HLK-PMxx protection: choosing the right MOV / fuse value
I am still working on a MySensors project which uses an HLK-PM03 module to provide 3.3V DC from mains to my circuit. This project was almost finished when I came across posts on forums recommending adding a fuse / varistor in front of the HLK-PM03 module to protect it.
So I dug that subject a bit more and I came across a recent HLK-PMxx datasheet in which this is mentioned: https://datasheet.lcsc.com/szlcsc/1912111437_HI-LINK-HLK-PM12_C209905.pdf
The fuse and varistor are really needed to protect the circuit ; the capacitor / inductance are apparently optional. In the datasheet, the constructor recommends using a 0.5A slow blow fuse and a 350V varistor (10D561K).
But I found a discussion on the MySensors forums with completely different recommendations. This one typically: https://forum.mysensors.org/topic/1607/safe-in-wall-ac-to-dc-transformers
The first post shows a diagram with a 0.3A slow blow fuse and a 250V varistor.
The choice of the "right" components really puzzled me recently, so here is my question: which varistor / fuse value should I select, at the end?
I'm particularly interested in knowing how varistors (MOVs) should be chosen, because I have read so many conflicting arguments on different forums...
A post of @skywatch I recently found (https://forum.mysensors.org/topic/9790/has-anyone-else-seen-a-varistor-mov-failure-in-a-power-supply?_=1597873012895) seems to confirm the choice of the manufacturer regarding the 350V RMS MOV.
Finally, note that I live in France ; mains voltage here is 230V AC +/- 10%.
Thanks in advance for your answers!
skywatch last edited by
@Encrypt The capacior and inductor may be optional for some useage, but they are there to reduce smaller spikes and interference on the power line entering the power supply and I strongly recommend that you use them for your application.
As for the mov, your mains voltage is 230V+/- 10%. This gives us a range of 207V - 253V AC. A 250V mov should start to conduct at 250V, so in this case it could blow even if the AC is within specification.
Usually the mov is there to protect against large spikes or surges from lightning strikes or faulty equipment supplying power to your area. In my opinion the 350V in the statsheet you have is spot on for this application.
Hello @skywatch !
Thank you very much for your answer! It is now clear for me!
Since one of my power supplies (the kind of black brick, like the ones that come with laptops) failed short because the varistor inside died, I no longer feel confident adding varistors to my designs.
@monte wouldn't that just blow the fuse?
@mfalkvidd it should, but in my case fuse somehow stayed intact, but UPS it was plugged into went into protection mode.
Now I am using Mean Well IRM-03 modules in my designs. They are only slightly more expensive than HLK, but have built-in fuse and overvoltage protection: https://www.meanwell.com/Upload/PDF/IRM-03/IRM-03-SPEC.PDF
Which kind of varistor did you use? What was its rated voltage?
As @skywatch mentioned, if you chose a varistor which value was "too close" to your mains line, it probably got triggered too much...
Also, if your fuse was "too big" (500mA or more), it's probably the reason why it didn't blow.
monte last edited by monte
@Encrypt I didn't choose anything, as I said it was a fuse and varistor in an enclosed PSU which came with some laptop. When I've opened it I've found blown MOV, after changing it and old main cap power supply became functional once again.
The thing I am saying is just tha MOV is a device conneted in short between live and neutral and with ability to fail short and not open it poses potential risks.
skywatch last edited by skywatch
The thing I am saying is just tha MOV is a device conneted in short between live >and neutral and with ability to fail short and not open it poses potential risks.
No it does not pose 'potential risks' - it removes a very real danger to the equipment and persons using that equipment.
A MOV is OPEN circuit when new. It only goes into conduction when a voltage greater than the value on the package is reached. When this happens it shorts live to neutral blowing the fuse and protecting the rest of the circuit and the person(s) using the device. This way it shorts the voltage surge to neutral (often ground) and in blowing the fuse it disconnects the power from the device.
Without a fuse in place it would be an issue, but that is why you ALWAYS use a fuse with a mov.
monte last edited by monte
@skywatch yes it does. The fact that you make effort to mitigate those risks, doesn't guarantee you've got rid of them completely. What if fuse blows second after your MOV caught on fire?
skywatch last edited by
@monte You will never get rid of all risk, it is a fact of life.
Considering that the diagram above shows a 500mA fast blow fuse, I doubt it will last a whole second with full mains across it. If worried about the fuse then use 2 or 3 in series.
As for the fire risk even the first link says "Under normal utility voltage conditions, this is not a problem."
The fact is that this is the best solution that gobal engineers have used for decades. It is an industry standard for protection. It is even used in Fluke multimeters to protect their input electronics from over voltage.
If you feel that there is a better solution with the same or better protection and lower risk, then please share it.
@skywatch do Fluke meters work unattended in enclosed cabinets or walls, using varistors for power supply 24/7?
As I've said I prefer using a better power supplies with built-in over-voltage protection. As for how it is achieved without a varistor you can find plenty information by yourself using google.