Elevator Design Safety Factors

[Perrins]

I’m currently designing and building a powered elevator / lift for my cottage. In recent years, the 130-stair climb up from the lake is getting challenging for my ageing parents. The lift will consist of a track at about a 32 deg. angle with a winch at the top to pull a car up and down. The car will have an emergency braking system that will automatically trip it if the cable tension drops below the weight of the empty car. The winch will be a simple cable and drum assembly mounted on self-aligning pillow blocks. The motor / gearbox will hang off the drum shaft with a torque arm to hold it in place. It will be powered by a PLC controlled 3 ph, 220 Vac, variable frequency drive with dynamic braking. Much of the design is being done on solid modelling s/w and FEA will be done on key elements. …OK I admit that designing stuff is also my hobby. This is actually my summer project at the cottage this year.

I’m not actually required to build it to elevator code, since it is a cottage / private application, however I plan to use common elevator design practices for guidance. Unfortunately I don’t have a copy of the relevant code for the area and would rather not buy one if I don’t have to. Does anyone have the commonly used safety factors for the following?

1. The gearbox on the winch drum. Is the SF on the working load, the maximum load (200% for the type I’m looking at) or on the ultimate load (I would need to contact the mfgr for the ultimate load or the internal SF’s)

2. The pillow block bearings on the winch drum

3. I believe the SF for the cable should be 10:1 for the working load limit. My cable load is only about 1000 lbf. I may even go for a 15:1 SF for the cable, since cable is relatively cheep.

4. I’ve seen values of 1/920th of span for the allowable track deflection. Does that make sense?


Thank you in advance for the help.

Rob

 

[gentcho]

Hi Perrins,
1. For a portable ski tow SF for cable is 4 - 4.5
2. SF for transmission - >1.5* Mdyn.max
3. SF for drum axe, pillow block bearings, fasteners and others >2.5*Max.dynamic Load
4.???
Success,
G.

 

[atlarson]

Rob,

I'm not familiar with specific industry standards for the design of powered man lifts, but if I were building one for personal use here is the approach I would take:

1) Determine the normal operating load - posted capacity, plus weight of the cart and cable.

2) Calculate the design load in the cable by dividing by sine 32 degrees and multiplying by a factor of safety to account for rolling resistance, intentional or unintentional overloading, miscelaneous friction, etc. A F.S. between 3 and 5 would probably be adequate.

3) Size your gearbox, bearings and motor to handle the design load based on the manufacturers recomendations. I doubt you will have much sucess in finding out the ultimate structural design safety factors for these items.

4) Once the motor , gearbox and drum size have been selected, calculate the tension in the cable produced by the stall torque of the motor. Size the cable, keyways, shafts, torque arm, and cart structure to withstand this load. Be sure to apply another safety factor , perhaps 2:1 on yield strength and 3:1 on ultimate strength.

5) Perform a proof test at `200% or more of posted capacity before loading your parents in the cart.

Regards,
Andy

 

[IRstuff]

not at expert, but am certainly paranoid...

I'd look into a track or roller system that does not allow the cars to jump track, ever, short of having an earthquake. The braking system would only work if the car stayed on the track. TTFN

 

[Perrins]

Alterson, thx.
as per item 1, done
Item 2, done
Item 3 Agreed, I would rather base the SF on the working or allowable loading of the components. I know that for elevator applications the SF tends to be considerably higher, since it is for people. I plan to use a SF of 10 to 15 for the cable, since it is cheap and at a higher risk of wear.
Item 4, good point. I haven’t finished sizing the motor. It will be either a 7.5 Hp or 10 Hp, depending on how fast I run the lift. This is partly cost driven.
Item 5, yes. I plan to test it by filling 40 Gal drums with water. The car will have a GVW of 2000 lbs. and the car will car about 1000 lbs. or four people.


Irstuff, thx

That has been accounted for. Some of the design is based on a roller coaster concept, so the car can’t actually come off the track

I’m also working through a failure mode analysis to outline the risks / mitigation.

 

[johnhanging]

Check out ANSI A10.4 and A10.8. In suspended scaffolding design the factor of safety is based on the hoist's rated capacity, not the design load. The hoist's rated capacity must be less than 1/3 of the stall load or cut-off load of the hoist. On a multi-layer drum hoist the load the hoist can apply to the wire rope for a given motor torque will increase as the number of layer of rope decreases. With the cutoff set at 3:1, a factor of safety of 4:1 will provide a margin of one. The rope safety factor is anywhere from 6 to 1 to 10 to 1 dependent on application.

