OkiePC
Lifetime Supporting Member
I am having some issues with triggers and message handshaking. I inherited an application which uses the AdaptaScan barcode reader to identify and report the barcode information from cases of sausage as they pass by on a conveyor.
The information is sent to a SLC via Devicenet polled mode. The SLC uses the barcode information to decide which of 4 "lane" choices to make for the product. Each lane is a different branch in the conveyor system and is further downstream. The lane assignment is pushed onto a FIFO stack which is unloaded by the photo cell prior to the lane 1 diverter downstream. The process continues for each lane for a total of 4 FIFO stacks. That part of the logic fucntions okay, but can get out of sequence when a photocell fails. I will revisit that section of the code at a later time.
The first photocell that detects the package in front of the adaptascan reader is wired to the SLC and debounced with a timer to begin the message triggering. If it's blocked for 0.4 seconds, then we have a real box and the adaptascan trigger is turned on for 1 second.
Here's where one of my primary problems occurs:
Now, when the trigger is turned on, according to all I've read, the adaptascan should respond with a message by toggling its output bit 15 to the opposite state. This should happen "When all fields are satisfied". According to the book, the device will send what it has when the trigger is turned off even if all fields are not satisfied. The latter is what is happening every time. So, I need to figure out why all fields are apparently not satisfied. Right now, the system only works because we are turning off the trigger bit exactly when the adaptascan is "on symbol" and it responds with the correct value.
Right now, I can move the package detection eye to the right or left an inch or two and immediately get nothing but "no reads" from the scanner.
Okay, no big deal really, except we are being asked to run more and more varieties of boxes and at some point it will be difficult to ensure that the adaptascan reader is actually "on symbol" when the PLC driven trigger bit is turned off.
So, I need to figure out how to make the adaptascan reader send a message as soon as it has one complete valid label decoded. The symbologies in use include, but may not be limited to: UPC-A, interleaved 2 of 5, and Code 128.
Another issue I had yesterday...My package detect photocell "got to floppin' around" when the nose-mount nut came loose and generated a bunch of ghost boxes that got the whole system out of sequence. I tightened it up and had the operator clear the line and reset the tracking, which clears all the FIFO stack data. The system was still out of sequence when we resumed operation. After several attempts, and checking all the other FIFO related photocells, I was about to give up and go get the laptop out when it straightened out.
Now, my theory is that there was buffered barcode data in the adaptascan, caused by the loose photocell, and we were receiving old message data in the SLC until we jacked with it long enough to clear out the adaptascan buffer...This is the basis for this, yet unproven, theory:
"Note: If DeviceNet is selected as the host (page 10-10), messages are queued until requested by the DeviceNet master (see Message Flow Control page 10-13)
The trigger controls when messages are placed in the queue."
That is from the manual "AdaptaScan Software (2755-ASN) User Manual".
It is the only mention of a message queue. The ladder logic devicenet example is extremely simple and it appears that the programmer used it unaltered when programming this machine.
Does anyone have more details about the adaptascan devicenet messaging using a discrete (devicenet) trigger or the message queue?
Thanks,
Paul
BOOK EXAMPLE:
MY CODE:
An overview of the line:
The information is sent to a SLC via Devicenet polled mode. The SLC uses the barcode information to decide which of 4 "lane" choices to make for the product. Each lane is a different branch in the conveyor system and is further downstream. The lane assignment is pushed onto a FIFO stack which is unloaded by the photo cell prior to the lane 1 diverter downstream. The process continues for each lane for a total of 4 FIFO stacks. That part of the logic fucntions okay, but can get out of sequence when a photocell fails. I will revisit that section of the code at a later time.
The first photocell that detects the package in front of the adaptascan reader is wired to the SLC and debounced with a timer to begin the message triggering. If it's blocked for 0.4 seconds, then we have a real box and the adaptascan trigger is turned on for 1 second.
Here's where one of my primary problems occurs:
Now, when the trigger is turned on, according to all I've read, the adaptascan should respond with a message by toggling its output bit 15 to the opposite state. This should happen "When all fields are satisfied". According to the book, the device will send what it has when the trigger is turned off even if all fields are not satisfied. The latter is what is happening every time. So, I need to figure out why all fields are apparently not satisfied. Right now, the system only works because we are turning off the trigger bit exactly when the adaptascan is "on symbol" and it responds with the correct value.
Right now, I can move the package detection eye to the right or left an inch or two and immediately get nothing but "no reads" from the scanner.
Okay, no big deal really, except we are being asked to run more and more varieties of boxes and at some point it will be difficult to ensure that the adaptascan reader is actually "on symbol" when the PLC driven trigger bit is turned off.
So, I need to figure out how to make the adaptascan reader send a message as soon as it has one complete valid label decoded. The symbologies in use include, but may not be limited to: UPC-A, interleaved 2 of 5, and Code 128.
Another issue I had yesterday...My package detect photocell "got to floppin' around" when the nose-mount nut came loose and generated a bunch of ghost boxes that got the whole system out of sequence. I tightened it up and had the operator clear the line and reset the tracking, which clears all the FIFO stack data. The system was still out of sequence when we resumed operation. After several attempts, and checking all the other FIFO related photocells, I was about to give up and go get the laptop out when it straightened out.
Now, my theory is that there was buffered barcode data in the adaptascan, caused by the loose photocell, and we were receiving old message data in the SLC until we jacked with it long enough to clear out the adaptascan buffer...This is the basis for this, yet unproven, theory:
"Note: If DeviceNet is selected as the host (page 10-10), messages are queued until requested by the DeviceNet master (see Message Flow Control page 10-13)
The trigger controls when messages are placed in the queue."
That is from the manual "AdaptaScan Software (2755-ASN) User Manual".
It is the only mention of a message queue. The ladder logic devicenet example is extremely simple and it appears that the programmer used it unaltered when programming this machine.
Does anyone have more details about the adaptascan devicenet messaging using a discrete (devicenet) trigger or the message queue?
Thanks,
Paul
BOOK EXAMPLE:
MY CODE:
An overview of the line:
