Wasting power in SMPS in high volume sales consumer goods.

[eem2am]

Hello,


I am working in the SMPS design and development department of a huge global consumer goods company (you will know the name, but I can’t say it).
The SMPS’s are offline, and of powers from 5W to 200W.

I am just designing a 42W offline, isolated flyback. The product is likely to be permanently connected to the mains, so it has to have a standby power draw of less than 300mW.

I had just finished the schematic , when the boss came over , and handed me a 10 Ohm, 2Watt, Wirewound, flameproof resistor (“Bathtub” type…you know the ones, they’re white and look like they’re made from the same stuff that they make space shuttle tiles out of )
…..the boss told me to put this resistor after the PFC Bus capacitor, (at 400V) in series with the flyback.
-At max load/minimum line, this resistor will constantly dissipate 1.8W of power, due to the high RMS current of the flyback.

-I am told that adding such a resistor is essential because if the flyback FET goes short circuit, then something could blow off the board and potentially start a fire……..however, with the said resistor in place, there will be no such fire, because the PFC bus voltage will be dropped across the flameproof resistor, which will harmlessly go open circuit.

-But the product has a plastic case, and I wonder how any device blowing off the board could possibly start a fire?

In standby Mode, the Boost PFC stage is not running, and the PFC Bus is just the rectified mains peak voltage of 330VDC.

Is it really necessary to add such a huge, resistor?
There is after all, a 5Amp, slow-blow fuse at the mains input.

It seems like a waste of power to me…..is it?

 

[montech]

I agree. Any"brute force" answer to current trip is outdated. Why didn't he just suggest a crowbar circuit?:razz: You and I know there are 100 fail safe circuits you can add to prevent such an unlikely event, that would respond in a millisecond or 2 instead of waiting for THE RESISTOR TO BURN OPEN. If the concern is high this resistor would BURN before opening. A better way is to use a FUSIBLE resistor in series with the DC voltage supply to the flyback primary, calculated for the power requirements needed. Or 99 of the other ways to economically do the same thing. :idea:s anyone?

 

[eem2am1]

Hello,

Yes i agree, but by "fusible" resistor....isn't that ging to be just the same thing as the 10R wirewound resistor?

...or do fusible resistors have Positive temperature coefficient?

 

[montech]

Here is a .pdf from Vishay

There are more out there selection depending on your application(current needs)

 

[mtwieg]

I've heard this conversation a few times before, and despite all the neat low loss protection circuitry you can use, odds are none of them can really survive a lightning strike, which is why completely passive solutions like fuse resistors and such still get used today. They just work, no matter what. But there's no reason it has to dissipate 1.8W. You might try less resistance at a lower wattage, so it dissipates less during operation but still fails during short circuit.

 

[eem2am1]

The thing is mtwieg, why bother adding anything at all......if the FET fails short, then so what.......the supply fuse will soon blow.

...as the original poster says, there is an enclosure for the product.....its not as if a huge , vast shower of sparks will fly all over the customers carpets until they ignite, burning down the house.

Adding such a series resistor is health-and-safety gone too far.

In these days where we try to save mW's of standby power, we should not be using series resistors like this..................

even using a lower ohmic value resistor would be pointless....and in any case it would offer less protection in terms of spark prevention.

These consumer items with such series resistors need to be reported to the authories immediatley, as they are burning up energy resources.