Fast Blow Circuit Breaker

[Sparweb]

Hello,
Puzzled by some odd behaviour in a circuit breaker, and hoping the forum can help shed some light.

We have a troubleshooting test box that relies upon fast-blow fuses to give an indication that a circuit functions correctly.
The cables from the test box are connected in series with a cable harness on the aircraft to "see" if an electric pulse is delivered.
The user knows that the circuit works, from power supply to ground, when the test-box fuse blows.

The fuses in the test box take the place of squib cartridges which activate the fire extinguishing bottles (engine compartment/cargo compartment/etc.)
Obviously, you don't want to discharge the fire extinguishers in the hangar.
To do the test, we disconnect the cables from the squibs, cover the squibs, and connect the leads to the test box.
When the technician pushes the button in the cockpit, the fuse in the test box should blow, simulating a firing of the extinguisher cartridges.
Works well enough, but you have to remember to bring fuses when you go to the aircraft, the aircraft may not be parked in a nice cozy hangar, it might be nighttime, -30 degrees, etc... you get the idea.

I came up with a plan to improve the test box by replacing the fuses with fast-blow circuit breakers.
I had encountered TE (Potter-Brumfeld) fast circuit breakers on a previous project, and I guess I was hoping to make myself a hero on this one!
A circuit breaker would be easier to see trip, and less hassle to reset it. Fine in theory, but we never could get one to trip.


Some numbers:
Firing circuit: Supply voltage 28V DC, 10A pulse for 50 millisecond.
Total circuit resistance (incl very long wires) 1.8 Ohm.
Squib cartridge load resistance 1.0 ohm.

Fuse p/n 22500002-HXP (Littelfuse)
Fuse rating 2 Amp
Fuse resistance = 0.05 Ohm
Dummy load (in series with fuse) 0.1 Ohm
Current pulse with fuse = 28V / (1.7+0.1+0.05) = 15 Amp
Fuse trip time 30 millisecond max @ 10A - which is why it works to detect the 50 ms pulse.

Circuit breaker Potter+Brumfeld (TE) W67-X2Q50-2
CB rating 2 Ampere
CB Resistance = 0.28 Ohm
Current pulse with CB = 28V / (1.7+0.1+0.28) = 13 Amp
CB Trip time 12 millisecond max @ 13 A - which should work well, but it didn't.
Datasheet

Replaced the first CB we tried, then the next, checked DC polarity... nothing could make it trip.
Eventually we got frustrated enough that we took the CB out, put it on a regulated 28V power supply, cranked it to 10 Amps, and still nothing tripped, even leaving it on for several minutes.
I think we returned 8 CB's back to the supplier, in the end. All apparently defective.

By the way, nothing's wrong with the aircraft. An all-up test of the bottles was conducted successfully last month.
The purpose of the test box is to allow the customer to test the system in the future, without firing the extinguishers.
The customer will be satisfied with using fuses, they don't know what they're missing.

Any ideas what went wrong?
I have a far-out theory, but I'd like to hear some other perspectives before biasing anyone's opinion.


STF

 

[itsmoked]

Hi SparWeb.

Wow. Pretty archaic - but positive.

Your breaker scheme didn't work because those breakers are really designed for day to day use of a specific type. They have to stay closed when you run 9.9A thru them all day - on a HOT day. The fast version of them require 80A thru a 10A rated breaker to trip 'instantly' and your puny 50ms pulse is never going to cut it. You'd have to screw around to find a breaker that might work. Like perhaps a 1Amp one that would get 800% of its rating during the test.

There are ICs that are electronic circuit-breakers that can do this precisely. It would not be too hard to roll your own.

If you want I could crank you out some little electronic test units that trip instantly at a precise current at all temperatures. Is this a tester others would desire too?







Keith Cress
kcress -

http://www.flaminsystems.com

 

[jraef]

I'll also add that manufacturing / testing / acceptance tolerances on circuit breakers are, as a general rule, much less stringent than on some fuses, and it sounds as though you were using some very tight spec fuses. So I'm not surprised.

Since this is 28VDC, you can experiment with something like this:


(image is a link)

There is a version that gives you protection for 4 2A circuits. Their intended use is to AVOID blowing fuses by acting faster.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[waross]

As I read the spec sheet, anything over 600% should trip the breaker in less than 12 milliseconds.
That's 12 Amps on a 2 Amp breaker.
Did you check the numbers closely?
If you were supplied with model W67-X2Q50-20
instead of model W67-X2Q50-2
It would never trip on 10 Amps.
One digit makes the difference between a 2 Amp breaker and a 20 Amp breaker.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Sparweb]

Waross,
You win the prize for guessing my theory about the problem. I personally think the manufacturer mislabeled the CB's, too.
We looked very carefully (at the outside) of the CB's which were clearly marked "W67-X2Q50-2" not "20"...
Discussed whether we should try to force it to trip by cranking the amps higher and higher (needed a new power supply to do it, though).
Eventually we decided we didn't gain anything from proving why the CB was faulty, it just was, and we went back to using fuses.
Glad I didn't use these CB's to protect an actual circuit!
(We don't specify breakers such as these on aircraft - not Mil-Spec type yada yada)

Jraef,
Those units are still pretty slow. The trip curve upper limits are on the order of several seconds.

Keith,
You think I'm archaic? Hey the aircraft OEM uses light bulbs to do a similar test on this circuit.
Our customer is going to think we're pretty slick with our test kit, in a Pelican case and everything!
We did try to get a 1-Amp version but none were available (even though they are printed on the datasheet).


STF