Car Alternators Make Great Electric Motors; Here’s How

The humble automotive alternator hides an interesting secret. Known as the part that converts power from internal combustion into the electricity needed to run everything else, they can also themselves be used as an electric motor.

The schematic of a simple automotive alternator, from US patent 3329841A filed in 1963 for Robert Bosch GmbH .
The schematic of a simple automotive alternator, from US patent 3329841A filed in 1963 for Robert Bosch GmbH.

These devices almost always take the form of a 3-phase alternator with the magnetic component supplied by an electromagnet on the rotor, and come with a rectifier and regulator pack to convert the higher AC voltage to 12V for the car electrical systems. Internally they have three connections to the stator coils which appear to be universally wired in a delta configuration, and a pair of connections to a set of brushes supplying the rotor coils through a set of slip rings. They have a surprisingly high capacity, and estimates put their capabilities as motors in the several horsepower. Best of all they are readily available second-hand and also surprisingly cheap, the Ford Focus unit shown here came from an eBay car breaker and cost only £15 (about $20).

We already hear you shouting “Why?!” at your magical internet device as you read this. Let’s jump into that.

These People Think Building Their Own Electric Vehicles is Fun!

One of the interesting facets of watching the UK Hacky Racer series grow from a bunch of friends making silly electric vehicles to something approaching a formal race series has been seeing the evolution of the art of building a Hacky Racer.  As the slightly grubbier cousin of the US Power Racing series it has benefited somewhat from inheriting some of their evolutionary experience, but that hasn’t stopped the Hacky Racers coming up with their own vehicle developments. They’ve moved from salvaged mobility and golf buggy motors to Chinese electric bicycle and tricycle motors, and now the more adventurous constructors are starting to look further afield for motive power. One promising source for an inexpensive decently-powered motor comes in the form of the car alternator.

Our Ford Focus alternator
Our Ford Focus alternator

Searching for car alternator conversions reveals a variety of pages, HOWTOs, and guides, many of which can be extremely confusing and overcomplex. In particular there are suggestions concerning the three stator connections, with advice to break out the individual windings and apply special wiring configurations to them. Based upon the experience of converting quite a few alternators this appears surprising, as all the various models we’ve converted have had the same ready-to-go delta configuration that needed no rewiring at all. Perhaps it’s time to present a Hackaday guide with a real alternator, and explode any remaining myths while we’re at it.

So, fired up by the prospect of a cheap brushless motor by the passage above, you’ve got a Ford Focus alternator on the bench before you. How does one go about converting it?

Wanton Destruction Of An Innocent Car Part

Removing the regulator/brush assembly
Removing the regulator/brush assembly

On the back of a modern alternator is universally a plastic dust cover secured by a set of bolts. These devices are designed to be refurbished so (perhaps surprisingly for a modern automotive component) they are usually very easy indeed to dismantle. If you take off the dust cover you’ll see the regulator, rectifiers, and brushes, sometimes integrated into a single unit, but more usually as in the case of the Focus alternator with the regulator and brushes as a separate assembly to the rectifier.

There is often a copious quantity of silicone sealant which needs to be cut away, but any nuts or bolts that secure the regulator should be able to be undone, and with care not to damage the brushes themselves it can be lifted clear in one piece. Then the rectifier unit can be removed, a process in which it is sometimes simpler to attack it with side cutters rather than try to remove it in one piece.

The rear plate of the alternator with the regulator and rectifier removed, showing the stator winding connections.
The rear plate of the alternator with the regulator and rectifier removed, showing the stator winding connections.

You should be able to identify the three bundles of thick enameled copper wires coming from the stator coils, and detach the rectifier straps from them. In some alternators they’re soldered, but some other particularly annoying designs they’re spot-welded. At the end of the dismantling process you should have a bare alternator with three sets of stator wires protruding and a bare shaft with two slip rings, whatever remains of the rectifier pack, and the regulator/brush pack.

The next step is to remove the regulator circuitry while preserving the shape of the regulator/brush assembly, and to locate and preserve the brush connections where they meet the regulator. Yet again there will be copious quantities of silicone potting compound to hack away, but eventually the regulator should be exposed. These are universally some form of hybrid circuit on a ceramic or metal substrate, with connections emerging from the moulded plastic surrounding them being soldered to pads on their edges. It should be relatively straightforward to identify the pair of connections for the brushes, carefully unsolder them, and push out the regulator circuit.

The completed motor.
The completed motor.

Finally, you should have a bare alternator, a brush pack with a missing regulator circuit, and the plastic dust cover. Simply solder three suitably large-gauge wires to the three sets of stator wires and cover them in heat-shrink, solder a pair of lighter wires to the brush connections, and reassemble the brush pack to the alternator. You may need to put some form of strain relief on the wires to the brushes. The rectifier pack doesn’t need reassembling, so on some models you may need to make a spacer to replace it in supporting one side of the brush pack.

Holes can be made in the dust cover for all the various wires, and the dust cover fitted with all the wires poking through. At this point you’ve converted your alternator, and all that remains is to drive it with something. Fortunately that is a surprisingly simple process with off-the-shelf parts.

Driving Your New Motor

Motor and controller, on the bench.
Motor and controller, on the bench.

A so-called brushless DC motor is simply an AC motor with a bundle of electronics that turns a DC supply into an AC one to run it. They have the advantage over brushed DC motors in reliability, efficiency, and ease of speed control, but at the expense of more complexity.

The good news for people converting automotive alternators into electric motors is that a whole range of brushless motor controllers can be had for not a lot of money, in the form of electronic speed controllers (ESC) intended for those Chinese electric bicycles and tricycles. They take a battery DC supply and produce a three-phase AC suitable to drive a delta-connected motor, and they work well with converted alternators.

ESCs have two modes, one for motors with Hall-effect feedback sensors, and one for motors without such as our alternator. Usually a wire link needs to be made to enable this, consult the instructions for your controller. We’ve found that an alternator drives well as a motor from a 36V or a 48V supply, and as long as a controller with enough power is used then they do so reliably. A quick AliExpress search for “brushless motor controller 1500W” turns up plenty of choice.

