Fanuc 31I-A5. Warning

[Eric@HorsepowerInc.]

I just thought I should pass this along as a warning to users of this control. Im in the process of running some extremely long programs that have to be ran off a card, and ran into a glitch in this control that I will be calling Fanuc on to see what they say. Control is a 31i a5 fully loaded with 1000 block look ahead, and believe look ahead is the issue. I can only figure the man above and rapid 0 saved my 7 hr part from the junk bin. The short version of the issue is I WOULD NOT HAVE A PROGRAM ON THE CARD AND IN THE CONTROL WITH THE SAME PROGRAM #. I never even though of this nor believed it could happen when running in tape mode, but the control decided to jump into the memory for a portion of the program. I duplicated this away from my part as well multiple times, so i believe in fact it is a glitch and just a user beware. I will update this topic after speaking with Fanuc. Just thought I should advise users of this potential crash issue.

[Newbeeee™]

Eric,

Were you running just drip feed, or as a sub with the main calling the prog on the card as an M198?

[Eric@HorsepowerInc.]

Newbeeee, running drip feed, directly from card. We have been doing this forever on other machines, just first time I've had an issue.

[Guest MTB Technical Services]

I just thought I should pass this along as a warning to users of this control. Im in the process of running some extremely long programs that have to be ran off a card, and ran into a glitch in this control that I will be calling Fanuc on to see what they say. Control is a 31i a5 fully loaded with 1000 block look ahead, and believe look ahead is the issue. I can only figure the man above and rapid 0 saved my 7 hr part from the junk bin. The short version of the issue is I WOULD NOT HAVE A PROGRAM ON THE CARD AND IN THE CONTROL WITH THE SAME PROGRAM #. I never even though of this nor believed it could happen when running in tape mode, but the control decided to jump into the memory for a portion of the program. I duplicated this away from my part as well multiple times, so i believe in fact it is a glitch and just a user beware. I will update this topic after speaking with Fanuc. Just thought I should advise users of this potential crash issue.

 

I'd have to see the program on the card before I would blame the control.

This is something I have never seen or heard of before.

 

I'm willing to bet that there is something in the program on the card itself that caused this, like an M99 or some Macro-B logic.

While it's possible there is a control error, it's not likely.

[Eric@HorsepowerInc.]

I'd have to see the program on the card before I would blame the control.

This is something I have never seen or heard of before.

 

I'm willing to bet that there is something in the program on the card itself that caused this, like an M99 or some Macro-B logic.

While it's possible there is a control error, it's not likely.

That's fine, take the info however you like it was just an FYI.  Not a new user here, thousands of high code programs ran here weekly and have never seen this happen. If it were only as simple as a M99.....Only program on card, no sub jumps, etc.          Like I stated I will talk with Fanuc and post their findings. 

[Guest MTB Technical Services]

That's fine, take the info however you like it was just an FYI.  Not a new user here, thousands of high code programs ran here weekly and have never seen this happen. If it were only as simple as a M99.....Only program on card, no sub jumps, etc.          Like I stated I will talk with Fanuc and post their findings. 

 

 

Are you using O word identifiers for all your programs or are you using the label within the arrows <PROGRAM ID EXAMPLE> ?

 

For giggles, You might want to try using a small main program in memory to make the M198 call directly instead of DNC.

This will at least give you more information as to where the problem might be.

It could be in the DNC protocol itself as opposed to the look-ahead buffer.

I haven't had any issues with the directories or folder structure on the 31i.

 

Out of curiosity, how large is the file on the card?

 

Is this a Mill, Lathe or Mill/Turn machine?

Manufacturer?

 

If there is a problem you can probably simply load the control update yourself.

I haven't had to do a 31i update in a couple of years but it looks like you might have found something here.

[Eric@HorsepowerInc.]

Are you using O word identifiers for all your programs or are you using the label within the arrows <PROGRAM ID EXAMPLE> ?

 

Out of curiosity, how large is the file on the card?

 

Is this a Mill, Lathe or Mill/Turn machine?

Manufacturer?

 

If there is a problem you can probably simply load the control update yourself.

I haven't had to do a 31i update in a couple of years but it looks like you might have found something here.

