From: aschoen@mo.net
Date: 30 Apr 1995 15:00:10 GMT
In <3nmh7e$3qd@newsbf02.news.aol.com>, kevins1202@aol.com (KevinS1202) writes:
>Hello my name is Kevin T. Shaw. I'm a little confused on the term HYSTERESIS. I had assumed it was like dead band or throttling range. Seems like everyone I ask has a differant definition. The application it is used in is DDC software. Some tell me it's time away from setpoint at certain conditions. Are there numerous types of the term "HYSTERESIS" I'm posting this as a challange to all you control gurus. I'm a building operator not an engineer so the challange is out to talk it DOWN to my level. :)
Kevin, if a control does not have hysteresis, it will give you the same exact result all the time given the same input signal(s). With hysteresis, the control does not necessarily provide the same exact result given the same input(s).
Mechanical controls generally have some amount of hysteresis because of resistance to the control's movement, i.e. internal friction, inherent in the control. Hope this helps.
Andy Schoen
aschoen@mo.net
Hysteresis is the measure of the actuator's inability to be perfect in its function. Ever watched a pnuematic damper actuator operate that has older linkage or the damper blades stick a little? Or a valve stem that someone has tighted the packing nut just a little to hard to stop the packing from leaking? These are good examples of were to find hysteresis.
An example would be to give a damper a 10% increase in output. When the increased signal is sent to the actuator, the damper doesn't start moving immediately. This is due to the initial static friction that must be overcome. Once it starts to move, the force required is less through the remainder of the travel. Now give it a 5% decrease in output. It may not move the damper blades at all due to sloppy linkage, worn bearings/bushings, etc. If it does move, odds are you are not going to end up with a full 5% travel in the opposite direction as you commanded. The difference is the hysteresis of the device.
A cooling coil valve in an air handler can be tested for hysteresis if you have a discharge temperature sensor that you can trend the leaving temperature. Starting at some known position (say 45% open and 56 F temperature), increase (open) the valve by 10%. Monitor the discharge temperature until it stabilizes for a couple of minutes (say at 50 F). Now decrease (close) the valve by 10% and monitor the temperature. If it did not return to approx. 56 F then the valve has hysteresis. The computer terminal may say that the valve is at 45% open, but there isn't as much water going though the coil as there was before.
In the limited DDC products that I have worked with, the functions that are available to compensate for device hysteresis add a little more signal to the device's commanded signal when a change of direction is about to be made. I have used them in the past when retrofitting an older pnuematic system with a DDC system.
Each piece of the control loop adds hysteresis (the electro/pnuematic transducer, the actuator, the linkage and the device). Testing the device in the way I mentioned above will take into account all of these at one time.
Hope this helps (better yet I hope I'm right!).
Jerry B.
Gresham, OR
Date: 28 Apr 1995 17:54:50 GMT
In article <3nq1pa$dg5@newsbf02.news.aol.com>, jerrybake@aol.com (JerryBake) writes:
|> A cooling coil valve in an air handler can be tested for hysteresis if you have a discharge temperature sensor that you can trend the leaving temperature. Starting at some known position (say 45% open and 56 F temperature), increase (open) the valve by 10%. Monitor the discharge temperature until it stabilizes for a couple of minutes (say at 50 F). Now decrease (close) the valve by 10% and monitor the temperature. If it did not return to approx. 56 F then the valve has hysteresis. The computer terminal may say that the valve is at 45% open, but there isn't as much water going though the coil as there was before.
Are you expecting the discharge temperature to be the same as when you started with the test? Well, I was thinking of the fact that, in the meanwhile, you cooled the supply air more. Therefore, the return air temperature is likely going to be lower, resulting in a different discharge temperature at 45% open than at the beginning.
From your experience, how big is this difference and is it negligible?
Thanks, Richard