Temperature gauge lag
Thread Content
I have a question for everyone. I have very high requirements regarding temperature control here; in particular, it’s necessary to be able to respond quickly to changes in actual temperature. We use PT100 thermistors for temperature measurement, and the readings are accurate. However, due to the fast rate of temperature increase, there is a significant delay in the temperature displayed on the gauge. When the temperature reaches 60 degrees, the heating stops, but the gauge doesn’t register this change immediately – it may take a minute before the actual temperature of around 70 degrees is shown. The temperature range we need is 0–150 degrees, and logically, thermistors shouldn’t cause such delays in temperature reading. Does anyone know what might be causing this delay in temperature measurement here? Thank you in advance!!
According to the owner’s technical requirements, it might be difficult to use a thermal resistor for this purpose; since the temperature range is from 0 to 150 degrees, it would be better to use a thermocouple! But ideally, I would recommend the following approach: 1. Use a CU50 thermistor for measurement; it’s best to choose a sensor with three wires, and this temperature value should be used only for monitoring purposes ； 2 Order a temperature switch from the manufacturer; the value set for this temperature switch corresponds to the temperature at which heating should be stopped as per the process requirements, and it provides a digital signal to control the reaction.
Thermal resistance sensors are used for point temperature measurement; the measurement process is not fast, and the speed of heat conduction directly affects the speed at which temperature changes can be detected. It is recommended that you try using armored thermal resistance sensors, opting for those with thinner wires, as this can help address the issue of heat conduction speed
The main issue lies in the measurement section; the thermometer probe you use might be problematic. If a thinner probe is used, with a diameter of 6 mm or less, the response time is only a few seconds. Or if there is a sleeve outside the thermometer, with air between it and the thermometer, then the heat transfer time will **be prolonged**. It’s better to give more consideration to the response time of PT100.
Do not use sleeves; instead, use armored thermocouples or armored thermal resistors. The reaction time is generally in the range of ten seconds or so. If a faster response is required, it is possible to arrange for custom manufacturing with the instrument manufacturer.
It’s not just the issue of thermocouple lag; there is also a lag in the heat conduction of the furnace wire, which may have an even greater impact. Thermocouples/thermistors can be made very thin. However, using thermocouples/thermoresistors with low heat capacity can only alleviate the problem, not resolve it completely. It is only possible to completely resolve the lag problem by using advanced control.
Based on your description, the fundamental issue lies not in the PT100 thermistor, but rather in the heating control mechanism. If steam (or hot water) is used for heating, an additional control valve can be installed to regulate the heating source; the signal from the PT100 thermistor at the site should be fed into a temperature controller equipped with PID regulation functions. Only in this way can the high-precision temperature control requirements be met. For reference only!
This is mainly due to the delay in AD acquisition in the detection section; therefore, a fast AD acquisition module should be used.
The response time of the thermal resistor is around 60 to 90 seconds, which is relatively slow. If the PID parameters are not set properly, it will further increase the difficulty of control.
I checked it out – it’s terrifying. It’s not so much a problem with the thermal response time itself. If its time constant is known, a (TsS+1) element can be added as compensation for the controller; most DCS systems have LEAD/LAG modules that can accomplish this. The fear is that when the measured temperature seems quite stable, it’s hard to know how wildly the actual temperature can fluctuate. However, it’s useful to do this kind of research as well, to know that there are millisecond- and microsecond-grade \"fast thermocouples\". The original poster had better use this. Haha. This post was last edited by RainWolf on 2009-2-25 at 14:15.]
Impact of response time: 1. For single-sample testing, fast-response fine PT100 or extruded thermistor is used； If you have money, use an infrared thermometer to take temperatures ； 2. The control loop utilizes differential compensation; if possible, conduct simulation experiments to determine the response time, gain coefficient, and delay time – these three parameters will facilitate the tuning of the PID loop. If there are feedforward variables available, they can be incorporated to use cascade control. 3. Control loop: The response time of the control valve generally cannot be changed, but it needs to be checked and maintained regularly. The focus is on determining the time it takes for the fluid to affect the controlled variable (temperature); if there are issues, adjustments need to be made to the process and equipment.
I think the reason for the delayed temperature response is that the temperature display is slow to react, has low accuracy, or the decimal point isn’t set properly; you can try using a new secondary meter.
That can’t be right; if we go by what you all say, then thermal resistance simply can’t be used. I think the original poster’s question is still related to product selection, measurement point locations, installation methods, and instrument settings, among other things.
I just attended the E+H training, and from what I remember, their new products have a response time that is seven times faster than conventional ones. Now that the specific basics are clear, hehe. Hope this can help the original poster
Some instrumentation books state that thermal resistors are generally used for temperatures ranging from 0 to 600 degrees, while thermocouples are used for temperatures above 600 degrees; it seems that most thermocouples do not provide high accuracy when measuring low temperatures. Will using a thermocouple here affect the measurement accuracy?
I think there should be no problem with the thermal resistor; the main issues lie in the installation and selection process, as well as whether there is any temperature variation. It’s possible that certain parameters, such as damping, aren’t set correctly. If the control system is used without taking temperature variations into account, it’s recommended to check the parameter settings and the installation
End reduction Φ5.3mm t50: Reduced-diameter tube ≤7.5s, straight tube: ≤18s