There are some problems with typesetting. It is recommended to browse and read at this website:
https://oshwhub.com/yANgZEN/buck-qu-do
When using the soldering station, we often forget to turn off the power and allow the soldering iron to heat up overnight.
Then there will be a waste of electricity, safety hazards, and oxidation of the soldering iron tip.
For these reasons, I set out to develop this soldering station.
This method is available in the ready-made DIY T12 handle and is simple to implement, but the T12 soldering iron tip is relatively expensive.
Existing products include a Hall sensor installed on the soldering iron stand to detect whether the soldering iron is placed on the soldering iron stand to achieve a sleep function, but the cost of customizing the soldering iron stand is also high.
Isn't there a simple and general method? Here’s my solution
First of all, let’s popularize what is [thermal resistance]?
First, let’s understand the principle of the radiator: it disperses the heat from the heat source so that the heat can be transferred to the air faster to complete the heat dissipation.
Then come back and look at the soldering iron tip. Is it equivalent to a radiator with a small heat dissipation area?
There is a parameter to measure the difference between radiators, called thermal resistance, and the unit is °C/W. In fact, the power device technical manual has instructions.
As shown in the figure above, the first line is the thermal resistance to the case. For example, if a good external heat dissipation is used, the internal temperature will rise by 0.58 degrees for every 1W of power.
The second line is the thermal resistance of the junction loop. It’s a mouthful, but I didn’t translate it. As the name implies, the internal thermal resistance to standing air.
That is, without external heat dissipation. For every 1W of power, the internal temperature rises by 62 degrees.
It can be seen that there is a direct proportional relationship between temperature and power.
Then there is the cost consideration. When measuring power, you only need to measure the current and voltage. Measuring the temperature is the job of the soldering station. No additional cost is required.
Of course, if the resistance of the heating core is relatively stable, Ohm's law can be used to convert the power. No need to measure current.
Generally, all circuits will be taught in teaching, but there are some very common circuits, so there is no need to reinvent the wheel.
The human-computer interaction part can be seen in my project from the last Lichuang Competition. Basically copied.
Let’s talk about what’s different.
First of all, if the soldering iron is driven by AC power, it is usually a thyristor, and if it is DC, a field effect transistor is used. This is a DC power supply, choose a field effect tube.
Then 936 has two heating cores. A1321 and A1322 respectively.
The former heating core resistance will change with temperature from 3 to 10 ohms, while the latter is 15 ohms.
The power mark of the former is 60W, while the power mark of the latter is 45W.
However, directly driving the 4 ohm cold resistor with PWM will cause a huge impact on the power supply. Causes power output protection.
What I thought of at this time was the BUCK step-down circuit. Since the inductor will limit the current, the impact on the power supply will be smaller.
The problem comes again. If the upper arm is driven by a charge pump such as IR2103, it cannot achieve 100% duty cycle, which means it cannot operate at full power.
what to do? We try to turn the BUCK circuit upside down, as shown in the picture below.
It can be seen that the entire output part is floating, and only one end is connected to the 24V power supply.
And it is feasible to keep T1 on all the time.
D12 and C24 are peak detection added to avoid installing C12 and C13. Trade-offs can actually be made.
ACS712 is a Hall effect current sensor. If you use a sampling resistor and differential amplification, you will need at least 5 resistors, and there are only 2 capacitors on the periphery of the chip. The size is reduced.
LCSC is for sale, it is recommended to read the technical manual. Foot-to-foot compatible and cheaper is the CC6902 .
There are two types of soldering iron tips mentioned above.
A1321 is a PT100 resistor used as a temperature sensor. Measuring resistance requires power, so U4.2 forms a constant current source circuit.
Because NMOS is used, the response will lag, so the phase compensation capacitor C19 must be welded, otherwise it will oscillate. (Adjust RV2 so that the output current is 200uA)
The A1322 soldering iron tip is a K-type thermocouple, 4096uV/100°C.
It can be seen that it is only 12mV at 300°C, which is too difficult for the ADC of the microcontroller to distinguish the specific temperature.
So U4.1 is needed for in-phase proportional amplification. As a trick here, three common resistors are used to create a 100-fold amplification.
Interested students can calculate it by themselves. In fact, there is a 9.9kΩ resistor. LCSC has it in stock, but I don’t have it in stock.
