Instead of having a fixed-weight counterweight, whose weight equal's the elevator cab's weight, plus 50% the expected average load, the counterweight would consist of a much smaller fixed weight, plus a large water tank, whose total weight equals that of the empty elevator cab.

At every level of the building, there'd be a water storage tank, and a pump. Whenever the elevator is requested to go from one level to another, we immediately pump water into (or out of) the counterweight, so that the water's weight equals the weight of the load in the cab.

If there isn't enough water in the storage tank on the counterweight's level, we either pump water from the level(s) below, or (if the counterweight is on or below ground level) use municipal water to fill the counterweight, or drain water from levels above.

(Draining water from above is usually the least energy efficient choice, even if we were to design each pump so it could double as a micro hydro turbine, and generate electricity from the down flowing water.)

After the counterweight balances the load, *then* the elevator's main motor moves the elevator up or down.

Or, the counterweight could be just enough lighter or heavier than the load, so that when the brakes disengage, the cab accelerates in the desired direction. In this version, no main elevator motor would be needed.
-- goldbb, Oct 08 2009

(?) Water elevator counterweight. http://www.american...8/5/1978_5_40.shtml
[spidermother, Oct 09 2009]

Same idea, but with quantized solid weights. Dynamic_20Counterwe...s_20for_20Elevators
My argument against this idea is that water takes time to pump in and out of a tank, whereas a sliding/weight-seleting mechanical device could function much quicker. [daseva, Oct 28 2009]

[+] Effectively you would end up with lots of water up top following the evening rush home, which would then be used to power the elevators for the morning incoming rush. Sounds like a closed system though, give or take; not sure why you have a microturbine in there... you could use wastewater which has already been pumped up to the floors (while fresh) to power it.
-- FlyingToaster, Oct 08 2009


Under normal use, we should expect that occasionally we will have an empty tank at some particular level of a building, and yet need to move water from that level into the counterweight.

How could that occur? The simplest reason might be someone regularly climbing up the stairs, then taking the elevator down.

Furthermore, suppose we want the system to have a fixed amount of water in it total, and never add water to it, or dump water from it.

If the empty situation occurs when there is no water on any level below the counterweight, then there *will* be water on one or more levels of the building that are above the counterweight.

If we merely drain the water downward, until it's at the counterweight's level, then we are wasting the (gravitational potential) energy of that water.

If we drain the water downward, passing it through a small hydroelectric turbine, we recoup some of the energy that had been expended getting the water up there in the first place.
-- goldbb, Oct 08 2009


ah, I see... but I think you could use computer control to do partial pumpings in either direction as the counterweight passes a floor. You're definitely going to need some sort of method to vary acceleration/velocity.
-- FlyingToaster, Oct 08 2009


Rather than small tanks and pumps at each level, it might be better to have one large pump and a tank at the top of the building, with a recoil or hanging hose to the counterweight tank.
If you also put a tank on the elevator car, and a hose (looping down to the bottom of the lift shaft) between the two, with a pump on both the counterweight and the car, the whole system could be 'live' and not need a motor at the top at all, just a pulley and emergency brake.
-- neutrinos_shadow, Oct 08 2009


"Baldwin invented an elevator whose suspending cables passed over a pulley at the top of the elevator shaft and were attached, at the other end, to a counterweight consisting of a large iron bucket. This bucket ran up and down in a tube, or standpipe, paralleling the shaft. By stepping on a pedal, an operator seated in the elevator cab could slosh water into the bucket, thereby causing the cab to rise. When he wanted to come down again, he stepped on another pedal that let the water out." (link).

This idea is at least a hundred years old.
-- spidermother, Oct 09 2009


spidermother, the article you're quoting is mistaken.

Googling for Cyrus Baldwin does indeed turn up an inventor of a type of elevator, and it's a hydraulic elevator even, but it's a conventional hydraulic cylinder and plunger arrangement, pushing the elevator car up from below.

Searching for "water balance elevator" turns up an invention similar to the one described, but its inventor is William E. Hale, not Cyrus Baldwin.

