Motion Instruction ERR = 3 (Execution collision) MAW instruction

Hi Guys,
I am having trouble with a program that I have running on two different machines (same logic). On one machine I am getting an intermittent error on my Motion Arm Watch instruction. The error code is 3 which means there is an execution collision. The only real difference in the machines is that on this one the conveyor runs much slower than the other and I think this may be causing the problem. The machine is a two axis pick and place picking up from a conveyor with an encoder on the conveyor as my third axis giving me my speed and positioning of the conveyor. I have checked AB's knowledge base but I can't make out the picture of the logic they are showing. Their knowledge base ID number for this is 27647. Their answer also applies to more than one MAW instructions (or seems to) and I only have one. Here is what it says:

Problem: Arming the Motion Axis Watch multiple times during axis travel causes an ERR (ERR = 3 Execution Collision).
Cause: Watch Event Armed Status is set on a course update period after the execution of MAW instruction in ladder scan. A second MAW instruction issued within this window will cause an ERR (ERR = 3 Execution Collision)
Verify: Logic tested and verified with ControlLogix V 12.01 software
Cure: Qualify the MAW input rung with a state machine along with qualifying the rung with WatchEventStatus and WatchEventArmedStatus.

I can provide my logic if you would like to look at it.


Does anyone have a good fix for this?

Thanks DoubleL

I don't think I will be able to help you but I bet someone else can. Do you have any part numbers on the hardware? AB drives Serco's? Motion Cards Encoder cards?

I know from experience that you get this error if you try to arm registration and start an axis home at the same time (MAR & MAH). I suspect that the MAW could be similar.
Whichever of the MAR, MAH, MAW that is activated second gets the error.

It is an AB Contrologix with Sercos M08SE with 1394 servo module.


Thanks for the fast replies. I hope to have this fixed before I leave here on Wednesday.

Don't do another motion instruction for that axis until the .Dn bit from the Arm command comes on. That pretty much goes for all the motion instructions (except for the merge move instructions), look at the .DN bit or the .PC bit before executing the next instruction for tht axis.

Having said that, I am not a motion gur, I've done just enough to be able to use the Kinetix stuff somewhat confidently, but this would be a good place to start looking.

I do have a PC bit in the rung after the instruction that won't allow the next rung to process but I still get the error. I thought this would fix it also. It is wierd that it happens on one machine and not the other. Only thing really diufferent aside from this one running slower would be firmware. It is newer on this machine.

DoubleL said:

I can provide my logic if you would like to look at it.

Click to expand...

Might be helpful if you're still stuck.

Logic

Here is a copy of my logic. Tried using a fix from Rockwell's knowledge base today but it caused the instruction to error the same way. It involved putting a watch event armed status bit before the MAW. Also talked to Rockwell and he suggested trying a one shot in front of the MAW. This did the same thing. It may be the way my logic is written. I am open to any suggestions.

I didn't see any logic for the "Depanner_On_Bit". Could you be getting intermittent, unexpected transitions of this bit?

One thing I noticed, which may or may not have an influence - in rungs 23 & 24 you've used the same control tag for two instances of MDW instruction.

suggested trying a one shot in front of the MAW

Click to expand...

Where did you put the one shot? If it wasn't immediately after the EQU, then try putting it there.

The logic for the Depanner on bit is in another routine. It is turned on via panelview buttons only but I guess it could be intermittent.


I just added the MDW instructions today and you are right I did use the same control tag for both. I didn't realize this. I will change that in the morning. The second one would only turn on in the event of a module fault so I don't think it has caused a problem yet.

I put the one shot right before the MAW instruction. Do you think it would make a difference?

BTW I didn't have any problems today. I don't know if it just didn't happen or if the MDW helped any.

But then again it is intermittent. Seems to happen more on night shift for what ever reason.

DoubleL said:

I put the one shot right before the MAW instruction. Do you think it would make a difference?

Click to expand...

The one shot will trigger for any transistion in the elements preceding it on the rung. If the problem is re-triggering the MAW, then putting the oneshot immediately after the EQU will prevent re-triggering by the subsequent bits. This assumes, of course, that the step number stays stable at 40.

That makes sense. I will give that try tomorrow. Thanks for the help.

Following on from http://forums.mrplc.com/index.php?showtopic=10855&hl==
as more people are helping here

I expect that the MAW is running and then MAW is re-executed

I think that you can force the error -
a)Stop the main conveyor (encoder axis)


b)Open depanner door (step 40 preferably)c)Close depanner door - Error occurs

Reason: Watch is still running when the door is open and you are trying to Re-enable the same instruction on the same axis

OR toggle Depanner_On_Bit when in step 40 with the conveyor (encoder axis) stopped

attempt at a solution
Aim is to only execute the MAW if needed and monitor the axis properties .WatchEventStatus for the trigger

Not sure if I understand everything. I need the MAW to trigger my pickup at the correct position of the conveyor. Are you saying that the watch event status would do the same? I'll give this a try in the morning and see what happens.

Also if I open the depanner doors it kills the servo module because that is the e-stop circuit. I could add a temporary bit in the rung and toggle it off and on and attempt the same thing couldn't I?