Recently, as part of my work on hybrid propulsion systems, I attended a demonstration where a large-scale lithium battery pack was deliberately pushed past the boundaries of existing experience. The intention was to run the system at 20 kW of output and 160 volts, creating a current flow of around 130 amps. Owing to a miscalculation, when we threw the switch and brought everything on line we achieved an output of something over 300 kW, drawing over 2,000 amps from the batteries.
Normally this would have immediately blown the main fuse. Unfortunately, in the rush to prepare for the demonstration, the fuse had been omitted. We tried to trip the relay to break the circuit, but all this did was splatter molten copper around the area with the relay arcing over. There was no way of shutting things down. A serious boat fire was avoided by hacking through the main cable with a fire axe, with a regular fireworks display going on all the while and smoke belching out of the boat. It was the most exciting systems demonstration I have ever attended!
I was immensely impressed by the power of these lithium batteries—they can deliver a truly astonishing amount of energy. But our accident also highlighted an emerging issue in marine electrical systems: safely containing the potentially dangerous amount of energy stored in ever larger and more powerful battery banks.
The primary mechanism for keeping batteries under control is a fuse. These simple devices are designed to “blow,” breaking an active electrical circuit when current on the circuit exceeds a given threshold. Typically, this is somewhere around 30 percent above the nominal rating of the fuse, but it is affected by ambient temperature and other factors.
It would seem that to safely contain a battery bank all you need is a fuse that will open the circuit if the amperage goes above the maximum desired level—in our case, 130 amps. Unfortunately, it’s not this simple.
AMPERE INTERRUPTING CAPACITY
Given enough current, fuses and other overcurrent protection (OCP) devices, including circuit breakers, can arc over and become ineffective. For any given device, the higher the voltage on the system, the lower the amperage at which the device may arc over. The arc-over amperage is described in something known as an Ampere Interrupting Capacity (AIC) rating. This tells us the highest level of amperage at which an OCP device will safely break a circuit at its rated voltage. AIC ratings are typically in the thousands of amps.
The internal chemistry of conventional lead-acid batteries limits the rate at which they can deliver energy, even in the case of a dead short (such as what happens when a wrench is dropped across the battery terminals). Even so, quite small batteries can deliver several thousand amps of current for a brief period of time (more than enough to melt that wrench). For this reason, the American Boat and Yacht Council (ABYC) and the ISO in Europe have always recommended that the primary OCP device on a boat have an AIC rating from 3,000 to 5,000 amps, depending on the size of the battery bank.
Unfortunately, the larger battery banks often found on cruising boats, even those with conventional batteries, can easily generate potential short-circuit currents above 5,000 amps, and with newer battery technologies, potential short-circuit currents are greater still. At present, there is nothing in the boatbuilding standards that addresses this issue.
NEW BATTERY TECHNOLOGIES
I have written several times about the Thin Plate Pure Lead (TPPL) construction employed in Odyssey lead-acid batteries. This design greatly accelerates the rate of energy absorption and release as compared to conventional lead-acid batteries. With a large TPPL battery bank we might see short-circuit currents of 20,000 amps or more. The required maximum 5,000-amp AIC rating in existing boatbuilding standards is grossly inadequate for containing the energy of these batteries.
Then there are lithium batteries, which is a whole new ball game. I have tried for two years to find out what the top potential short circuit currents are with lithium batteries, but to no avail. Part of the problem is that large lithium battery banks are relatively new, as opposed to the much smaller batteries long found in cell phones, laptops and portable tools. There are also few laboratories that can do the necessary testing to establish this data. With larger lithium battery banks, I have a hunch we could easily see short-circuit currents of 30,000 amps or more.
Then there is the question of voltage: again, the higher the voltage, the greater the likelihood of arc-over. A fuse that can manage a 12- or 24-volt battery bank may not do so when voltage is raised to 144 volts, and especially not the 500 volts now seen on some marine hybrid propulsion systems. Also important is whether we are dealing with DC or AC circuits. Typically, AIC ratings for a given OCP device are considerably lower on a DC circuit. This is because AC current cycles from positive to negative many times a second, declining to 0 amps in between, which helps suppress any arcing, whereas DC current remains constant, so there is no arc suppression.
Finally, there is the question of response times. Fuses are designed to blow either quickly or slowly. In many marine applications, slow-blow fuses are preferable. For example, on circuits powering electric motors, especially winches, windlasses and bow thrusters, there can be momentary “inrush” currents many times higher than the normal operating current. A fast-blow fuse on such a circuit would be a nuisance. However, when it comes to a dead short across a battery bank, you definitely want a fast-blow fuse.
KEEPING THE LID ON
Currently, the most cost effective and practical way to control large battery banks on boats is to install what is known as a Class-T fuse as close as possible to the positive battery post. These fuses, which feature a high AIC, are readily available, relatively compact and inexpensive. They come in a wide array of voltage and amperage ratings, and typically have an AIC rating of 20,000 amps.
In our runaway electric-propulsion demonstration, the electric motor was holding the current down to 2,000 or so amps, so a fuse with an AIC rating as low as 3,000 amps should have safely broken the circuit. A Class-T fuse—there was one in the technician’s pocket!—would certainly have done the job.
But let’s say we have a dead short across the battery cables, for example, if someone had put a screw through a bulkhead and hit both cables. I’m not at all sure that even a Class-T fuse would safely shut things down, in which case, but for a handy fire axe, the almost inevitable consequence would be an out-of-control boat fire.
I am continuing to study the short-circuit potential of these new battery technologies. It’s a subject that will become ever more important as an increasing number of these amazing energy storage devices find their way onto boats.