choosing fusible resistors

[larryg]

Hi all,

I was wondering if anyone out there has any experience specifying fusible resistors for use as circuit protection in a 3 phase watt hour meter. I've looked around on the web a good bit and have not found any definitive information for how to properly use these things.

Here are my particulars:

VAC input: 45 - 528
3 phases + Neutral
Peak Primary load: 10W (this is not constant, but rather pulsed during cellular radio transmission. Pulses are 1.154ms out of 4.616ms)
Nominal Primary load: 2.1W

This will be feeding an offline flyback circuit.

I know I need to have one in each phase + neutral, but I'm not sure what size to choose and what timing I would need.

Thanks for your assistance.

Regards,

Larry G

 

[Warpspeed]

I am not sure why you want to do this. Resistors will run hot, smoke and smell and destroy everything around them, long before they actually fail open circuit.

Conventional fuses can be purchased to have a wide range of overload characteristics from extremely fast blow, to very slow blow. They can also be soldered onto a circuit board just like a resistor. Something like this perhaps :

http://www.goldmine-elec-products.com/prodinfo.asp?number=G15924

 

[VEBill]

I'm confused (easily, but that's another story) by the specs. Using 3-phase AC, but only a 10 watt load? Is there another external load ? It isn't clear.

If the fused load is really only 2 watts average, then how do you fuse a current in the low mA range?

The other issue is using fuses to protect a 3-phase circuit. When one goes, what happens to the other two phases? Do they go berserk and sequentially blow their fuses before something is damaged? CBs for 3-phase are ganged so they all pop at the same time. Hard to gang fuses...

 

[larryg]

To help clear things up...here's a bit of background.

This circuitry will be in a 3 phase watt-hour utility meter. Not a lot of power will be consumed during the normal operation of metering. However, during the daily transmission of kWh data back to the server, small bursts of power will be used to support the radio transmission. There are super capacitors on board to help deal with the burst and provide hold-up in the event of power failure.

As for the use of resistors vs. fuses, I really don't have a preference. In single phase applications that I have seen, both fuses and fusible resistors have been used. Of course, I'm guessing here. All the info I've seen in the single phase example says "a fusible resistor is used to limit in-rush current and comply with safety agency requirments to blow open in the event of a catastophic failure in the power supply". In-rush current won't be much as I only have two 33uF caps in series.

Another thing I'm not sure about is the voltage rating of the resistors. As this circuit has the potential to operate at over 500V, that could be a problem with the fusible resistors available. The ones I've seen are only rated at 3W 500V max. So...

Not really sure where to go next..

Thanks,

larryg

 

[VEBill]

I guess that you need to power your meter circuitry from all three phases because you don't want people abusing the meter to get free 1- or 2-phase power. The messages back to base should probably include error codes for missing phases, paralleled phases, and other such conditions that might arise when the power company isn't looking.

Some X-10 modules (those AC remote control gadgets) use a fine gauge wire (soldered directly into the PCB like a jumper) to provide ultimate protection. But that is for the full several amp load, not just a few mA. In other words, whatever you do, you have to make sure that virtually all (or all) of your failure mechanism will draw enough current to reliably pop the fuse.

I guess lightning protection of your circuitry is a whole other major consideration. Adds a nice layer of complication when you try to predict the failure modes.

I suspect that this little power supply is not a trivial design exercise. (Insert nightmare vision of power meter with the glass cover blackened from the inside...) I hope that you've budgeted accordingly.

Do any of your power company customers have any qualification requirements and test procedures? If so, then they might provide some useful clues about the best way forward.

Or peek inside similar products. Or check the patents (very easy these days).

 

[moira]

LarryG: Your posting is lacking much info; as noted by all! First thing Larry, is NO FUSES IN ANY NEUTRAL!!!! EVER!!!!! If you expect three phase power in nominal 480 VAC feeders, you must fuse each phase ( 3 fuses), one in each line for at least 175 % of caculated amperage load. Again I don't Know what you have: a self manufactured supply or direct from a utility. You must protect people and circuit from ground fault and short circuit with 600 VAC rated fuses or circuit (3 phase) breaker if you expect to be using the 500+ volt power you mention. If you only have 208 VAC mains; then, 250 max rated fuses/ breaker will be enough. I get some kind of impression that you are mixing up sections of your circuits. Your basic power supply is one aspect. Having taken care of that ; you want to construct your integral and regulated power supply. Thru what ever, lets say a silicone controlled rectifier network, you will create your multiple lower DC voltages for your true electronic equipments. In your main DC section or sections, this is where you would ,probably, employ fusible resistors. These are high quality and design items. Most usually in 1/4 to 1/2 watt sizes, most usually in very low ohamage; 0.1 to 1.0 ohm. Basically acting as a low resistance connector, they disipate heat generation very easily and let thru their pre, factory calibrated amperage. UNTIL a serious overload accurs! Then this resistor goes OPENED! Their intent is to protect people and equipment from damage or injury from excessive current flow. Like, maybe, your wide screen TV burning up in your home! Their good design forces a repair. They usually are soldered into the circuit. The down stream problem must be found and repaird and the fusible resistor must be replaced. Can't just snap another in; like a glass fuse and going "pop" again! Always think first about how to turn circuits off and providing for overloads to open, safely all your circuits and you will be very succesful. Best regards, MOIRA

 

[Skogsgurra]

Hmm...

Gunnar Englund

http://www.gke.org

 

[larryg]

Moira: Let me see if I can clear up the confusion -

I am designing an off-line flbyack supply that must run off of 3 phase power. This supply will be inside a commerical watt-hour utility meter. Therefore, the 3 phase will come directly off of the pole (xfmr). I have seen similar designs of this sort (from meter mfg's) and they typically utilized fusible resistors in all lines (3 ph + N)after the MOVs but before rectification. I have never seen fuses for this application. The values of the resistors are typically in the 10-20 ohm range. In addition, the Neutral is also rectified in the event one of the phases is miswired. My question has been: Why are fusible resistors used instead of fuses in this application and if resistors are used, how do you size them?

As I metioned before the load is very light for the majority of the time (25mA). Only during radio transmission does the load significantly increase and even then, it is on 25% DC and supported by 3F of capacitance.

I think perhaps the reason that resistors are used instead of fuses relates to the operation requirements. Since this is a meter that measures kWh and will indicate how much power to charge for, we probably don't want fuses to pop due to transients or the like and prevent the power company from missing any revenue. Using resistors would allow the meter to continue to function except in the event of a catastrophic failure.

Does this make better sense?

Regards,

LarryG

 

[Skogsgurra]

I think that you were reasonable clear from the beginning. The reason you use fuse resistors is simply because they limit current. An ordinary fuse (the little glass type ones, for instance) does not. A short after the fuse produces more current than the fuse can break. So there will/can be a standing arc that can cause lots of damage to the surroundings. A short after a fuse resistor is limited in current (10 - 20 oms resistor) so that the fuse can interrupt the current when it breaks. Philips has had a range of "flameless" resistors that, I think, was developed for this application. But there are others as well.

Gunnar Englund

http://www.gke.org