Team:BIT/Temperature control

Click on other areas to close the home directory

Temperature control
Abstract
【Objective】The temperature control module can provide different temperature conditions for different needs in the biological part. It can provide miRNA linkage with 85℃ 2min+37℃ 5min; LAMP with 67℃ 15min and CRISPR/Cas12 with 37℃ 20min. The temperature is constant during these processes. This module can help us make a small, low-power, high-control precision instrument.
【Method】The temperature control module controls the average power of the heating film by controlling the duty ratio of the PWM signal output to the heating film. At the same time, the semiconductor refrigeration chip and the fan are controlled at a fixed time, so we can control the temperature rise and fall of the sliding chip. They realize the rapid switching between different temperatures. In the control process, we use negative feedback regulation and PID control algorithm to achieve accurate temperature control. Accurate temperature control can help to achieve high specificity and sensitivity of nucleic acid amplification and CRISPR / Cas12 reaction. This module mainly uses MCU as the main controller to achieve temperature acquisition, heating, cooling and serial communication.
【Results】The heating rate of the module is up to 2.26℃/s. It can get a good temperature uniformity performance, a high control accuracy and a small steady-state error. The key indicators such as average heating rate and uniform temperature performance can meet the requirements of GB/T 1.1-2009 technical indicators. The results tell us that the module can realize accurate temperature control. It can help us achieve our requirements for high specificity and sensitivity of nucleic acid amplification and CRISPR / Cas12 reaction.
System mathematical model
The system mathematical model can describe the relationship between the system input, output and internal variables. It reveals the inherent relationship between the system structure, its parameters and its performance. We calculate the mathematical model of the system and use it to guide the design of temperature control system.
Temperature of slide chip is controlled object of temperature control module. Control variable is the PWM signal loaded on the heating film and the cooling plate. PWM wave acts on the heating plate so heat is produced or absorbed by Joule effect, Tomson effect. It means the temperature of sample block rises and falls down.
The fig2 is our variable temperature system. The controlled objects are the real-time temperature of the sample block, the heat flow rate of the semiconductor refrigeration sheet flowing to the sample block during heating or refrigeration, and the heat flow rate of the heat exchange loss between the sample block and the surrounding area.
In the variable temperature system, the heating film produces heat, the semiconductor refrigeration plate absorbs heat and the system composed of sliding chips and test fluids absorbs or releases heat. Assuming that the temperature distribution of the system composed of sliding chip and test fluid is uniform on the cross section, that is, the temperature distribution is linear in the vertical direction of the system, one-dimensional heat transfer simplification of the system can be carried out. So we can simplify the one-dimensional heat transfer of the system. So we can assume that the temperature on the same section is equal, assuming that the total heat capacity of the system consisting of a test tube and a test solution is C. Then according to the heat transfer equation we can get:
    Tc——Temperature of sliding chip (Assuming that the temperature of the slide chip is the same as that of the surface of the heating film)
        Q——Heat flowing to sliding chip in unit time during heating or cooling
        Q0——The heat loss caused by heat exchange between the sliding chip and the surrounding area unit time
Thermoelectric Effect of Semiconductor Refrigeration Sheets is Combined by Different Effects. Under constant current, the pure refrigerating capacity of semiconductor refrigeration plate is:
The pure heating capacity of heating film is：
Qp is the heat absorbed ( or released ) per unit time by the Paltie effect. is the heat produced per unit time by Joule effect and Fourier effect. aab is the seebeck coefficient which is related to the material. R is semiconductor refrigeration sheet resistance. K is the total thermal conductivity of the cooling plate. And I is equivalent current which is controlled by duty cycle.
