Fail-safe VFD drive in hoist application (Power black-out) 2

[ken777]

Hi,
I'm a mechanical guy but would appreciate some basic explanation on how to achieve what our team needs to design so I can follow up with a design consultant / vendor who will manage the VFD part of the deal.
In a crane hoist application with electric motors driving the hoist drum via VFD (with PLC control over VFD), is it feasible to include a fail-safe stop for a case where there is a power black-out? Design case would be a load lowering condition where the load is being driven by gravity and needs to be arrested upon mains power shut-down.
We typically do it hydraulically so I just need to know there is a VFD equivalent. Typical stopping time would expected to be ~500ms ramp time to (almost) full stop condition. The parking (static) brake cannot arrest a moving, runaway load condition.
I was told the motor being driven by the falling load, would act as a generator and power the DC bus on the VFD. This could possibly be used to slow down the load to a stop. I would need a bit more clarity on that concept and if it can be applied to what I have described.
Thanks.

 

[itsmoked]

A drive remains powered as long as it's being driven by the load.

However if it was left to stop the load as soon as the load was stopped there would be no more power and it would trip off-line.

I would expect the static brake to be absolutely capable of stopping any and all expected loads. Perhaps only once but at least that.

Drives fail. Their power output stages can fail catastrophically and do. That would guarantee an out-of-control dropped load if the static brake is anemic.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[BrianPetersen]

Electric motor with a spring-applied mechanical brake. No power ... spring applies the brake. Nothing to do with the VFD although the VFD may have an output to control the coil for said brake.

Although those mechanical brakes are meant to be "holding" brakes and aren't meant for repeated "braking", for the once in a blue moon situation where the power fails while it is in motion, the ones that I've seen will serve the purpose.

 

[GroovyGuy]

The 'holding' brake must be able to stop the winch during a power failure or a drive failure.
I experienced that once when a dc luffing hoist was raising a shiploader and the power-system to the facility failed during the lift. What made this scary was that this lift was part of a Coast Guard acceptance test and not only were we lifting the shiploader, but a 50T dead weight as well, The 50T weight consisted of a huge block of concrete. The shiploader free-wheeled about 20ft, rather quickly I might add, before the disc-brake caught hold and stopped the descent. It finally came to a stop about 5 ft above the deck of the barge [that ferried the weight out to the shiploader]. I was on the shiploader at the time this happened, I can tell you that it was brown-shorts time for a lot of us.

ps the disc-brake was located on the motor side of the luffing gear-box and was of the electric energize-to-release type.

"I have not failed. I've just found 10,000 ways that won't work." Thomas Alva Edison (1847-1931)

 

[ken777]

Hi Guys,
Thanks for the comments. Just to be clear, the governing specifications and rules require that the static brake be sized to 150% torque of the drive unit (motor) so they are quite sturdy. However, the same rule requires that the "dynamic brake" bring the load to a stop and the static brake then act as "parking brake".
The same spec / rules require that we cut off the supply power to "simulate" a black-out condition during certification tests and prove that the load is stopped. This works perfectly fine on our hydraulic systems designs but now we have to come up with an electric drive hoist design.
Having had a 88MT load come down on me in free-fall in a hurry while load testing a project, I found the hard way that the static brake did not stop it on time. I am trying to make sure my vendor is giving me a good solution to this problem (I don't want to run from a free-falling load again and deal with the fall-out). Being an offshore project, the requirements are a bit redundant and have to be proven on paper and on site while certifying.

"itsmoked" mentioned the drive would be powered so long as the load was moving and that would work if the load can be stopped within 500 ms and then the static brake would take over. Will research further.

 

[Parchie]

Ken,
I have seen a 5-ton, ship unloader grab bucket freefall unto a coal barge belly when the dynamic braking system failed and only stopped when the mechanical drum brakes acted to stop it! The whole machinery room smelled of burnt disc! These things happen. When I first worked in that company, I kept wondering why that hoist motor is installed with so big a brake drum and restraining coil springs as big as my big toe in diameter!

 

[waross]

How about an air pressure released disk brake?
What size is the motor?
In the smaller sizes there is no problem finding a suitable brake.
I remember one very large hoist drum. It was powered by four 1300 HP DC motors. Each motor had a very large drum brake mounted on the non drive end.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ken777]

Will look at additional "mechanical brake" arrangements, as suggested.
Each hoist will have 2x Electric motors, each rated at 1500kW. Motor will have a static brake at NDE shaft end and the drive mechanism needs to have a dynamic brake as well, in addition to static brake.

