- Semiconductor BusinessHOME
- Products and Services of Macnica,Inc.
-
technical information
-
Events and Seminars
- Handling Manufacturer
- Support
- Inquiry
- Click here to purchase products
- Semiconductor business e-mail magazine registration
Narrow down by specifying conditions
現在2149件がヒットしています。check
Hello to everyone reading this blog. FAE of Gassan Shu is.
In previous articles, LT8609A We have introduced the features of the , ideal operating waveforms, and measurement results of the evaluation board and the homemade board. last time I compared the measurement results of the evaluation board and my homemade board and experienced the gap between the ideal and the reality. So this time, I will focus on the board design aspects such as board layout and wiring.
At the time, when I designed the board, I didn't understand the key points of layout design. As a result, the layout was far from ideal, and I believe that's why I wasn't able to observe the waveforms I wanted. In this article, I'll explain the basic points of layout design and specifically explain the problems with my board.
Layout design points
First, let's look at the important points in layout design using the circuit diagram and layout diagram.
circuit diagram
Layout diagram
Using the diagram as a reference, we will introduce some important points to aim for when designing a layout for the LT8609A.
1. Place Cin (input capacitor) close to the input pin
Because the input hot loop has the largest current change, placing the input capacitor as close to the IC as possible can help suppress parasitic inductance components and noise.
2. Peripheral components such as inductors should be placed on the same surface (surface layer) and connected with thick, short wiring without vias.
Connecting vias increases the parasitic inductance, making it more susceptible to noise. Connecting in a straight line on the surface leads to stable operation.
3. SW (switching) nodes should be thick and short.
Using thick wiring over the shortest distance will help reduce EMI noise.
4. Place the Cout (output capacitor) as close as possible
If the output capacitor is located far away, it may cause a decrease in load response and an increase in ripple noise.
5. Proper GND
It is recommended to connect the IC's GND pad directly below to the solid GND on the board.
With these points in mind, let's take a look at a circuit board I actually designed.
The board layout I created in the practical class
From here, I will introduce the board layout that I actually designed. The part framed in red in the diagram below is the power board. (Sorry it's a bit messy...)
Power supply board (front)
Power supply board (back side)
Next, we will look at specific problems with this board layout.
Layout issues
The input capacitor is on the back
Having the input capacitor far from the IC causes parasitic inductance components and increases noise.
・ The wiring of the switching node is thin and long
The parasitic inductance component increases, and the noise also increases.
・ GND wiring is thin and long
The wiring path from the IC's GND pad to the PCB's solid GND is thin and long, which also causes parasitic inductance and increases noise.
We have listed some problems with the board in relation to the layout design points mentioned above.
In this practical lesson, we used a universal board, which made it difficult to achieve the ideal layout design. In particular, because the power supply IC was mounted on a DIP conversion board and then soldered to the universal board, it was not possible to place the input capacitor and inductor in close proximity to the IC. In addition, the LT8609A does not have a GND pin, so the only GND for the IC was the thermal pad.
As a result, the only way to connect the GND pad to the GND plane on the universal board was via the pins on the DIP conversion board, which resulted in a long wiring path to the GND plane. Furthermore, due to board space constraints, some of the components had to be placed on the backside, forcing the wiring to go through vias. We believe that these layout issues were the cause of the waveforms that were far from ideal.
Summary of the training
The board I made this time was not ideal, and as I mentioned in my last blog, it did not have a stable output. However, I managed to get it to work during the training period.
In this workshop, we created a maze escape machine and were able to actually run it. It's not a pretty shape, but I'd like to briefly introduce it to you.
A photo of the vehicle is attached below.
Body (top view)
Body (side view)
Next, here's a video of the actual run. The robot was able to reach the finish line while avoiding the walls.
Through practical training
We had three months to create it, but I'm glad we were able to make it into a reality. Although I felt a sense of accomplishment, there were also some parts that felt incomplete. This practical training really made me realize how difficult it is to design something from scratch. In particular, the quality of the power supply design was not satisfactory, and I don't think I even read the datasheets properly.
This blog has been divided into three parts and has provided a basic explanation of power supply design, including stories of my own failures. Thank you to everyone who has read this far. I hope that my articles will be of some use to those designing power supply ICs. I am still an inexperienced FAE, but I will continue to work hard every day to improve.
I hope to see you again somewhere.
List of articles about making maze escape machines
Introduction to the LT8609A for beginners
・ LT8609A operating waveforms that even beginners can understand
・LT8609A board layout design and practical training summary