Dual Voltage (12v dc / 120v ac) Small Electric Water Heater

[Chance]

Originally Posted by boywonder

Chance: We are disagreeing with you because modifying water heaters can be very dangerous business. The fact that you are downplaying the potential things that can go wrong is worrisome to me.

I am presently working on a project that involves designing a small pressure vessel that operates between .5 torr and 1200 psi at 680 deg F, so I am quite familiar with the potential failure modes and relevant safety requirements of this sort of stuff. I remind the engineers working on this project almost daily about the potential dangers of this sort of energy.

Your comments about just using the water heater as-is and plugging into an inverter are good ones; that is a very safe approach......and it won't void the warranty!


.......also the 12V 300W element has arrived, I just need to find some time.....gotta fix the garage door opener this weekend.

I'm not downplaying anything, and if you read all that I wrote I made it clear that people can get hurt or killed. But you know what, a compressed air tank can also rupture and cause damage or kill someone, and I personally wouldn't call that an explosion either while many would. Again semantics.

Safety is very important to me. Right out of school I worked in a chemical plant where we manufactured products that were explosive, and due to this I took extensive seminars on how to design equipment and control rooms to better withstand the effects of explosions.

I've also designed, purchased, and installed many ASME coded vessels so I'm more familiar with the code than most guys. In one case the working pressure was so high that liquid water was compressible, making even hydrostatic testing dangerous.

Safety is important to me and I'm not downplaying anything. At the same time I'm not going to exaggerate the danger either by blowing it out of proportion.

[boywonder]

I got a little more accomplished on the water heater today.

I got the 12V element installed and removed the 110V power cord and installed 10AWG wires. I also wired in a relay in the van to provide the needed 25A to the heater.

I need to order a bunch of fittings and tubing from McMaster-Carr hopefully tonight; I thought I had more in my "inventory".....I have very little for plumbing this......

1440W 110V and 300W 12V elements. The new element is nickel plated for additional durability....we'll see.....



The 12V element screws right in, as expected.

[boywonder]

OK, I finally got the 12V setup tested. I ordered a few incorrect fittings and I'm still sorting out the final configuration, but I managed to get a basic setup up and wobbling.

The test setup is: 7 gallon water supply (Aquatainer)-shurflo pump-2 gallon expansion tank (used as a reservoir tank also)-water heater. The expansion tank is Tee-d into the line between the pump and the water heater to minimize pump cycling and to increase the water capacity of the system.

After taking temperature rise and battery usage data, I dialed in the air pressure of the tank to maximize the holding capacity of the tank when used in this application. This resulted in a tank holding capacity of 5 quarts with 3 quarts of air on the other side of the bladder/membrane; that's the best that I could get out of the 2 gallon tank.

For testing I monitored battery voltage, voltage at the heater element, and temperature at the output of the water heater by opening a valve and dispensing about 4 oz or so of water into a cup and measuring the temperature of the water with a thermocouple every 10 minutes.

Here is the test setup:











...and here is the temperature rise data....

The water started out at 62 degrees F and it took about 90 minutes to to rise 40 degrees F. Over the course of testing the water samples probably totaled about a quart or so, so that quart of sample water was replaced by cold supply water at every test interval, making the temperature rise a bit slower than not flowing anything.



I ended up setting the water heater thermostat to about 105 deg F, as that seemed sufficiently hot to shower with, although when sprayed through a sprayer, I may dial it hotter.

Shortly after starting the test, the battery voltage dropped to 12.2V and stayed there for the entire test. The voltage at the heater element was 11.3V, reflecting a .9V drop through the 15' of 10AWG wiring. So the battery drain was roughly 25 amps*90 minutes=37 amp-hours.

[boywonder]

...for readers with a keen eye (or for those who actually read the earlier posts), one will notice that my adiabatic (no heat loss) calculations from a few weeks ago were almost exactly 1/2 of reality, so....off by a factor of 2, ie it took 90 minutes for a 40 deg F temperature rise compared to the calculated ideal condition of 45 minutes.

This implies that a 600W 12V element will heat 2 1/2 gallons of water in this pint-sized water heater in about 45 minutes.

[E350]

Congratulations! First night shower success!

This is exciting personally, because I have always sought the goal of a diesel electric camper van.

Integration into solar for second night boondocking showers next?

Thanks for showing your homework, I may actually absorb something about electrical engineering...

(25 amps*1.5 hours=37.5 amp-hours)

[boywonder]

E350: Yes, you got the math right!

A few notes about the heat losses....

I had the thermostat cover off the whole time with a bare section of the tank exposed, ie no insulation. The factory fiberglass install job doesn't look all that great either; there is masking tape squishing the fiberglass around much of the heater diameter.

...and my wiring is burning up .9V so P=VI so P=.9V*25A=22W lost heating up wire.

So with larger wire, the thermostat cover in place and a water heater jacket installed, we could possibly shave 10-15 minutes off of the warm-up time....or just use a 600W element and MUCH larger wire.

Running the engine or connecting to shore power/charging will also speed things up since the batteries will be at 13.2V or more.

[1der]

Thanks boywonder, great info.

I was thinking about a few things:
1. Where did the loss of efficiency occur from your orig 19.25 amps? Perhaps the vessel mass itself acting as a heat sink which also has to be brought up to temperature, plus some heat loss (minor) through the insulation? (some to wire loss as well)
2. Viability of heating water from 40 degrees ambient to 100 (60 degree vs 40 deg rise) - would require 37*1.5 = 55 amps.
3. Is it more efficient to start heating the water when it is already 62 deg ambient and maintain the 102 for a few extra hours? It would be interesting to know what the required amps are to maintain 102 for a couple/three of hours. especially if those hours are when temps have dropped to 50.

