Turning an old car into a powerful generator

Modern cars are powerful. Today there's no car on the market today that doesn't have enough power. The cheapest cars you can buy, small sub $20k hatchbacks, are all around 100kw. They all have performance figures that beat sports cars from the 1970s.

The limiting factor on early acceleration (0-60kph) today for all cars is now tyres. Beyond that air resistance might make more engine power matter but in general it's all about the tyres. Every car can spin their wheels at the lights if they want to. If you feel the need to floor it at the lights you're showing off your tyres not your engine power (and also wearing them out).

And in many cases, that much power transports a single person and occasionally a few bags of groceries.

edit: By point of comparison, a good cyclist produces about 100 watts, probably less if they're on a leisurely cruise. That's about 2% of the wattage of a car engine.

The continuation of this thought is to realize what an immense amount of power there is in a gallon of gas. It is basically free energy compared to using serfs or beasts of burden to do the work.

This is very visible in EVs. They can power a house for multiple days, but last only few hours at highway speeds.

If you actually run it at that power constantly it's not going to last long though. It's peak power. You're not going to drag race your car, most of the time it's just cruising.

This is a big difference to (piston) airplane engines because those are designed to give near 100% a lot of the time.

petrol is roughly 9.5kW/litre (~35kWh/US gallon) if that's a better reference for consumption (of which maybe 30% will make it to energy through a modern engine?)

Its why having an EV hookup is so funny: your house is now a tiny percentage of your power bill/load dimensioning.

Also why I'd really like to have my car battery be usable by the house: storing solar or winter/night consumption all require a battery. You'll never need a bigger one for your house than the one already in your EV.

The rediculoussness of the energy use of transport is especially noticeable when you are driving a Tesla which by default shows you a dial of power consumption and generation in kW.

When rolling down a hill with regenerative breaking, the car is generating 50kW.

You can literally power many average homes with that.

> By gearbox I mean anything to change from one RPM to another

I thought you meant the car's gearbox

I’m very much not an engineer, but a below and pulley changing the ratio seems the most efficient.

This is a reasonable desperation setup for emergency power, but not a great off-grid solution.

Highway cruise for a compact car needs around 7 to 20HP, so a reasonable target output is in that range. That's 5 to 15 KWh. Seems small, but as others have pointed out, auto engines have peak outputs far beyond their continuous rating.

1800 RPM 60Hz generators are available, and larger generators tend to run at 1800 RPM. Or you could do something with belts or gears to keep the engine RPM down, as others pointed out. Running at low RPM is good if you want to run for a long time.

Probably a good idea to have the system disconnect output power until the frequency reaches at least 50 Hz, because this thing needs quite a while to reach operating frequency. Bringing up something like a refrigeration compressor (a likely emergency load) from 0 Hz to 60 Hz over the course of a minute may burn it out. Under-frequency operation is very bad for AC motors; they draw way too much current and overheat, because the inductance of the motor isn't able to oppose the lower frequency. Put an ohmmeter across an AC motor and note how low the DC resistance is.

> A cars cooling system is designed around that and the case where you need all 150 horsepower for 5 seconds, then drive for 30 seconds at 20 horse power before slamming on the brakes

My old 2004 Opel 1.6, 105hp could deliver almost full load (car fully packed with passengers and baggage, with roof trunk also packed, going 140km/h uphill with fully open throttle, fuel usage was reported as 16L/100km, typical road usage was 8L/100km) for about half an hour without any problems.

>A cars cooling system is designed around that and the case where you need all 150 horsepower for 5 seconds, then drive for 30 seconds at 20 horse power before slamming on the brakes (6 horsepower to run the AC) for the next red light where you wait for 10 seconds (again 6 horse power for the AC).

I'm no expert, but I can't imagine that this is true. I agree that is "average driving conditions", but there are plenty of times you're driving way outside of those conditions. That can't be how the cooling system is designed. I've never seen a car even tick up in temperature, under all kinds of tom foolery.

> max power means your engine needs a full rebuild every 20 hours of operation

Top fuel drag racers need a rebuild after every run, which typically lasts 4-5 seconds. Their engines can produce 10,000 HP, which is about 7.5 MW.

https://en.wikipedia.org/wiki/Top_Fuel

Depends on… well, everything, but towing a 1-tonne trailer up a long hill is far more taxing on a car’s cooling system than this. I’d expect any decent car to be designed to handle continuous running at peak torque, if not peak power.

(Why peak torque? Because that’s peak efficiency, for petrol engines at least. Makes sense if you think about it.)

"Your car engine can typically deliver 150 horse power to the wheels, but after 6 seconds of that you are at freeway speeds and then your engine is only delivering about 25 horsepower."

Unless you're in Colorado doing 80 mph, uphill for ten miles, fully loaded with cargo - then you engine brake (ie dissipate your KE into the engine block) on the way down. Or you take you car to the track. In both situations your cooling system is still expected to deliver, and does, since its not that big an issue to size it properly and a massive reliability issue if it isn't.

25hp is probably enough to power your household (18kW or 170A at 110V). It's certainly well over the mean summer usage, but peak usage is probably higher in hot climates (I couldn't find any typical peak usage numbers).

