Elevator Rail Stop Analysis

[a7x1984]

Have a bit of an unusual little project to analyze elevator rail stops, which are used as a barrier for the elevator techs to test the pressure in their hydraulic system.

Attached is my analysis attempt. It is the impact analysis only, not the analysis of the connection strength. Any critique is welcome, as I have not performed a calc like this before - well maybe since dynamics course 6 years ago.

I do have one question. I assume that St. Venant's principle is very important to note here; therefore, the first through-bolt will absorb a majority of the loading, in my opinion. To make life easier, I propose to transfer 3/4 of the impact force to the first bolt and conservatively 1/2 to the 2nd bolt. Thoughts?



In Russia building design you!

 

[wannabeSE]

Have you looked at ASME A17.1, Safety Code for Elevators and Escalators to see if it specifies a design force for the stops. I just skimmed the calculations and the 480 kip force needed to stop a 3 kip elevator seems high.

 

[a7x1984]

I haven't looked at that publication. I agree, something is suspicious about that force value. We were told this was a temporary stop for testing equipment. Would that be applicable to ASME A17.1? I'll have to take a peak at it if I can find it.

In Russia building design you!

 

[Archie264]

I'm no expert on this but to get a stopping force you have to resolve kinetic energy over the distance by which it dissipates. Which, it appears, you did, by using the stiffness of the rail plates. Therefore it seems to me that your force calculation is likely to be correct. But that's an exceedingly small distance over which to absorb the energy. Would something such as heavy springs and/or heavy duty rubber bumpers be able to provide you a longer stopping distance and thereby a much small force to be carried by the bolts?

 

[dhengr]

A7x1984:
You have to know much more about what the elevator people are trying to do with this test. Why don’t they have a std. detail for this, they know how their equipment works and is designed. What pressure are they trying to test to? What size ram or rams push this elevator? This should give you a force, less the weight of the cab. And, your design force should be some reasonable multiple of this, 2, 3, maybe 4, but not 78 or whatever. Certainly, it can’t be more than the strength of the tee rails as columns or the strength of their attachment to the bldg. or you’ll be fixin that too. I would think that 2.083 ft/sec is mighty fast for a hydraulic elevator, that’s 4-5 sec., floor to floor.

Who suggested you use 1/2"x2" bars and 3/8" all thread for the bolting? What is the shear and bearing cap’y. of the 3/8" all thread in this application? Do they really expect an instantaneous stop, in zero distance? Therein lies part of the problem with getting such high forces. These just don’t make any sense, you will rip the elevator cab and its guide hardware apart, you will literally push the tee rail through the roof, you certainly will shear the 3/8" all thread. And, this assumes they/you have enough hydraulic pump volume and pressure capacity to make things happen that quickly. Finally, normal tolerances in drilling the tee rails and the stop bars almost guarantees that you won’t load all four stop bars or all four bolts at once, and you won’t know which bolt is going to go first, that’s a tolerance problem. You will more than likely just cleave one bolt at a time in quick succession.

This type of problem isn’t my daily cup of tea either, so I’ll have to think a bit about this problem and your calcs. so far. I think your calcs. essentially represent a perfectly elastic system, and you don’t have that, can’t have that. You should be thinking in terms of impulse and momentum, not kinetic and elastic energy. In fact, the stopping force (impact force) is actually a function of the time to stop, and distance traveled in stopping. Remember the example of the car hitting a bridge abutment, and the damage done and impact forces, vs. the car at the same speed hitting a cable net type barrier, and the lesser forces involved over longer time and distance. And, your design has to manipulate t & d, and absorb some energy, to bring the magnitude of the stopping force down, or the forces go through the roof, with the tee rails. There are a couple people on E-Tips who have a better handle on these types of problems than I do, but until they come along, the above is my two cents worth.

 

[briancpotter]

My thoughts are similar to those above: your calculated force is unreasonably high, your stopping distance is unreasonably low (I work it out to be 0.01 inches based on your numbers), and your assumption of elasticity is probably inaccurate. My understanding is that impact factors are often in the range of 2-3, but a 480 kip force for a 3 kip elevator is an impact factor of 160, which seems...unreasonable. It's not as if the cab is travelling particularly fast (2 ft/sec is 1.36 mph), so a half a million pounds of force doesn't seem right.

If you can't find any other references (and I'd look hard for some), at the very least you can use a more reasonable stopping distance, based on perhaps the bending of the bolt.

Brian C Potter, PE

http://simplesupports.wordpress.com

 

[a7x1984]

dhengr: Why don't they have a standard detail? Who knows. If they did, they wouldn't be asking us to analyze it. The 125 fpm velocity was what they wanted us to analyze it for. They performed a previous test of the stops at a much lower velocity, i.e. that assembly with the 3/8" is already in-place. Who knows, who verified the connection at that lower speed - it may have been a standard detail.

