load sharing

JeffKiper

Lifetime Supporting Member + Moderator
Join Date
Jun 2006
Location
Indiana
Posts
2,447
I have been asked to quote a few jobs lately that have conveyors with 2~4 motors running the same section. These are explosion proof motors and the size is the problem for where they are physically located. We can only put a 10 hp motors in there. It is just tight.
The drives change from customer to customer. I try to use a baseline drive like a AB PF70 ,ABB ACS550, Baldor VS1, so just about anybody's drives. Siemens will be soon I am sure.

My question is how do you handle load sharing? I was thinking about having all of the motors running in Sensorless vector and speed regulating the first one. The rest of them would run in torque regulation mode. The torque signal would be the torque feedback of the main motor that is running in speed mode.

A few questions with this method.
1. Am I stupid for thinking of such a simple approach?
2. Lag on startup. From the time the main drive gets a speed ref and replies with the torque feedback it will be in overload condition for a short time. Right? Or is that such a short time that it will not matter.
3. The speed regulation is not tight at all. A V/H drive has done this with the correct room so we can get the right sized motor in there. Would I have a huge fluctuation in the speed. I don't think so because of the sensorless vector mode.

Any pointers?
 
A few answers or just observations
1. Stupid?? Nope not at all. You are not the first to parallel prime movers. Now you may be a man willing to take on a challenge but not stupid. Counter idea - move the motor (just one 40HP) to a separate room and drive conveyer with a drive shaft ??

Motors have to be very carefully matched, RPM torque etc etc.
May also be a very good application for hydraulic drive. Control pump speed or swashplate angle to get desired conveyer speed. Explosion proof is not cheap for sure and not an off shelf item (??)

Do not forget overload protection for each motor.

I would guess with properly sized VFD (chosen not by HP alone but as Dick DV says amperage lag should not be a problem for the VFD. I assume you will use only one encoder.
 
How are they mechanically coupled together? Is it just the belt? Are they chained together? What speed range do you have to run them in?

I almost think you would be better not to run them in vector mode. Provided all are coupled identically (same ratios, ramps, speeds, etc)V/Hz would possibly allow each individual motor's slip to vary such that it balances out the load. I could see the drives in vector mode fighting each other where in V/Hz the drives won't be trying to regulate anything. If the speed regulation is not required to be tight then I don't see an advantage here in using vector mode.
 
Thinking a bit more on this. The VFD regardless of V/Hz or sensorless will be sending out a "specific" frequency to all four motors. The "synchronous frequency" and "synchronous speed" are all the same. However I think problems will come about when we consider slip. Assuming slip at full motor load is 5% and if all perfectly matched then they will all be running at same speed including slip and all should be good. However if there is a 10% difference in slip between two motors one could be at 5% and other at 4.5 or 5.5 % slip. The one with less slip will be putting out more work and one with more is slacking on the job ie producing less work (I think I got that right).

I think all things considered this is a good application for hydraulic drive. One nice thing about hydraulic is if the drive motor stalls it does not burn out.

Dan Bentler
 
Thinking a bit more on this. The VFD regardless of V/Hz or sensorless will be sending out a "specific" frequency to all four motors. The "synchronous frequency" and "synchronous speed" are all the same. However I think problems will come about when we consider slip. Assuming slip at full motor load is 5% and if all perfectly matched then they will all be running at same speed including slip and all should be good. However if there is a 10% difference in slip between two motors one could be at 5% and other at 4.5 or 5.5 % slip. The one with less slip will be putting out more work and one with more is slacking on the job ie producing less work (I think I got that right).

I think all things considered this is a good application for hydraulic drive. One nice thing about hydraulic is if the drive motor stalls it does not burn out.

Dan Bentler

Hi Dan, Are you looking at this from the standpoint that all four motors are being controlled by one single drive? I interpretted it as it was four separate drives and four separate motors. I think a big key here is how the motors are coupled mechanically. This would affect how much any one motor might get overloaded and where I was aiming with some of my original questions. If everything is mechanically tied in fairly rigid fashion I can't see you having a difference in slip any where near 0.5 to 1.0%.

Sounds like a good question for DickDV. Hopefully the title won't throw him off from reading it.
 
Jeff, We use serval motors on converors all the time. These however are roller type not belts. We use 1 drive sized for all motors with indvidual overload and disconnect for each motor. I have one app I am working on one now where I will be attempting to match the speed of about 12 ac motors with 6 hydrolic ones on the same conveyor. I will be using encoder feedback from the hydrolic to control the ac motors with an encoder on the ac set to check for error.
 
Jeff,

I haven't done an application like this, but I wonder if you would see stability issues with three drives trying to follow one drive's torque f/b - would it settle or oscillate?

Also, be careful running drives in sensorless vector mode for torque control. It was explained to me by an AB guy that sensorless vector mode does not provide true torque control, rather it is current control. I can say from experience that the two are not the same - especially at zero and near zero speed.

I think the idea of running all in volts/hertz may be worth considering.

Of course on some of the higher end drives, you can effectively do an electronic lineshaft, but for dumb conveyors it seems overkill to not use a simple drive.
 
Four motors on one drive I have done no problem but not for an EX environment. This works best if the actual load is shared equally between the motors so they have the same torque requirements. I have also done load sharing between vector drives coupled to the same load but that seems overkill for a conveyor.

Would 4 motors on one drive pass muster in terms of the explosion proof environment? One motor might stall and very quickly overheat.

I've recently done a refit of a Polytype coating machine in an explosive environment and had to use dual channel thermistor relays with one channel linked to the drive and the other wired to the emergency stop system.

Maybe DickDV is on vacation today.

Nick

Edit: One more thing. Ex motors with encoders fitted are very rare items!
 
This application is a long geared conveyor carring barrels of liquor. Multiple drives. I would use drive to drive hardwire to eliminate the scan delay in passing it through a PLC.
If we could cram a larger motor in there we would physically there isn't room. Concrete tunnel conveyor has increased in width to the point there is no more walkway next to it. You power down the system and walk on it.
 
Multiple Vector Drives:

Drive one would be the master and do all the ramping etc. The actual speed from drive one is passed to all other drives as a speed setpoint where 5% is added. The torque feedback from drive one becomes the torque limit for the other drives.

The other way to do it is to put A PID loop in each slave drive with the setpoint as the torque feedback from the master and the feedback as the torque feedback of the slave drive.

Nick
 

Similar Topics

For a stone crusher, which is driven by 2 identical motors (each 160kw) has VFD (SCHNEIDER ATV930C22N4C) for 1st motor and 2nd motor starts DOL...
Replies
27
Views
7,786
I have two motor that are rigid coupled and drive a belt. What is the best way to share the load between both. A need a controlable speed for the...
Replies
5
Views
3,152
I have several long conveyor systems (6" chain) powered by multiple 7.5 HP DC motors. DC drives are AB 1370 Series B and are very old and no...
Replies
5
Views
7,866
I am currently commissioning a system of DC drives that I would like to have load sharing capabilities. I am using a L63 ControlLogix controller...
Replies
5
Views
5,105
Hi everyone, I got a Parker HMI, CTC PA2. I tried to download program to it but it froze in the process and could not finalize. Now, when I try to...
Replies
0
Views
25
Back
Top Bottom