Encoder Use with VFD

TPress

Member
Join Date
Apr 2024
Location
Iowa
Posts
9
Sorry in advance for the long post, but this requires a little back story.
  • I work in a facility with a couple hundred VFDs
    • Most are in a web handling application
    • Some are driving an extruder (plastic)
    • Drives are primarily AB (1336 and PF755) and Parker/SSD (690+ and 890)
    • Sizes are from 2HP - 500 HP. Most are in the 5-20 range
  • Most drives are running closed loop with either a load cell or pressure transducer for feedback
  • Many (most?) of the drives have the feedback instrument wired directly into the drive. The PLC provides the speed reference, and the instrument is used for trim
  • Almost every drive has been outfitted with an encoder
  • We have found that this is a bad deal for several reasons
    • Stuck in older/obsolete drive hardware because upgrading involves engineering work
    • Obsolete drive software requiring older OS for programming & troubleshooting PCs (our IT department hates this)
    • Having the control divided between two components with two different operating environments (the PLC and the drive) makes troubleshooting a PITA
  • All of our upgrades are PF755 drives with the PLC controlling the drive 100% over Ethernet/IP.
    • Makes the drive as "dumb" as possible so it can be replaced or upgraded easier in the future
    • Keeps all of the PID in the PLC where it is easier to manage
    • There's no perceivable effect on the process when doing this vs having the feedback hooked directly to the drive

Now for my question (s).....
  • Do I even need the encoders?
    • What/how much advantage do I get?
    • If PLC scan time plus Ethernet/IP comms are fast enough for the process, is the encoder helping at all?
  • There are some drives that have fixed setpoints (the line-speed "master"). How much is the speed control in these improved with a fixed/open-loop control?
 
Not an answer but a bit of background since I worked in the printing industry years back.

The 1336's and the Eurotherm drives were the pinnacle at the time.
You could also use the Eurotherms as a mini PLC having build in blocks for control and PID's and what not.
It was all about reaction time and thus better control of the process.
Thus, they used the darn things in the printing industry pretty much everywhere.

I'm not saying it cannot be done (PLC's have gotten to the point were they are ridiculously fast and with huge amounts of memory)
but I would be very cautious about going down that route
without spending some serious time studying the system and it's functions.
 
How tightly do you need each motor's speed/torque to be controlled? Oversimplifying it a little, the encoder lets the drive close the loop and maintain very tight control of the motor's speed/torque. I'm not in your world but I imagine you need that in at least some of your applications, especially the web handling.
 
My instinct tells me replacing the encoders would be a bad game to play, just because a surplus of data isn't a bad thing as long as that data is used expressively. It sounds to me like the biggest friction points with the use of those drives is in the maintenance of those systems, what with them having different OS and such. I think you should way some options and find out if taking out the encoders is a viable option given the PID system (hopefully) optimizing the mechanics of how the system behaves. nothing wrong with an experiment in the short term as long as risks are managed and intent is communicated, right?
 
Not an answer but a bit of background since I worked in the printing industry years back.

The 1336's and the Eurotherm drives were the pinnacle at the time.
You could also use the Eurotherms as a mini PLC having build in blocks for control and PID's and what not.
It was all about reaction time and thus better control of the process.
Thus, they used the darn things in the printing industry pretty much everywhere.

I'm not saying it cannot be done (PLC's have gotten to the point were they are ridiculously fast and with huge amounts of memory)
but I would be very cautious about going down that route
without spending some serious time studying the system and it's functions.
Yep.... that's exactly how we got here. At the time, the Eurotherm/SSD/Parker drives were all the rage. We have successfully upgraded several of these lines with full PLC control of the drives with no issues. That said, we are running plastic film, which is a bit more compliant than paper. The integrator we use (that specializes in web handling) does this pretty widely and have developed some pretty clever FBs for winders, tension control, etc.
 
After reading some of the replies and thinking about how to respond, it dawned on me that our integrator is actually here on site right now (doing a drive upgrade), so I asked their design engineer.

He said that at higher motor speeds, we probably don't need the encoders... that the drives could control well enough. BUT... at lower speeds, the drive has less capability to control the motor to a set (calculated) speed, and that the control would get a bit "wonky" as the feedback oscillated and the PID kept trying to correct.

After thinking about it more, the places where we've disabled encoder feedback out of necessity (parts, downtime, etc.) and gotten away with it were applications that ran at relatively high motor speeds and/or had long tension zones (where the plastic film could absorb some of the imprecision).
 
Keep in mind that the big money pit on a press is start up and shut down.
When you get it running it's smooth sailing.
Usually slow speeds (not full speed) during the 2 phases above.
I bet your IT won't be there for you when the bills keep piling up.
 
Keep in mind that the big money pit on a press is start up and shut down.
When you get it running it's smooth sailing.
Usually slow speeds (not full speed) during the 2 phases above.
I bet your IT won't be there for you when the bills keep piling up.

I agree, and we'll stick with the encoders. I wasn't visualizing the second closed loop I between the drive and the motor making sure the loop we are controlling is actually doing what we tell it to do.
 

Similar Topics

Hi, Is there a VFD that can do SLS without an encoder to ANY motor. I know siemens has it but you have to have a special motor from them. /Tim
Replies
4
Views
753
Do I need an encoder (slave) axis for my project? I am using a Powerflex755 and Incremental Rotary Encoder (scaled into linear movement axis)...
Replies
1
Views
2,699
Hello everyone, I am currently working on re-writing the PLC programming from scratch for a glass tempering machine. I used to do mostly process...
Replies
6
Views
11,371
I've got a machine tool that runs a servo, along with a variable speed drill. The servo input card is being damaged when the VFD fires up from...
Replies
8
Views
5,155
I have an application using an incremental encoder and then I convert it to degree (0-360) using calculation program. For a while, the calculation...
Replies
8
Views
357
Back
Top Bottom