A much more efficient alternator
 

Alternators are basically unchanged over the last 60 years. Still the same crappy diode rectification with 10% volt drop across them. Still the same crappy method of rotor current regulation. They are 50% efficient and nobody seems to care. Then we throw it into a lead acid battery which is in itself only 50% efficient. Oh wow, 25% efficiency. It's 2020 for gods sake.

Replace the diodes to Schottky and halve the rectification losses. There's active rectification, but that's actually less efficient at high current.

How about eliminate the rectification entirely? Why have AC produced in the first place? Rotor current is DC. Instead of alternating the current, just alternate the poles in the rotor. Use AC in the rotor. It doesn't even have to be sine wave, square wave is probably better. I'm sure a hall sensor and a few simple components could synchronise the rotor poles with the stator, thus producing DC. Basically BLDC in reverse.

I2R losses. Buck converters are a thing. Stop making low voltage. Make high voltage with low current, and convert it to low voltage after its left the alternator. With no rectification losses, and much lower I2R losses inside the alternator, the stator can handle much more current. And, a cooling fan is no longer required, further improving efficiency. I've played around with circuit simulators and a 50A alternator is capable of putting out 3kw with these modifications. But why not use a 24v one? Half the current again, much lower I2R losses.

And forget star or delta. Put the 3 windings in SERIES. Reverse one of them, and now you have a sine wave of higher magnitude. Now you can reduce current even more.

In summary:
1. Use a 24v alternator instead of 12v
2. Remove the rectifier
3. Change the windings to series
4. Convert the rotor to AC
5. Remove the fan
6. Run the high voltage through a buck converter
7. Use a lithium battery

It should now be possible to run the alternator ONLY when braking, and have sufficient energy recovery (like 400A) to keep the battery around 80%.

One problem with your suggestion is that you cannot induce DC indefinitely.

Voltage is induced by the change of magnetic field, voltage will be zero even with a constant magnetic field present.

And voltage will be negative while the field is removed, even without the field ever being reversed.

Hi Crashy.
Electric and hybrid cars must use something like this now for regeneration. I assume they use active rectification (?) I think we will see this soon in conventional cars too. Auto makers are always looking for lower production costs and higher performance, including better efficiency. I assume it will look like an alternator with internal electronics. Probably operate at a few hundred volts, with a buck converter down to 12. Or 48 if that "standard" ever gets adopted.
I don't think we'll see lithium batteries as the 12 volt source until their low temperature performance and cost are better than lead.
-mort

24v vs 12v
 

It has been years since I did serious work with electronics, though I graduated from Elkins Institute as a TV-Radio tech. I have always wired my cars from scratch so I know how it is routed. Now that we have electric power steering, I have read companies are looking at electric Air-Conditioning to remove the compressor drag on the engine. I am still debating if I should install a standard Air-Conditioner in my Sunbeam or hold off for total electric AC. My fear is it will be a 24v system...though I can convert if necessary. Just a thought.

Prius has electric air conditioning. Has electric everything actually as there are no fan belts.

Scrap the alt altogether, replace with a similar sized BLDC motor, run regen when you need to or it's most beneficial, run assist for the same, converter to 12v off the battery pack, if you have one.

Hmm...sounds familiar.

ecomodder.com: Controller mods or build for E-assist altermotor

DDG: heat pump in electric car with buck converter.

Have you heard of something called money?

I looked into the price of active rectification. It would cost around 70 dollars in parts to retrofit an alternator with active rectification. An OEM could probably do it for 50 (Bosch has done this already on some alternators).

24V means adding DC-DC converters to power switches and lights and such. You would need to spend around 100 dollars.

They could also use thinner laminations to gain a little bit more.

The fuel economy difference would probably be well under 1% for all three combined, and no one would ever notice, so the manufacturer chooses to pocket the extra 200 dollars.

Who are you talking to?
Did you even read the thread before commenting?
I said NOT to use active rectification, instead use AC rotor current rectification.
I said use a 24v alternator and convert it to 12v. I never suggested changing everything in the vehicle to be 24v.
The fuel saving will not be 1%, we already know alternator delete saves 10%, and an alternator is 50% efficient, so a 100% efficient alternator would save 5%. But you have made the silly assumption that fuel economy is the only incentive. It can pump out 400A when required.

