VFD vs Servo Lifting Application

[mcis99]

I have a design application involving the vertical positioning of a horizontal pipe. One of the design inputs that I've been given is that the positioning has to be within plus or minus 1/4" accurate and repeatable. There are various vertical positions based on pipe diameter sizes. The different sizes also means that the load will change as well between each type of pipe. We will be using a rack and pinion drive system and traveling at 2 in/s max through a 70:1 gearbox. 1 revolution of the gearbox output shaft will lift the pipe 6.28".

The original plan was to use a servo to do the positioning, but we are exploring the idea of using a vfd and ac motor (1800 RPM) instead due to cost reasons. I'm looking at a closed loop vector control VFD and 2500 PPR encoder. We will ramp down to a slower speed of 1 in/s approaching the target position before stopping. The plc will read the encoder position on the drive via fieldbus. Upon stopping the motor when the target position is encountered, an output from the drive will actuate the motor brake.

I think this is viable to do, and based on my calculations using the above parameters I think it's definitely feasible. Personally, I would rather pony up the money for a servo system. For only a few thousand more we can have a packaged servo system that's a lot more flexible and precise in my opinion. Plus I'm wary of the plus/minus 1/4" tolerance spec and feel that we may need to be more accurate. I'm also skeptical of what type of delay/problems the fieldbus communication may cause in the grand scheme of things.

Has anyone had any experience trying to be as accurate as 1/4" using the above method? Your thoughts and opinions will be greatly appreciated.

 

[DickDV]

That is certainly acheivable with a sensorless vector drive but I would suggest one that has output relays that do level detection and time delays.

It's not quite as simple as stopping and putting on the brake or vice versa.

It takes time, for example, to magnetize the motor and, if you release the brake at the same time you start the drive, the load will drop a bit before the motor develops enough torque to carry the load. Setting a relay to detect torque level and then release the brake works much smoother.

 

[mcis99]

Yes, you're exactly right. That's the route that I plan on taking, except I felt that I may need to close the loop using an encoder instead of the using sensorless vector. Maybe not though...

I had a previous application using a yaskawa drive on a vertical hoisting application using the level detection relays that you speak of. I didn't have to run an encoder back to the drive in that case. I set the level detection output relay to a low frequency, and upon ramping down the output relay forced the brake. I had no issues with load dropping.

My only concern is that some of the mechanical design team feel that I need to be within thirty thousandths as opposed to the 1/4". In that case I want to go with a servo. Like I said previously, I'm extremely wary of the plus or minus 1/4" that others have cited. A certain level of "tolerance" has been built into the tooling (which we insert into the pipe after it is lifted) to allow for this greater variance supposedly.

 

[tygerdawg]

Some years ago I had a similar question and asked the SEW-Eurodrive guy. At the time he stated that "an AC motor coupled with our drive and an encoder" give me equivalent performance to a servo rig.

I was skeptical, but at least it's worth a phone call to see what their Applications Engineers say. Or Nord, or Leuze, or Falk, or Reliance, or ...

Your situation seems to be completely dependent upon your speeds and loads, and should be evaluated very closely.

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[Skogsgurra]

There was a similar case in thread237-220699
The OP needed a drive for a hoist application and found that the Invertek Optidrive could be used without any encoder and that it did hold the load.

The OP actually came back with a very positive answer: "Gunnar, Thanks for the recommendation of the OptiDrive Plus 3G. We picked one up. We also bought an encoder module for the ABB ACS350 drive we are currently using. Wow! So far, there is absolutely no comparison in performance. The ODP can start from a dead stop and hold the hoist there at our maximum weight. That's very impressive. We're talking with ABB's engineers now. It's hard to belive that a sensorless vector drive can beat out a closed loop vector drive in our hoist application. I really wanted to go with the ABB, but everyone who's seen the two picks th ODP every time."

But don't take my word for it. Read the whole thread.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...