topic Re: Temperature control, inertia. in 8-bit Microcontrollers https://community.nxp.com/t5/8-bit-Microcontrollers/Temperature-control-inertia/m-p/453578#M21620 <HTML><HEAD></HEAD><BODY><P>PID is right way to go. "Trouble with inertia" sounds like anti-windup is not implemented. What happens without anti-windup:</P><P>Real world actuator always has limited range, for example PWM duty settings from 0 to 100. But I-term of PID, due to slow rise/fall of temperature, may quickly accumulate and make PID asking actuator&nbsp; to throttle to 1000, 10000 or 1E10 for example, instead of available only 100... After temperature crosses target temperature value, I-term starts falling, but it will take quite long until it falls to negative I-term values. And T keeps rising, temperature oscillates wildly...</P><P>What you can do to fix problem? For example, when PID gives out of range actuator value, set I-integral to value, which would make PID formula pointing to margin actuator value. </P></BODY></HTML> Sat, 04 Jul 2015 06:46:04 GMT kef2 2015-07-04T06:46:04Z Temperature control, inertia. https://community.nxp.com/t5/8-bit-Microcontrollers/Temperature-control-inertia/m-p/453577#M21619 <HTML><HEAD></HEAD><BODY><P>Hello everyone. I'm working on a temperature control project that prescisa to be very stable. I'm having trouble with inertia.</P><P>Does anyone know of any control algorithm to help me?</P><P>I'm working with PT60. I tried to implement a PID control but could not.</P><P>Thanks.</P></BODY></HTML> Fri, 03 Jul 2015 20:53:58 GMT https://community.nxp.com/t5/8-bit-Microcontrollers/Temperature-control-inertia/m-p/453577#M21619 eliezeroc 2015-07-03T20:53:58Z Re: Temperature control, inertia. https://community.nxp.com/t5/8-bit-Microcontrollers/Temperature-control-inertia/m-p/453578#M21620 <HTML><HEAD></HEAD><BODY><P>PID is right way to go. "Trouble with inertia" sounds like anti-windup is not implemented. What happens without anti-windup:</P><P>Real world actuator always has limited range, for example PWM duty settings from 0 to 100. But I-term of PID, due to slow rise/fall of temperature, may quickly accumulate and make PID asking actuator&nbsp; to throttle to 1000, 10000 or 1E10 for example, instead of available only 100... After temperature crosses target temperature value, I-term starts falling, but it will take quite long until it falls to negative I-term values. And T keeps rising, temperature oscillates wildly...</P><P>What you can do to fix problem? For example, when PID gives out of range actuator value, set I-integral to value, which would make PID formula pointing to margin actuator value. </P></BODY></HTML> Sat, 04 Jul 2015 06:46:04 GMT https://community.nxp.com/t5/8-bit-Microcontrollers/Temperature-control-inertia/m-p/453578#M21620 kef2 2015-07-04T06:46:04Z