We can know from formula 2 and formula 3 that there is a certain relationship between Q、I and Tc. Althoughmay generate the increment of ΔI, it can be considered that Tc does not change because of the large lag characteristics of the system. Using Taylor expansion to handle the formula 2 and formula 3, we can get:
In the formula4, the f is the is the equation between Q and I. Because in the process of computer control, the sampling frequency is very high and the change of sampling interval between each two times is very small, we can consider the f as a very small amount. So we can approximate the formula 4 as：
According to the principle of heat transfer, when the thermal resistance of the sample block is θ, the thermal loss of the sample block in unit time is :
With the formula 5 and formula 6，we can get a new formula 1：
Formula 7 is the differential equation of variable temperature system. After mathematical processing, we can get a transfer function of a variable temperature system :
And K=K0θ,T=Cθ，so after simplification and approximation, the temperature change system can be regarded as a first-order system.
For the first-order system, we can use the negative feedback control method to control and construct.
Hardware design framework
There are three steps：linkage（85℃，2min；37℃，5min）、LAMP（67℃，15min）and CRISPR/Cas12a reaction（37℃，20min）. The purpose of thermal control is to ensure that the test temperature in the sliding chip changes according to the expected temperature gradient curve.
The users inject the sample into the slide chip through the injection port for amplification reaction in the main reaction chamber. In this process, the temperature control module provides the cyclic temperature change operation of amplification reaction and real-time feeds back the temperature value of the temperature sensor data after calculation to the host computer to monitor the temperature control process and conduct temperature control adjustment. Temperature acquisition module and semiconductor refrigeration / heating drive module are the most important parts of the system. Their performance is directly related to the final temperature control speed and accuracy. We design the circuit board schematic and PCB, and make four-layer circuit board.
Control algorithm design framework
We use closed-loop negative feedback to control the temperature change. The principle is shown in Fig.5. The system uses the temperature sensor to measure the existing temperature, through analog to digital conversion into the controller so it can complete the feedback control. Through this closed-loop control system, we can make our equipment reach the required temperature.
Circuit Board and Selection of Key Devices
We have designed our own PCB circuit board, and the chip is welded on the PCB circuit board to connect external devices at the same time. The PCB circuit board is as follows：
In the table 1 we show the key components of the temperature control module.
The above design is assembled and the temperature control module is constructed. The results are as follows:
Core Controller - MCU minimum system
The minimum system includes MCU, power circuit, reset circuit, oscillation circuit and program download circuit. OSC1 constitutes a oscillating circuit and JTAG is the JTAG download interface，The minimum system frame diagram of the MCU is shown below.
Temperature collect module
Selection of PT1000
The temperature sensor is attached to the bottom of the slip-chip. The temperature is converted to voltage by the temperature measuring bridge and measured by the ADC module of MCU. There are three main types of temperature sensors : thermocouples, thermal resistors and thermal resistors. The thermocouple has a wide temperature measurement range. The temperature measurement accuracy is 0.5℃－5.0℃. However, its poor repeatability is easily affected by environmental interference signals and the temperature drift of the preamplifier. Thermal resistors such as platinum resistance temperature sensor has a good stability and good linearity ( approximate straight line ), small error, temperature measurement accuracy in 0.1℃－1.0℃. But the price is expensive. Finally, thermistor is small in size, fast in response time and low in price. The temperature range is -50℃－200℃, and the temperature measurement accuracy is 0.05℃－1.5℃. So thermistor meets for our requirements. Therefore, we use PT1000 thermistor as temperature control sensor.
The temperature measurement range of PT1000 is -200℃ – + 650 ℃. The resistance value is 1000Ω at 0 ℃, and the resistance value is about 2120.515Ω at 300 ℃At the same time, in the range of 0 ℃ ~ + 100℃, the relationship between temperature and resistance is close to linear, so we use the least square method to fit the relationship between temperature and resistance.
In the required temperature range of the system, the output resistance of Pt1000 has a good linear relationship with temperature. It has a high sensitivity of 3.8506Ω／℃. The output characteristic curve simulation diagram is shown in Fig.9.
Temperature measurement circuit ( ADC sampling module )
In order to realize the linearization of the temperature measurement circuit and higher accuracy and sensitivity, we convert the temperature signal and adjust it by the temperature measurement bridge plus amplification circuit. The actual temperature measurement circuit is shown in Fig.11 and the circuit can well realize the zeroing and filling of the temperature measurement circuit.