Given a controller, there is one more requirement for our alternator to become a motor, it must have a DC supply to its rotor winding. It needs to have about 2 or 3A flowing through it, for which a current-limited PSU module performs the task admirably. Having to use that power makes the motor a bit less efficient than a permanent magnet one, but the cost of a scrap alternator is hard to beat.

The motor featured in our pictures is destined to be one of a pair providing traction in a new car for an assault on this year’s races. Personal experience with SMIDSY the Robot Wars robot would lead me to give them forced-air cooling, but unlike the electric tricycle motors these do seem to cope well with getting hot. An alternator motor might not be the one-stop solution to whatever your small-scale traction needs could be, but even so it’s worth being aware that they are an option without unexpected wiring rituals. If you convert one for a project, please make sure to write it up and send it to our tips line!

163 thoughts on “Car Alternators Make Great Electric Motors; Here’s How

  1. this only applies to alternators with an internal regulator

    older ones that use an external regulator do not work like this

    they have a shitload of coils, each on has it’s own “tophat” diode

    that is one of the reasons older cars were so electrically noisy

        1. Why would someone want to hack parts for older cars into something that is not a car part? 1990 is a long time ago now. (making myself feel old). There should be plenty of post-1990 cars at any junkyard. Leave the pre-1990 stuff alone. Somebody might want it for keeping their pet older car running!

          1. Yeah, but people who want to replicate this need to know from the start that there ARE alternators that don’t work. When I go to source components, I chose the cheapest one that I think will work. If I didn’t know the “needs to be from a post-1990 car” I’d be annoyed if that leads me to spend money on a non-working part.

          2. “Why would someone want to hack parts for older cars into something that is not a car part?”

            You do realize you just asked this question on hackaday right? That’s like asking why anyone would want to drive Toyota on a Toyota forum.

        2. Yeah, but this article isn’t about keeping older cars and the alternators that go with them, this article is about turning certain types of alternators into electric motors for other projects.

          I get you keep older cars, but it’s irrelevant here.

        3. “oh no. some scrappy inventor type will staele my car alternator from my 1970’s antique.” Stop with the stash of Qualudes you are working with. No one cares about some ancient piece of crap for new stuff. Enjoy the 5 gallons to the mile and zip it.

        1. @Jenny
          In all this fascinating discussion one thing I haven’t said is:
          Great article Jenny. Thanks for sharing it and your knowledge and expertise. It has inspired me to maybe go back to my own alternator experiments. Maybe the low speed torque issue is just all talk! :-)

    1. I’ve worked on a lot of car alternators and the only real difference between internal and external regulated is where the regulator is located (internal or external). Both designs have the same 6 diode pack to change their 3 phase alternator output to dc and both designs have the same smaller 6 diode pack to feed the regulator circuit from the 3 phase alternator once it’s producing power. There are a few single phase alternators out there that claim to do a better job but they are really just a manufacturing cost savings.

      One thing to know is all of these alternators are designed to put out power after a certain RPM is reached, for example older alternators did not put out appreciable power until they reached 2000 RPM, so the same should be true when driving one of these in reverse. You should fine an RPM where torque will ramp up.

      1. that’s mainly the regulator, they are really crude

        Elecktor published a great design for a “proper” regulator, it ramps up the field coil rather than just switching it on

        my LJ50 Suzuki 4WD had a “boost” switch for over taking, a relay that killed power to the field coil

        1. Today some engine ECUs do the same: disable the alternator or the A/C in case of high power requirements (overtaking). Some also try to maximize battery charge/alternator power (voltage) in “engine brake” situations to recover some of the kinetic energy of the vehicle.

      2. I have an 1940’s vintage Allis Chalmers tractor. Outside of it being 6V and positive ground, and having a generator and not an alternator it has one other really neat feature that I wish modern cars had. It has a knob you can pull out that changes the charging rate. One position is about 2A which is about what the ignition takes, and the other is about 20A. The 20A position is really for running the headlights, but it is also nice if you just fire it up for a few minutes as it puts a pretty significant charge on the battery pretty quickly. The fly in the ointment of course is that you have to remember not to cook the battery with it. It also has a manual choke that you have to remember to turn off. Luckily, the tractor just starts to cough and run rough if you leave the choke pulled out, and that will also remind you to push the generator knob in with no other ill effects. I have an old motorcycle that has a manual choke and if you take off and forget to push the choke on that back in, by the time it starts coughing and running poorly, the plugs are fouled and it is not going to start again if you turn it off. You learned early on with that bike to always carry a spare set of plugs with you in the tool kit. You also learned that it was smarter to reach under the tank and lift up the choke arm on the carb assembly by hand and not use the choke knob. By the tine the head got hot enough to make holding the choke lever painful, you no longer needed it.

        1. I was noticing the starter generators on the Case 220 I was thinking about purchasing and restoring more-so to have a Hydriv system. Interesting design that seems practical I guess other than the flywheel mechanical effect of the engine as a rotor. Kind of humorous was the transition to mounting a pulley on the flywheel and just pull starting research that I did recently, that turned into where I found an opposed twin Briggs and Stratton stock pulley for pull starting when hunting for the horizontal shaft version.

          That train of thought got me thinking when reading the comment that there is an older electric vehicle plan I invested in back in the early 2000’s that used a surplus jet engine starter for the motor. Not sure about those details or where those plans are now days.

          Wondering how practical gas turbine starters and other starters might be to convert into a motor?

          I have been researching and reading into the Weber State University series on hybrid vehicle CVT’s and controls since I’m planning to take the horizontal shaft 694cc 18HP Briggs and Stratton engine I have and ultimately make into a hybrid vehicle engine/generator that I guess is termed the “assist” style.

          I’m thinking in stages like more likely first the ECU/EMS for the EFI conversion, then making an alternator-generator from one of the motors (lately thinking on a garden tractor platform testbed as an engine welder plasma cutter compressor and hybrid garden tractor also… so neat to read the alternator to motor conv.), then implementing in the I think Dodge Ram with the stock CVT and drivetrain, then eventually ideally bypassing most of the drivetrain and adding 4×4 hub motors with regenerative braking.

