That’s Not A Junker… That’s My Generator

If you live somewhere prone to power outages, you might have thought about buying a generator. The problem is that small generators are cheap but — well — small. Big generators are expensive. [Jake von Slatt] had an idea. He has a “yard car” which we thought might be a junk car but, instead, it is an old car he uses to drive around his yard doing tasks. It has a winch and a welder. Now it has a big generator, too. You can follow the project in the three videos found below.

The project started with a scrap generator with a blown motor. Of course, the car has a motor so — in theory — pretty simple. Remove the generator from the motor and graft it to the car’s motor. But the details are what will kill you.

The first video shows the teardown of the generator unit and some planning. Some custom parts that were recovered from the scrap pile replaced some of the parts on the generator, including a custom casting leftover from another project that took a bit of machining to repurpose.

By the second video, he had a proof of concept working, but it had a small problem. He was controlling the speed of the motor with a stick on the accelerator linkage. So, he modified an idler from an AC compressor to limit the speed of the generator. The final video uses an Arduino to drive the cruise control to keep a steady RPM to the generator.

Overall, this is an interesting project both because it is unusual and useful, but also because of the wide range of skills [Jake] demonstrates — he knows his way around a machine shop, a car, and the Arduino. Pretty impressive!

We’d rather reuse a car engine than drive the generator ourselves. For some reason, that comes up repeatedly.

104 thoughts on “That’s Not A Junker… That’s My Generator

  1. Nice! I’ve considered doing something similar with my old diesel Jetta. I thought it would be neat if it was like a dynamometer, where you drive the vehicle up onto something with rollers under the wheels. Unfortunately the gearing is a little too much for me at this stage in life at least.

      1. You would have to charge MORE OFTEN as there is no free energy to be had in such arrangements! Study your physics. Mechanical losses, and energy lost through sound, vibration, electrical resistance and heat will prevent such a system from even making as much power as it would take away from the electric car!

      2. Right, That’s first thing I’m thinking then have Solar panels on the entire top, part of the car out of the new material just like the bottom of the car is all batteries sheets of batteries you make the top over the sun, of course one compliments, the other, and you eliminate gas fuel

    1. Buy a big generator with no motor put a car rim or big pully on it. Than get and old diesel motor tractor or truck with a simple injection pump easy to fix put a filter system on the fuel system use a radiator with a auto trans cooler in it run your fuel through that to heat it also keeps the engine cool. Remember diesel run on compression use waist oil kerosene vegetable oil all this is free and people throw it out use your old engine oil any oil as long as it’s runny

    2. That’d work, but then you have drivetrain losses (~10-15%), transfer losses between tires and dyno wheel (I don’t know, but it would be very significant), and then any gearing losses between dyno wheel and the generator shaft. Not at all saying those losses wouldn’t be worth it, just tradeoffs to keep in mind if you ever did decide to pull the trigger on the project.

      1. It would be more efficient if not at idle if the engine size is appropriately matched to the load. Could run a much larger generator that way too, although it would need a much bigger (toothed) belt or direct drive.

  2. Regulating the power & speed from the engine would probably be pretty interesting. You wouldn’t want the engine to bog down too much when a heavy electric load goes on, or race when said load cuts out.
    Idk, maybe that’s in the video?

  3. Fortunately, I have a hybrid and simply connected the high voltage battery to a solar charge controller that works with up to 500 volts and the hybrid easily generates 7500 watts into 120/240 volts, very quietly and efficiently

      1. They never “know why”, or “act surprised”, those are anthropomorphisms. The circuit simply reads instantaneous voltage/capacity and calculates data based on current state-of-charge.

    1. There used to be the same for farm tractors too I believe. My family ran some cow barns with such a hookup, if the power went out they would back a tractor up to the building and install a linkage so it would turn a generator. That was many years ago though.

      1. They still make them – I’ve seen them from 10 KW up to probably 25 KW or so. The economics are borderline. Depreciation on my tractor is very, very roughly $10/hr. It might be $5, it might be $20, but in any case it’s an expensive way to make energy.

        1. It might come out to that if you simply divide the new price with the hours ran. But does using it as a generator shorten its useful lifespan nearly as much as actually working with it?

          The economics may still be between awful and terrible, I just don’t think depreciation would play that much of a role.