Trust, me I do not agree with that number the least bit, guys. That is why I was concerned about my procedure. A perfectly elastic response doesn't seem right at all. But, as my first 'stab' at the calc, I wanted to be as swift and conservative as possible, as I will tell you what the "real" problem is - it is being asked to complete this task in 2 hours. That is nearly as unreasonable as the 480 kips. So, dhengr, I will tell you that often times, I will ask all of the sorts of questions you describe - they are excellent. However, I often get squashed by the time frame to get it done. I don't like it, but I am not the PE and I am not the project manager determining how much we will bill.

On a side note: I had thought about combining the stiffness of the rail stops and the stiffness of the elevator cab. But, given no information about the cab, the best I could do was assume the same stiffness as the rail stop - by combination of springs in series, that only brought the total force down to three hundred or so kips - still no good.

I am going to check out some elevator-specific references and ASME ones as well. Workin off the clock again...





In Russia building design you!

 

[a7x1984]

I found ASME 17.1 2007. There was no mention of the specific procedures related to rail stops when testing the hydraulic system. I only noticed the requirements for the mechanical devices to be used in service.

In Russia building design you!

 

[chicopee]

I am not quite sure this is the wisest thing to do. I have never seen anything like this before and I have done " a ton" of many elevator inspections in my younger days. As a matter of fact this has all the elements of a dangerous situation.

 

[wannabeSE]

I don't know the elevator code lingo well enough to find things easily. Last time I looked at it was when I was asked to review shop drawings for attaching a rope gripper used for emergency stopping(the forces in the submittal seemed low). The code covers the forces but it is was not called a rope gripper and I had a hard time finding it.

The change in length of bars is probably insignificant when compared to the elongation of the rails where they are attached (area of 4 plates = 4in²; area of 2 15#Ts = 8.81 in²). Without guidance from the manufacture or elevator code, you need to follow the load path.

 

[a7x1984]

chicopee: I do not know how that helps me at all. I am instructed to perform these calculations. Are you suggesting I say that I say "no" to this? EIT's are somewhat at the mercy of their project managers. Besides, even if I were to say, "I don't feel comfortable performing these calculations." You know what is going to happen next? My project manager will do it by themselves. In both situations, our company performs the work, and it is signed and sealed by the PM.

If you have ideas to improve the analysis, I'd love to hear, otherwise, I don't know what else to say to your comment.

wannabeSE: I did see those in the ASME code. That rope gripper device is used in-service, so it makes sense that it would be in there. I spent a half-hour searching through the testing section to look for something that could have been called something different than "rail stops" - no luck. Again, the testing portion of the code doesn't seem to go into a ton of detail into the means and methods of testing - mainly the required outcome of them. Even a simple Google search of 'elevator rail stops' seemed to give me little to work with.

Thanks to everybody else for your comments.

In Russia building design you!

 

[TXStructural]

I don't understand. You describe a situation where the elevator tech wants to run the car up to the stops and then do what? Pressurize the hydraulics and see what happens? Or suddenly stop the moving car, under force from the hydraulics, and at the risk of ripping the rails from their anchorages? I'm not so sure the rails for an elevator are designed to take an upward force of that magnitude. If they simply want to run the car up, stop it against the temporary stops, and adjust the systems so it stalls correctly, I might not worry about the dynamics as much as making sure nothing was going to come loose under too much force by the lifting system.

 

[a7x1984]

TXStructural et al: Here is an excerpt from the email forwarded to me:

Attached are the pictures of the guide rails we discussed with steel plates in place.

Empty car weight is 3090 lbs.
Speed of car 125 fpm

The tee guide rails are pinned at the top 4 brackets, 2 on each rail stack with 3/8” bolts and whiz nuts to stop upward movement of rails.
The plates on the side of the rails are ½” flat plate through bolted to the rail blades as shown on attached drawing by 3/8” all thread.
The purpose of these blocks is to stop upward movement of the elevator to allow pressure to build in the hydraulic unit to activate our valve bypass pressure to insure all components of the hydraulic system withstands 500lbs. of relief pressure.
The test is usually conducted by running the car to the top landing and then run on inspection speed 12 fpm –stops are 3 ¾” above- to hit stops and the pressure builds in the hydraulic system. Inspector wants to make sure running at 125 fpm the blocks are good. We have tested these devices in place as they are at 12 fpm.

Attached is one of the photos he sent (And, yes he sent a photo of a photo on a cell phone). Maybe it will look familiar to somebody. This is everything I had to work with.

In Russia building design you!