Problem with using VFD in hoist application

[M_McFly]

I'm looking for input from any who has experience getting a VFD drive to work correctly in a hoist or winch application.

We have a small hoist rigged up with an ABB ACS380 drive operating an asynchronous motor. The motor has an electromechanical brake on it and is connected to the hoist drum through a gearbox. This gearbox ratio is low so it will readily backdrive under load. Both the drive and brake are controlled via PLC (brake is operated by PLC directly, not through the ABB drive).

Here's the problem: when the motor is started under load (ie - weight is present on the cable), the motor will backdrive a bit as soon as the brake is released. If the load is high enough this backdrive will run away from the motor and it wont be able catch the load. This is a classic problem in hoisting applications, and the solution is generally to start the drive with the brake closed to build up some torque, THEN release the brake to eliminate the backdrive. For the life of me I cant get the motor to start building torque prior to the brake releasing. I can see a consistent 200 ms delay between brake opening and the motor building torque. This happens every time, regardless of brake-opening delay or any other sort of brake/motor timing. The drive absolutely refuses to build torque if the motor speed is exactly zero (ie - brake engaged). I'm running the drive in vector mode and dont have an encoder on it.

Thoughts on what we're doing wrong? Would the presence of an encoder allow us to build torque prior to brake release? This is certainly a solved problem but for some reason we cant seem to employ it.

 

[Illinoyance]

Some elevator companies use a load cell on the brake. Th motor controller applies torque until the torque on the brake is zero, then the brake opens.

 

[MwTech Inc]

i would not have a hoist in my shop that static would not hold a load........

Really..........???

 

[Garwood]

Winches and hoists often use a sprag with the brake. Then the brake only operates one direction and the motor can drive through it.

 

[M_McFly]

i would not have a hoist in my shop that static would not hold a load........

Really..........???

Yeah, neither would we. This is a specialized piece of kit that operates under very specific parameters, so it's not a safety thing.

 

[M_McFly]

Winches and hoists often use a sprag with the brake. Then the brake only operates one direction and the motor can drive through it.

This is a good idea. Thanks.

 

[MwTech Inc]

ok.... you may not be able to share "secret" stuff , but I can see no reason to have a runaway hoist doing anything..
What if the power goes off???? PLC burps???any number of things .

You say it's not a safety thing??.........Unless the load and/or what's below it have no importance if something fails?
Not being a smart butt but why not just use a "normal" hoist and be done?

Of change the gearbox and use the VFD to spin er up?

 

[john.k]

Probably fixed by including a block and more falls of lift wire......ie four falls of wire ,two sheaves in the hook block.

 

[john.k]

I once pulled an ancient hoist apart,and there was a long spiral spring inside a drum that acted as an automatic brake .....drive torque wound the spring up and freed it ,(lift rope) torque expanded the spiral into the drum which was locked to the frame ,providing an automatic brake to hold the load...............my old Coles crane has DC motors driving hoists with solenoid brakes ,and can lift 1 mm ,and hold ....its the most rock steady crane ever .......Ive worked mobiles where the hoists would drop near 150mm (6") as the brake s and hoist wernt able to be sychronized.

 

[M_McFly]

Thanks for all who weighed in on this.

For anyone that comes across this in the future, this particular drive is absolutely capable of building torque prior to brake release (building torque at zero motor speed). Our troubles were traced back to a poor PLC implementation that prohibited us from effectively coordinating brake release and motor energization. After correcting our mistake, we suddenly had the ability to energize and pre-magnetize the motor well before the mechanical brake released. Loads are now controlled immediately upon release, even in the absence of a self-locking gearbox.

The mechanical implementations offered by others would have worked as well. Thanks for everyone's thoughts.

 

[johansen]

I think the issue is that your vfd is in vector mode and is too slow at determining the motor's load, and as such it runs away. this could be a failure of the vfd algorithm to properly deal with a motor running backwards, overpowering it. (its actually operating as a generator, if the system is efficient enough)

second issue may be that the volts per hz at the vfd is too slow at ramping up, or starts from too few volts.

what i would do is take the vfd out of vector mode and program it for 20% boost volts starting at zero hertz. then release the brake. the motor will backdrive very slowly, but as soon as the vfd ramps the frequency up the motor will immediately have torque available, because the motor already has a magnetic field present in it.