By the way, this GS8552 is one of the sponsors of the last competition, cough cough cough~~~ I have said too much.
Let's compare it
Although I really want to use the op amps sponsored by Fuman, Fuman does not have a product line of precision op amps.
As can be seen from the figure above, the input bias voltages of the two differ by a factor of 1000.
[Realistic op amps are not ideal. During normal operation, there will be a voltage difference between the non-inverting and inverting input terminals. This is the input bias voltage]
If LM358 is used for amplification, the parameter estimation based on the technical documentation may deviate by 100 degrees.
Moreover, this op amp also has a feature, which is rail-to-rail input and output, which can make full use of the ADC dynamic of the microcontroller.
Will LCD1602 light up? ? ? I doubt [causing war]
If you want to use it flexibly, first create a ring buffer
uint8_t lcd_buf[256];
uint8_t lcd_load_head=0;
uint8_t lcd_load_tail=0;
Then store the data to be sent
void lcd_write(uint8_t dat,uint8_t cmd)
{
lcd_buf[lcd_load_head++]=(dat>>4)|(cmd?0x10:0);
lcd_buf[lcd_load_head++]=(dat&0x0f)|(cmd?0x10:0);
}
Then make regular calls to send data. Since I have key reuse here, I read the keys first and then send them.
void lcd_maintain()
{
lcd_dat_load(0xff);
GOIO_input_init();
KEY_ODR=0;
asm("nop");asm("nop");
asm("nop");asm("nop");
key_load();//读取按键
KEY_ODR=1;
if(lcd_load_head==lcd_load_tail)
return;
EN_ODR=1;
GOIO_output_init();
lcd_dat_load(lcd_buf[lcd_load_tail++]);
asm("nop");asm("nop");
asm("nop");asm("nop");
EN_ODR=0;
}
After reading this, do you still dare to say that you knew how to order LCD1602 before?
void pid_maintain(float TEMP_tiger,float POWER_tiger)
{
static PID pid_P = {0.45 , 0.15 , 0.01 , 0, 0, 0, 50};
static PID pid_T = {3.75 , 0.02 , 0.02 , 0, 0, 0, 50};
static float temp_P=0,temp_T=0;
static uint16_t PTC_temp[4]={0};
static uint8_t PTC_num;
static float pwm=0;
if(TEMP_tiger==0||POWER_tiger==0)//输出为零,清除所有参数
{
Tim1_pwm_set(0);
pid_P.ek_1=0;
pid_P.ek_2=0;
pid_T.ek_1=0;
pid_T.ek_2=0;
PTC_temp[0]=0;
PTC_temp[1]=0;
PTC_temp[2]=0;
PTC_temp[3]=0;
pwm=0;
return;
}
if(PTC_temp[PTC_num]==0)//平均值预加载
{
PTC_temp[0]=
PTC_temp[1]=
PTC_temp[2]=
PTC_temp[3]=
adc_get_ptc();
}
else
{
PTC_temp[PTC_num++]=adc_get_ptc();//4次平均,消除噪声
}
PTC_num&=3;
temp_P=PID_Increase(&pid_P,0.0001f*adc_get_uout()*adc_get_iout(),POWER_tiger*10);
temp_T=PID_Increase(&pid_T,(PTC_temp[0]+PTC_temp[1]+PTC_temp[2]+PTC_temp[3]+2)/4,TEMP_tiger);
if(temp_T400)
pwm=400;
if(pwm<0)
pwm=0;
Tim1_pwm_set((uint16_t)pwm);
}
Note that this is a relative PID.
The two variables here limit each other, and the core algorithm is to take the minimum value to output while stabilizing power and temperature.
Someone wanted the code for the buzzer music before, but the decoding part is too lengthy, so you should read the code yourself.
There is also a lot of relevant information on the Internet that can help with understanding.
Because this series of microcontrollers is very popular, there are many chips that are foot-to-foot compatible for power supplies and peripherals, and they are all relatively cheap (before price increases).
In order to learn more, I will carry out transplantation between multiple microcontrollers.
I have also written the code for the STM8 platform, but the 8KB flash is not enough for me to waste.
Therefore, the complete function is only available in HK32F030MF4P6, and the corresponding code is included in the attachment.
