Max speed for a gantry crane before it toples

[sacem1]

I am posting in this thread as I can not see other more suited, We have this problem:
A gantry crane 10 m high, 40 m wide and with a load carry capacity of 30 MT sits on railway trucks wich have 4 wheels resting on a railway ( 2 at each end + 2 midway)the distance between end wheels is 6.5 m.

Right now the gantry crane moves along the 90 m long railway propelled by 4 motors one at each end of each support truck at each side, the motors are started by variable speed controls wich acelerate and decelerate in 5 seconds from stop to max speed (only one speed) and from max speed to standing stop in another 5 sec.

The actual speed of the gantry crane is 15 m/min and I reckon it is a little on the slow side (high on the safe side) but my customer wants to raise the speed to the 60 m/min level and I do not know if at that speed a sudden stop (caused by any foreign object on the runways, remember they are rails at floor level) can make the whole structure to stop suddenly, and because it carries most of its weight at 10 m of height tople down and crash with heavy material and possibly human losses.
The crane's own weight is 55 MT or the top bridge, 14 MT in the legs and 12 MT on the rail truck total 81 MT plus almost 3MT for the hoist itself sitting on the top rails and the live load of 30 MT that in the most unfavourable condition would be raised to the highest level (8.5m).
At what speed would we expect that with a sudden stop the gantry crane would tople and fall down, that would have to be cut to a half at least to find the safe max speed that can be achieved.

My fear of something blocking the wheels is not so unreasonable if you take into account that this gantry crane operates in a stockyard full with rebars, angles, bars, etc wich are piled and sorted all the time and some of the lift trucks which also share the same working grounds can leave something blocking the railways and causing a disaster.
Thanks for any help you can give.
Regards
SACEM

 

[MikeHalloran]

The crane trucks should be equipped with 'rail sweeps'; plates welded to the truck ends that are shaped to clear the rail by just a mm or two, and extend down into whatever grooves are provided in the floor, partly to reduce the risk of having the incident you have conjectured.



Mike Halloran
Pembroke Pines, FL, USA

 

[GregLocock]

If it comes to a dead stop instantaneously then it will topple if the gain in potential energy as it rotates around the front wheel exceeds the kinetic energy of the thing prior to the stop. Working that out is high school maths. This approach is robust in that any real system will see some friction etc which will tend to remove some of the KE.

Even at speeds below that, in the problem as described, it will still lift off the rails, which is probably not acceptable, or realistic. The problem with that analysis. is that the stop won't really be instantaneous, and there are compliance effects that tend to keep the wheels on the rails.



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[MikeHalloran]

The problem is a little messier than 2D kinematics would suggest. If only one of the trucks gets jammed, the other is still free to keep going, twisting the jammed truck's tower, with an assist from the inertia of the beam, the hoist, and the load.

Crane structures are scary flexible in torsion.




Mike Halloran
Pembroke Pines, FL, USA

 

[GregLocock]

Mike's complication is absolutely relevant in sme circumsatnces... so we have two alternatives - either carry on from 'reasonable' assumptions, or else you need to find someone who does this stuff for real.

I would add that my worst case scenario is still robust in terms of the problem as stated- both wheels tripped at once is worst for tip-over, if the CG is approximately midway between the rails.




Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[evelrod]

The problems suggested and the possible results are likely real enough, at least in theory. However, in the real world of gantry cranes and ship loaders (L.A. and Long Beach harbors), some have "cow catchers" but most do not. Some have sensors that detect objects, but most do not. Some have automated drives, but in those cases, workmen are present (at least by the union and harbor safety rules). 15m/min. is a reasonably accurate speed, at least as I have seen it. In the several years that I was around these type of cranes, I have only seen one failure as described. That was in Orange County in the late 80's when wind blew a crane to the end of the tracks and it STOPPED...like NOW! It did not tip over or crash, but really bent things up. It took us weeks to shore it up and repair it. This was in a plant that made vaults for underground projects and vaults for caskets. Not the biggest of cranes, but I think it addressed the concerns expressed in this thread.

The last of the ship loader gantry cranes were from Brazil and installed in the L.A. Harbor in 1997. I did not work on them, but one of my crew did. He said they had all the 'bells and whistles' thus complying with all the new safety rules.

Rod

 

[sacem1]

I want to thank all your replys in such short time, well my customer does want to double the speed but has agreed in accepting the increase in costs that any requiered device that increases the safety of the gantry crane.

