Page 392 - Boiler_Operators_Handbook,_Second_Edition
P. 392

Controls                                                                                            377








































                                                 Figure 11-10. Simple valve positioner

               boiler cooled because it had to overcome the pressure  the controller has done all it could to restore the pro-
               of the feedwater trying to open the valve. While the  cess variable to setpoint. Unless that’s an intentional
               boiler came up to pressure the actuator pressure didn’t  condition (it could be) the output will eventually get
               change. (Why should it? It was happy because the valve  the process variable going in the right direction and it
               position matched the control signal). When the boiler  will return to the setpoint. It won’t stop there though,
               starts making steam and the water level drops the level  it will continue right on past because the output hasn’t
               controller will raise the control signal a little indicating  changed. With a reset controller it can’t because the out-
               the valve should open a little. In normal operation the  put is in windup, or wind down.
               valve would respond rather quickly but this first time,     Once the error is in the opposite direction the output
               after a shutdown, it won’t. It’s because the positioner  will start to change. During the period when the control-
               has to bleed off all that air that was compressed into  ler is building up from zero or dropping down from sup-
               the diaphragm chamber to raise the pressure enough to  ply pressure the controlled device, a valve for example,
               keep the valve closed against the high feedwater differ-  doesn’t respond because it only responds to signals
               ential (that’s now gone). That may explain why you’ve  within the control range. The result is always a long delay
               been surprised at the lag in response when operating a  (seconds, not hours—although sometimes it seems like
               valve manually. I know I was, more than once, because  hours) before the output gets back into the normal control
               I thought I had opened the valve a little and got no  range and, as a result, the process variable is swinging
               response so I raised it a little more and the next thing I  all over the place. That’s the effect of reset windup. You
               knew the water level was at the top of the glass because  won’t see most modern controllers doing it because the
               once it did start to open, it opened!                control manufacturers have designed the controls to elim-
                    The original pneumatic controllers did the same  inate it. You may still encounter it with valve positioners
               sort of thing and introduction of live zero made it hap-  and damper actuators. If you do run into it, at least you
               pen at both ends of the control signal range. A generic 3  will know why it happens.
               to 15 psig controller could wind up to an output equal     Reset windup is not the only problem that I had
               to the standard 18 psig supply pressure or wind down  to deal with and you probably will not. At some point
               to zero output. When that happens we’re out of control,  in time you will hear the terms “procedureless” and
   387   388   389   390   391   392   393   394   395   396   397