That old invention does indeed use water in a moving tank as a variable weight counterbalance... but it's operation is quite different. In that device, whenever the water is let out, it's dumped out to a discharge tank at the bottom of the counterbalance's shaft, and from there pumped all the way up to a tank on the roof. Furthermore, when water is added to the counterbalance, it's always dropped in from the roof tank.

Throwing away the gravitational potential energy of the water, by dropping it, is quite wasteful.

Even neutrinos_shadow's anno (where water is dumped from the roof tank to the counterweight, and sent back via a return hose) is more energy efficient.

The goal of my idea is for a way to never throw away energy, if it can be at all avoided.

FlyingToaster, partial dumpings as the counterweight passes a floor is definitely better than using the brake, both to increase the longevity of the brake, and because it should end up with more of the water in the tanks at higher levels. And computer control is absolutely a good idea, considering how tricky Hale's water balance elevator was to operate.
-- goldbb, Oct 11 2009


//The goal of my idea is for a way to never throw away energy//
You can't win.
You can't break even.
You can't quit the game.

Those are the rules.
-- AbsintheWithoutLeave, Oct 11 2009


I'm neutral. The idea of a dynamic counterweight is innovative but does this really solve a problem? Would elevators perform any better or more efficiently as a result of always being equally balanced? You can never get the exact weight of an occupied cab because the movement of the people will cause the weight to fluctuate until it can stabilize itself once the movement stops. This is comparable to having to stand still on a bathroom scale.
-- Jscotty, Oct 11 2009


//the article you're quoting is mistaken// My bad for trusting a single source. Thanks for the correction.
-- spidermother, Oct 27 2009


There is a way this can work - if the occupants of the elevator can be persuaded to urinate into a flexible hose which leads to the hollow counterweight.
-- hippo, Oct 27 2009


Is the counterweight usually sized for 50% of the average load, or 50% of maximum? The latter would minimize the required motor size. Note that many types of electric motors (AC as well as DC) will work as generators, so some of the energy spent raising an overly-heavy counterweight will be regained even if the weight descends when it's heavier than the car.
-- supercat, Oct 28 2009


Right. I recently thought lifts could use "regenerative braking" - that is generate electric power through the motor when the lift-counterweight system is going the way gravity wants it. I suspect they don't though, since it's cheaper to build a simple system which is happy to waste power. Though...does anyone know how it's controlled when it's going with gravity ? The energy must go somewhere...

[+] anyway. Of course it doesn't have to be water. The counterweights could carry low priority parcels. Automatically - just add your box to the queue on your floor, and it should make to the destination floor eventually. One day.
-- VaquitaTim, Oct 28 2009


Actually, I suspect many elevators and escalators do use regenerative braking in some measure when descending (though probably not when decelerating at the end of a move). The rotor on a typical motor will try to run at a certain speed related to either the AC line frequency (in the case of induction motors) or the drive voltage (in the case of commutator motors). If nothing is trying to make it run faster or slower than its unloaded speed, the motor will neither take in nor put out current. If something is trying to make the rotor run slower than its target speed (even if just friction in the bearings), the motor will draw current from the line. If instead something is trying to make the motor run faster than the target speed, it will feed current into the line.

No fancy electronics are required if the motor is connected 'directly' to the power source; the feed-back of power will happen automatically. If a motor is electronically controlled, the feed can only happen if the electronics are designed to allow for it, but if an elevator simply uses relays to switch the motor current, regenerative braking will naturally occur.
-- supercat, Oct 29 2009


AbsintheWithoutLeave, while you can't get ahead if everything which goes down the elevator came up in the first place, I seems to me that if a sufficient number of people walk up the stairs, and ride down the elevator, we should be be able to do better than break even.
-- goldbb, Nov 08 2009


//goldbb: ...walk up the stairs, and ride down the elevator...//

Or combine the two--an x-story elevator with internal stairs, which stops only on floors nx+1.
-- ldischler, Nov 09 2009


I suspect this elevator would be terribly slow, waiting on the water level to rise.
-- RayfordSteele, Nov 10 2009



random, halfbakery