The calculation formula of output voltage V0 is：
Using the formula 9, the resistance value of R3 can be calculated by the output voltage V0 collected by MCU. And the actual temperature value can be calculated according to the linear relationship between PT1000 temperature and resistance value.
Use the ADC12 module of MCU to collect the voltage output by the temperature measurement circuit and complete the analog-to-digital conversion. The conversion results are shown in Formula 10. The AD conversion of MCU supports 12-bit precision analog-to-digital conversion. And the sampling period can be controlled by software. It has four different sampling modes. The reference voltage can be selected as 2.5 V or 1.5 V or external voltage and a series of functional characteristics such as clock source can be selected.
Design Temperature control component
Selection of Heating Film
Silicone rubber heating film use the soft thin silica gel so it can effectively paste the heating surface. It is simple installation, safe and reliable.
To match the size of our slide chip, we used a 20 * 90mm heating film. To achieve the desired heating rate, The heating film has a rated voltage of 12V and a rated power of 25W.
Selection of Semiconductor Refrigeration Plates
Working principle of semiconductor refrigeration sheet is based on Paltier principle. It means when the circuit composed of two different conductors A and B is connected with direct current, some other heat will be released at the junction except Joule heat, while the other junction absorbs heat. The heat absorbed and released is proportional to the current intensity I.
The typical structure of Paltier shown on Fig.12. Paltier is made up of many N-type and P-type semiconductor particles aligned with each other, while N and P are connected by a general conductor to form a complete circuit, usually copper, aluminum or other metal conductor, and finally sandwiched by two pieces of ceramic.
We use the semiconductor refrigeration chip to accelerate the heat conduction of the sliding chip, so that the temperature of the sliding chip can be released faster when the temperature is cooled. With the structure we designed, we used two maximum refrigerating capacity of 25W, 20mm * 40mm cooling plate. They have high reliability and are fully capable of fast cooling.
Fan selection
We use miniature cooling fan. It has the characteristics of hydraulic bearing, all environmental protection materials, static noise ratio up to 25 dB, and the speed can be in the range of 6000-17000RPM. The rated voltage of the fan is 12V, the rated power is 2.1W, and the size parameter is 40*40*20mm. After our inspection, we can ensure that this fan with semiconductor refrigeration chip can play a cooling and cooling effect on our PCR instrument.
Drive circuit design
Adopt photoelectric coupling chip and MOS tube circuit as driving circuit.
The structure of the optical coupler is equivalent to encapsulating the light-emitting diode and the photosensitive (triode) tube together. The light-emitting diode converts the input electrical signal into an optical signal and transmits it to the photosensitive tube to convert it into an electrical signal output. Since there is no direct electrical connection, it not only couples and transmits the signal, but also has the effect of isolating interference.
According to the characteristics of NMOS, the drain D is connected to the power source, and the source S is grounded. If the gate voltage Vgs is greater than the conduction Vgs given in the parameter manual, the conduction between D and S can be achieved.
Fig. 12 Optocoupler wiring
Fig. 13 MOS tube wiring
Fig. 14 Photocoupler and MOS module
Communication circuit
The design of communication circuit is the premise of man - machine interface. In order to see the real-time change curve of temperature, users need the MCU to transmit the real-time collected temperature data to the host computer (smart phones and computers), and then process the data to obtain the real-time dynamic curve through the application software installed in the host computer (WeChat small program or some else).
In the debugging process, in order to facilitate debugging, we use serial communication with the computer for data interaction. MCU single has two universal serial communication interfaces. USB interface is the most common and convenient data transmission port used by upper computer. RS232 to USB data line has high reliability when transmitting data in short distance, but because the level of serial port signal ( TTL level ) and RS232 signal level is not consistent, the two must be level conversion.
In the process of using, data exchange is carried out with mobile phones through Bluetooth communication based on WeChat small program platform. Bluetooth is a small range wireless connection technology (generally within 10 meters) radio technology, which can realize convenient, flexible, safe, low cost and low power data communication between devices.
