LNK306 on compact design
Because I need a very compact design, so I choose LNK306 as the AC-DC converter to power a 3V MCU, I following the reference design (high side buck direct feedback) shown on the datasheet, and since I can’t find a SMD fusible resistor, I’ve replace it with a fuse and a regular resistor(with same value), at the end side, a 3.3v LDO (AMS1117) and some bypass capacitors were added, together they power everything on the broad(see the schematic).
A RF module(CC1101) is also on the broad in which the MCU control. It’s a 6x6 cm two layer broad, the power module sit on the bottom layer, the MCU and the RF module are on the top layer, all components are SMD.
As I connect the broad to the grid, I instantly got a small band on the resistor(it become fully conduct, sometime after replace it won’t band, don’t know why), but the power module manage to work and I got 3.3v at the end side.
The MUC is work but the RF module is very unstable(sometime work, but mostly not), so I use an USB TTL(CP2012) connect my PC to the MCU through UART, where I’ve write some code to control and monitor the broad(like a console), but the word(or characters) come back from the console are garbled, like you give a command and it execute properly but the return message contain queer characters which is no sense, seem like they have been disturb by something.
If I use the computer USB as power source, everything will back to normal and all these queer things will disappear.
On the last trial I got a band on the resistor again, but this time I burn my PC, this was a month ago, now I’ve brought a new computer.
Please help me, any idea/advice/guidance would greatly appreciate
Comments
The resistor got band(blow up) is R1(see on the uploaded sch), it’s intend to replace the fusible resistor(the RF1 on reference design), because I cannot find a SMD fusible resistor, I’ve replace it with a 250v/0.5A SMD fuse and a common SMD resistor with same value. It could be the surge voltage blow up the resistor, as this situation(blow up) does not always happen, and after it blow up the resistor become no resist(check by multimeter), but the power module still work. Anyway, on the end side(DC side) the output voltage has a small continually fluctuation(as check by multimeter), this voltage fluctuation does not happen, if I use computer USB as the power source. So in my conclusion, the RF module doesn’t work well is because of the unstable power source, and the MCU UART got garbled is a result of interference of the power module.
I wonder is there anyways that I can smooth and stabilize the power source, just like the power that comes out for the computer USB.
On the application note(AN-37) there’re some topology shown, some of them are using optocoupler, can the power be more stable if I use optocoupler topology.
Can I replace the RF1 fusible resistor with just a fuse(without the resistor coz they keep blowing up)
Don’t worry about the burn pc, it is not a safety issue, the word burn is just a metaphor, it is not really catch on fire, just the mother broad got shock and stop working, nothing really burn. It may cause by the surge voltage or the testing broad short circuit, also there will be no any connection to the system except power on the final product. Isolation power circuit usually need a transformer, this will increase the cost size and it’s a through hole component.
Great appreciate for your help.
Hi,
You will still need a resistor to limit the inrush current during startup, when the initial voltage on the capacitor is zero voltage. I know that you see different behaviors using using this power supply compared to the power from the USB port. Do you have the waveforms on the output voltage, that I can understand what exactly "the smal continually fluctuation"?
Sorry, I cannot provide this waveforms, I’m not a professional like you, I don’t have the equipment, all I can do is following the reference design. The voltage come out from the power module is fluctuate(changing) from 0.1 to 0.3 volt , this is observed by multimeter, and it doesn’t happen on the power that coming out from PC USB.
I was hoping that you can give me some tips or advices, something like add more capacitor, indictor filter …etc in a right place, or using other topology, so that the power can be more stable as close to the PC USB power.
Thank a lot.
Hi,
Understand the situation. The difficulty is if there are some issue, we really need the waveforms to find out what really happen to the power supply. Looks like you just need something that works for you, you really do not have much concern about the cost. So Here is my suggestions:
1. Design the power supply with higher voltage output, maybe 10V. Use a post regulation like the LM7805 to generate the 5V.
2. Looks to me, you want to get similar performance of output voltage as the PC USB power. You may need to consider an isolated power supply. As the PC USB is supplied from the battery of the PC, which is already isolated from the line voltage (high voltage) through the adapter. You could design a isolated flyback power supply or purchase a power module from the market to get 5V.
Hi,
I read through the post, it looks like it is a system level configuration issue, as you got the right output voltage through the power supply, but you had some communication issue of the coding. I am not sure I fully understand the issue, but if you see some specifical issue related to the power supply, it may help me to understand it and help you on that.
Would you please help me to understand what does it mean " I got a band on the resistor", which resistor are your talking about? I also saw you burn your PC when you connect the system to your PC, is it a safety issue, do you consider design a power supply with isolation insteadd of non-isolation buck circuit?