Author: Bubu Wolong will bear.
Recently, I encountered a strange thing. When debugging a DC/DC circuit, it is clear that all DC/DC circuit parameters are exactly the same as the previous version, only 4V instead of 5V. If you encounter this kind of problem, everyone will do it! The following describes the debugging analysis process of our Bubu Xiong:
First, the hardware engineer’s design circuit diagram is as follows: Original Today’s headline: Wolong will IT technology.
I saw a DC/DC chips of BL8033CB6TR, which was searched online.
Go to the chip official website to search for it.
appears as shown below:
Click on the blue BL8033 to appear the corresponding page to download the datasheet is the area shown in the picture:
After downloading the manual, the 3.3V output reference circuit is provided. The reference circuit is as follows:
Calculate the formula according to the output voltage in the manual:
Can calculate the design output voltage = 0.8 * (1 + 40.2 / 7.68) = 4.9875 (V) very close to 5V, it should be said that very reasonable parameters, that is 啥 is 4V output, the previous version is also this parameter ah, Careful comparison of the reference circuit found that the BST capacitor recommended is 1uF and this example is 0.1uF, holding a try to change a 1uF welding finish or 4V, NND, this will do!
Try to replace the 10K of the loop with 0R. Anyway, it does not affect the voltage. It only affects the loop stability. It is replaced by 0R. The boot is 10V. Why is this, the resistance ratio is very large? It’s really hard to get old bears. Look carefully and see the problem. Original Today’s headline: Wolong will IT technology.
It turned out to be a bad thing for PCB layout engineers. He must have been fascinated with his neighbors last night, and his work is absent-minded! The old bear was so busy that he was going to cure him for a long time
pcb board as shown:
From top to bottom is R46/R48/R45.
And the silk screen on the PCB is as shown:
The original R48 and R45 silk screens were reversed, that is, when the pcb was patched, the patch manufacturers were reversed according to the silk screen stickers. The R45 should have been 10k 7.68k, and the R48 was originally 7.68k posted as 10k. The wrong output voltage = 0.8 * (1 + 40.2/10) = 4.106V, is indeed the previous 4V, and the 10K to change R45 actually changed the resistance value of R48 to make it 0R, so due to amplification The multiples are infinite, so the input 12V output is also close to 12V, so test it up to 10V, so the culprit finally finds that the pcblayout engineer reversed the silkscreen.
Lao Xi believes that the discovery of this problem comes from the analysis of the phenomenon! It’s bumping like a headless fly, you can’t figure it out, you have to analyze it! With more analysis, the experience is coming. However, it is also necessary to talk about this fascinated neighboring sister, even a screen printing will be misplaced PCB engineer. Such carelessness, which sister can see you, haha. Original Today’s headline: Wolong will IT technology.
Lao Xi believes that as a hardware engineer must have the details of the phenomenon, the ability to analyze carefully!
Here is the suspense of the second story: The circuit is correct. If the load is turned on, the power detection chip will be burned regardless of the load. The debugging process will be described in detail in the next story, and continue to pay attention to the old bear article!
Feel like, like a piece, collect it, welcome attention, forward! ! ! Thanks for the support
Original: Wolong Club Bubu Xiong.
Plus me more flash: wlh_it.