48V as main DC voltage on boat - anyone?

[Lepke]

For many years, large boats used 32v systems. It was only the reduced demands of newer 12v products that led to most 32v systems going away. I ran 32v most of my life and now run 48v for my inverter. I've never heard of a 32v fire from bad connections, and in asking other old time mariners, the group hadn't either.

My previous 12v inverter needed cables the size of my thumb, and the batteries close by. With 48v my cables are smaller and I have some banks 20' away from the inverter. I love my 48v system.

[nebster]

The auto industry is starting to move towards 48V nominal systems again, mostly because of the opportunity to power small traction motors in hybrid vehicles.

Even USB-PD, the standard we'll be using for at least the next ten years, has bumped up to 20V for efficient delivery.

Insurers will write coverage on boats and other vehicles with 48V systems, just like they have for years, as long as the market demands it. If there is an increased risk with those systems -- and I am skeptical there is, once manufacturers enter and we move away from the occasional floating science project -- insurers will price it in rather than back out of the market entirely. Just like they try to price in everything else.

[ramblinrod]

Originally Posted by Lepke

For many years, large boats used 32v systems. It was only the reduced demands of newer 12v products that led to most 32v systems going away. I ran 32v most of my life and now run 48v for my inverter. I've never heard of a 32v fire from bad connections, and in asking other old time mariners, the group hadn't either.

My previous 12v inverter needed cables the size of my thumb, and the batteries close by. With 48v my cables are smaller and I have some banks 20' away from the inverter. I love my 48v system.

P = E^2/R.

This is one of those immutable laws.

Therefore, everything else equal, a 32 Vdc and even more so a 48 Vdc system HAS TO BE more prone to start an electrical fire than a 12 Vdc system, where and when ever a high impedance connection occurs.

To deny this is to effectively deny that poor electrical connections ever occur on boats.

Anyway, the OP's query was about a 2019 modification to a 1990ish 37' sailboat; and I believe all of my advice on that subject is perfectly sound.

[BlackHeron]

Originally Posted by ramblinrod

P = E^2/R.

This is one of those immutable laws.

Therefore, everything else equal, a 32 Vdc and even more so a 48 Vdc system HAS TO BE more prone to start an electrical fire than a 12 Vdc system, where and when ever a high impedance connection occurs.

To deny this is to effectively deny that poor electrical connections ever occur on boats.

Anyway, the OP's query was about a 2019 modification to a 1990ish 37' sailboat; and I believe all of my advice on that subject is perfectly sound.

You are using the equation wrong.

A watt is a watt. If a light, a heater, or a pump needs x amount of power to work it is going to use that much power to do that given amount of work.

Voltage and Amperage are two ends of a sea-saw. One side goes up and the other goes down. You can put the sea-saw pivot on R, but that doesn't show the whole story. R is NOT constant between 12v, 24v, 32v, or whatever equipment because it can't be.

R is tailored by the engineer designing the equipment so that watts are watts at the end. If you need a 1HP pump then you need 746watts of power regardless of system voltage. The resistance of the coil in the motor will be different for the 12v version of that pump and the 24v one. Ohms law demands rhat it be twice as much , otherwise there would not be balance within the Force.

Watts are watts!

[ramblinrod]

Originally Posted by BlackHeron

You are using the equation wrong.

A watt is a watt. If a light, a heater, or a pump needs x amount of power to work it is going to use that much power to do that given amount of work.

Voltage and Amperage are two ends of a sea-saw. One side goes up and the other goes down. You can put the sea-saw pivot on R, but that doesn't show the whole story. R is NOT constant between 12v, 24v, 32v, or whatever equipment because it can't be.

R is tailored by the engineer designing the equipment so that watts are watts at the end. If you need a 1HP pump then you need 746watts of power regardless of system voltage. The resistance of the coil in the motor will be different for the 12v version of that pump and the 24v one. Ohms law demands rhat it be twice as much , otherwise there would not be balance within the Force.

Watts are watts!

Well, I wouldn't want to contradict Ohms law, so...

Resistance is resistance is resistance.

A fixed resistance (such as carbon in an arced DC connection) does not change because of the voltage applied to it.

If a bad electrical connection has 10 ohms of resistance, it has 10 ohms of resistance regardless whether 12 Vdc or 48 Vdc is applied to it.

The voltage drop across it is dependant on the total circuit resistance, but assuming a load and cable having negligible resistance...

A derivative of Watts Law (actually using Ohm's Law) is...

