Philldpapill
06-27-2009, 07:44 PM
Does anyone have some experience with Proportional-Integral-Derivative Control systems with regard to controlling the cell temperature? You can skip the next paragraph if you don't want the extra info.
For those that don't know what this is, a "PID" controller is something that monitors a system, and keeps it stable - like your cruise control. With cruise control, if you start going up a hill, your speed decreases, so your car increases the fuel supplied, thereby increasing your speed so that the speed is constant. These controllers monitor the "sum of the error"(e.g. the total time your car has been going UNDER the desired speed), how much the error is off proportional to a set limit(e.g. how far below the speed your car is moving at a given moment), and how fast the error is changing(e.g. how fast your car is slowing down when it should be going at a constant speed).
Hence, Proportional, Integral, and Derivative control. Most control systems are easily tuned, but I am having a hard time trying to tune mine.
The purpose of this control, is to monitor the temperature of an HHO cell, and reduce/increase the amount of current supplied, in order to maintain a constant cell temperature. This will be part of my HHO Control system that I posted about in a previous thread. The controller has a ton of features for collecting data, and controlling HHO production based on a number of inputs.
An HHO cell works most efficiently at higher temperatures. This is because some of the energy needed to break the H20, is already provided in the form of heat. Due to this, less power is needed to produce the same amount of HHO. The downside to this, however, is that at higher temperatures more water vapor is produced. The water vapor can be taken care of through a bubbler, but increased water vapor production means that some energy is wasted in the form of evaporation of the water. Basic thermodynamics explains why, so I'm not going into that.
The problem with the control, however, is that it is very difficult to maintain an ideal temperature of the cell. If the temperature is too low, the PWM duty cycle is increased, allowing more current to flow. This increase in current will inherently increase the temperature. This increased temperature will cause the controller to cut back on the current. This oscillation is normal in most systems as it will be dampened out. However, because the changes in temperature are sooo god-awful slow, it's hard to dampen out the temperature fluctuation. I'm thinking there may be a simplier way to control the temperature/current flow WITHOUT a PID control
Does anyone have a novel approach to this situation? Remember, I'm trying to keep the cell temp at a fairly precise value, and the amount of current provided to the cell, is an inversely proportional function of the temperature.
For those that don't know what this is, a "PID" controller is something that monitors a system, and keeps it stable - like your cruise control. With cruise control, if you start going up a hill, your speed decreases, so your car increases the fuel supplied, thereby increasing your speed so that the speed is constant. These controllers monitor the "sum of the error"(e.g. the total time your car has been going UNDER the desired speed), how much the error is off proportional to a set limit(e.g. how far below the speed your car is moving at a given moment), and how fast the error is changing(e.g. how fast your car is slowing down when it should be going at a constant speed).
Hence, Proportional, Integral, and Derivative control. Most control systems are easily tuned, but I am having a hard time trying to tune mine.
The purpose of this control, is to monitor the temperature of an HHO cell, and reduce/increase the amount of current supplied, in order to maintain a constant cell temperature. This will be part of my HHO Control system that I posted about in a previous thread. The controller has a ton of features for collecting data, and controlling HHO production based on a number of inputs.
An HHO cell works most efficiently at higher temperatures. This is because some of the energy needed to break the H20, is already provided in the form of heat. Due to this, less power is needed to produce the same amount of HHO. The downside to this, however, is that at higher temperatures more water vapor is produced. The water vapor can be taken care of through a bubbler, but increased water vapor production means that some energy is wasted in the form of evaporation of the water. Basic thermodynamics explains why, so I'm not going into that.
The problem with the control, however, is that it is very difficult to maintain an ideal temperature of the cell. If the temperature is too low, the PWM duty cycle is increased, allowing more current to flow. This increase in current will inherently increase the temperature. This increased temperature will cause the controller to cut back on the current. This oscillation is normal in most systems as it will be dampened out. However, because the changes in temperature are sooo god-awful slow, it's hard to dampen out the temperature fluctuation. I'm thinking there may be a simplier way to control the temperature/current flow WITHOUT a PID control
Does anyone have a novel approach to this situation? Remember, I'm trying to keep the cell temp at a fairly precise value, and the amount of current provided to the cell, is an inversely proportional function of the temperature.