          1. Slap a generator on the PTO of a Ford 8n tractor. 70 years old, so no depreciation, they last forever, easy to fix, perfect to have around for emergencies also great for pumping water, sawing lumber, pulling things, plus , you know, tractor stuff. Fun to drive too.

          2. When you have cows that need milked and you have 50 or 150 or possibly even more and you have no electricity you have to do this. It’s not really a choice. It’s not something you’re going to calculate see what the price of electricity is. You have to milk the cows.

        2. What sort of tractor, or are you worried about resale value based on engine hours? Stationary use at part load from time to time isn’t usually going to shorten the lifetime much or add wear to most of the parts; a buyer might be convinced to take into account what you used it for rather than just the number of engine hours.
          For anyone who’s not planning to sell, I wouldn’t worry about that at all. I’d be more interested in seeing what my burn rate was versus a real generator as far as running costs. Real diesel generators of equal reliability&power as small tractors aren’t particularly cheap; if it’s only for backup use I think there’s a case for these things.

    2. My IHC Scout 800A, essentially the love child of a tractor & a truck, had a PTO shaft. One end was used to drive the front wheels, the other end was available unless you had the optional winch.

    3. Land Rover transfer cases had this feature right up to around 2016 although in later Range Rovers and Discoverys there was frequently stuff in the way that would’ve prevented you bolting a large takeoff on there – it was/is also used for overdrive/underdrive units as well as the more traditional hydraulic pumps, fire pumps, generators, and agricultural takeoffs sometimes fitted to earlier models.

      Then again Range Rover still had a hole in the front bumper for a starting handle into the 1980’s…

      1. Per their website that solution relies on your car’s alternator to generate electricity. It’s basically just an inverter in a fancy box. I could be wrong but I don’t think most alternators are rated for the 1000w that they say their product can pull.

        1. I think you may, in fact, be wrong on that. Even a Honda Civic from 2016 comes specced with a 95A alternator and it’s going to run somewhere between 12-14.4V so that’s a pretty decent output and that’s an older/smaller car (at least in the US).
          Granted there are inefficiencies that come into play and you wouldn’t want to run the car at 6000+ RPM all the time to get the alternator’s max output, but 1000W isn’t that unrealistic of a number for a modern alternator IMO.

          1. You would NOT need to run the car at high RPM, as alternators on cars are normally spun about 3 times FASTER than engine RPM due to the effective gear ratio created by the pully size differences, usually engine speed below 2000 RPM will give very good alternator output.

        2. Modern car alternators are often 100 amp output, so ~1200W, and it doesn’t take much more than 10A to run a car even with fuel injection etc.

          120-150A is not uncommon on more luxurious models and I’m sure 200A has been done at least a couple of times.

          Having the car battery there to absorb large demand spikes you could probably run a 2 or 3kW inverter with an average 1kW load quite happily even accounting for losses.

    1. They make power inverters that connect to your cars battery/altinator and convert the 12vdc to 120vac. Harbor Freight (for example) sells one rated for 5000W continuous 10000W peak for $400

      1. It’s pretty trivial to install an inverter in most older vehicles. I’ve done it in 3 of my trucks, the most recent being a harbor-freight 1000/2000W under the hood (NOT recommended) of my 88 s10. Lots of room for airflow, so I’m not too worried about heat.

        Note, though – DO NOT use a lawnmower style starter solenoid for your power relay. They do not like a contiuous load. I’ve got mine hardwired and run off the inverters power switch, but a more elegant relocation would use a heavy-duty solenoid (plow, hydraulic motor) or contactor.

        . . .

        What I’ve always wanted to do, though, is have a “yard truck” with a modified transmission that would allow for fully decoupled power takeoff – If you live in a rust-belt area, industrial size generator heads come up for sale stupidly cheap every few months – think 100-250kW for under a grand. Yeah, overkill, but how much power is a standard full-size truck capable of? Well within that.

    1. That doesn’t make sense as a unit. Watts is a unit of power, it is joules per second, so Watts per gallon would be joules per second per gallon. Watts per gallon is pretty much like saying horsepower per gallon, it just doesn’t make sense.

      An appropriate unit would be joules per gallon but even then you are mixing metric and imperial units which is never a good thing, even better would be joules per litre.