Using solar to heat the water will require 37 to 55 amps depending on when the the heating cycle starts. Since we are looking at a 300w panel/ 30 amp MPPT (perfect world numbers), realistically, the solar dump into the water heater after battery replenishment would be marginal. I am being conservative in estimating non-water heater energy usage at 70 amps per day (fridge, Led Lights, charging devices,music, fan for space heater, water pump, no mwave) I think I would be lucky to get 100 amps of solar energy per day which leaves 30 for hot water. Does all that sound about right?

What are thoughts around viability of this approach for off grid water heating? ( let's focus on just this concept/approach and required sizing of battery bank, solar, etc., no need to compare to Propane, Espar, generator, etc.)

[boywonder]

Originally Posted by 1der

Thanks boywonder, great info.

I was thinking about a few things:
1. Where did the loss of efficiency occur from your orig 19.25 amps? Perhaps the vessel mass itself acting as a heat sink which also has to be brought up to temperature, plus some heat loss (minor) through the insulation? (some to wire loss as well)
2. Viability of heating water from 40 degrees ambient to 100 (60 degree vs 40 deg rise) - would require 37*1.5 = 55 amps.
3. Is it more efficient to start heating the water when it is already 62 deg ambient and maintain the 102 for a few extra hours? It would be interesting to know what the required amps are to maintain 102 for a couple/three of hours. especially if those hours are when temps have dropped to 50.

Using solar to heat the water will require 37 to 55 amps depending on when the the heating cycle starts. Since we are looking at a 300w panel/ 30 amp MPPT (perfect world numbers), realistically, the solar dump into the water heater after battery replenishment would be marginal. I am being conservative in estimating non-water heater energy usage at 70 amps per day (fridge, Led Lights, charging devices,music, fan for space heater, water pump, no mwave) I think I would be lucky to get 100 amps of solar energy per day which leaves 30 for hot water. Does all that sound about right?

What are thoughts around viability of this approach for off grid water heating? ( let's focus on just this concept/approach and required sizing of battery bank, solar, etc., no need to compare to Propane, Espar, generator, etc.)

Ray:

1. Certainly, the mass of the steel vessel has to be heated as well (I did not include this in the earlier ballpark calc), and steel has a fairly high specific heat capacity. For fun I should weigh the heater and estimate the mass of the tank. The heat loss through the insulation is not insignificant, it's pretty much everything. If that were zero, we'd just have to heat up the water and not add any more heat....forever....

2. Correct, except your units need to be amp-hours

3. It would only be more efficient to pre-heat and maintain temp if your are plugged in or driving when heating from ambient, since the net energy loss will be greater since the vessel temp is elevated (and losing heat) for a longer period of time. I thought of monitoring the duty cycle to maintain the set temp but that is excruciatingly boring without some sort of data acquisition system to monitor the voltage. The manual method requires a timer and continually looking at the voltage of the element.

Based on your next paragraph.....I'm screwed........

I have a microwave and only 130W of solar with the same batteries as you but I'm hoping it's not that bad. To date, the house batteries seem to support a couple of days of camping no problems, even without the solar panel. If I can get 30 amp-hours of solar per day I'd probably be happy, that would cover the times I need hot water for a shower.

There are a few things that we can do to minimize heat loss, like set the water heater temp to "just hot enough" where you are not mixing in any cold water during usage, and putting additional insulation around the tank as well. The water heater is rather shoe-horned in my install but I can probably get one layer around it.

So far, it seems like it beats a solar bag for convenience especially on cloudy days.

[E350]

Don't worry boywonder, you are not screwed. You must design the system first, then through use you will find creatives ways to use it.

For instance, we stay at an RV park in Draper, UT when we ski Snowbird and have stayed in a nearly boondocking (but with elect hookups) RV park outside Jackson Hole. Your set up would be perfect in those conditions. On drive days between or to ski resorts, the alternator can assist. And this is entirely without use of solar which can only help. I know you are likely a sand SMB'er rather than a snow SMB'er, but your set up will definitely help us snow SMB'ers.

Thank you for your good work. Don't stop.

[boywonder]

E350: 99.9% of my camping does not involve shore power so I may be partially screwed anyway, although maybe not......one test is worth a thousand expert opinions........


Just for fun, the Home depot website says the 2.5 gallon water heater weighs 22 lbs. I don't know if that is gross or net weight, so let's assume that the steel tank weighs 15 lbs.

Specific heat of carbon steel is .12 BTU/(lb-deg F)

15 lbs steel*(.12 btu/(lb degree F)*(40 degrees F)= 72 btu

Refining the earlier calc, 1 gallon of water weighs 8.34 lbs, so

2.5 gallons*8.34 lbs/gallon*(1 btu/(lb degree F)*(40 degrees F)=834 btu

so the total BTU requirement is 834+72=906 btu

906 btu*(.29 watt-hours/btu)=262 watt-hours

the power through the element was a bit reduced based on a voltage across the element of 11.3 volts, so let's use ohm's law to calculate the new power going through the element.

Applying Ohm's law....

P=Vsquared/R, so R=Vsquared/P=(12*12)V/300W=.48 ohms

so the actual power dissipated in the element at 11.3V is approx P=Vsquared/R=(11.3*11.3)/.48 ohms=266W

262 watt-hrs/266W=.98 hrs, so about an hour to heat up 2.5 gallons 40 deg F, assuming no heat loss through the jacket/insulation.

The extra 30 minutes of heat-up time is due to heat loss through the insulation and the heat lost from taking water samples during the test......