P = E^2/R

If E = 12 Vdc and R = 10 ohms, P = 14.4W

If P = 48 Vdc and R = 10 ohms, P = 230.4 W

In the real world, the circuit wiring and load would also have some resistance and the corresponding voltage drop across the high Z contact would be less (to some extent) but make no mistake, with increased system voltage, the power dissipated in heat at a bad connection, increases exponentially in accordance with Watts Law, every single time.

What's worse, higher voltage is more likely to cause di-electric breakdown of the air in an open connection, causing the arcing, that causes the carbon build-up in the connection.

I hope this causes you a holy $%^ moment if you have been planning increasing your vessel's DC system voltage.

[BlackHeron]

There are no "fixed resistances" in the electrical world when comparing systems of different voltages. You just can't take a 120v heater and hook it up to. 220v (or 208v 3-phase for that matter) without reengineering the circuit because they are just not compatible.

If you are comparing a fixed resistance carbon coil uwted on one voltage to another identical-resistance coil on another voltage they will be drawing different amounts of power in their respective circuits. A 100w heater on 12v is not the same item as a 100w heater on 24v...

Resistance is NOT fixed when comparing between different voltage systems. Your central premise is flawed.

Holy-s moments or not, this is ridiculous.

As for dielectric breakdown in low voltage insulation due to the actual voltage differences give me a flipping break. The difference between 12v and 48v is infintiesimaly minimal. In the trades we don't even start to worry about this ever on any low voltage systems (and we don't consider anything under 600v to be anything but "low voltage ")

At one time I was certified to do medium-voltage terminations unsupervised and untested up to 20kv. I went to school for years to become a commercial electrician and specialized in electrical distribution centers, running them as a foreman. This entire conversation about 48v being so intrinsically unsafe is laughable to me. Stop making a fool of yourself.

If anything, because of I-squared*R heat issues the greater amperages and elevated voltage drop properties of lower voltage systems such as 12v marine and automotive/RV distribution and branch circuits are more prone to fires and arcing than 24 or 48.

[ramblinrod]

Originally Posted by BlackHeron

There are no "fixed resistances" in the electrical world when comparing systems of different voltages. You just can't take a 120v heater and hook it up to. 220v (or 208v 3-phase for that matter) without reengineering the circuit because they are just not compatible.

OK, so you are really going to make me do this.

Lets do a circuit analysis for a 12 Vdc system and a 48 Vdc, both with a 180W load, and a 10 ohm contact resistance.

Circuit A:

For 180W @ 12 Vdc, I = P/E = 15A, R = 0.8 ohms.

The cable with 105C insulation outside an engine compartment must be 16 AWG minimum (unless larger required to minimize cable voltage drop based on length and load type).

RT = 10 + 0.8 = 10.8 ohms

ET = 12 Vdc

IT = 1.11 A

P contact = I^2*R = 12.3W

The contact gets warm.

Circuit B:

For 180W @ 48 Vdc, I = P/E = 3.75 A, R = E/I = 12.8 ohms.

The cable with 105C insulation outside an engine compartment must be 16 AWG minimum (unless larger required to minimize cable voltage drop based on length and load type). No change from the 12 Vdc system.

RT = 10 + 12.8 = 22.8 ohms

ET = 48 Vdc

IT = 2.1A

P contact = I^2*R = 44.1 W

So the contact gets considerably warmer, maybe enough to start a fire.

Now one doesn't normally go up to a 48 Vdc system to run 180W loads on 16 AWG wires; they increase system voltage to run higher loads on smaller wires. So if the reason for upping to 48 Vdc was to operate even just a 360W device (ridiculous), here's watt (teehee) happens...

Circuit C:

For 360W @ 48 Vdc, I = P/E = 7.5 A, R = 6.4 ohms.

The cable still has to only be 16 AWG (depending on load type and distance).

RT = 10 + 6.4 = 16.4 ohms

ET = 48 Vdc

IT = 2.9 A

P contact = I^2*R = 86 W

This is likely hot enough to let the magic smoke out.

So...

Lesson 1: When one increases system voltage to reduce load device current, everything else equal, any amount of contact resistance is more dangerous (and more likely to occur).

Lesson 2: If one increases system voltage so they can operate higher loads, any amount of contact resistance is even more dangerous (and even more likely to occur yet).

There is a very good reason why most small rec boats (~12 to 50 ft) were designed with 12 Vdc systems...

...because risk of fire increases with system voltage.

[tanglewood]

Originally Posted by ramblinrod

OK, so you are really going to make me do this.

Lets do a circuit analysis for a 12 Vdc system and a 48 Vdc, both with a 180W load, and a 10 ohm contact resistance.