The Bluetooth module is a PCB board with integrated Bluetooth function, which is used for short-distance wireless communication. It is divided into Bluetooth data module and Bluetooth voice module according to function. The Bluetooth module refers to a collection of chip basic circuits that integrate Bluetooth functions for wireless network communication. The Bluetooth core protocol is shown in the figure below：
In the temperature control process, the MCU transmits the collected real-time temperature to the Bluetooth module through the serial port, and the Bluetooth module sends the data to the mobile phone through the Bluetooth protocol. Relying on the Bluetooth related functions in the mobile phone WeChat applet, the temperature control curve accepted in the mobile phone Results as shown below：
Fig. 16 Temperature Control Curve of WeChat Mini Program
Fig. 17 Bluetooth module
The design of the connection mainly uses the flexible control function of the serial port of the MCU. But also in order to use the powerful calculation and display function of the mobile phone and the computer, so as to realize the real-time acquisition, monitoring and processing of the effective signal on the spot, which is one of the ways to realize the good human-computer interaction interface.
Control algorithm design-PID control algorithm
principle of PID control algorithm
The accuracy and stability of temperature control have a significant impact on the results of LAMP amplification, if the temperature control accuracy or stability is not good, it will lead to the reduction of amplification efficiency, non-specific amplification and so on, thus affecting the experimental results. In addition, the heating and cooling rate is also an important aspect of the performance of the temperature control system. If the heating and cooling speed is not ideal, the reaction time of the whole process will be greatly prolonged. This is not acceptable in applications such as on-site detection and rapid detection. Therefore, it is necessary to study the temperature control algorithm in order to improve the performance of the whole temperature control system.
In engineering practice, the most widely used control algorithm is proportional, integral and differential control, referred to as PID control, which has simple structure, good robustness and easy to implement. It is one of the most classical and most simple and effective control methods. For any control problem, PID control is the first choice. When the controller works, the proportion, integral and differential of the error signal are linearly combined to form a control quantity to control the controlled object, so it is called PID controller. The control law is as follows:
Digital PID control algorithm
The positional digital PID control algorithm can be obtained by discretizing formula 11, that is, the discrete PID expression of k sampling time can be obtained by using a series of sampling time points k to represent continuous time t, rectangular numerical integration approximation instead of integral and first order backward difference approximation instead of differentiation.：
Integral separated PID
When the deviation between the controlled quantity and the set value is large, the integral action is cancelled to prevent the stability of the system from decreasing and the overshoot increasing due to the integral action; when the controlled quantity is close to the given value, integral control is introduced to eliminate the static error and improve the control accuracy. The integral separation control not only maintains the function of the integral, but also reduces the overshoot, so that the control effect is greatly improved
For integral separation control, first set the separation threshold E
1) when | e (k) | ≤ | E |, that is, when the absolute value of deviation is small, the control accuracy of the system can be ensured by using PID control;
2) when | e (k) | ≥ | E |, that is, when the absolute value of deviation is large, the amount of overshoot can be greatly reduced by using PD control. The integral separation algorithm can be expressed as
In the formula, α is the logical coefficient，when it is 1，otherwise it is 0。
Parameter setting
In the above formula, each term is preceded by coefficients, which are called the parameters of the PID controller, and the parameters are adjusted to achieve the desired control performance called parameter tuning. Parameter tuning of PID controller is the core content of control system design.
1) first, pre-select a sampling period short enough to make the system work；
2）Only the proportional control link is added until the response curve with fast response and small overshoot is obtained；
3）Add integral control. First reduce the scale factor selected in step 2) to the original 50-80%. Then adjust from large to small, observe the response curve, eliminate the steady-state error on the basis of maintaining the stability of the system, adjust the proportional coefficient accordingly, and finally determine the parameters of proportion and integration；
4）Add the differential control, gradually increase the differential coefficient, and change the proportional coefficient and integral coefficient accordingly, repeatedly try to obtain satisfactory control effect and PID control parameters.
