wacky

[mkd]

I've always had trouble with special characters on Mori/fanuc

surprised you got away with it at all