Powerflex 755 occasionally missing the speed reference

Hello all,

I seem to have exhausted the limits of my knowledge, thought I might give yours a try. I've got a machine that wraps hose before the vulcanizing process. I've got 2 reel-ups to gather the hose onto reels. I take the speed of the wrapper, and calculate how fast I need the reel-up to go to keep up. I send the same speed reference to both reel-ups. Only the one I'm using gets the "start" signal. Every once in a while, (maybe 3 times a shift...) The same reel-up will not start when the operator starts the machine. The drive will say "running at speed." "0 RPM". The operator can push the stop and start button to re-start, and it will pick up and run like nothing happened. I've got 6 of these machines with the same program, and I don't have this issue anywhere else, so I don't think it's my programming. I've swapped drives to see if I can move the problem. It didn't move. I've tried new ethernet cables, and different ports. I've checked and double checked parameters in the drive. Of course, when I'm online the machine runs with no problems. I'm not sure where else to look.

Thanks in advance for your help!

Jeremy

Welcome to the PLCTalk forum community !

Usually when the symptom is "the drive doesn't start when I command it to start" the issue is the asynchronous nature of the I/O connection and the failure of the program to give a low->high transition of the Start command while verifying that the Stop command is already low.

But in your case the drive says it's running, just at the wrong reference.

Are you using the 755's as CIP Motion devices, or just as ordinary speed-controlled VFDs ?

Is there anything in your design or program that might switch the PowerFlex to using a different reference than the network signal ?

If you aren't using the discrete I/O, tie all the discrete inputs down to DC common so that stray voltage can't inadvertently select a different speed reference.

You may want to monitor the Reference Owner to see what device is currently giving the 0 Hz reference to the drive.

Thanks for the welcome Ken, and the suggestions.

This 755 for the reel-up is an ordinary speed controlled VFD. I've got 3 in the wrapper that are CIP Motion, but not the reel-ups.

Parameter 545 is Port 13 (Ethernet)

I am using the I/O to catch the jog/start buttons and run the lights at the remote operating stations by each reel-up.

Math is where I calculate the speed for the reel up, and run is where I send it.

That looks to me like you're sending a Reference of 0.0 Hz, and the drive is happily following that reference.

If your speed reference is based on the Actual Speed of another drive, and that speed is zero... then your speed reference is going to be zero.

Clearly I'm overlooking or misunderstanding something. Are you getting a zero speed feedback that you don't think you should be getting ?

They aren't currently running the machine. That is where all the zeros come from.

When the machine runs the Cat_IPM_Scaled is constantly moving. It's maintaining the pitch on the nylon wrap going on the hose, or coming off. The machine wraps, and unwraps running forward or reverse. My Calculated_Letoff_RPM is also constantly moving. So my LETOFF_1:O.Reference is constantly moving. So even if I sent a 0 speed reference when the reel-up starts, it's constantly updating.

I've got 6 other machines that work fine. I'm not sure what the issue is with this one.

See the top quote:

https://www.goodreads.com/author/quotes/2448.Arthur_Conan_Doyle

You have swapped drives, you've changed Ethernet cables, you swapped Ethernet ports. You know you have effective comms because the drive starts. Sooner or later you have to admit that you have confirmed all the hardware. Either the selected speed reference source is changing to something you don't want or the correctly selected speed reference is 0. "...whatever remains, however improbable, must be the truth."

Keith

I think that Parameter 925 (Ref Select Owner) and Parameter 930 (Speed Reference Source) would be worth monitoring, to see if the drive is somehow getting a command to refer to the HIM or an analog or discrete preset instead of the EtherNet/IP reference.

Bouncing off of what Ken said about the "asynchronous nature of the I/O", is it possible that your calculated speed command is going to zero and then bouncing back up every other scan or so?

I'm not sure how a drive would actually react to this but it's a thought.

Can you show us your logic for start and stop?

If you are in fact getting the Drive_Active bit high when the problem occurs and the reference is in fact 0.0 HZ then your logic somewhere is telling it to run at that speed.

If you are telling yourself "it cant be me" then take another pass at it. I know this from my own ego.

Does the wrapper start before the reel? Or do they start at the same time?

Here is what it looks like when it's running. I'm watching parameters to see if anything jumps out at me.

For a tiny instant in between the second MUL and the ABS instruction pictured in the "Math" rung, the value of the "Calculated_Letoff_RPM" tag is negative.

Because that tag is used to calculate the output command and a zero is substituted for the Letoff_1:O.Reference tag if it's ever calculated to be negative, then it's possible that this is the source of your zero value.

I can't tell the execution order or task assignments of these various routines from the screenshots, of course. If they're all being executed in the continuous Task, I wouldn't expect that zero value to persist for more than one I/O scan, if at all.

But the fact that somebody put a "if negative, then substitute zero" rung for a value that's supposed to be calculated as an Absolute value suggests that they did it because the execution of their code sometimes allowed that to occur.

A diagnostic CTU counter placed on that "if negative then zero" branch would capture such an event.

Some things I learned today... I can jog the reel up and get the same result, and it's probably easier to troubleshoot that waiting for the operator to run that reel up. Now knowing this, I was watching parameter 2 Commanded SpdRef. I push the jog button I get a value of 500 in there. Sometimes the reel up will spin... Sometimes it won't, but every time parameter 2 had 500 in it. I had the operator run it. I had a value of 190 in parameter 2, but no rotation on the reel up. Operator pushes stop/start... value of 190, and the reel up spins.

It would seem that initially I thought the speed reference wasn't getting there because I saw running at speed 0 RPM on the him. What I saw today though tells me the speed reference is getting there, it's just not handling it right. Is it something in the motor? There is no encoder for feedback.

Check parameter 594 when you try this. It is really close to the end of the line in speed selection in the drive. If you are seeing appropriate values there then it would seem to be something odd with the motor control. If you are not seeing appropriate values, we are back to speed selection.

Keith

This sounds more and more like a "missed Start bit transition" to me.

Can you ZIP and post the program itself ? Examining screenshots is unsatisfying.