          A little ambitious, though great article to get me (and others) thinking about using alternators as motors and also reminds me of the hybrid CVT’s since most look like they have two motors with some beefy and high power stators and impressive rotors that work as alternators that are 3-phase also and are way powerful.

          1. @jack “The vehicle you’re thinking of was in Mother Earth News in the late 70’s I believe.”

            Seems that’s where I ordered the plans from or was turned onto by, good call. That and a book on biodiesel and running diesel engines on a wider range of fuels and plans for a wood gasifier. I recall being corrupted (or inspired to investigate) after seeing a VW Diesel Hybrid that I’m thinking was based on the Lupo back in 2001, though I’m not finding anything on the net now other than the VW XL 1: https://en.wikipedia.org/wiki/Volkswagen_1-litre_car.

            I recall my Dad, having worked on the KC-135’s and other jet engines, not being sure the long term reliability of using as a motor and he thought wasn’t the best… though can work.

            I’m excited about the Prius mainly… and I guess other hybrid CVT’s, starting to show up in salvage yards. I mean, that’s crazy pricing for a CVT if you can get on the 40-50% off holiday sales. Even if having to pay more for the inverter/converters. Especially since most have two motors and are higher energy also.

            Here is a gentleman (Damien Maguire) hacking them out too with some open source code work done as a bonus: https://www.youtube.com/watch?v=qB2tn1myorc

    2. Did you not see the 1963 patent illustration, showing the exact same circuit? (well, okay, it’s wye-wired instead of delta.) The “shitload of coils” you refer to were the voltage-sensitive relays used to drive the rotor to regulate the output voltage. The electrical noise you refer to comes from those relays switching on and off to supply different amounts of current to the rotor. But that has nothing to do with the alternator itself. NONE of that is needed for using that as a motor, as Jenny describes. The only part you need is the alternator itself. Three field coils, one rotor coil, and the slip rings connecting to that rotor coil. No diodes, no regulators, no “shitload”.

      1. AFAIK normally it was just ONE voltage sensitive relay with a normally closed contact, bridging out a series resistor to the field coil and thus increasing the excitation current. A crude PWM step down circuit relying on the inductivity of the field with about 200Hz of switching frequency.
        Normally there was a second relay, an overcurrent protection.

    3. So for an old type with external regulator just ignore the section about removing the regulator and connect to the brushes directly.
      The 3 phase stator is exactly the same with 6 diodes as a rectifier. Yes I once replaced a broken alternator of a car with an older used one with external regulator, so I know it.

      1. I tried this hack myself a while ago using an ESC intended for a drone. The low speed torque was disappointing even with high field coil currents but turn the field coil current down and the speed setting up and OMG you can get these things to run at fantastic speeds! I suspect the torque holds through the higher speeds as well and this where the high power is; at high revs.
        If anyone knows otherwise please share because I’m thinking this hack produces a high speed high power motor and not one with appreciable low speed torque.

          1. @Jenny
            I was thinking to roll one myself because I suspected what you have just confirmed, the drone ESC are not designed for low speed torque (seems fair given the application :-) I could probably hobble a crude (low speed) controller together pretty quickly to confirm the point. I may believe this can be done fairly quickly only because I have not done it before. I reserve the right to change my opinion. ;-)
            Thanks for tip Jenny.

        1. This matches what I found when testing auto alternators – very low torque, but very high RPM. They are essentially useless below 2000 RPM, but have no trouble reaching 10K-15K RPM.

          Efficiency also stinks. It’s around 60% at low speeds, and never gets better than 80% at the high end. Nobody knows how to build inefficient motors like the auto companies!

          Note that you need to drive it with very high frequency AC to get high RPM. The field laminations are cheap; one reason for the low efficiency.

          Larger truck and bus alternators are built much better, and so are more worthwhile to “hack” into useful motors.

          1. @Lee
            I can only agree about them being useless below 2000rpm unless the application was high speed with little torque required to get started (like a drone). I didn’t measure the speed but would not be surprised to find it was far higher than 15000rpm. About half way up to top speed it sounds like a jet engine at full throttle… and still the speed increases… scary… very scary.
            I need to think about the laminations of the field coils a bit more. I’m not sure there is significant AC magnetic flux happening there as the field coil current is fairly constant. And as there is an air gap in the magnetic circuit it is likely to be the bulk of the magnetic reluctance and therefore the bulk of the magnetic field energy will be there in the air gap… but more thinking.. rotational things… more complicated than transformers… hmmmm….

  2. An interesting additional neat aspect of this is that you can adjust the balance between Kv and Kt by adjusting the field coil current (depending on your alternator, generally in the 500mA – 4A range).

    The higher the field coil current the slower the motor will spin at any given voltage, but with greater torque. The regulator that comes with the alternator takes advantage of this to efficiently keep the output voltage in the 13.5-14.0v range regardless of engine RPM and electrical load.

    As electrical load increases (say you turn the headlights on) the voltage starts to drop and the regulator increases the field coil current until the voltage is back in spec. Similarly, if the load drops or the engine revs up and the voltage starts rising the regulator will start backing off the field coil current until it’s in range.

    1. This makes it what’s called a “Separately Excited Motor” and it gives you the possibility of using the field coil current control to extend the useful speed control and torque control range across 2x or 3x the range along the speed range while producing useful torque.

      Even just adjusting the field current to get your motor dialed in for the range you want for your application and leaving it is as simple as turning the trim pot on your current-controlled power supply which is a kind of flexibility you’re not going to find with anything other than a separately excited motor (like a converted alternator).

      Also, you can get automotive alternators rated for 200A @ 13.5v (2700W) and at a dity cycle that would make most electric motors blush (or catch fire).

      1. This. This is very neat. It essentially gives you a stepless electronically controllable “transmission”. Need a “lower gear”? – just increase the field strenght. Need a “higher gear”? – well, just decrease said field. And, like you said. Even if you use a fixed field current this kind of motor is tunable to your needs. There has to be lots of applications where this is useful.

