#The3rdLiChuangContest#Portable CNC power supply for debugging

1. Name of contestant (required):
Fan Ang
2. Name of company or school (optional):
Siemens
3. Current position or title (optional):
Supervisor
4. Name of entry (required):
Portable CNC power supply for debugging
 
5. Briefly describe your idea and work (required):
Working range: Input AC220V +-10%, output 0...30V adjustable, current 0...0.8A adjustable
Adopts isolated flyback switching power supply structure, high efficiency, primary and secondary isolation
Small size, about the same as a mobile hard disk, easy to carry
Built-in user-friendly HMI, LCD display, encoder adjustment
With additional and secondary development functions, such as battery charging, VI adjustable curve output, 4..20mA current source, industrial field signal source, etc.
 
6. Materials to be used in LiChuang Mall (required):
STM32F051C8
PC817C
Rotary encoder switch EVE-VCGJL016B
KF5N60F
 
7. Materials not from Lichuang Mall or other supplements to be used (required):
1.8-inch LCD module
EFD25 transformer
8. Name of EDA tool software to be used (required):
LCEDA
[Contents to be filled in when the work is officially published (enter the design stage after successful registration)]
 
1. Introduction of the work
For example, it can include but is not limited to the following:
1. The appearance picture of the whole machine or the welded and assembled PCBA picture;
Appearance concept diagram:
Internal structure concept
Internal structure concept
 
 Special reminder here: The safety distance between the primary and secondary sides is not enough in the current state! Please do not imitate!
 
  
2. Research background, purpose and function, and market application prospects of the work;
Currently, most debugging power supplies on the market use a structure of power frequency transformer plus power tube linear price reduction, with a power output range of 0~30V*3~5A, and focus on accuracy; but the disadvantage is that it is large in size and heavy in weight and is not convenient to carry.
I need a small power of 0~30V*1A, not particularly high in accuracy, and a portable power supply for on-site debugging. Similar products on the market have large power and single functions.
Therefore, the key features of this entry are set as "multi-function", "portable" and "switch type".
The market prospects are mainly for engineers with needs like mine.
 
3. What are the highlights of the work in terms of innovation, fun, practicality and even public welfare?
The human-machine interface uses TFT LCD instead of the traditional seven-segment digital tube, which can provide more information (such as curve display). After opening all information, it has stronger secondary development potential
. The operation uses a rotary encoder, which is convenient and fast. The parameter adjustment is proportional to the rotation speed (the faster you turn it, the faster the parameter changes, showing an exponential curve), and the user experience is better.
The shell is made of aluminum alloy profiles, and the LCD screen is also a standard size, which is convenient for procurement. It will be very convenient for everyone to DIY or imitate in the future.
 
2. System architecture diagram
Use a flowchart or mind map to describe the composition structure of your work, that is, the scheme diagram.
 
Principle structure block diagram:
 
3. Description of the hardware part
1. Attach pictures or source files of the schematic diagram & PCB physical diagram (the official recommends that you try to upload the source file). If it is a picture, please make sure that the picture is clear and legible;
2. Use text to roughly explain the implementation principle of the work and the working process of the system.
3. Indicate the name of the EDA tool software used and attach the design link.
 
This design uses LCEDA, which is very easy to use and extremely suitable for the design and implementation of this work.
Attach the project link for everyone to communicate:
 
https://lceda.cn/arokh/dcp3001
 
Schematic diagram:
PCB:
http://club.szlcsc.com/article/downFile_C05AA4E03ACF07C3.html
A few considerations:
1. This power supply is only a CNC power supply rather than a digital power supply, so the feedback loop is still in analog form. The CNC part completes the voltage and current settings and calibration.
2. In terms of safety, I was entangled in the grounding problem. Considering the heat dissipation, I want to use an aluminum alloy shell. If the grounded power socket is very large. Consider the internal double insulation for the time being, and then talk about
 
the block description if it doesn't work:
 
1. Auxiliary power supply part:
  
Considering that the characteristics of the auxiliary power supply are low power, multiple outputs, and isolation, the flyback topology is specially selected.
 
The solution uses PowerInt's LNK62x, which has a simple structure and low cost, and the solution is mature and widely used in home appliances.
 