  4. I hope and pray that everyone will realize that we, the whole world, will have to break our addiction to fossil fuels and buy a bunch of used solar panels – 250 watts each, fifty dollars each. Then get a decent controller and batteries. An inverter to get AC, and never, ever have to buy any fossil fuel. Please reuse and save the Earth.

    1. A solar panel produces 50% of its capacity compared to when new at the 10 year mark. If you purchase solar panels new (actual price not a tax payers funded discount price) . You will never produce enough energy to cover the cost of panels. It is a money loosing venture.

      1. Using your preferred search engine, you should find similar results. Of course, there is no harm in reading the articles in their entirety, but here are some sections that address the actual measured degradation rates. You’re welcome.

        See the section: How does solar panel efficiency change over time?

        https://www.cnet.com/home/energy-and-utilities/solar-panel-efficiency/

        See the section: What Is Solar Panel Degradation Rate?

        https://www.forbes.com/home-improvement/solar/how-long-do-solar-panels-last/

        See the section: Measuring solar panel degradation

        https://www.consumeraffairs.com/solar-energy/how-long-do-solar-panels-last.html

      2. And this is just false. There is a grain of truth, output does decline. But it’s more like 80-90% after 10years, way more than 50%.

        And payback is usually between 2-7 years, depending on electricity price and distance to equator.

        If you take a loan to install, then payback is in practice 0, since you trade your electric bill into paying for the panels instead.

      3. degradation rate is when a solar panel has reduced its power output and is considered a consistent risk for your solar power system. On average, solar panels’ energy production will decrease by 0.5% annually. However, this is a very slight decrease over the years. For instance, after 20 years, your panels should still work at about 90% of their original output.

      4. You’re just trading one environmental factor for another. Just think about all the toxic and heavy metals that are involved. Every EV produced is a rolling environmental disaster and people like you are blind to the truth. Climate alarmists are telling you things as fact when it is not a fact but only their belief. It’s their new religion. They believe it therefore it must be true. But they have no answers when confronted with real evidence contradicting their beliefs.

        1. I do believe you have all over looked the simple rule of every action has an equal and opposite reaction. Hence. A stàrter is engaged a electric magnet pulls a lever. + Lever with dual pivots could and would throw the starter gear down the bindex causing the actual starter to turn just a little. Then the return spring and inertia feom already moving fly wheel would turn some more then when returning to its original position would pass through same field that launched it to begin with. So with proper bindex on starter shaft fly wheel is kept at a steady speed by the thrust of a single electrical action and surplus electricity is then used for boost motors heating. Cooling. So on

    2. Used solar panels are sold for a reason. If they were still worth using then why would they be sold?

      Your 250 Watt rating is probably peak rating when new. Most solar panels will never produce that amount of power more than just instantaneously.

      You say to get batteries but batteries, especially lithium aren’t good for the environment either and have a pretty limited lifespan.

      Your “solution” only covers home use and only when there is enough sunlight to produce power which for a lot of the world isn’t a lot of the time. It also never mentions anything about one of the largest other everyday uses of fossil fuels, transportation. Most people cannot afford an electric car and the range of electric cars decreases quite a lot over time, hence reducing the usefulness of a used electric car.

    3. And really long extension cords so we can power our transport with those used solar panels. And good thing we are building so much used solar panel capacity into our economy. 9 billion whole world people are going to need a whole lot of used panels.

  5. I have used a bunch of different equipment to power my off grid home and business.
    Best part is no recuring bills.
    Currently the bulk of my power comes from a pv
    array,and it is completely reliable,even with mechanical
    damage to some of the pannels.The whole off grid sysyem has also payed for itself,many times over.
    I have built little generators,and have a number of them that will end up as doners,for someone elses projects.
    here is the best web site devoted to all things to do with
    self reliance

    https://www.lowtechmagazine.com/

  6. Any Van Slatt project is bound to be interesting and informative, but always entertaining and inspiring. I don’t know if he reads Hackaday, but I did have reaction to a few of his design choices. His project works, so my remarks may not be of value to him now, but…

    A 7812 is not a good choice of regulator for his circuit, because it needs a few volts of overhead to function stably. There are LDO alternatives, or he could choose a 7809, instead.