 

[Speedy]

I'd consider the fail-safe braking system used on elevators if I were you.... also run the cable within the under side of the track where possible, wire cables can be lethal when they snap. Speedy

"Tell a man there are 300 billion stars in the universe and he'll believe you. Tell him a bench has wet paint on it and he'll have to touch to be sure."

 

[harrisj]

If your winch drum is multi-layer, there is a change in velocity ratio as johnhanging says. You should also consider that wire rope can jump as it its tension forces it into gaps in the second and subsequent layers. This isn't particularly dangerous (although it's not good for the rope) but it's pretty alarming for your pasengers when being hauled up a long slope and the car lurches back an inch or two. Might even trip your emergency braking system.

You can get round this by using a capstan winch for hauling, and having a secondary powered drum for the (low tension) take up drum. Or you can lay the rope precisely using a reverser screw to feed the wire onto the drum one layer at a time with no gaps between the rope.

Have you thought of having a counterbalance weight travelling in opposite direction to the car? This means more track (between the rails of the existing track?) but the rope handling problem is easier. Also less power and more safety.

Who makes the system stop and go? Passengers? If so, how do you control your motor? Please, not radio control for safety critical stuff. Some kid may be playing with his RC boat in the lake .......

Finally, hauling up slopes is sometimes easier than controlled descent. Are you controlling speed using dynamic braking only? If so, what happens to the energy you're creating? Gotta go somewhere!

Best of luck with the project - sounds fun

Regards - John

 

[Perrins]

Johnhanging, thx, I'll take a look

Speedy, thx,
Yes the elevator will have two independent clamping systems, which will clamp the track, similar to a table saw kickback. The weight of the cart will clamp it tighter. The clamps will engage if the cable tension drops below about 800 lbs. I may add a shock absorbing system to it but I'm still thinking about the added complexity. I also plan to drop test the brake system.

Good Quote, but I prefer, "If necessity is the mother of invention, you can be assured that laziness is the father" Rob Perrins '92


harrisj, thx,
1 & 2. The winch drum will be a single wrap with a machined grove. The drum will be machined from a 12" pipe so it will be less than 20" wide for the 140 ft cart travel. The angle of misalignment will be very small, so I won't need to guide the cable onto the machined drum groove. This is a very simple system.

3. Yes I liked the idea and the related benefits, but a counter balance system seemed to be a little too complicated. As it is I'm laying 2000+ lbs. of steel down a mountainside for the track. I've already poured 15,000 lbs of concrete, buy hand, for the counter weight and footings.

4. The Up, Down, Stop & E-stop are located at the top, the bottom and on the cart. The E-stop will be hard wired through a relay and the others will be wired up to a PLC for the logic. The wiring to the cart will be done with a coiled cable riding on a 3/16" wire rope.

5. I'm still sourcing a surplus or used Variable Frequency Drive with dynamic braking for the downward deceleration). The VFD will ramp the speed down with the dynamic braking. They usually have resisters (heaters) to dissipate the deceleration energy.

 

[harrisj]

Sounds like you're way ahead of me.

On the last topic, the issue is not deceleration, it's the change in potential energy from top to bottom. I work out (very roughly) that if your car takes 100 seconds for the descent, you're dissipating 3.5 kW of potential energy. That's quite a resistor you need.

Of course, to haul up at the same speed, you'll need the same, plus losses (say 5 kW) produced by the motor. But I'm sure you've done all that stuff.

Wire the motor into a water heater and your passengers will have about 2 hot cups of coffee each when they reach the lakeside!

John

 

[Perrins]

John,

You're in the right ballpark, but I'm not quite as patient, ...the travel time is only 50 seconds. I plan on using a 7.5 Hp motor. The conservation of energy approach is certainly valid (although, I can't say I thought of working it out that way), but the majority of that energy shouldn't make it back to the motor and VFD, on the descent. I plan to use a worm drive gearbox. As a general rule of thumb, if the gear ratio is >40:1 the winch won’t be able to back drive the worm gear. I plan to use a 3600 rpm motor to bump my gear ratio up to about 80:1. The heat generated should be mostly in the gearbox, between the worm and gear. This is a common application for a worm drive. The motor should be mostly idling on the descent.

rob