Electronics > Projects, Designs, and Technical Stuff
Toggle switch for small aircraft charging system in case of failure
ITman496:
Hello!
I'm trying to figure out the best way to have a toggle switch to temporarily disable the charging system of a small 440CC engine.
The charging system is a shunt regulator on a single phase stator and permanent magnets on the flywheel. It's a 4 wire regulator.
The theory, as far as I can tell for how it works, is the regulator directly dumps into the battery after passing through the rectifying diodes in the regulator, and the regulator has SCRs connected across the stator winding BEFORE the rectifying diodes, and it will intermittently shunt the current of the stator winding to short them out when the voltage at the battery starts to rise too high. As load increases, it shorts them less, and as load decreases, it shorts them more often, until zero load where theoretically it's holding the winding shorted as to not charge anything.
My question is this. I'm building an airplane, and I know from past experience with friends with dirtbikes/motorcycles/atvs/snowmobiles, which all have this charging system, that regulators commonly fail, and their failure mode is to start just pumping power into the 12V system, overcharging the battery, boiling it over, and frying whatever is on the 12V system if it cant take voltages north of 17V or however high it decides to go.
My ability to disconnect the regulator has two benefits.
One, on the ground, I can easily turn it on and off and see, hey, I'm charging when its on, and not when its off, due to the voltage dropping from 13 something to 12 something and then rising back up again.
Two, if the regulator fails, I can turn it off and just cruise home on battery. The capacity of the battery is such that it can run the aircraft even with the backup electric fuel pump switched on for around 2-3 hours. It's oversized for the demand because I need the CCA for the starter motor. It's a big boy. But when the aircraft is flying, normally, the only electrical demand is the flight computer, which draws 1.5 amps. The fuel pump is mechanical (crank case pressure pulse driven) and the backup draws 1.6 Amps.
If I can't turn a failed charging system off, then my issue is under my seat, in my wooden airplane, is a battery that's having up to 130W (I have no idea what the overcharge curve of an AGM lead acid battery is. How does the current absorption work as you go over the limit?) being shoved into it. This turns an OK turn and head back for home situation into a land ASAP situation, which is far more annoying. Also, I'd rather like to not find out just how high the unregulated voltages can go, and fry my flight computer.
So what I am asking here is... How can I disconnect the charging system safely while not damaging it due to the aforementioned testing idea.
So far I've devised 3 methods.
A: Disconnect positive wire from regulator from battery, let it do whatever.
Pros: Simple, the wire for it even passes right where the switch would be located.
Cons: Nothing to sink the rectified voltage from the stator into. Voltage skyrockets, I've heard upwards of 100V p2p can be sourced from those stator winding at high RPMs!
B: Short stator winding on engine with toggle switch, cut it off at the source.
Pros: As this is how the regulator works anyway, it shouldn't harm anything, right?
Cons: Unnecessary parasitic load on the engine. It won't be 130W, as while the stator can supply a relatively constant amperage across the rpm range, the voltage will rise and fall with rpm, and if you short it out, you reduce the voltage across it and the back EMF winds up reducing the resistance on the motor. I think. That's what I read, anyway. The idea being that if your regulator on your motorcycle shorts the winding out, its not putting the full load of the electrical system on the engine, just a little bit. (The numbers I saw were that a 400w charging system on a motorcycle will experience 60W of heat generation when the regulator is completely nullifying the output due to the battery being fully charged) But the switch that shorts it will still have to flow considerable current. How much does a stator winding supply at short circuit?
C: Run one stator winding wire through the switch before it goes into the regulator input.
Pros: Completely open circuits the stator, removing its ability to generate any load at all, can't overcharge the battery, can't load down the engine, can't get hot.
Cons: Open circuit voltage of the winding can skyrocket north of 100V on some engines at high speed, or so I've read. Is this bad? Does the winding care? The other catch I can think of is when you reconnect the winding back to the regulator input. If you are unlucky, you'll close the switch contacts right at the highest peak voltage, exposing the regulator/rectifier to incredibly high voltages, blowing it out.
D: Do nothing. I'm overthinking this problem. Get rid of the switch and just move on.
Does anyone have any experience with any of this?
ITman496:
Embedded here is an example circuit of how the system works that I found on google. The example is 3phase, however, and mine is single phase, but you get the idea.
EDIT: However I would like to note that my regulator does not get an ignition circuit power source. Fed from its own output. Also, it is old, and this is an old design, so the regulation is likely not an IC, but a zener that triggers the SCRs when the voltage is too high. It is not.. very exact regulation.
Nusa:
In general, the answer would be A. As for cons, what do you think will happen in the zero-load case? And how is that a problem?
However, if you've already got a proper fuse or circuit breaker panel, the answer may be D. You can pull the breaker or pull the fuse to kill the circuit.
If it's not a fused circuit, it should be, like everything else electrical in an aircraft. You can't afford to have the wire itself be the weakest link.
ITman496:
My worry for A is that with nothing to sink the load into, the voltage coming out of the regulator will spike to whatever the p2p voltage of the stator winding is. Which may be above the rating of certain components in the regulator. I've seen evidence of this when friends have ran ATV's without batteries in them. It tends to melt the wiring harness from the surging voltages shooting through, the lights tend to run a lot brighter and pulse at idle.
It does have a breaker, but that will not really help because the issue is it not regulating. When a regulator fails, its typical failure mode is to simply do nothing. Which leaves the stator to dump freely through the rectifying diodes and charge and charge and charge the battery until it boils over and there's nothing left. The charging current will never exceed the usual current you'd see through the circuit, it just will never stop.
Unless anyone has a better idea, I'm going to simply put a toggle switch that is normally always switched on that links the output of the regulator to the battery. I'll then put a red flip cover on it so it can't be disturbed and is actively held down, conducting. If the computer sees the voltage rise above 15 or so volts, it will throw an alarm, and if I determine its overcharging, I'll pull the cover up and flip the switch to isolate the rectifier. Then I'll go and land and replace the unit, since at that point its already broken and it doesn't matter if it breaks more.
Nusa:
If you don't have a complete circuit, there is no load. It's current that melts things, not voltage. So your anecdote about the ATV isn't terribly relevant. Yes, the charging circuit needs the battery to act like a capacitor and smooth the voltage, so no surprise the voltages are too high and pulsing. Visible pulsing at idle is simply because the frequency of the AC being converted to DC is very low when engine speed is low. It pulses at high speeds too, but it gets too fast to see with the human eye.
Most aircraft-type breakers are also intended to be used as switches, in which case you've already got what you're asking for. If you have a different kind of DC breaker, then never mind. I never said the breaker/fuse should pop in this instance, merely that one should be present for safety reasons.
Navigation
[0] Message Index
[#] Next page
Go to full version