Going Off The Grid - Mobile - Part I

1/30/06

The idea: To build a complete, self-contained AC power system that can power at least 2-3 computer systems as well as several lights into a vehicle that can

be lived in and be recharged from the vehicle's own power system as well as from shore or eventually solar power.

A friend and I had the idea that to beat the high cost of living, one could build an RV into the back of a U-Haul or FedEx-type truck and thus eliminate monthly

expenses such as rent, power, water, etc. Such a truck modified as an RV would certainly be large enough for one person to live in if it were modified to be a living area.

But what about power? Could such a vehicle have a self-contained inverter system installed such that it could power several lights, a few computers with displays as well as

maybe a TV & DVD player or stereo? We've started a project to determine if it's possible. We're just getting started with some rough experiments building the power system,

measuring it, and testing it. But will it work?

In the future, we'll get a truck and install it, powering the electrical setup by adding a second alternator to the vehicle's engine. That way one alternator can drive the engine

components, and a second dedicated one can charge the entire inverter system for "in-house" power.

How It Works:

Any good mobile inverter system faces several problems which must be overcome in order to function properly, to protect the fragile electronic equipment that is plugged into it,

and to prolong the life of the system's components which will eventually wear out even under the best conditions. Those problems can be summarized as:

1) Preventing overloading a single input power source, and thus causing it to fail prematurely.

2) Preventing flowback current from flowing from the charger/inverter system back into the input power source while the system is running.

3) Providing adequate protection against corrosion between the system's components in harsh environments such as near oceans or in extremely hot environments.

Excessive heat causes premature component failure and so should be avoided.

The Components:

Any such system at a minimum should include:

1) An input source such as an AC alternator or optional AC shore power to charge the DC batteries.

Solar may also be used as an input, but solar is less efficient and more expensive currently than most other sources.

2) A DC battery array to be charged from which stored power will be drawn and inverted back to AC to power devices.

3) An inverter to convert the stored DC battery power to AC current that can power end-user devices.

Ideally an inverter/pass-through charger should be used so that existing AC power can be passed through directly when available.

4) Ideally some kind of diode flowback box which prevents current from flowing in the wrong direction during charging.

5) Some kind of minimal circuit protector between the final inverter and devices to prevent device overload in the event of a

spike or system failure.

An Early Test System:

Ok! Let's set up a quick test system and see what happens!

Box of parts.

In the final system we'll use an array of 2 or 3 deep-cycle marine batteries in parallel.

For this test we'll use a high-end Hawker 12V 60-amp/hour battery which has

been charged from house AC current. In any such system the ultimate goal is to balance the battery array

such that it can handle the complete load without discharging at a rate that is too fast and without the

output voltage of the system dropping as the system is used. Too many devices drawing too much power at once,

or an inverter that is too small for the device load will both cause the batteries' voltage to drop and also generate excessive heat

in the inverter which will cause its components to fail prematurely. In extreme cases the inverter or even the end-user devices

might be destroyed. We ideally also want a battery array that can be charged reasonably quickly - ideally overnight or while driving the

vehicle for a few hours. But batteries are heavy and expensive - we want to design the system such that the battery array is just slightly

larger than the load and inverter requirements. Most systems overall lose about 10% energy going in and out, or only 80% efficiency.

Such systems also work at their peak efficiency when they are run at 80% of their capacity. So ideally we want to size the system so that the

maximum power drawn by all the devices is never more than 80% of the maximum output of the entire system. In simple terms this means

using a battery array and inverter that have about 20% more capacity than the maximum that the entire system would ever draw. This is

easy to calculate: Total wattage needed == ( (Each device's max Amp rating X 120V) X # of devices). So for example for a 3A max device at

120V, the power system must be able to deliver 360 watts. Keep in mind that is the maximum required and that msot devices run

normally at well below their maximum rating. Problemmatic devices are ones that draw huge amounts of current or require

large startup surges to get the devices going such as air conditioners or refrigerators. If those devices are to be used in the system,

it's better to drive them off a completely spearate subsystem from small-current devices. Ok, so now that we have the basics

let's put together a test system!

Two inverters: a 300W car plugin style and a whopping 700W one with battery terminals,

and some heavy gauge battery cables.

First test: a 13W light bulb powered by the 300W inverter connected to the battery.

No problem. The light bulb stays lit for hours with no problem and no drop in voltage.

Let's insert a multimeter in between the battery and inverter to measue the current drop.

Since we are only driving one or two small devices we can use the small inverter for now.

Now for a slightly heavier test: a sacrifical PowerMac 6100/60 at 3A - something we don't mind

losing if something goes wrong. It seems to work fine but it's only one device.

Tomorrow we will try a bigger test: 3 complete PCs with monitors, a power strip and several lights

all at the same time. In theory 2 or 3 60 AMH batteries in parallel and a 700W inverter should be more than

enough to drive everything for at least 3-4 hours. 2 Deep-cycle marine batteries should be able to drive

a similar system for close to 8 hours. A complete system in a vehicle would have a second alternator feeding into

a flowback box, whose output would feed into the battery array to charge the batteries which would power the inverter

which would drive the devices. An even more ideal addition would be some kind of dashboard box with indicators

monitoring each component in the system so that one could see the status of each component in case something is

about to go wrong.