Electrical failure / 503

[Captdill]

I have been running my 503 with Ducati charging system for a long time. For the last two or three years it has been with a Lithium-ion battery. The other day it spun my starter with ease and 10-15 minutes later I was on crosswind when electrical bus went dead. Volt meter to zero, and all electrics dead. Landed and found the battery with zero volts. I also found a fuse blown between the battery and the voltage regulator. There is a protection circuit that kills the battery with certain faults, and a reset button brings it back. After discovering that and resetting, the battery came up to 12V, but needed a charge before it could crank the engine. I replaced the fuse and ran the engine. I really couldn?t tell if the charging system was working. I guess I normally see that comparing voltage before and after start, but the battery voltage after charge was north of 13. So, I guess my real question is a chicken or egg thing. Did my battery fail first, resulting in the blown fuse, or did my blown fuse allow my battery to discharge just enough to start me up but then fail on crosswind?

[Navycposd]

Glad to know your charge circuit is working well with a Li battery!
The output of the 503 is just the pulsed AC .... strongly doubt that is the issue.
The little voltage regulators are always suspect. My analog gauge said voltage was fine. When I added a radio that had a voltage display, I found my regulator was all over the place.
I would suspect the protection circuit in the battery did it's job, preventing an overcurrent that could cause a failure of the Li battery.
As to which component is really at fault? Which let the smoke out?

[Eric Page]

If resetting the battery brought it back to 12V (nearly dead for a lithium battery), then it's likely that the BMS protection tripped and disconnected the cells from the output terminals. The question is, why? Did it disconnect due to over-discharge (i.e. the fuse blew first, for some unknown reason, then the battery was depleted until BMS cut-off occurred), or did it disconnect due to over-voltage (i.e. the regulator failed and caused a high voltage condition, the battery absorbed excess current long enough to blow the fuse, then the battery depleted until BMS cut-off). Since you've run it successfully since replacing the fuse, the second option seems unlikely, unless, perhaps, you have a temperature-related regulator problem.

I haven't played around with one of the aviation lithium batteries yet so I'm not very smart on their behavior. If the BMS is allowing the cells to charge to 100% then the resting voltage of a fully charged battery would be around 13.6V (3.4V per cell, four cells in series). The BMS may not do that (the cells will last longer if it doesn't). Some battery manufacturers de-rate cell capacity and only charge them to 80 or 90% to extend service life. Whatever the case, your regulator's output will have to be higher than the BMS's top charge voltage to fully charge the battery. If that happens on a 503, it's pure luck, as lithium batteries didn't exist when the engine was designed.

I took a quick look at Rotax-Owner.com for the 503 manuals. They have an operator manual and a repair manual. Only the operator manual says anything about the charging system, and it's both brief and confusing. It looks like there were two options for voltage regulators: one that regulated at "between 11 and 12 volts," which the manual seems to suggest is AC (it refers to an RMS voltage, which is an expression of an AC voltage's DC equivalence), despite calling the part a rectifier/regulator. The other option is described as a 3-phase regulator providing "13.5 to 14.5 volts," but the diagrams all show a single phase system with two coils wired in parallel. As I said, confusing. I presume you have the latter type since you're running more than incandescent lights.

So, that's a lot of unhelpful blather. Here's what I would do:

Remove the battery and check it outside the aircraft. First, charge the battery fully, let it rest for an hour, then rig up some kind of load. A 60W automotive halogen headlight bulb would work as a makeshift ~5A load. Note the voltage with the battery at rest (all measurements at the battery terminals), then measure it every ten minutes or so for an hour with the bulb connected. It should drop a few tenths of a volt at the start, then hold fairly steady for the rest of the hour. If it drops off sharply and/or the BMS cuts it off, then the battery is probably worn out. Unless you're running a very small battery, it should have no trouble with this test.

Do not use a carbon-pile load tester on a lithium battery. The testers' control knobs are very crude and it will be easy to overload the battery. Best case you'll trip the BMS protection again; worst case it'll damage something.

Disconnect the generator/stator wires (probably yellow) from the rectifier/regulator. With a multimeter in resistance (ohms) mode, check across the two wires. I don't know exactly what number you should get, but probably something like 2-3 ohms. If you get a dead short, or no reading (open), then the stator is bad. Also check between the stator windings and the engine case (ground). This should be open; if you get any ohms reading, a winding is shorting to the case.

If the battery tests OK, and the stator windings are OK, that just leaves three options: the regulator, the fuse or the wiring. Regulators are very difficult to test without a purpose-build testing rig since they're almost always fully potted with resin or silicone to keep out moisture, so you can't access their internal components. All you can do is test their performance in situ.

It's possible that the fuse was simply mechanically worn out. Two-strokes aren't exactly turbine smooth, and fuses are by nature made of fragile wire elements. Over many hundreds of hours, the fuse element can be fatigued by vibration in a way that causes it to either break mechanically or develop a weak spot that fails electrically at less than its rated current. If that's the case, then there might not be anything wrong and you'll get another few hundred hours out of the replacement fuse.

Finally, check all of the wiring from the regulator to the battery. If the path from the regulator to the fuse or from the fuse to the the battery shorted to ground briefly, it could have blown the fuse, blown open the momentary short, then depleted the battery until its BMS disconnected. Also pay attention to the wire connections at the fuse holder. A loose connection can cause high resistance, which causes heat. An abnormally high ambient temperature can make a fuse trip at less than its rated current.

[Captdill]

I think it was the fuse. I just flew it and watched the volts more carefully before start, after start, after loading various circuits, and after shutdown. Battery bus measured 13.0 V after being off the charger since yesterday, then went to 13.5 after start. I think I blew the fuse months ago when I was fiddling under the dash, I know just when. While I haven?t flown it that much in those months (a couple times since December) it probably sums to at least 1.5 hours. So, I am thinking the Lithium battery holds its voltage and cranking power right up to the end, and first sign of it dying is its death. Yesterday, before I charged it but after pressing the reset button, I thought I had a short circuit because flipping on the battery switch dropped the volts from 12+ to 8-. During that period voltage went to zero and I had to reset a couple more times. Finally, instead of chasing down a short, it occurred to me that I should charge the battery even though it would measure 12 V.