 

File size on card is over 15 MB   over 750,000 lines of code/filtered

 

Machine is a Doosan DMN 400HS  31i a5

Table/Table 5ax

 

The program runs flawless when a program# in the control does not match a # on the Card

[Guest]

Tape mode sucks @$$. I NEVER use it. It's old. I ALWAYS run M198 to card. I've run 200MB+ programs that way wit no issues since the advent of the I Series control.

 

I've done calls 3 ways;

 

M198 P1234

 

M198 <TEST1234.TXT>

 

M198 <TEST1234>

 

I suspect "Tape" mode has something to do with it. I NEVER touch that old $#!+. I'm sure many people use it with no issues. That's fine do what you want. My way is 100% rock solid.

 

Set the following;

#138.7=1

#3402.2-1

#6030=198

 

 

And be done with it.

[Eric@HorsepowerInc.]

Tape mode sucks @$$. I NEVER use it. It's old. I ALWAYS run M198 to card. I've run 200MB+ programs that way wit no issues since the advent of the I Series control.

 

I've done calls 3 ways;

 

M198 P1234

 

M198 <TEST1234.TXT>

 

M198 <TEST1234>

 

I suspect "Tape" mode has something to do with it. I NEVER touch that old $#!+. I'm sure many people use it with no issues. That's fine do what you want. My way is 100% rock solid.

 

Set the following;

#138.7=1

#3402.2-1

#6030=198

 

 

And be done with it.

. Excellent, that is good info. We sent the issue to Fanuc and will wait to hear what they say, in the mean time I will give this a go. Thanks

[civiceg]

What Fanuc controls are supported with that <> trick. 16i and newer???  Always done M198Pxxxx

[Guest]

What Fanuc controls are supported with that <> trick. 16i and newer???  Always done M198Pxxxx

30 Series Controls  .

 

15/16/18/21 Controls

[Eric@HorsepowerInc.]

Foghorn, just out of curiosity why do you feel running DNC from card vs sub call to card is so different?

[Guest]

Foghorn, just out of curiosity why do you feel running DNC from card vs sub call to card is so different?

Because it's old. It was designed for drip feeding through RS-232.

 

CF (I/O Channel 4) = Fire Hydrant

USB (I/O Channel 17) = Fire Hydrant

RS-232 (I/O Channel 0 and 1) = Coffee Straw AT BEST

 

Not much more that I can say other than that.

[?Mark]

LOL! Pretty visual description there James 

[Eric@HorsepowerInc.]

Because it's old. It was designed for drip feeding through RS-232.

 

CF (I/O Channel 4) = Fire Hydrant

USB (I/O Channel 17) = Fire Hydrant

RS-232 (I/O Channel 0 and 1) = Coffee Straw AT BEST

 

Not much more that I can say other than that.

Must be slightly different on our control then, we are at fire hydrant status LOL( card runs thru channel 4)  We never use RS-232 ports on machine  only thru PCMCIA card or USB

[Eric@HorsepowerInc.]

^{Also here is what we received from Fanuc thus far.}

 

^{First, does the program include a "G65" Command?  When it changes to CNC Memory from DNC Memory, do you get an alarm or does the machine's movements indicate the program has changed?  Do you notice the mode of the machine changing from REMOTE to AUTO?  For tracking purposes, I recommend activating operation history and monitoring DNCI (G43.5).  This will indicate two things.  First, it will indicate if this signal is transitioning during the program thus changing the mode from REMOTE to AUTO.  Secondly, this will indicate the date/time the transition occurs helping to locate when the mode changes.  For your convenience, I started a service request of "XXXXXXX".  If you have any questions, please contact FANUC at 888-326-8287 (Press 2 for "CNC" and 2 for "Service")}

[Guest]

Must be slightly different on our control then, we are at fire hydrant status LOL( card runs thru channel 4)  We never use RS-232 ports on machine  only thru PCMCIA card or USB

I think I mis-communicated... My apologies.

 

The "TAPE" mode function was designed from the beginning to support RS-232 and literal tape. It was not designed to support the fire hydrant. That doesn't mean it can't, it just means it was not designed to support it.