Over voltage required for EarthX battery

[madhun]

Originally Posted by DanH

My own regulator appears to control peak voltage with the battery disconnected, but no one seems to know exactly how well. EIS indications suggest output voltage oscillating like an erratic yo-yo. However, it does not trip the crowbar, so it's not getting real high. I've picked up a new toy (a Dataq DI1000), so I'll soon have hard numbers (peak, average, etc) on extended no-battery alternator output voltage. Later I'll look at load dump, the bus voltage peak just as the battery disconnects..



Because when the regulator failed, the resulting high voltage burned through the internal BMS disconnect. Se post 28.

That's a pretty nifty little toy you have there. For anyone that might be looking for an alternative, eLogger V4 from Eagle Tree Systems is an incredibly reliable unit, and almost unbreakable, from the RC & FPV world. With 4 hours of datalogging capability it would suit most tests. Pretty cost effective also.

[Blain]

So I am using an EarthX in my ship. Now I'm a bit concerned about the runaway. I understand what a crowbar circuit is but have no idea where or how to acquire one. Do we build the circuit? Is it a commercially available product?

[erich weaver]

Check in with B&C Specialty

[tspear]

What about using a clamp to prevent over voltage?
That is what we used in data centers. (I barely know the difference, but from what the EE explained the clamp does not require all power cut to turn off. It just caps the voltage, required when you did not want to turn off the computers)

Tim

[StickNRudder]

Originally Posted by tspear

What about using a clamp to prevent over voltage?
That is what we used in data centers. (I barely know the difference, but from what the EE explained the clamp does not require all power cut to turn off. It just caps the voltage, required when you did not want to turn off the computers)

Tim

That is kinda the job of the voltage regulator which controls the field current only generating enough energy to satisfy the load. To clamp a runaway alternator you would need the clamp to dissipate all the excess energy that the alternator is producing. Let?s say you have a 60A alternator and your aircraft is consuming 10A. Than means the clamp would have to dissipate 650W in heat (W = V * A).

[EarthX Lithium]

Originally Posted by DanH

Perhaps it would be best if Kathy explained the voltage limitation of the MOSFET used for overvoltage disconnect.

Per EarthX's engineering: the over-voltage protection does use MOSFETs. MOSFETs have a NPN junction, so it is like two diodes back to back. You can google how they work, but OHMs law does not apply here. They do however behave similar to diodes in respect to reverse breakdown voltage. That is at a certain "high" voltage current will flow "backwards" through the device. The reverse breakdown voltage is 100 volt.

Now we also have to talk about the time durations. Voltage spikes or surges are typically in the micro-second or milli-second range. The EarthX battery as well as any aircraft equipment is designed to handle these spikes up to and exceeding 100 volts, not sustained. Voltage spikes are not long enough to cause reverse current to flow in the MOSFET. And they are not long enough to damage aircraft electrical equipment.

Unloaded alternators have also been discussed here. An alternators output without a battery and or unloaded, as Dan pointed out, has a high ripple voltage. But the voltage ripple is also in the hundreds of hertz, so only in the milli-second time period. So again, not long enough time to trip over-voltage (crowbar), or damage any electronics. Plus, even without a battery, it still is not unloaded. Any little bit of aircraft electrical load with it's capacitor and inductors help to stabilize the alternator output


As far as Boeing, the 787 uses lithium cobalt chemistry and has over 20 million hours of trouble free operation with a lithium battery

[ChiefPilot]

Originally Posted by DanH

I kinda thought we all had one these days...the ubiquitous EFIS. Mine is set to display an alarm at about 13v on the low end and 15v on the high end.

I know, right? Yet the aircraft that started this whole thing either didn't have one or the pilot did not or could not flick off the alternator.

For someone who doesn't have an EFIS/EMS, the standalone unit is a great option - cheap and easy.

[rv7charlie]

Some points of info, in no particular order:

OV annunciation is OK, but with OV protection that removes the offending source, low voltage annunciation effectively serves the same purpose while in flight, by letting you know you've got a charging system failure. With OV protection in place, all you really need to know *while in flight* is that you've got a charging system failure & you need to appropriately plan to get on the ground.

Load dump: As others have pointed out, some new alternators/regulators have load dump protection built in. But many older alternators/regulators do not (I've experienced a load dump failure.) If you have a charging system *without* load dump protection built in, and the battery has the ability to disconnect itself, then it has the ability to cause a load dump failure of the charging system.

Bonus load dump data point: If the battery is fully charged, then a battery disconnect would not cause a load dump because it wouldn't be presenting any load to the charging system.

Clamps: Technically, yes; practically, not so much. OV protection with typical external regulators simply crowbars the ~5A field line to ground; easy to do with light weight, cheap components. To clamp the B lead output, you'd need to either insert a variable resistance in line with the B lead (failure mode issues, a lot of heat when activated, heavier, expensive for the high current device, etc), or shunt the extra current to ground with a variable resistance (again, all of the above).

[DanH]

From post #56, this thread:

Originally Posted by EarthX Lithium

Per EarthX's engineering: the over-voltage protection does use MOSFETs......The reverse breakdown voltage is 100 volt.

Now we also have to talk about the time durations. Voltage spikes or surges are typically in the micro-second or milli-second range. The EarthX battery as well as any aircraft equipment is designed to handle these spikes up to and exceeding 100 volts, not sustained. Voltage spikes are not long enough to cause reverse current to flow in the MOSFET. And they are not long enough to damage aircraft electrical equipment.

From post #1, this thread:

Originally Posted by EarthX Lithium

He did not have over voltage protection on his plane.

During his flight, his regulator failed and his voltages climbed up and remained at +29V and 40+ amps for many minutes. (After approximately 6 minutes, his alternator failed too). A couple of times the voltage spiked above 60V per the data from the EFIS. The over voltage protection for the EarthX batteries is a maximum of 60V, even though the FAA requires protection only up to 19.8V. When the voltages spiked, it was enough to cause a cell rupture as the pilot smelt a funny electrical smell and with continued high voltages and amps, the battery did go into thermal runaway.

Sustained voltage 29+, spike above 60v, BMS failed to protect cells.

Look, I realize some readers will take the position that the BMS should be able to fend off anything an unrestrained alternator might do, and obviously it can't in its current iteration.

Personally I think it's really dumb to not have fast, fully automatic alternator OV protection (with lithium or lead acid), and with it, I don't think a 100v, or a 60v, or even the FAA's 19.8v protection limit is unreasonable for the EarthX's BMS. I just want to know for sure how it works. Think about it. Don't you want TCs, A&Ps, and DARs to be able to speak with authority when they insist on alternator OV protection with your battery?

[tspear]

Based working on IT systems for building control and data centers. I tend to think in big over complex solutions with five and six nines of uptime. e.g. Clamp, capacitor systems for automatic voltage regulation....
Consider the size of the electrical system in the plane; I keep coming back to the idea of using a dual battery solution. A small lead acid battery to act as the buffer for the primary electrical load variation and in case of alternator failure (over voltage or over current); then a EarthX type battery to provide ship power and starting juice.
For such a solution, how big of a lead acid battery would be required? And at that point, is a more sophisticated Li battery still worth it?

Tim