Thanks to Hangshun Chen Gong for his technical support. I turned in my homework relatively early. If you can't drive the peripherals, you can learn from my code, but don't forget me when the code is open sourced for the second time. At least post the link to the web page.
Be careful not to solder C15 when programming HK32F030M, because the ADC and programming port are reused, and the filter capacitor will bypass the programming data, so remember this.
The above is a sketch of the logic function. Let’s explain it by comparing it with the 1602 screen.
C19 has been corrected in the production files, so you can use it with confidence. Because the stock of electrolytic capacitors is insufficient, the remaining capacitors are soldered to important positions. It has little impact because several of them are connected in parallel. D3 also does not have SS56, but has two SS36s on top. Pay attention to the 12V fan interface. Because we do not intend to PWM the screen backlight, only the fan part is welded.
As for the installation diagram, the shell here is too thick, so the nut cannot be installed on the aviation seat. The model has been corrected, and the 5mm wall thickness is a bit thick. In the future, a thinner shell will be redesigned to reduce 3D printing costs. If it is PLA, it is recommended to use this thickness, and the filling degree can be reduced. R26, R27, and C10 are reserved for STM8. No welding is required for non-STM8.
The air duct design is specially reserved at the bottom, and the heating element is placed in one corner.
(If you are pursuing efficiency, you can replace the inductor with an iron-silicon-aluminum magnetic ring of the same size and specifications, which has a larger rated current and lower iron loss.)
The other side is the op amp, thermal sensor, and buzzer to ensure that it is not disturbed by temperature as much as possible while allowing the buzzer sound to come out through the grid.
EAGLE9.6 is used here. It’s not that I don’t want to give the source project. Even if I give it, I can only look at it and not move it. [It is recommended to regenerate the PCB here in the EDA schematic diagram to draw]
The annual fee for the genuine software is US$500, and the free version is only 8x8cm. To apply for the educational version, you need a school seal to prove it.
The pirated software is still stuck in the V7 version.
My project here is 10x7cm. Therefore, amateurs can only change the software and draw again.
In addition to the built-in packaging library, it looks good and has smoothness.
I have never experienced the professional version of LCEDA, so I don’t know if the smoothness is comparable. But the built-in packaging is definitely not as refined as EAGLE.
The 936 soldering iron handle is available on Taobao for 11 yuan with free shipping. Find it yourself.
The soldering iron tips are available for 2 yuan each with free shipping, or for a set of 10 yuan with free shipping. It is recommended that you try to find the one that suits you and then buy more expensive soldering iron tips, which are less likely to oxidize.
My recommended order: knife, horseshoe, chisel, point. The tip has a small specific heat capacity and often cannot be welded. And it’s not easy to hang tin.
Here is a power calculation formula: P=U^2/R. That is, at 30V, 15Ω, the maximum power is 60W.
Therefore, the input voltage can be increased to increase the power. The limit input voltage of this circuit is 35V.
Especially the A1322 heating core is only 40W at 24V.
It is recommended to buy a 24V3A iron-cased industrial power supply with free shipping for 20 yuan, which is safe and guaranteed.
If you must buy a disassembled bare-board power supply, it is recommended to first check whether the high-voltage filter capacitor has 1uF per watt. Otherwise, the power is likely to be false.
There is a maximum power limit on the software. Assuming your power supply is 24V2A, set the maximum power to 50W to prevent power protection and reduce the output voltage.
The power is close to 20W during normal use, mainly because it will heat up to the set temperature at full power in the first few dozen seconds after startup. If it's not stuck in a box, convection heat dissipation won't be a problem.
V02:
All reference designs on this site are sourced from major semiconductor manufacturers or collected online for learning and research. The copyright belongs to the semiconductor manufacturer or the original author. If you believe that the reference design of this site infringes upon your relevant rights and interests, please send us a rights notice. As a neutral platform service provider, we will take measures to delete the relevant content in accordance with relevant laws after receiving the relevant notice from the rights holder. Please send relevant notifications to email: bbs_service@eeworld.com.cn.
It is your responsibility to test the circuit yourself and determine its suitability for you. EEWorld will not be liable for direct, indirect, special, incidental, consequential or punitive damages arising from any cause or anything connected to any reference design used.
Supported by EEWorld Datasheet