Do-More and CTRIO2

[DLTimmons]

I have a very simple index app that I tried to use a Do-More and a CTRIO2 on I switch programs for indexing and jog base off a bit on a touch panel.I have included interlocks to keep the program from exit when the command stage is active.

The CTAXTRAP and CTAXJOG exit to the fault stage with no error showing in $Danly.errorcode but there is a warning that show up in desginer have a screen shot of it but the file to large to attach. What strange is after jumping to the error stage the enable and suppend show to be active jump back to the command stage and enable is showing inverted to the input with suppend off it then jumps to the fault stage again.The only way to get operational is to cycle power that not going to fly cycling power ever 2 or 3 parts on a punch press.

Minium I need a way to rest with out power cycling.

I have screen shots and the program that I can not download.

AD support said there was a new firm ware and the web site but the link down load a 9/5/12 version 1.0.0 not version 1.0.1

Donnie

[franji1]

To get the latest firmware, make sure you have access to the Internet.  Then from within Designer, do Help->Check for Updates on the menu, and this should prompt you to download the latest H2-DM1* firmware (even CTRIO2).  This hits the Host Engineering web site's firmware files, which are always the latest/greatest available.

[DLTimmons]

You have this listed on the firmware page.

1.0.1   15-Apr-2013 H2-CTRIO2 Fixes a problem where the CTAXLIMT profile was not correctly reversing direction on a limit.

Fixed a problem where the DIR input was not being properly evaluated.
 

I did the check for updates and it still shows 1.0.0 as the current CTRIO2 firmware and 1.0.9 for Do-More firm ware. Did find 1.0.1 in the All_HostEng Firmware_Files.Zip

the interlocks seam to have fixed the error for the most part but I do need a way to stop and breakout of a CTAXTRAP before the steps have completed or a way to block the output pulses. There has to be a way to reset the CTRIO error without power cycling.

[BobO]

I'm not completely understanding your problem, but no, you should not need to reset the CTRIO2.

When you say that CTAXTRAP and CTAXJOG are exiting to the fault stage, how is that happening? They should never go to the error stage under normal conditions?

And can you tell me what the warning says?

[DLTimmons]

I have have three screen capatures that will show what I am talking about one has the error page but they are to large to attach to this site. Will send an e-mail to the support box

I think what was happing was a DLC bit from the touch screen was droping out for a scan froceing the program with the command to exit in the middle of the move. I sense add several interlocks to try to stop this.

I still need to be able to stop a move and kill it then restart a fresh move.

[BobO]

I still need to be able to stop a move and kill it then restart a fresh move.

Just drop the box enable.

[DLTimmons]

Just drop the box enable.

That what I remember starting all my problems dropping the enable and leaving that stage of the program before the move was completed.

[Greg]

DLTimmons, I just sent you an e-mail in response to the pics you sent to tech support. Looks like to me you may be abnormally terminating a driver used by CTAXJOG in Stage 2 of JogStepper program code block. If DLC4 is turned ON and then DLC5 is turned ON, then the program will exit and abnormally terminate that driver (see pic). If this happens, then it could leave that driver in a weird state which, in turn, causes your other stuff (e.g. CTAXTRAP) not to work properly.

[BobO]

That what I remember starting all my problems dropping the enable and leaving that stage of the program before the move was completed.

The abnormal termination warning that you were getting is generated when you have shut down a program or stage while an instruction is still active. In the code that you sent us, there was an EXIT instruction in the stage with the jog. If that EXIT is enabled while the CTAXJOG is running...or hasn't completely terminated followed you dropping the enable...that will result in abnormal termination. The better way to exit is to have the CTAXJOG transition to another stage which contains the EXIT instruction. It is slightly more work that way, but it will always work correctly.

As a general rule, you should never see the abnormal termination warning and you should always treat it as a programming error. While in most cases we do an ok job of cleaning up when abnormally terminated, we have missed a few things in the past and possibly in this case as well. I will look into it to see if there are some values in the structure that need to be cleaned up...I suspect there are.