A car alternator is typically 2kW-3kW at most, and is very roughly 50% mechanically efficient, so it uses 4-6kW of mechanical power. Ie: 1/18th of the 150hp (111kw) engine's capacity...about 5%.

Cars are already designed to run A/C and alternator almost continuously, and the AC not only generates load, it pumps a lot of heat into the air coming through the radiator because the condenser is in front of the radiator. They're designed (if the manufacturer did their environmental testing properly) to do that even in ~110+ degree weather.

Methinks you should stick to things you know something about.

PS: Many cars, even regular passenger cars - can handle being driven around a track, where you can go through a full tank in under an hour's worth of driving, and are either on the gas or braking during most of the session.

There are also things called hills and mountains, which may take minutes to climb, or more. Plenty of cars make it up the Mt. Washington auto road (where the challenge is making it back down without overheating one's brakes; engine braking must be used.)

There's something called "towing", which lots of people do the world around with minivans and passenger cars (not just pickups and SUVs.)

The fan in cars is typically not enough to keep it cool while stationary under load, but in typical home applications you won't have that big load anyway. Big load for a car is >10kW. You can also pretty cheaply get a bigger external fan, like those used on dynamometers.

One nice side effect of using car as generator - free heat in winter if you extend cooling loop to your home.

The fans cars have are not sufficient to cool the engine at high load.

> 3. remove the mechanical thermostat on the engine, so that it takes longer to get to operating temperature.

Modern engines are designed to work at preferred temperature. They have two cooling loops, one small which is constantly cooling cylinders and one big (with radiator) which cools the small one to keep engine at proper temperature, not too high and not too low, by mixing some of small loop coolant with large loop coolant.

Point 2 is the best solution.

Install a larger radiator is the correct fix if this is a concern. While the stock fan is not the most airflow, if it isn't enough you want the bigger radiator.

I read "gearbox" as, well, an automotive gearbox, which would hardly be surprising in the context of the article.

2:1 could easily mean crank the gearbox into a gear that approximately produces that ratio.

> You've tried to inject something on your own accord for possibly a misunderstanding??

WTH is that supposed to mean? If I misunderstood anything it was a honest mistake. GP suggested a RPM reduction for fuel efficiency reasons, it would be counterproductive to offset that efficiency gain with a high-loss mechanical device. There was neither ill intent nor hidden agenda.

Yeah, they didn't mention many specifics; I also imagine that they're just pulling the power off the engine using the accessory belt connected to their existing alternator -> inverter setup.

Similarly -- the concerns about how to cool the device if it s putting down 100hp -- that's 74570 watts; it's unlikely they're using a harbor freight generator's alternator to generate that kind of power.

> Are we talking about a "slushbox" automatic without a locking torque converter, or an all gear manual transmission - there is a big difference.

Correct. IIRC again (numbers from 2010ish, probably there were some improvements since but I bet physics limit those) the best stick were 15%-20%; slushboxes were 25%-30%.

Such loss is justified by making the car able to operate on a much wider range, from standstill to highway speeds, and solve the ~zero torque at <1000rpm problem. IOW without a gearbox the car is useless.

Without that requirement a simple reduction would be much more efficient.

One of the huge benefits of going electric is that - in addition of having stellar electric to mechanical energy conversion efficiency (90-95%, vs a very lousy number I can't recall for chemical to mechanic energy) - the engine can basically be put on direct drive because torque is immediately available starting at 0rpm and all the way to max rpm, so no additional loss.

> I didn't specify what type of gear box because there are many different options. Belts of the front pulleys would work. You could rig up a chain drive system as well. there are pros and cons to each.

Totally fair! I didn't want to sound dismissing, only complementary.

Automotive gearboxes prime goal is to:

a) expand the workable range of the IC engine (1000-5000rpm for Diesel, 1000-9000 - at best - for NA petrol, turbocharging dials it down)

b) solve the zero-to-low rpm problem that causes stalling due to nonexistent torque

It results in a complex device where many parts are still spinning even on the non-engaged gear. Look at a cut out of a manual transmission. The irony is that an epicyclic automatic transmission is mechanically efficient, up to the torque converter which uses fluid coupling.

Efficiency is a secondary goal to the ones above, without which the car would be useless.

If the torque converter doesn't have a lock that is a big loss. Those same transmissions also tend to have low efficiency hydraulic pumps internally to select the gears. Modern automatic transmissions tend to be much better than older ones.

>all they need electricity for is the engine starter and, depending on age, the cooling fan.

If you get an old enough diesel engine or can do the mod yourself you don't even need an electric starter. The first diesel engine I worked on used a gasoline engine as a "pony motor" to spin the diesel engine flywheel enough to generate the compression needed to start the diesel engine. From memory (probably wrong haha) you cranked the gasoline engine and brought it up to high rpm and then you started the diesel engine using a lever that gradually engaged the diesel engine flywheel bringing it up to operating speed. Once the diesel engine was running you backed off the lever and killed the gasoline engine.