    2. For the alternator to work as a motor it has to be driven by a 3 phase source, and in the case of this article that source is a cheap Chinese controller. And this means the motor speed will be locked into the speed set by the controller.

      Changing the field coil current won’t change the motor’s speed. Changing the field coil current will just change the available torque.

      1. Tom Hargrave: Yes, the speed is locked to the excitation frequency in a synchronous motor, and brushless DC motors (which is what Jenny is making here) are indeed synchronous motors. BUT, in any electric motor, the torque drops with speed due to the back EMF generated within the motor, and there is a maximum speed that the motor will run, which is that speed at which the back EMF equals the applied voltage. If you drive the motor faster than this speed by applying external torque to it, then the back EMF exceeds the applied voltage, and the motor becomes a generator.

        If you look at the specifications for any BLDC motor, you will find a “KV” rating, known as the motor constant. This is the number of RPM it can achieve per applied volt, assuming you’re driving it at the appropriate frequency.

        1. @BBJim,
          Well said, Succinct and clear, just one caveat though; AFAIK the drive on most BLDC motors is at a voltage far higher than the motor can actually handle without some PWM style current limiting. Just looking at the windings in a typical drone type BLDC there just isn’t enough turns in there for the back emf generated to counter the applied voltage and the wire guage is large enough for that to mean some extreme currents but the wires are not large enough to carry that extreme current. (At least that is how it looks to my uncalibrated eyeball.) Hence the PWM to limit the current and it also means a sensorless controller can sample the back emf as part of the magic that is sensorless control of a BLDC.
          @Tom,
          I have done this albeit with a drone ESC and I can confirm that turning down the field col current with an ESC driving the motor, the speed does very definitely pick up. It will of course have an impact on the torque available but also the speed. My motor was running unloaded at the time. It was just an ESC for a drone after all ;-)

  3. I’ve been wanting to do this for some time; thanks for the article! Looking forward to the follow-up with details on how to use and/or hack the Chinese controllers (controlling with a microcontroller, details about the sensorless mode, etc.)

  4. Sometimes these are converted to wind generators by dismantling the armature and replacing the coils with ring shaped neodymium magnets. That way you don’t have to energize the rotor at all. The same would apply to using it as a motor.

    1. I was thinking about a relative of this: using this as a wind powered generator and using line AC (stepped down) to drive the rotor winding. Then it should generate voltage in phase with the line, shouldn’t it? (Assuming not using the rectifier circuitry, just getting AC out.)

        1. The turbine needs to run at some optimal tip speed ratio, so the speed of the generator should change with the speed of the wind. Running an alternator as a synchronous generator basically fixes the rotor speed, so when the wind is blowing less it’s working as a big fan, and when the wind is blowing more it works as a terribly inefficient wind turbine.

          1. Having actual experience and not just an opinion: I’ve worked on a wind turbine that was in fact “just” a 2-phase motor, with very shallow blades.

            When the microcontroller detected suitable wind, it would engage the contactors to motor the blades up to speed and then disconnect the contactors to see if the blades started turning faster, and if they did, re-engage.

            The turbine was incapable of starting itself due to the shallow blade angle – in high winds, the blades stall and a relatively small brake is sufficient. But the shallow blade angle also made it more efficient at its operating speed.

            Educate yourself, then speak.

        2. AFAIK it was done in some wind turbines. But you can not drive the armature with line frequency, you have to drive it with the difference of your rotational speed and the desired line frequency. Like an asynchronous generator where you apply a defined “slip frequency”.
          But you can not use the alternator for this application. It has only a single phase field/rotor winding. You need a 3 phase rotor – similar to an asynchronous motor with slip ring rotor (not “squirrel cage” construction)

          1. We are talking about a car alternator here right? And they have three phase outputs, right? And just one field coil to make the generator magic happen.
            The only problem I see at this very distant view is the line voltage compared to generator voltage is going to need some adapting via a transformer (in principle).

      1. Subject to a suitable gear box to match the generator speed to the ideal turbine speed This is how the very large scale generators work as I understand them. But. There is a caveat required to accommodate varying wind speeds, which is that the arms of the propeller can be rotated on their long axis and so present a more or less aggressive angle to suit the wind speed. Put another way (because I am not confident that explanation is of any use to anyone) the angle of the blades should be adjustable so as the propeller turns the blades will cut a thicker or thinner slice through the air. (Still not confident ;-)
        In the hope that the above made sense then the only other issue is the gear box. The propeller will have a ‘sweet’ range of speeds and so will the generator. The gear box is the matching component of the two speeds.
        If there is doubt about the very large commercial units operating as I described I can only add that I got the first clue by watching a wind farm and noticed that irrespective of the location of the generator every one was turning at exactly the same speed. While there may not be a great deal of difference anyway given they are all in the same area that still doesn’t explain why the speeds were so very very identical.
        I don’t know what the algorithm for controlling the blade angle would be (asking the internet would be a good start and probably a good finish as well) and there is still the question of the field coil current which would also have to be controlled and then it might also be a good idea to have some form of current or power delivery limiting in case of strong winds.
        Almost forgot, you will also need a rotating mount so you can keep the blades perpendicular to the wind direction and an isolation transformer would also be wise even if not functionally necessary.
        This would be a great project but not a simple one. I look forward to seeing posts for it if you go ahead with it.

        1. Thinking about this a little more and based on the ideas that the blade angle is likely to be adjusted to suit the wind speed and the field coil current adjusted to control power delivery (the field coil current would normally be used to regulate voltage as the voltage developed in the generator windings is a function of the rate of change of the magnetic field and therefore a function of the field strength and rotational speed of the generator but for this application where the output voltage is set by the grid you would need to use the field strength to control output current which will add the winding resistance as the factor that allows you to use the field current to regulate the output current of the generator).
          So my thinking is that an adaptive algorithm that assesses the output power of your generator, makes small adjustments to these two variables (blade angle and field current) and re-assess power and adjust accordingly. This is very similar to a MPPT for solar applications if you are familiar with those. Not exactly sure how you would optimise both the angle and current at the same time but someone on here may have some light to cast on this topic?

