Fluid Power Journal

Elevator Stopping

figureitout-logoA simple circuit was designed to lift a service elevator up from the basement to the third floor. The HPU had a pressure-compensated pump and a three-position proportional directional valve. A proximity switch was used at each floor to signal the small PLC that, in turn, would signal the valve amplifier card to start de-acceleration until the elevator stopped at each floor.

The elevator had a hard stop at the bottom floor and relied on the proportional valve and proximity sensors for the stopping position at the upper floors. The valve’s amplifier card had four separate ramp adjustments so both acceleration and de-acceleration could be adjusted independently of each other in both directions.

An average weight normally loaded in the elevator was used to set the ramp speeds and stopping positions. The proportional valve was ordered with an LVDT spool feedback option to assure spool position accuracy.

During installation and de-bug, the engineer discovered that if a load with a different weight was used, the elevator cylinder’s stopping point at the upper positions would vary and cause the elevator floor to stop short or over-shoot at all three floors. Heavier loads caused the elevator to stop short of the desired position and lighter ones would over-shoot the desired stopping point, leaving the three floors with an unacceptable offset.

The engineer could adjust the de-acceleration ramps, fixing the current problem. But as soon as a different weight load needed to be lifted or lowered, adjustments were required again. If the proportional valve spool would go to the same open position with ½% accuracy due to the LVDT option, why did the stopping point change as much as 4″ when the load weight changed?

What do you think is the problem?

See the Solution

Many engineers mistakenly think the use of a proportional valve for speed control gives precise and consistent speeds. They feel the spool positioning is very accurate and repeatable when LVDT feedback devices on the main spool are used.

However, a costly proportional valve is no better than a plain, old-fashioned needle valve. The flow changes if the delta “P” across the valve changes. When loads of different weights are used, the load pressure on the outlet of the proportional valve changes, causing the flow across the valve spool to change.

The addition of a pressure-compensating “Hydrostat” module placed under the proportional valve solved this problem. The “Hydrostat” makes the proportional valve pressure compensated, ignoring the load changes, just the same as a pressure-compensated flow control valve. The stopping window was reduced to less than 1 inch total, which was fine for this service elevator.

If this were a people-carrying elevator, additional accuracy could be obtained by using some type of cylinder feedback device, such as a linear transducer. The LVDT on the valve knows exactly where the spool stops, but has no way to knowing where the cylinder actually stops. A feedback device on the actuator would know the cylinder’s position through the entire stroke and feed the information to the PLC.

By Robert Sheaf, CFPAI/AJPP, CFPE, CFPS, CFPECS, CFPMT, CFPMIP, CFPMMH, CFPMIH, CFPMM, CFC Industrial Training

Share this information.

Related Posts

One thought on “Elevator Stopping”

  1. Rance Herren says:

    The varying elevator load weight results in a varying load induced pressure. Given the pump is pressure compensated (constant supply pressure when the proportional valve is throttling, irrespective of the load) the differential pressure across the proportional valve yields increased flow and therefore actuator speed with lighter loads and less flow with heavier loads. Since the proximity sensors were used to detect the elevator stopping position, they are dependent on a constant and consistent cylinder velocity which is not possible with a varying load. The LVDT is strictly for positioning the valve spool to increase repeatability, reduce positional error and hysteresis in the valve itself but not the actuator it is controlling.

    There are some relatively simple solutions to this problem such as adding a meter in compensator to the proportional valve or changing the pump control to a load sensing type both of which operate on the principle of maintaining a constant pressure differential, resulting in a relatively constant flow across the proportional valve as the load pressure varies. The most accurate and precise solution is a closed loop positioning system with velocity overlay but the controls complexity and cost is significantly higher than a strictly hydro-mechanical solution.

Leave a Reply

Your email address will not be published. Required fields are marked *