An obvious disadvantage would be the necessity to maintain stocks of two fuel types. Other than that it is a near fool-proof way to handle cranking a diesel engine. The electric starter is replaced by the gasoline engine and flywheel linkage. You effectively roll-start the diesel using the gasoline engine.

I've tried with Opel from 2004. I had a very modern version (with everything over CAN, even gas pedal) and I had to disconnect A LOT of sensors in order for it to no longer start (don't ask why). Modern diesel cars are a little worse for this, because of all the emissions regulations requiring adblue and dpf filters to work correctly. If you run out of adblue in some cars, you have to get it towed to dealer for checkup and reset.

Most household requirements are about 13kW. So not much different than your suburban bungalow needs.

230VAC flat screen TVs tend to be powered by 14VDC at the motherboard level, might well function if fed 12VDC from an external power source, bypassing the built-in step-down converter entirely.

Having wired up a couple travel trailers and a boat (and, incidentally, being an EE who can theoretically run those calcs), I agree with the parent: It would be better if the industry switched to 120V/220V AC for more parts, and away from 12V.

The problem is that the ampacity in a 12V DC light circuit is 10x that of an equivalent-power 120V AC system. And worse yet, power dissipation in a resistive element is equal to I^2 R, so you're heating the wiring harness with power losses are 100x worse than in a 120V system and almost 200x worse than a 220V system. So you have to oversize the wires to reduce R, which adds weight and cost. And if you drop 2V to line losses in your 12V system, your device is seeing 83% of the nominal voltage, while if you drop 2V in a 120V system (you won't, because Ohm says voltage drop is equal to IR, and I is 1/10th that of a 12V system), but still, you're 98% nominal.

They're running 12V because that's the voltage of a classic 6-cell lead-acid battery, and you probably didn't want 220V in your Model T, even with felted asbestos insulation, but I'd be confident enough to lick a modern XLPE wire carrying 220V. A century of using those lead acid batteries (and worse, 3-cell 6V systems...can't endorse converting your classic car to 12V enough...) created now-entrenched economies of scale for lamps, and switches, and pumps, and radios, and fans, and most of the other things you need in a boat or camper. The only advantage of using those 12V parts is that their manufacturers put a little more effort into making them efficient. No one (sadly) will notice or care if the AC-DC converter that runs the clock on your stove burns 5W at idle even though your wall clock can run on one AA battery for years, but they will notice and care if their camper battery is dead after leaving it parked for a couple weeks.

Today, I recommend using LiFePO4 batteries in whatever series cell arrangement gets you the required watt hours, regardless of that output voltage. Then run a modern, high-efficiency (high-frequency) digital inverter to bump the voltage up to whatever your local AC grid runs at. You can go 220V AC in the US if you really want more efficiency and smaller wires, and are willing to deal with the hassle of finding the right lightbulbs and international power cords and so on.

Yes.

Small internal combustion engines are super simple, and small engine manufacturers like those that make generators have an equally impressive dealer, spare parts and repair network to automotive manufacturers.

In this case it was a harbor freight generator. They have lots of replacement engines available for very affordable prices (starting at less than $200) https://www.harborfreight.com/generators-engines/engines.htm...

I always say, why buy a Toyota when a lawnmower engine will do do? It's not like Toyotas are more common.

You don't really need an RTOS to do something like this safely, but agree you need more firmware knowledge than "tinkers with Arduinos" to stay clear of possible unfortunate failure modes. You definitely need to build in some interlocks and/or other sanity checking.

Similar with the PCB's, many of the better dev boards are pretty decent, the problem comes with poor connections designed for easy access, etc. This can be resolved without spinning a new board though, or perhaps designing a simple extender.

Arduino is single threaded on bare metal — you don’t need RTOS in this case, afaict

The engine is for sale if you want to buy it. Most engine will last longer than the car if you take care of it. Back in the 1950s engines were a lot more likely to fail and so people would keep an old car running by buying a junkyard engine. Now that isn't very popular (it still happens, but it isn't popular)

A high powered alternator and a carbureted engine with mechanical timing and nothing else on the engine but a starter, exhaust, flywheel, and cooling seems ideal.

Something like this: https://www.powerbastards.com/proddetail.asp?prod=Fitzall-22...

That way you can set the starter idle for high RPMs to get the system up to speed with the resistance from the alternators and then set the idle speed at the ideal speed for the alternator to produce the most power.

But even on a 4 cylinder engine you're going to have horsepower to spare, so you may want multiple alternators wired in series with beefy transfer bars, so build a custom mounting plate for as many as it takes to almost bog down the engine at 1200 rpm or so.

These $350 alternators produce 220 amps at 14.6 volts at 1200 rpm, or 3200 watts each. I imagine you could run at least 3 of them on a properly set up 4 cylinder engine. That's getting close to 10 kilowatts of power before conversion and you would probably still not be taxing the system.

On the other hand, at this point you've spent $2,000 or so and a month of backyard engineering time to build an 8000 watt generator when you can buy a 13,500 watt generator at lowes for $1,300 dollars.

If you have a good motor you can run on a stand and a bunch of cheap or free alternators, then you just need the mounting system and inverter. Typical alternators put out about 40-80 amps, or 580-1200 watts. That at least has a chance of being cheaper.