        2. The wind turbines are all turning at exactly the same speed because they are AC synchronous generators, so they don’t have much choice. Instead of controlling the speed, the pitches of the blades control how much power a given turbine contributes, by either leading or lagging the phase of the grid. This allows for turbines exposed to greater wind to generate more power without changing speed.

          This applies only to synchronous wind turbines – there are also variable-speed turbines that use any of a number of types of generators.

          1. Wind turbines are generally not AC synchronous generators, otherwise it would be impossible to get any energy out at low wind speeds.

            They use an inverter to drive the AC output.

            They all turn at the same speed because, for a given wind speed, there is an optimum rotation rate (tip speed) to extract maximum energy from the air, so their control systems run them at that speed. It’s analogous to MPPT control.

          2. Whether it is generally true or not, some wind turbines actually ARE synchronous generators. I was answering the specific question of why the turbines on a given wind farm were turning EXACTLY the same speed, and this is the correct answer. The blade pitch is used to vary the torque, and therefore the power delivered to the grid, so your claim of “impossible” is wrong. And in fact, many wind farms go off line when the wind speed is too low for them to generate significant power.

            The original question was why they would all be running at exactly the same speed, since the wind speed would vary over the area of the farm. Your answer does not answer the question.

          3. @BB Jim
            I think this is kinda what I said ;-) but nice to get confirmation thank you.
            @W
            We might be debating semantics without knowing it but BB Jim and I are not that far from the truth in saying that wind turbines can be AC synchronous generators as is any polyphase generator. The designation of type generally seems to have more to do with how they are connected and controlled.
            The wind farms I have observed very definitely had turbines in sync. The explanation that they use the same power extraction maximising technique could not provide the speed sync I observed. Several turbines over a wide area and after several minutes, no appreciable shift of phase angle between them is no fluke.
            I think the confusion here might be that you assume a synchronous generator would mean a fixed rotational speed that would preclude energy harvesting from a significant proportion of the wind energy on offer from mother nature because the wind is too slow.
            If I am right about that then there are a few factors you may be overlooking. First, the assumption that the slower wind has an energy worth harvesting. My understanding is the power is a function of the square of the wind speed so low speed may be very low power. Second is that the generator can be wound to provide a low rotational speed for a given line frequency. Third is that the angle of the blades determines the wind speed to rotational speed ratio (all other things being equal) and that ratio can be significant. Ask anybody experienced in sailing (wind propelled, not motor boating ;).

          4. the VAWTs I’ve built with car alternators just turn as fast as a VAWT can, if you’ve messed with VAWTs you’ll understand

            I just varied the field coil current

          5. Just to be clear, again: I was not saying that all wind turbines use synchronous AC generators. I was only speaking to the specific case where a whole field of turbines are precisely synchronized. Stephen Walsh described a case where over a wide area, a number of turbines were turning at exactly the same speed, and this explanation applies to that case. Yes, turbines that DON’T have variable-pitch blades (such as VAWTs) have to power generators that can operate over a range of rotational speed, but that is not the case with airscrew type turbines. And also to be as painfully clear as possible, I am NOT saying that all airscrew wind turbines use synchronous AC generators.

          6. @BBJim,
            LOL
            So what you are saying is they can be synchronous generators but they don’t have to be synchronous generators ;-)
            LOL
            Your post almost sounds like a Monty Python quote “and just to be absolutely clear! Nobody is stoning anybody until I blow this whistle”

          7. @Cyberteque, were the VAWT’s you have experience with configured as synchronous generators pumping directly into the grid? I ask because that is what is being debated and a VAWT of this type cannot simply run as fast as it likes without a mechanical gearbox or electrical dissociation such as an inverter to pump into the grid and thereby removing the sync requirement between generator and grid.

    2. Doing that would completely undo the ability to regulate the output. You would need to disconnect the generator below a fixed turbine speed and hope the load was heavy enough to keep the output voltage (and turbine speed) within load voltage limits in strong winds. Why not just leave the original regulator in place and use the alternator as she was designed? Replacing the field coils does sound like a lot of bother and a loss of a useful function.

        1. @Martin,
          I wonder where the balance point is here. One the one hand there is a field coil current and the other a power converter (I assume) to match generator volts to load volts. There is also a potentially much larger issue with power factor. A diode bridge is simple yes but offers no PFC function so 10% or more of the available power is reactive power that contributes nothing to the load but does add to the heat by increasing the current in the generator windings. And should I raise the spectre of the dreaded 3rd harmonic? :-)

  5. I highly encourage you to read the rules for the Hacky Racers series, it’s highly entertaining. Some highlights:
    – Do not half-ass this. Use full ass.
    – This penalty will most likely be made up on the spot, will cost you considerable ire and outrage and lifting this penalty will require considerable bribery
    – Vehicular weaponry is banned.
    – Impacting the barriers is not an acceptable method of braking for the brake test

  6. If you want something clean, a new Chinese made alternator can be picked up for as little as 20 dollars. The front bearing on alternators are massive, very attractive. While the internal electromagnet does reduce efficiency and add slip rings to Thr wear and tear list. They do handle heat extremely well, while rare earth magnets typically die at about 80 degrees celcius. At mak power an alternator can easily provide 1kw. So for 24 w of rotor excitation, that’s less than 3%. Who cares about 3%?. One final thought. The outputs from the coils are not very sinusoidal. Its more like triangular, which probably aids in generating power at a wide range of rpm. Does this complicate the optimal ESC drive pwm shape, probably, something to look into if one wants to aim for the optimum.

        1. I have no personal experience of samarium cobalt magnets (only what I have read) but I would speculate that if the motor was running at (or above) 250C then the enamel in the windings would start to smoke.

          Also I do have experience of buying neodymium super magnets and I can attest to the fact that there is a lot of rubbish being sold as A1 product.

          1. @Mr. Scott

            “Rubbish in what way, what would distinguish good from rubbish?”…

            (1) field strength varies greatly from what is advertised and is much less than that of a known A1 magnet (bought 4 in one batch from same supplier and measured strength varies by as much as 75% from strongest to weakest – and strongest is only 50% of that of A1 magnet)

            (2) plating does not adhere well, easily comes off.

