UCC2891: Inrush current protection at the AC input

Serdar Köse

Part Number: UCC2891

My PSU design is based on PMP4447. I experience time to time fuse breakage at AC input at the start-up. That's why I'd like to have an in-rush protection that is low cost and smaller size. Could any expert advise which solution is best to use?

PSU powers CC2650 based board a 12V module that consumes about 0.5 Watt at max.

Here is my schematics:

Datasheet of the fuse: www.mouser.com/.../MF2410 Datasheet2013-247737.pdf

Bridge rectifier: http://www.mouser.com/ds/2/149/MB8S-1010921.pdf

Input capacitors ( C1, C2) http://www.farnell.com/datasheets/1879709.pdf?_ga=1.156179451.1778970300.1468686107

The fuse I use is a fast acting one, I am considering to change it with a resettable-fuse, e.g., a PPTC fuse. However, it is an outdoor application and the devices are primarily used in Nordic environment where the outdoor temperature can get -20C or lower, and performance of PPTC is based on the temperature. I wonder if it would cause problem that it can allow more current to flow at the low temps and can block the current when it gets warm at normal conditions. I am considering to use LVR055K http://www.farnell.com/datasheets/1902602.pdf?_ga=1.88011992.1778970300.1468686107

Another issue about how fast it reacts:

Do I understand correct that it takes about 90 sec to cut the power if 1.25A is applied? (it is too slow if it is the case)

Other option is to use a NTC Thermistor at the AC input, however, I again have the same temperature concern if the resistance at the cold weather could be a problem? I am considering this option B57153S330M http://www.mouser.com/ds/2/400/B57153_S153-525574.pdf

I'd highly appreciate your comments

over 8 years ago

0 Peter Meaney over 8 years ago

TI__Mastermind 31580 points

Hi Serdar

The UCC28910 evaluation module uses this part number as the fuse FKN1WSJR-52-4R7 which is a fusible resistor and should be very robust during start-up conditions.

Most fuses for this type of application where there is a relatively high in-rush current versus the nominal input current use an anti-surge or slow blow type fuse.

Regards

Peter

0 Colin Gillmor over 8 years ago

TI__Mastermind 25570 points

Hello Serdar

Normally one would use an anti-surge fuse at the input of an AC/DC converter like this. This is to ensure that the fuse doesn't blow when it sees the inevitable inrush current. I see that the bridge you are using has an 12t (fusing) rating of 5A2s so the anti-surge fuse should have a lower rating than this so that you can guarantee that the fuse blows before the bridge goes s/c. The next thing is that at this power level (< 0.5W) the input currents are going to be << 1A. I would suggest that you go to an anti-surge fuse rated at <5A2t with a rating much lower than 1A.

You are correct, the fuse you are using now will take 90sec to blow at 1.25A. Don't forget that the kind of event that the fuse is protecting against is a catastrophic failure of some component - the diode bridge or bulk capacitor going short circuit for example. Events like this will drive very high currents through the circuit and the fuse will blow quickly. You also need to make sure that your PCB tracks can carry this fault current - you won't get a safety agency approval if the PCB traces melt !.

As for inrush current limiting - at this very low power - have you thought of simply having an regular resistor ? 1 Ohm or so ? - I haven't done the calculations so I'm guessing here. The other alternatives involve a resistor with arelay (or solid stage device - MOSFET, TRIAC etc) bypass which is turned on once the inrush event is over.

This component is a safety critical one so I'm not going to give any specific advice because I don't know exactly what the circuit design is intended to do, but I'd expect that one or other of the general ideas I've given here would work in your case. Just be sure to have your solution verified during the safety certification process.

Regards

Colin

0 Serdar Köse over 8 years ago in reply to Colin Gillmor

Genius 3360 points

Hi Colin and Peter,

Thank you for the valuable information.

I see that I made a mistake by using a fast blowing fuse. I am planning to replace it with one of the below options:
- Slow blowing fuse: 3403.0165.11 www.mouser.com/.../typ_umt_250-793149.pdf
- 4.7 ohm/1W, fusible resistor: PR01000104708JR500 www.mouser.com/.../pr010203-239785.pdf

Would you recommend to use one over another?

As I see the fuse is smaller, doesn't cause heat and relatively more expensive (still decent price), suitable for P&P. On the other hand, the resistor requires larger (size and require some distance from the other components due to heat), cheaper, less power efficient, needs manual or wave soldering. Is there anything else to point out about the pros and cons? Is it possible to say if fuse or resistor would be more reliable?

Thank you,

Serdar

0 Colin Gillmor over 8 years ago in reply to Serdar Köse

TI__Mastermind 25570 points

Hello Serdar

This component is a safety critical part and as such I'm not able to recommend a particular part. I don't know which product and / or generic safety standards you are trying to meet but I would suggest that you have a chat with someone who is familiar with them and take their advice when selecting the part you eventually use.

I would make the general comment that I would probably choose a fuse rather than a fusible resistor if this were my design - (and then I would talk to an expert to choose the correct part !)

Regards
Colin

Power management

Power management forum

UCC2891: Inrush current protection at the AC input