[DLTimmons]

What happens if you have a trapezoid move in process and you drop the enable input? Does it ramp to a stop then exits the box as success? Or does exit as an error because you did not reach the target? or does it just waits till the enable comes high again and finish the move?

I seam to have more problems with the Asynchronous command than anything else. I still think you have a very good CPU I just have to take my game to that level

[BobO]

Dropping the enable terminates the move immediately, without decel, and triggers the Success condition. The Error condition is for when there was an error loading or running the profile.

If you need to decel, you'll need to use the Suspend input and then drop the Enable after decel. There is a struct field that tells whether it is still active or not...but I'm not in front of my PC and I don't remember the field name.

The key thing with asynchronous instructions is to be sure to let them run to completion...you must make sure they run until the Success or Error bit turns on (or stage transitions). In the case of the CTRIO, when you drop the enable, it may be many scans before the command gets to the module and the module replies. By waiting until box signals completion, you are guaranteed for that process to have completed properly.

[DLTimmons]

Dropping the enable terminates the move immediately, without decel, and triggers the Success condition. The Error condition is for when there was an error loading or running the profile.

This info on how the enable input functions would be good info to have on the help page. I'm sure there will be others that have no ideal how it works. I was thinking that the only way it came out of a move with success was by completing the move.

I do like how the suspend works see many possiable uses for that.

Again thanks for putting up with me on this.

Donnie

[Greg]

DLTimmons, I noticed in your $Main, you are also using HALT instructions to terminate (abnormally) JogStepper, IndexStepper and Press code blocks (see pic). Since these code blocks have asynchronous instructions in them (e.g. CTAXJOG, CTAXTRAP), then halting them like this could cause the "abnormal termination" messages you are getting. It is better to allow those instructions themselves to terminate normally by jumping to a stage (i.e. On Error/On Success JMP to Stage) and then logging any error (if necessary), and then use the EXIT in that jumped-to stage. This way all things are cleaned up normally. You should only use the HALT (under most circumstances) as an E-STOP or something similar (i.e. an emergency).

We are adding more clarification in the Help file at your request.

[DLTimmons]

DLTimmons, I noticed in your $Main, you are also using HALT instructions to terminate (abnormally) JogStepper, IndexStepper and Press code blocks (see pic). Since these code blocks have asynchronous instructions in them (e.g. CTAXJOG, CTAXTRAP), then halting them like this could cause the "abnormal termination" messages you are getting. It is better to allow those instructions themselves to terminate normally by jumping to a stage (i.e. On Error/On Success JMP to Stage) and then logging any error (if necessary), and then use the EXIT in that jumped-to stage. This way all things are cleaned up normally. You should only use the HALT (under most circumstances) as an E-STOP or something similar (i.e. an emergency).

We are adding more clarification in the Help file at your request.

I have the rungs you have pictured now controled with interlocks where it cannot drop out while the stage with the Asynchronous comamad is active is in only after it comes out of that stage does let me change programs

Based on your e-mail I remove the halts from Main and added exits to every stage other than the stages with the CTAXJOG and CTAXTRAP.

[franji1]

Based on your e-mail I remove the halts from Main and added exits to every stage other than the stages with the CTAXJOG and CTAXTRAP.

Make sure that "every stage" has no other instructions that have side-effects of "exiting".  This includes any other async instructions (not just the two you mentioned).  Async instructions are identified with the yellow, red, and blue triangles in the corners.

Just ask the question, "are there any instructions or operations this stage is doing that can be aborted at any point in time".  This includes at the system level, and at the instruction level (e.g. timers won't time, counters won't count, even something simple like an output bit might not get SET or RST, etc.).  Usually, mid-stage flow exiting is for e-stop type operations, but mode changes (e.g. auto to manual or manual to auto) involve more of a "programatic" transition (e.g. homings, clearings, initialization, etc.) back to a "base" state before the OTHER mode takes over.