            (3) plating has microscopic holes (not visible to naked eye) which soon forms blister as corrosion develops underneath.

  7. Always had the curiosity about if could be possible to use the alternator as a electric motor inside the car like the microhybrids cars do nowadays. It would be amazing if we could install a kit in our old petrol car for convert it to a microhybrid car making it to work a as a generator when the car was not consuming energy absorbing the kinetic energy and switch to electric motor when the accelerator pedal was pressed.

    this could make to save some extra fuel in our old cars with a maybe easy and cheap solution (or maybe don’t worth it).

      1. This was supposed to be the next big thing, changing all automotive electrical systems to 48V. It was supposed to cut way down on the amount of copper needed in the wiring harnesses.

        1. I did read something this year about a retrofit hybrid system for vintage Porsche 911’s that would be mounted in place of the flywheel. Something like that for my 80’s BMW is on my wish list.

          1. Probably not that much. 12V are already enough to cause electro-chemical damage, if present on the wrong place.
            Although a 24V “live” PCB immersed in ordinary tap water looks pretty bad after only a few minutes: exposed tin plating mostly gone copper starting to vanish. That was a test for waterproofness, were we used the wrong test sample – with a hole for a connector :-(

    1. It seems like GM did something close to this on their early “mild hybrids”. It struck me as more of a “compliance car” than a real thing, but technically it was a hybrid. I suppose technically it could drive in the carpool lanes.

  8. Several horsepower?

    This alternator is rated at 90 Amps. At 13.8 volts that would be 333.1 duckpower.

    Going from star to delta configuration would increase that to 566.3 duckpower.

    Definitely not several horsepower.

    If you tried to get 5 horsepower out of one of these then you would end up with about 7 CFM of smokepower.

    Sorry about the non-SI units I have plenty of ducks but no horse.

        1. @Murray,
          That seems very low, 960RPM @ 12V? Are you pumping huge currents through the field coils? I have an old Yamaha alternator (I think) unit and with a couple hundred milliamps in the field coils this thing at 12V would challenge a coreless motor for speed. ;-)

      1. 1 horsepower = 300 duck power
        several = more than 2, less than many

        You’d have to get to 601 ducks to equal several horses, even using an incomplete horse to get to several.

          1. Just convert that to lift power and see if it makes sense for 42 to be the answer to everything, or not.

            420 would be more realistic, maybe it is a smaller duck. But ~300 is based on a Mallard, already one of the largest ducks, so getting to 42 seems doubtful.

      1. That’s true for a generator.

        In a star alternator the three outputs are one duck sin(0), one duck sin(120) and one duck sin(240).

        You can choose to drive a motor differently.

        In delta you have points A B C.

        If you drive one duck between A and B, one duck between B and C then you have minus two ducks between C and A which will contribute to the rotation force by using repulsion rather than attraction.

        So while your max current can’t change as its a limitation of the copper, you can drive at 1+sin(60)

        1. What you say is true FOR A GIVEN VOLTAGE. What you gain by having more current available, you lose by having fewer turns in series, i.e., less voltage. Ask yourself what would be the case if each coil was driven separately, completely disconnected from the other coils, from separate batteries. This is essentially like the parallel case, where the center point of the wye is grounded. There is no magic by which you get more power based on how you connect the coils. Sorry, but you’re just wrong.

          1. You’re talking about how the motor is connected and I’m talking about how it is driven.

            Say you have a simple 4 phase stepper. Phases A, B, C, D that you can also drive in reverse -A, -B, -C, -D

            You can sequence A B C D A etc

            Or you can sequence AB BC CD DA

            Or you can sequence AB-C-D, BC-D-A, CD-A-B, DA-B-C

            If you have star connections the drive circuit is more complex.

            If you have three phase delta then the third phase is automatically drive reverse by the other phase voltages.

            Have a look at how a three wire three phase delta hard drive spindle motor is driven with a scope.

        2. I have no idea where you are getting these positive and negative ducks from or what you are trying to say but I can tell you that, all things being equal and 100% efficient, power in = power out, changing from delta to wye or vice versa may change the output voltage but not the output power without changing the input (mechanical) power.
          In delta and in wye you have A, B and C and in a generator, the resistance to turning is always a function of magnetic repulsion and in a motor the force turning the motor is attraction (or, more correctly, in a motor the forces act to maximise the inductance of the magnetic circuit and the opposite in a generator, the forces act to minimise the inductance). There is no magic about “C and A” that makes them any different to A and B or B and C and definitely nothing that means one phase current will act differently to any other phase current.
          If I have missed your point, please enlighten me. In your last line about “1+sin(60)” I assume you mean the angle of the currents applied to the motor? If so then this would represent enormous ‘slip’ in the motor I think. Have I got this right? Is this what you mean?

          1. @Rob,
            I should have addressed this to you sorry, my bad. I am still curious to know what your post about the ducks between A & B and B & C adding up to negative two ducks between C & A and the mysterious comment about 1+sin(60) actually mean.

          2. [Stephen Walsh]
            He probably doesn’t know what the duck he is talking about…
            It is obvious in his mentioning ducks in wye or delta configuration.
            EVERYBODY (else) knows that people need to get their ducks in a ROW!
            B^)

          3. Ducks in a row is on the ground, to move them to market. Or home, tell the ducks I said move them home. M A R K E T, “home.”

            For duckpower you have to harness them while flying.