In addition, the supplier of this solution provides a complete automatic design tool PIExpert, which is very convenient for design optimization.
 According to the structural limitations and the number of output paths, the transformer uses EE13 core and 5+5 skeleton. The transformer design output by the software is for reference only. On this basis, I have further optimized the primary and secondary creepage distances and pin allocation. See the schematic diagram for detailed allocation.
 Attached design engineering file:
LinkSwitch-CV_PIDesign1.rar (Downloads: 1121)  
2. Main power stage:
 This part still uses the classic flyback structure. The main controller is also the classic UCx845. I believe that friends who are familiar with power supplies are already familiar with the x845 IC, so the detailed design description is not repeated.
The following is a screenshot of my homemade flyback design gadget (for various reasons and copyright reference issues, it is inconvenient to share the source file, please forgive me).
If you need to use the UCx845 automatic design tool, I recommend TI's Webench, which is also very convenient!
Input and output parameter settings:
  
transformer parameters (except for the incorrect air gap, other parameters are almost the same). There are actually some problems with the RC absorption circuit, which need to be corrected:
use EXCEL to draw the power level Bode diagram
to adjust the compensator, as well as the compensated Bode diagram (not yet fine-tuned).
I personally think that the switching power supply is 30% design and 70% debugging. Although the design stage needs to be done well, there are actually quite a lot of variables, and more effort is needed for later debugging.
 
3. Voltage and current feedback loop:
This part of the circuit is my "Spring Breeze Power Supply" adapted from the predecessor Chunfeng (I would like to thank the predecessor here!), and I made some changes based on my understanding.
The upper part is the voltage loop, and the lower part is the current loop. The two parts drive the optocoupler after being "wired-OR" through the diode.
The structures of the two parts are actually the same. After the voltage and current are sampled, they first pass through the first-level 0dB drive, and then output to the second-level TYPEII compensator for compensation and optocoupler drive.
Not all capacitors in the figure are required, and the parameter values ​​are still to be refined and adjusted.
It is particularly important to note that the first-level TYPE I capacitor and the subsequent TYPE II compensation should not exist at the same time. Otherwise, there will be a situation where multiple poles are superimposed (>-40dB attenuation)
 
 
4. Bill of Materials (BOM List)
List the main components (key components are sufficient) used in your work, such as microcontrollers & ARM chips, application-specific integrated circuits (ASICs), sensors, functional modules, etc.
If the listed chips are from our LiChuang Mall, it is best to write the product number of the device or attach the corresponding purchase link.
 5. Description of the software part (optional)
If your work involves software, please list the corresponding software workflow diagram, as well as the routines and source code of the key parts of the work (if you want to open source, please upload all the source code).
   DCP3001_V01.rar (Downloads: 1779)  
VI. Work Demonstration
Please upload the functional demonstration of your work to Tencent Video and edit it to this post (or attach a video link). Uploading the video as required will earn you 10 points. For details, please refer to the activity rules.
 
 
If it cannot be played, please click the link:
http://v.qq.com/x/page/j0757r8o0ep.html
 

 
VII. Summary
For example, your experience in completing the work, the technical problems or debugging experience you encountered, the future plan of the work, and your suggestions and opinions for our organizers.
 
-->When designing the transformer with PIExpert, the parameters were changed but not re-optimized, resulting in incorrect transformation ratio, high reflected voltage, and stuck for two days;
-->ADC sampling seems to have encountered aliasing problems, and low voltage cannot be measured. Time is too tight this time, so let's do the function first. If we improve it later, we will add a low-pass filter to filter out the switching frequency;
-->The feedback loop does not oscillate with the value calculated for the first time. In order to verify it, the compensator was removed, and the Vds waveform shook decisively. I will go online and adjust it carefully when I have time. The purpose of building a loop experimental platform this time has been achieved;
-->The schematic diagram will be updated later, and the time node cannot be given at present;
-->As a middle-aged greasy uncle, there are too many trivial matters, and it is difficult to calm down and do some hobbies. I don’t know if I can participate next time. Anyway, I have tried my best this time; For
 
 
more project details, see the link: http://club.szlcsc.com/article/details_16056_1.html
This project belongs to the Lichuang community "arokh"