The crane IS now in operation at 15 m/sec and it does have the sweeps in front and back of all the wheels like MikeHalloran suggested. Greg has a point in that both wheels need to get blocked, really not the wheels but the rail trucks themselves, in order to tip over the gantry crane, but that would be very difficult to happen, unless you hit full advance the end stops. Probably a derailing of the crane would be most likely because of the flexingwhich is big in this type of cranes like Mike said.

Evelrod makes a good point in that the cranes are not equipped in a similar way and that they usually are that slow but having to deal with the speed increase this is what I have thought:

At actual speed the deceleration has to go from top speed to zero in less than 0.052 seconds to tip over the crane.

At double speed the time requiered is doubled to 0.104 sec, thats with out considering flexing or derailing so what I have to achive is a 0.5 sec? delay after initial hit in order to avoid the tipping over.

One idea is to fit some crushable structure at both railtruck ends like the car fronts now that collapse in a secuential way to protect the car cabin, the other is to fit a proximity swith that detects any obstacle in front of the crane runway and cuts the advance, the third to install in the motor variator control a braking resistance so that higher deceleration rates can be achieved once the sensor trigger a blocked-way alarm.

Besides the above a support located at the other end of each rail truck is going to be welded in the eventual case that everything above fails and the operator drives the gantry crane to its end stops and crashes, then as the other side tries to raise itself to tip over the welded stops will hold and stop the crane from tipping further.

If any one has a better idea I am open to them completly

Thanks again

Regards
SACEM

 

[GregLocock]

Just one additional point. In the more likely case of a single wheel jamming the tendency is for the crane to pirouette around that wheel. In order to derail itself it merely has to lift all wheels bar the front pair clear of the track by the height of their flanges - a much smaller task than physically toppling the crane.

All of your suggestions sound sensible - how about a spring /damper system to act as a buffer instead of the crushable structure? There are advantages in both approaches.








Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[GregLocock]

This may amuse you. The physics are accurate, within the constraints of my assumptions. This is a speed of 2 m/s or 120 m/min, as you can see it might derail, but it does not look likely to tip.

The critical speed seems to be more than 3.5 m/s but less than 4.

Obviously there could be some nasty interactions between winch lateral location and the height of the load.




Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[evelrod]

Sorry, I must disagree about the 'one wheel stopping' scenario. I have seen several loaders with one wheel jammed and still operating...in fact, the operator did not even notice (remote coke loader, Long Beach coke barn)!

Collapsible structures? Many rail stops are of the 'energy absorbing' type. Use one of the already proven technologies...in conjunction with---

Proximity switch? I would prefer the more modern tech of motion/radar detectors. I've seen them used on conveyors and they work well and are not affected by dirt or moisture.

Rod

 

[GregLocock]

Rats, I meant one wheel absolutely blocked, not merely jammed. Anyway, it turns out that for my model it is much easier to trip the crane if you block both wheels, blocking one wheel can derail it, but the extra degrees of freedom reduce the impact forces.

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[sacem1]

Thanks again for your responses.

I agree with Gregg that if only one wheel locks most probably it would derail instead of tipping over, also I seems logical that if one wheel is locked from turning most likely what will happen is that the wheel blocked will start sliding and not tip over the crane.

The radar sensor are not availiable around here and importing them and not having someone to back them with service is not a good way to go.

My worst case scenario is not a blocked wheel but a blocked rail truck, that is the over the rail scrapers will not allow any object big enough to block the wheel to get near it, the problem is if lets say a 24" high WF beam is pushed back as to block the rail in a 3000 ton pile of them, it has already happened with small angles and they were bent, but picture a situation where the whole car crashes into a protruding beam and there starts all the problem-

The spring is a reasonable option, probably instead of the crushable structure.

Thanks again

Regards
SACEM

 

[sacem1]

Sorry Greg forgot to comment on the speed limits:

My number are a ñittle more conservatives and I have agreed with the crane user to raise the speed from 0.25 m/sec to 0.75 m/sec and addind the safety devices already said.

Regards

SACEM

 

[GregLocock]

How do you control the stops at the ends at the moment - are there buffers?


Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[sacem1]

Right now as usual in these installations there is a solid steel vertical stop welded and grounded to a solid deep foundatio in concrete.

SACEM