        3. @Rob
          Against my better judgement and since you won’t answer my questions I have read through this one more time to try to make sense of what you are saying. I have a pretty good idea of how all these motors and generators work but half of what you say just doesn’t make sense and finally I have worked out why; because it just doesn’t make sense.
          First up, let’s dispense with the stepper motor discussion as that is just a distraction: steppers are a reluctance motor, which is not the same thing at all. The motor we are discussing is essentially an AC synchronous motor in one sense and otherwise and more commonly considered to be an electronically commutated DC motor. DC motors use the fact that as the motor turns a voltage is induced in the windings that acts to oppose the current in the windings. This is how the speed of an ordinary brushed DC motor is determined. The opposing induced voltage counteracts the applied voltage to the point where the difference leaves just enough current flowing in the winding resistance to maintain the speed of the motor against the mechanical losses that would otherwise slow it down. Apart from the differences in commutation, this is exactly the motor we are talking about and a basic principle upon which it operates.
          Your statements about application of ducks to phases and how the third phase somehow magically gets two negative ducks with forces flipping from attraction to repulsion is just plain wrong but in part only because you have mistaken an effect in the windings as somehow causal rather than caused. The effect is this: if you drive two of the three phases of a motor of this type with steady state conditions established before the third phase is removed then the motor will act as a rotating transformer and reconstruct the waveform of the missing phase by virtue of the induced voltage already mentioned. This is I think is what your minus two ducks is about. This is also what can make it difficult to detect a missing phase on an induction motor and why missing phase is usually detected by current and not voltage. I think you may have come across this somewhere and misunderstood what was going on and assumed that the voltage waveform meant there was a current flowing which is also why your explanation talks about delta rather than wye because you think that is how the third phase currents will flow. They obviously could not flow if it were wye. You may not believe this, but in either case the fact is that the current does not flow as you describe because for the most part it does not flow at all in the third phase.
          The reconstructed phase is not in itself useful and otherwise has little meaning. It cannot be exploited for driving the motor AFAIK but the missing phase could cause the motor to overheat (common problem with induction motors) if the load on the motor is unchanged then the power required from the two remaining phases must increase to compensate for the missing phase.
          As for “driving differently” there really isn’t that much room to maneuver in this respect. The motor will dictate the driving requirements and doing anything else will only reduce the efficiency. The driving forces generated by any two phases (all things being equal) will be exactly the same as the forces generated by any other two phases. There is no magic flipping of forces or any such on the third or any other phase either. As I said, all phases work the same way drawing the same currents, generating the same forces and the same magnetic fields just with a 120 degree shift one to the next.
          Your final statement about 1 + sin(60) still makes no sense to me.
          @BrightBlueJim have I covered it ok? Your thoughts?
          @Lee & @Rob power, torque and speed are all closely related. The confusion is comparing incomparable things and fundamental misunderstanding of both.
          @Ren obviously, I agree ;-) on all points. Just trying to do my part to stem the flow of misinformation, and to investigate this new duck theory I keep hearing about. I should talk to a quack about it maybe?

          1. Stephen Walsh, you went into WAY more detail than I was willing to spend the time on, but I don’t know how effective it was. I once tried to explain to a refrigeration “expert” on why you can’t just exhaust the condenser cooling air into the room you’re trying to chill, but it was just wasted time, because he had thirty years’ experience in commercial refrigeration.

          2. @BBJim,
            As it turned out it was way more time than I wanted to spend too. ;) I do appreciate the moral support though, so thank you. I hope I didn’t get anything wrong or badly unclear in all that. I skimped the editing somewhat.

            So much of my professional career has been spent dealing with the myth and legend that passes as facts that I just can’t let misinformation happen unopposed. But given the silence from this guy now I wonder if he is a prank poster just trying to wind us up? I guess… quack!.. we may never…. quack!… know. Quack quack!

            Long live duck power! :-)

        1. Lol,

          That was the point that didn’t really stand out in my comments.

          I thought horsepower was last used by the ancient Romans so you can imagine how confusing horsepower is for me.

          So a quick google showed that one horsepower was roughly equivalent to 300 duckpower.

          I use SI watts which any engineer would be comfortable with.

          But this is a US site and your units are all over the place as if SI units don’t exist.

          There’s an old saying … If it can be done 14 different ways then there will be 15 standards.

      1. But how much torque can you get out of them from stationary?
        5HP x2 is more than enough to couple to teh rear wheels of a FWD and make it a soft hybrid, think moving in traffic but I’m assuming that’s at a high RPM.
        So from 0 RPM can we get the car moving ?

    1. @Rob,
      Ok now I understand you and where the confusion lies.
      You are correct in that going from star to delta without changing the applied voltages would generate sqrt(3) times more power and also require sqrt(3) times as much current to make that happen. So discounting the forward bias voltage power losses in the now bypassed rectifier from when this thing was an alternator we would have about 90A x 12.5V (?) in star configuration and 90A x sqrt(3) x 12.5V in delta configuration which is 1,950W which is 2.63 horse power and none too shabby I say. :-) Definitely in the ‘several horsepower’ range, or at least in the ‘number of horsepower’ range but not, I admit, in the ‘many horsepower’ range.
      I’d state this in ducks for you but I prefer to use, if not SI units, then at least established units that have the advantage of not being just silly. What’s the point of units that no one understands? Horsepower as a unit has been around a long time and it is not that bad. Give it a go. You might like it. Most of the planet understand it.

      1. I think that the principal reason for expressing something in ill-defined or undefined units, is to leave oneself open to backpedaling, saying, “no no no, I didn’t mean many horsepower – I clearly said ‘ducks’.” But by later using ducks as a measure of either current or voltage (I’m not sure which it was supposed to be), he undermines this. It’s like expressing speed in parsecs, a la Han Solo.

  9. The next step up would be a variant of the so-called “belt alternator starter” system from a variety of manufacturers. GM had a very nice ~20kW water cooled unit (made by Continental, IIRC), which was basically an induction motor with some beefy bearings on the ends to take high belt loading so it could actually transfer 20kW of energy back and forth. You can find them online in the $100 range. There are different units from different manufacturers, and they can handle considerably more power than your average alternator for not significantly more money on the used market. I wish people were utilizing these more before they get sent to the crusher.

      1. There were at least two generations of this system. The unit I’m referring to is from the 2nd. It was a “mild” hybrid and was used in a variety of GM cars and trucks, but I don’t remember off the top of my head which ones. Search for “GM BAS” or something similar and you should be able to find it.

    1. They were used as a “hybrid” system on some trucks, with a small battery pack. The main use was accelerating from a stop to give a bit of extra boost to save fuel. Only did any real good if you drove the truck in stop and go traffic a lot. In long haul highway use the battery was just extra weight to carry.