Different results of setting different PID parameters at 67 ℃ are shown below
Parameter time-varying
If the parameters of PID are kept fixed in the working process of the whole system, there will be problems in the control of the system. During the heating process of the control system, the heating film is turned on, the fan is turned off, the fan is turned on during the cooling process, and the semiconductor refrigerator works. In addition, the time variation of some material properties in the system is obvious. These factors cause the system characteristics of the variable temperature system to change with the reaction, while the adaptability of the control parameters to the change of system characteristics is limited. When the system characteristics change. The original set parameters do not adapt to the changing system characteristics, which makes the control performance of the system worse. in order to solve this problem, it is necessary to improve the control so that the parameters can be adjusted on-line according to the changes of system characteristics. This project sets different PID parameters for different heating and cooling stages.
Performance test of temperature control
Set up the above temperature control module to build a principle verification platform, as shown in the following figure, to test the performance of the module.
Heating rate test
Open the host computer software, observe and record the real-time temperature curve and data of the host computer software interface. The real-time temperature should be changed according to the setting temperature and program setting, such as connection reaction (85 ℃, 2 minutes 37 ℃, 5min), LAMP amplification reaction (67 ℃, 15min) and CRISPR/Cas12a reaction (37 ℃, 20min).
The room temperature is recorded as TA, the target temperature is recorded as TB, the time from TA to TB is recorded as t, and the average heating rate is calculated according to formula (15).
The heating rates of the three places are 2.26C / s (85 ℃), 0.36C / s (37 ℃) and 1.68C / s (67 ℃), respectively, of which 85 ℃ and 67 ℃ can meet the GB/T 1.1-2009 standard.
Module temperature uniformity test
When the temperature is controlled to 67 ℃, the Fluke 51 Ⅱ thermometer is used to measure the temperature of the six main chambers of the sliding chip.
Repeat the experiment three times, take the average temperature of each chamber, the temperature data is shown below, it can be found that the overall temperature uniformity is relatively good, only the temperature of the hole on both sides is slightly lower than that of the central four-chamber, which can be related to the position of the sensor and the gap between the heating film and the sliding chip.
Module temperature control precision
The temperature control module is controlled to 67 ℃. After the constant temperature for 10 seconds, the time is 30 seconds, and the maximum temperature and the lowest temperature are recorded. Half of the difference between them is Δ temperature. Delta elephants (i= 1, 2, etc.). 6) the absolute value of the difference between the measured value and the set temperature should be less than 0.5 ℃.
Temperature accuracy
Set the temperature control module to control 67 ℃, keep the temperature constant for 10 s, time 60 s, and record the temperature every 10 s. 6) the absolute difference between the average temperature and the set temperature should be less than 0.5 ℃.
Biological function Test of temperature Control Module
Biological Function Test by LAMP Reaction
The LMAP reaction reagent was labeled with fluorescent dye and put into four chambers in the center of the sliding chip together with the negative control group. The temperature control system was heated to 67 ℃ to maintain 15min and provide the temperature needed for LAMP reaction.
The result of fluorescence microscope as shown in figure 26 shows that obvious fluorescence can be observed in the chamber where LAMP reagent is located, and the difference in fluorescence brightness is small, while there is no fluorescence in the chamber where the negative control is located.
The fluorescence intensity was detected by enzyme labeling instrument, and the results were shown in figure 27. It was found that higher fluorescence content could be detected in the chamber where LAMP reagent was located, and the difference was small, while there was no fluorescence in the chamber where the negative control was located.
This shows that the temperature control module built for this project can meet the requirements of biological detection. Temperature control module provides appropriate reaction temperature for biological reaction, and can achieve the desired detection effect. At the same time, the reaction conditions (temperature) of biological reaction in different reaction chambers are basically the same, which can ensure the parallelism of biological experiments.














Creat BY Beijing Institute of Technology

Contact us

Follow us

E-mail:igem_bit@outlook.com