    The speed interface seems way overkill. Transformer isolation makes sense, but with an optocoupler, a few resistors, and a few diodes, you can read the AC output directly while switching the 5v rail (from the Arduino) into the speed measuring pin. In fact, with an optocoupler it might be possible (and better) to intercept a tach signal or ignition coil drive on the
    engine compartment harness someplace and control his generator’s freq on the basis of one of those signals.

    As to the control algorithm, I concur with a Youtube commenter who suggested that a PID scheme might be best. However, a simple change to the OP’s existing code could afford significant benefit at the expense of 2 or 3 lines of code. Instead of hard-coding discrete motor timing constants for different magnitudes of frequency error, write a simple equation to generate those “constants” on the fly. Subtract desired freq from measured freq, times a scaling value you determine empirically. The greater the error, the harder the motor is driven to correct. This gets you the P in PID.

    I employed a scheme like this many years ago to control an environmental control heater on a a very sensitive pressure transducer, but with one difference… instead of a linear relationship between error and corrective response, I squared the error (second order expression.) If you plot the correction response of the linear system, it looks like a VEE. If you plot the second-order correction, it looks like a parabola.

      1. LDO – Low Dropout
        AC – Alternating Current
        Arduino – open-source electronics platform
        freq – frequency
        PID – Proportional-Integral-Derivative
        OP – Original Poster
        (Generated by an LLM ;)

  7. Not related to generators per se. But this reminded me of something me Dad told me many years ago. He said they used to make some sort of kit where you could replaced half of V8 engine head with an air compressor head or something like that. Or maybe it was half of an air cooled Volkswagen engine? I’ve wondered about doing something like that to have a neat, little, compact, everything sort of utility engine for making compressed air, plus alternator for welding, etc.

    1. There’s some big air compressors back in the day that used a Ford V8 and converted one bank to compressors. There was one at the junkyard I worked at when I was a kid, it was old by that point and didn’t run.

    2. There was an industrial version of the VW aircooled flat-four that used one bank as an air compressor. Not sure what it was called.

      The was also a thing called an “engine-air” if i remember correctly. This used a hose to connect to a spark plug socket on your VW, and used the pressure to drive a secondary piston, which drove a compressor. Probably a bit of a fire hazard, but not as much as filling your tyre directly with fuel/air from the cylinder! I haven’t seen a real one, but read about this in “How to Keep your Volkswagen Alive” by John Muir.

  8. Circuit could benefit from a dedicated hardware safety circuit… If the frequency gets too high then a dedicated frequency counter could cause the cruise control clutch to disengage…. Let’s say disengage at some really high number for example 68hz. Would be a shame of complex hardware/software failure would blow the engine or overspeed the house appliances….

  9. Some reasons why perfectly good solar panels might be available used:
    — panels may have been fully depreciated
    — new panels may qualify for a new round of depreciation
    — incentives may be available to upgrade a solar farm
    — all upfront investments in land, permits, politics, racking, transmission infrastructure have already been made.
    — New panels are cheaper and more powerful than the old ones were, even when new. More salable output from the existing support infrastructure. Fresh depreciation. Etc. Investors can calculate things differently than the rest of us.

    1. Knock freeze plugs out. Draft all intake air through the water jackets and cool the oil. Water pump becomes a supercharger. What !!!!!!! Yeah I just said that. Water in engine block. Stupid. Roo cold freezes breaks things. To hot. Too much pressure blows and breaks things. Too small of a window to be as good universal as back up or emergency may require

  10. Great story. May not be practical at the end but certainly interesting in terms of what someone with the intellect, abilities and right equipment might accomplish. Thanks for presenting this adventure 😄

  11. Back in the mid-1970s you could buy a kit from J. C. Whitney to get AC from the car alternator. That is if the alternator didn’t have an internal voltage regulator, you could use the AC voltage it produced. It wasn’t 60 Hz, and the engine had to be running fast to get a full 120 volts from it. Current was limited to what the alternator could produce without burning out. It was simple enough a person could do it without buying a kit.

  12. Nice project well done, but Hackaday messed up the subject heading
    The Arduino is “controlling” the engine speed ,. Not driving the generator head,
    The generator head is driven aka powered by the cars engine

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.