      A good hack would be to adapt a car’s stock alternator to be able to switch back and forth from alternator to motor.

        1. You must have a really big swimming pool when you think about a 20kW pump. At my parents pool (40m³) a 1kW pump is really oversized. It forces the water that fast through the solar absorbers, that the temperature differential is tiny (<1K) and I had to compensate the controller for sensor tolerances to make it work properly.
          The reason for that oversized pump was confusion between input power and output power of the motor – already 2 times when the pump had to be replaced. a 500W pump would be sufficient.

        1. ROFLMAO Yes we will. And our Prime Minister will be on holiday as will any minister with a climate related portfolio or a constituency particularly badly affected (burnt to the ground, flooded Noah style, ripped from the surface of the earth by a hurricane or just generally demolished by storms).
          How good is Australia!!!??? How bad is our government? :-(
          Apologies for off topic. Our pollies inspire me to ranting. Hope you understand.

    1. Yes, “could,” and breaking would indeed be the most likely outcome, or rather, burning. Braking could also be achieved, but you’d need to add some control circuitry.

      You could probably make use of a starter solenoid as a relay for it.

    2. That’s actually a point for having a rotor winding – you can drive it at a low level to get speed when driving the vehicle, and then raise it to go into braking mode. The problem with regenerative braking with permanent magnet motors is that the voltage they put out usually has to be boosted in order to charge the batteries. ‘Course, you may lose as much power in that rotor winding as you do in a boost converter.

      1. I think the usual reason that the “overcomplicated” guides tell you to re-wire (or rather, swap some connections around), is to reconfigure your alternator to wye configuration so that you can get reasonable power out of it at lower voltages. If you leave it as delta, you’re going to need a ~48V supply.

      1. Thanks Murray. Do you know how many volts it can take without smoking? I would love to use one if I could get up to 3000rpm, so the calculator says I need 37.5V. But I guess these were designed for 12V.
        Anybody?

        1. The motor constant (the number of volts per RPM) for a commutated motor (whether using a mechanical commutator or an electronic one) depends on several factors – the number of stator poles, number of rotor poles, number of wire turns on the poles (both rotor and stator), and – most relevant to this case – the rotor current. If you blast the rotor with whatever it will draw at 12 V, you get a very low “KV” rating, but reducing the current increases the KV number. I suspect that the stated low KV number was with the rotor connected to unrestricted 12 V.

          One of the nice things about a wound rotor (as opposed to a permanent magnet) motor, is that you can vary the field current. At higher currents you get higher torque but proportionally lower speed, and of course vice-versa. But if you want to know what speed a generator/motor will be sure to operate at, look at what it was designed for. In rough numbers, most automotive crankshaft pulleys are in the range of 8″ diameter, while the alternator pulleys are around 3-4″. Sorry for the King’s units, but the ratios are all that matters = this indicates that automotive alternators run at about 2-3 times the crankshaft speed, so at cruising speed, the alternator is running somewhere around 6000 RPM, and this would be where it is designed to work well. So I would expect it to work well as a motor in this speed range as well. But NOT with 12 V applied to the rotor – it will be much slower then.

    1. Like the other replies say: it depends, and you control the torque/speed curve via the rotor current.

      However, consider their usual use case: attached to a car engine that’s probably mostly running around 1000-4000 RPM via a belt that gives about 4x speed multiplier on the alternator. So the alternator is designed to run around 4000-16,000 RPM most of the time, probably to provide peak power output somewhere around 12,000 RPM, and will survive 25,000-30,000 RPM depending on the engine.

      You can expect efficiency to drop off at the higher speeds due to both windage and eddy current losses.

      If you were designing a system around one of these, you would want to characterise the motor in detail, but that should give a starting point. The RPMs are going to be annoyingly high for a traction engine, i.e. you’re going to need a monster gear ratio to drive a vehicle from one of these efficiently. Yes you can crank up the rotor field current to get more torque at lower speed, but you’ll be limited by heat in the rotor and losses from driving all that power into the rotor.

      As usual – motors are way more complicated than people first think, and there’s a crapload of work in optimising one for your application.

  10. There was some speculation in F1 2011-2013 that the Red Bull team along with their engine supplier Renault was using their alternator as an extra Hybrid drive. I believe they cooked about a half dozen alternators over that time period, and it raised lots of questions, but nothing was ever proven.

  11. Want some grunt, Use a starter motor. Mind you finding a speed controller for one is not an easy task. But they do put out a bit of torque. not a good idea to dangle your bits in one to see if the gear is trowing out either.

    1. The starter motor is not constructed for efficiency, as it is only for short term operation. It has to be lightweight and cheap. And yes, controlling a series wind motor is also not that easy.

  12. I’ve been tearing alternators apart since the 70’s. Never yet seen one wired Delta. Nipon denso, delco, fomo, dodge, motorola, etc. . Most tap the star point and halfwave (three diodes) to run the regulator/ field circuit. Just as in that diagram. Challenge you to find one circuit diagram, from any auto maker, that uses delta. I could be wrong, but I’m open to being shown so…. Industrial motors are often wired delta, but there are problems in that there is no neutral point, and they like to find easy paths to ground especially above 600 volts. You also generate more voltage with star, for a given amount of copper, because they are in series. sort-of (X 1.73 etc).

    1. Notice the Ford one in the picture, each set of field coil wires has two wires. Separate then and you can check, they’re wired as delta. Could it be a newer versus older thing? A Euro vs American thing? All the ones we’ve done have been delta.

      The Bosch patent is star, but it’s from 1963.

    2. It does not make a difference, if delta or wye configuration. This is just a question of the nominal voltage of the single coils. Also the field rectifier is not really a halfwave configuration: It makes use of the “lower” diodes of the main rectifier. The extra excitation diodes are necessary to avoid battery drain, when the engine is not running. Theoretically you could do this with an extra diode in series with the line to the battery – but this would be a >100A diode and you have two times the forward voltage in the main circuit. Nobody wants this, so you use three small extra diodes to separate the circuits.

Leave a Reply

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