Circuit A:

For 180W @ 12 Vdc, I = P/E = 15A, R = 0.8 ohms.

The cable with 105C insulation outside an engine compartment must be 16 AWG minimum (unless larger required to minimize cable voltage drop based on length and load type).

RT = 10 + 0.8 = 10.8 ohms

ET = 12 Vdc

IT = 1.11 A

P contact = I^2*R = 12.3W

The contact gets warm.

Circuit B:

For 180W @ 48 Vdc, I = P/E = 3.75 A, R = E/I = 12.8 ohms.

The cable with 105C insulation outside an engine compartment must be 16 AWG minimum (unless larger required to minimize cable voltage drop based on length and load type). No change from the 12 Vdc system.

RT = 10 + 12.8 = 22.8 ohms

ET = 48 Vdc

IT = 2.1A

P contact = I^2*R = 44.1 W

So the contact gets considerably warmer, maybe enough to start a fire.

Now one doesn't normally go up to a 48 Vdc system to run 180W loads on 16 AWG wires; they increase system voltage to run higher loads on smaller wires. So if the reason for upping to 48 Vdc was to operate even just a 360W device (ridiculous), here's watt (teehee) happens...

Circuit C:

For 360W @ 48 Vdc, I = P/E = 7.5 A, R = 6.4 ohms.

The cable still has to only be 16 AWG (depending on load type and distance).

RT = 10 + 6.4 = 16.4 ohms

ET = 48 Vdc

IT = 2.9 A

P contact = I^2*R = 86 W

This is likely hot enough to let the magic smoke out.

So...

Lesson 1: When one increases system voltage to reduce load device current, everything else equal, any amount of contact resistance is more dangerous (and more likely to occur).

Lesson 2: If one increases system voltage so they can operate higher loads, any amount of contact resistance is even more dangerous (and even more likely to occur yet).

There is a very good reason why most small rec boats (~12 to 50 ft) were designed with 12 Vdc systems...

...because risk of fire increases with system voltage.

Where is the 10 ohm contact resistance coming from? Looking at a couple of switches, contact resistance is in the milliohms, so 3 orders of magnitude lower than your model.

[BlackHeron]

This has gone beyond the level of an SNL skit's ridiculousness.

I'm out...

[ramblinrod]

Originally Posted by tanglewood

Where is the 10 ohm contact resistance coming from? Looking at a couple of switches, contact resistance is in the milliohms, so 3 orders of magnitude lower than your model.

Good question.

High impedance (Z) connections are very common occurrences on boats.

It could be due to a corroded wire, a poor crimp, a loose fuse holder tang, a broken wire, a loose screw on a terminal block or any number of other high Z connection possibilities.

A high Z connection is much more likely to happen once one gets arcing across those contacts from trying to switch a higher voltage like 48 Vdc.

Note that a standard toggle breaker (e.g. Blue Seas # 7210) that is rated to switch 240 Vac is only rated to switch 32 Vdc max. voltage.

Now lets look at a standard toggle switch, (e.g. Blues Seas #4150)

These are rated to switch 15A @ 125V AC, 10A @ 250V AC, or 15A @ 12V DC.

For systems higher than 12Vdc, one has to change up to a Blue Seas # 8204, having ratings of 15A @ 125V AC, 10A @ 250V AC, or 15A @ 32V DC.

Still not suitable for 48 Vdc.

Why is this RamblinRod?

This is due to the tendency of DC to arc across contacts, creating carbon, and causing high Z.

The higher the DC voltage, the more prone to arcing, carbon build-up, higher voltage drop, resulting heat, and possible fire.

This is just one more example of...

…"What one doesn't know CAN hurt them."

Stay Safe.

[rom]

Originally Posted by ramblinrod

OK, so you are really going to make me do this.

What about using voltage drop instead of circuit voltage in your equation ? You will find that voltage drop e.g. in a cable is lower with higher voltages EEE because of lower current, and so is the dissipated power in this cable (or contact if you prefer).

[ramblinrod]

Originally Posted by rom

What about using voltage drop instead of circuit voltage in your equation ? You will find that voltage drop e.g. in a cable is lower with higher voltages EEE because of lower current, and so is the dissipated power in this cable (or contact if you prefer).

This is true (as mentioned many times by myself and others before in this and other threads).

But as demonstrated here, for other than "high current loads", it is really a non-issue.

For high current loads like inverters and stuff at the bow, it can make a difference, but it is often far cheaper, easier, AND SAFER to use fat cables (when appropriate) than changing house bank voltage.

AsI mentioned previously, the break point (and I have analysed this for many installations, is around 45 ft+, where it starts to make sense putting a dedicated battery in the bow.