 

[jraef]

As a general rule, at 1500kW there is no practical way to keep the motors energized to facilitate dynamic braking. But you indicated 500msec as the maximum braking time? That alone may be unrealistic, but if it is possible to stop the motor in 500ms, then it's possible to hold up the motor power for 500ms. How dynamic braking works is to keep the motor energized so that flux remains in the windings at a frequency lower than the relative rotor speed. The now over hauling load represents kinetic energy, which then makes the motor an induction generator. When the power flows back into the VFD, it raises the DC bus, then that kinetic energy, as voltage potential, is shunted off into a resistor bank and converted to thermal energy. But because the first step is for the VFD to create flux in the windings, when line power is lost, the VFD dies. No braking. On small drives you can add a battery bank to the DC bus to keep it capable of flux creation, but I'm pretty sure 1500kW is over the top for that, at least practically.

Now let's address the 500ms. The quick rule on electronic braking is that cannot possibly brake a motor to a stop electronically any faster than the motor can accelerate it to full speed when across-the-line. So if your motor can accelerate the fully loaded hoist motor in 500ms or less, then in theory you can electronically stop it in that time. In other words your negative torque capability of that motor can NEVER exceed the positive torque capability. So start there. If the load takes 2 seconds to reach full speed if started across the line, it's pointless to discuss electronic braking any further without relaxing the stop time requirement. At that point, the only viable option is mechanical braking.

Generally, a hoist mechanical brake is sized to handle this, but not used for this in anything other than an emergency situation, i.e. total power loss.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[LionelHutz]

Regardless of what is specified, do you not have a general overall duty to make the hoist safe?

If you are going to use a VFD, then you have to recognize that the VFD could fail or trip and quit powering the motor at any time. this means the hoist MUST have a mechanical brake that can ALWAYS stop the load in whatever time is required. This means your current parking brake setup is not suitable. Having a properly capable brake is the only way to meet the general expected requirement that the hoist will always fail safely.

Technically, it may be possible to make the VFD decelerate and stop the motor during a power outage, but there is no point chasing that possibility since it will not make the hoist capable of always failing safely.

 

[ken777]

Thank you all for the very relevant insights. Will note these points down for the discussion with the vendor design team that will deal with the VFD drives.
As mentioned, we have these issues already worked out and practically tested / certified on our hydraulic drive hoists (in-house design). Will incorporate the mechanical concepts to address all the risks / concerns raised.
There will also be a 3rd party design review / approval process, so the overall design will have a comprehensive review for safety concerns raised.

 

[GroovyGuy]

Hi Ken,
VFDs have been used extensively in hoisting applications, and I see no reason why it cannot be made to work for your application. I suspect that what you are looking for is a 4-quadrant vector drive with tach/position feedback. This type of drive can smoothly de accelerate a load, as well a hold a load at full-torque at 0 rpm. I have done this once before on small indexing hoist for a drywall stacker. It worked really well.
GG


"I have not failed. I've just found 10,000 ways that won't work." Thomas Alva Edison (1847-1931)

 

[waross]

A comment on a motor mounted brake for fail safe operation.
Some years ago, there was a failure of a construction elevator that resulted in two fatalities and serious injury to two more workers.
The elevator consisted of a tower up the side of a high rise office tower. (Scotia Bank Tower in Toronto Canada.)
A cage ran up and down the side of the tower.
This was used to transport both men and material to the upper floors.
There was a rack gear extending the height of the tower.
Drive was by way of a pinion gear meshing with the rack gear.
The cage was counterweighted to approximately 1/2 the working load.
When the drive failed there were only 4 men in the cage and no great weight of material.
The weight of the counter weight was greater than the weight of the cage and contents.
The counter weight descended accelerating the cage upwards.
The cage hit the top and stopped about 70 stories up.
The workers continued to rise until stopped and injured or killed by impact with the top of the cage.
The drive failed when the pinion shaft broke.
The elevator cage was equipped with safety brakes designed to grip the tower and stop the cage in the event of cable failure.
The safety brakes were normally held off by the tension of the supporting cables.
As the failure was a broken pinion shaft and not broken cables, the safety brakes did not engage.
The point is that a motor shaft mounted brake may not provide fail safe operation in all failure modes.



Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jraef]

In some types of hoists, they use a tapered motor. The rotor is tapered and fits into a similarly tapered stator. When energized, the motor magnetic force the rotor out axially against a spring force, then it rotates when free. If power is ever lost, the magnetic force is lost and the tapered rotor jams back into the stator, stopping it. Do a search for a "Demag" motor if you want to investigate that option.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[controlsdude]

I thought this link on google was worth a thousand words. I did not realize that everyone does things differently. Some people have good imaginations


I just put in google search = crane hoist safety brake , select images

 

[ken777]

Just to recap, we were discussing if the VFD DC Bus would allow to slow down / stop the hoist load via the available regenerative current in a power shutdown scenario, before the static brake could tackle the runaway load.
The static mechanical brake (crane hoist safety brake) is ancillary to the dynamic brake (VFD braking) and is already available in the system.
Thanks for all the interesting posts. I have noted down all relevant points for follow up investigation.