Best way to coordinate two drives

Sun TC

Member
S
Join Date
Aug 2013
Location
Bowling Green, KY
Posts
5
Hello,

I am having an issue trying to get two motors to work in unison, particularly during accel/decel. I have attached a .pdf with the logic controlling the two motors. The pull roll drive pulls sheet material through a series folding plows and rollers, then pushes it into the cutoff assembly that cuts the sheets into even lengths. They work pretty well together steady state at a given rate, but during transitions from one speed to another, we get short sheets. When running at low speeds while jogging, all the sheets are short. Typical run speed is around 2000 sheets per minute, which translates to 400 cuts per minute on the cutoff assembly, as we get five sheets per cut. The system is pretty stable when steady state above roughly 900 sheets per minute.

Hardware being used is a controllogix L71S processor with Powerflex 525 25B-D010N114 Cutoff drive and the same drive as the pull roll drive. Both motors are 5 HP Induction motors. The Cutoff drive has a working encoder, the pull roll has one wired to the drive, but doesn't seem to be using it.

This line was upgraded from a SLC500 with Magnatek drives to the current setup about a year ago, by a third party integrator. We just found out about this little bug-a-boo this week. Is there a better way of coordinating these two motors so they stay together during transitions?
 
Sun TC said:
Hello,

I am having an issue trying to get two motors to work in unison, particularly during accel/decel. I have attached a .pdf with the logic controlling the two motors. The pull roll drive pulls sheet material through a series folding plows and rollers, then pushes it into the cutoff assembly that cuts the sheets into even lengths. They work pretty well together steady state at a given rate, but during transitions from one speed to another, we get short sheets. When running at low speeds while jogging, all the sheets are short. Typical run speed is around 2000 sheets per minute, which translates to 400 cuts per minute on the cutoff assembly, as we get five sheets per cut. The system is pretty stable when steady state above roughly 900 sheets per minute.

Hardware being used is a controllogix L71S processor with Powerflex 525 25B-D010N114 Cutoff drive and the same drive as the pull roll drive. Both motors are 5 HP Induction motors. The Cutoff drive has a working encoder, the pull roll has one wired to the drive, but doesn't seem to be using it.

This line was upgraded from a SLC500 with Magnatek drives to the current setup about a year ago, by a third party integrator. We just found out about this little bug-a-boo this week. Is there a better way of coordinating these two motors so they stay together during transitions?
Click to expand...
One issue is that the 525, although possessive of "Flux Vector Control", is specifically "velocity" FVC, not torque or position. There is no torque loop or position loop inside. That might explain why the error looms more prominent at low speeds; at higher speeds the small errors are lost in the larger values but at slow speed the same errors become a greater proportion of the travel. The Magnetek drives were likely full FVC and were being used in Positioning mode. You would need the 755 drives for that.
 
I'm not familiar with the 525. Does it have have PID control capability? Can you set an analog output to output torque, and feed it back into an analog input as the PID PV, set point would be the torque output from the "master"
 
I got that from your first post. Does it have a PID loop? If it does you can make a torque control loop using an output setup for torque wired back to an input that is the PV of the loop. SP equals the output torque from the master
 
Trying to get two drives to track during a speed change when one has a tach/encoder and the other does not is going to be difficult at best. Do you know whether the Magnetek drive setup was using electronic line shaft software (one would have had a PG-W2 dual encoder card or similar if AC VFDs)? Did the old drives both have functioning encoders, being used?
 
In theory, you should be able to get real close if the encoder from the Master is the speed reference of the Slave. Reduce the accel/decel of the Slave to 0 or as close as possible, and just use the sccel/decel of the Master.

I have no idea if the 525 can use an encoder as a reference, but many drives can.
 
Peter Nachtwey said:
PLCs are not fast enough for torque or current loops. That should be handled in the drive.
Click to expand...

I am trying to describe this, though probably not very well.

The drive has control capability, I am saying use the output torque of the master drive as it's SP, use the speed as its CV, and use an analog output from the slave drive setup for output torque as the PV into the loop via an analog input of the slave. The loop needs tuning of course.

This will work fine for simpler load sharing applications

If the drive is using flux vector, or any form of vector control for that matter it is odd that it can't do torque control
 
It seems that this is not a load sharing application, it is a motion control application. Peter is quite right that most PLCs aren't cut out for motion control. I once wrote a gearing function for and S7-400 PLC that worked up to a point but you won't get enough bandwidth for good position control. How accurate do your lengths need to be?

It sounds to me like you should be speaking toe the Integrator and asking them to sort it out.

Nick
 
Just change the drives to PF755 (assuming you want to stay with AB) and use the encoders on both motors. The 525 is a fine drive, but it's not cut out for your application.
 
Thanks for all the replies guys. We're rapidly coming to the conclusion that doing this with VFDs as is, is not going to feed the bulldog all the time, even if we can get it right some of the time. We have it somewhat corrected by adjusting the gear ration in a curve and moving the new gear ratio value in to the calculation every 2.5 Hz, but it is still a bandaid. We are working with integrator to possibly upgrade to Kenetix servos or maybe even an actual mechanical gearing solution.
 
Sun TC said:
Thanks for all the replies guys. We're rapidly coming to the conclusion that doing this with VFDs as is, is not going to feed the bulldog all the time, even if we can get it right some of the time. We have it somewhat corrected by adjusting the gear ration in a curve and moving the new gear ratio value in to the calculation every 2.5 Hz, but it is still a bandaid. We are working with integrator to possibly upgrade to Kenetix servos or maybe even an actual mechanical gearing solution.
Click to expand...
Changing to Kinetix would likely work, but it means changing the motors too. You can do this with the motors and encoders you have by changing to 755s. It's the 525s that are the issue here. They don't have what you need.
 
jraef said:
Changing to Kinetix would likely work, but it means changing the motors too. You can do this with the motors and encoders you have by changing to 755s. It's the 525s that are the issue here. They don't have what you need.
Click to expand...

+1

I just recently commissioned a project that was using 755's with encoders and setup in a master/slave configuration like Lord Farquaad suggested above.

One thing new to me in this particular project (I didn't build the original program) was that they had all of the drives setup in Motion Groups, and most of the control was done from the PLC side rather than the drive side. It seemed to work though.
 

Similar Topics

D
Studio 5000 best practice for multiple inputs controlling one output.
Compactlogix controller, program has 28 conveyors that use TON's to start the conveyors. The TT sounds a warning horn during start and the DN...
Replies
10
Views
485
James Mcquade
J
T
The best way to control 48 PID temperature control
I have S7 1512C controler for controlling 48 PID temperature loop, the output is PWM. Please I need the best, most efficient way to write the...
Replies
13
Views
600
tarekrslan
T
R
Best Ethernet/IP HART bridge to use?
I am going to need to use HART multi-drop in order to handle a series of Vega Radar units. There are a lot of options and I'm wondering what...
Replies
3
Views
253
joseph_e2
J
B
Best Practice - What controls recommended for a 2-position turntable
Out of interest, I'd like some thoughts on what would be considered best practice with regards to a 2-position turntable control scheme (see...
2
Replies
17
Views
1,131
Peter Nachtwey
Peter Nachtwey
E
The best Access Point to work in industry
Hello colleagues Anybody knows, what could be the best or a better option to bring with you when you are in service at factory to connect via...
Replies
1
Views
261
Cheeseface
C
Back
Top Bottom