Once one gets much above 3 HP (thruster) it makes sense to put a 12 Vdc or 24 Vdc battery near the load, rather than run cables back to the house bank.

If the house bank is midship, it favours the bigger cable decision.

For inverters, it is often easy enough to mount it very near the house bank (though I like to mount them just on the opposite side of the bulkhead where practical).

But that has absolutely no bearing on the position that I am presenting, that...

"Higher system voltage results in higher voltage drops across high impedance connections, and these created additional heat, and can lead to a boat electrical fire."

It does.

Anyone who argues differently is simply incorrect and denying the laws of physics as they exist.

It's one thing to hack a vessel electrical system, and hope it works, and hope it doesn't hurt anyone.

It is another thing to design, install, and warrant them, professionally.

It just has to be done right.

Increasing system voltage where not warranted (just to save a few bucks on cable) while increasing risk of fire, is not doing it right IMHO.

[evm1024]

Originally Posted by ramblinrod

You seem to be missing part of the equation.

Transmission line voltage in Canada is 725 kV.

Why don't we bring that into homes?

Such an irrelevant straw man.

The Safety Clearance Distance for 725 kV is 33.33 feet....

Apples and Volcanos

[nebster]

Originally Posted by ramblinrod

"Higher system voltage results in higher voltage drops across high impedance connections, and these created additional heat, and can lead to a boat electrical fire."

It does.

Sure it can. Which is why we engineer the connections appropriately, and we use rated switches and OPDs. None of it is hard to do, there shouldn't be many of them anyway, and power systems do generally require care regardless of voltage.

It just has to be done right.

You seem to imply that doing it "right" is only possible for professionals and that it's especially difficult. I would say that neither is true.

[evm1024]

There are many advantages to higher generation and storage voltages in boats (and cars for that matter). And when coupled to point of use reduction of voltage (along with the inherent increased current) offers a desirable "next generation" solution to boats power system design.

The real "problem" is not fire hazard, or shock hazard, or even DIY hazard but rather the design of switches and breakers commonly used in our boats.

Oh, No carbon build up in the contacts in use in switches now days. The contacts in those switches and breakers are made from copper, copper alloys or silver alloys. Pitting and welding are the "pitfalls".

As was noted these switches and breakers have a much higher voltage rating for AC verses DC at any given amperage. This has to do with the arc generated when a switch closes (lessor) or opens (greater).

This is explained quite well in this quotation from an article in EDN.

Imagine you have two circuits, each carrying the same current—one is an ac circuit and the other is a dc circuit. When you switch off power to an ac circuit, a voltage spark (or arc) is created inside the switch that’s quickly extinguished—a desirable condition. This is because an ac sine wave is naturally at zero amps twice per cycle. So, there’s a 50% chance that the power to the circuit will not be at peak levels when the power is switched off.

However, this isn’t the case in a dc circuit, where both the current and voltage are constant. When power is switched off to a dc circuit, it can take much longer for the voltage arc inside the switch to extinguish itself. This is why the switching speed (how fast the switch contacts open and close) of a switch in a dc circuit is so important. The objective is for the switch contacts to separate as fast as possible when turning off power to the circuit. This helps minimize the arc’s time to develop and extinguish itself.

More info can be gathered from: http://www.aeroelectric.com/Referenc...ing_Manual.pdf for any that are interested.

But back to the issue at hand - The typical switches used in out boats are rated for less than 30 volts DC. This presents a problem with higher DC voltages. But 24 volts is "fine"

Oh - I should note that circuit breakers are not designed to be used as off-on switches. But we do tend to use them that way.

When using circuit breakers as switches or using switches at a DC voltage higher than their rating (higher AC too of course) the problem shows up primarily as greatly reduced life spans.

The switch contacts pit and burn and in many cases weld closed. Internal heating has resulted in melted plastics and metal parts in the switches in some cases due to extended arc times. It should be noted that the deformation typically results in an increasing arc gap that will help extinguish the arc. The deformation often (by design - duh!) also causes the switch to no longer be able to make a contact.

Fires caused by resistance heating within the switch body are rare to non-existent during make or break. Fires in switch bodies are more likely to be caused by over current conditions.

The bottom line for me IMHO (well not so humble) is that going to 24 volts DC base voltage in our boats is doable with exiting hardware (switches etc).

Going to an elevated voltage above 24 VDC has advantages but requires that the the entire system be designed to have component safe working voltage specs that meet or exceed the elevated voltage (e.g. 36 or 48).

Non trivial? Of course!