Have you heard of money?
I'm willing to bet that my design would be cheaper to manufacture. If you kept the same 50A output, you could do it with far less copper for a start. No fan, less aluminium, smaller bearings etc. It would be 1/4 the size

I got your post Crashy in my e-mail about alternators. I did not post the argument against your suggestion, and I am not sure what the discussion is about, although I do understand the alternator in cars started in the 1960's with Chrysler. The reason for switching from a generator to an alternator is the ability to charge a battery even at low speeds and idle. I did not read your total post so I am not sure of the argument.

Okay fair enough, an AC rotor works, but it costs more to have commutation with brushes...probably a lot more. Diode losses are half as much at 24V though, so it might not even be worth the expense.

Uh, I did say 24V and implied conversion to 12V. That's what a DC-DC converter is for. They're expensive! It's really not a surprise that 24V hasn't caught on, with alternators being as cheap as they are.

Personally, one day when I have some spare time, I'm going to run a sensored BLDC motor with a controller and high voltage battery to make a mild hybrid. Thin laminations from a high quality motor takes care of the efficiency issues with a permanent magnet, but alternator makers are too cheap to use them.

Commutation is another way I guess, but it could be done electronically using existing slip rings and a hall sensor. There's also the option of only doing half wave rectification at double the current, it should make no difference to the heat, but it will only have 0.3v drop across one Schottky at something like 40v at 1000rpm (0.75% lost). I played around with inductor capacitor combinations in parallel, and was able to get AC converted to DC without any active components, but it would not work once rpm changed.

A buck converter has only 3 components, so should be cheap, even at high current. It does not need fancy filtering, remember that the output from an alternator is very dirty

What happens if you run two alternators in series? Half the efficiency?

Er, wait you can't turn the 3 phase summed power into DC if the rotor is switching? You would have to ditch at least one phase to get AC which you could then convert to DC by switching the polarity of the rotor.

Half wave rectification is an interesting idea; as you say it's not heat limited but winding resistance still limits power. Using half wave rectification on an oversized alternator for a car with low electrical demands could be a quick way to boost efficiency by a few percent. You could trim the cooling fan to reduce parasitic load further.

24V half wave rectification would keep power around the same, but raise efficiency at least at lower speed (iron losses will double). This actually seems DIYable, though a 12V alternator field coil running at 24V isn't great.

There are both 12-volt and 24-volt systems already available, they're mostly targetted to the RV market and long-distance truckers.

The Insight stock engine has a very large permanent magnet 3 phase motor bolted to the output shaft on the engine. For a while I had this connected to a Schottky rectifier and then to a DC-DC converter with the internal rectification bypassed. This was connected to a bank of large supercapacitors and a small lithium battery. It worked well, but there were a few things to consider:

-At idle, the motor produced around 65v, and at redline it was around 400v. Both of these extremes were outside of the range of voltage my DC-DC officially operated at, though I never had any issues with its protection tripping. That said, it wouldn't surprise me if a vehicle that wasn't so electrically frugal at idle might have overloaded it under certain conditions, e.g. defroster, A/C + blower, headlights, maybe heated seats or something running. Not an unrealistic situation.

-No more bump starting. An alternator basically starts producing electricity right away, the DC-DC did not.

-The Mean Well unit I used would likely not have worked in very hot climates. Imagine leaving your car parked in the sun, only to find it wouldn't charge your 12v battery until it cooled down.

-Electrical loads were typically under 20 amps. Reducing electrical loads probably *realistically* goes farther than increasing efficiency of it's production.

-Supercapacitors and lithium batteries can be dangerous under certain circumstances.

Hey! I got one of those...with one of those...and two of those on the shelf...and all the rest...

...Nah.

This is a good point, which is why I never got the motivation to go tinker with the alternator. Once I had switched out all the lighting to LEDs on the car, it just didn't feel like a good use of time chasing down a few watts.

There's also the fact that the usual 1.5-2 ton car really doesn't recuperate very much energy from just an alternator pushing a few hp into the battery. On a Lotus Elise or Honda Insight, a few hp counts for more.