thermoelectric power, no alternator, for $10000

[bzipitidoo]

Removing the alternator would be awesome if there's a decent replacement. Here's the thread about just how much FE is to be gained: http://www.gassavers.org/showthread.php?t=1074

But replace with what? Grid power has a lot of trouble with wearing out even deep cycle batteries and being inconvenient. Solar power won't work in the dark. For wind power, if it didn't cost more in drag than an alternator, a turbine would have to be where? Hung in front of the car? And probably there's a minimum speed under which a turbine wouldn't generate enough power.

One thought was, what about electricity generated from heat, as thermocouples and Peltier devices can do? I found a company experimenting with this idea here:
http://www.hi-z.com/websit07.htm Sounds like it can work great. The biggest immediate problem is cost. Their generator used 72 Peltiers which I thought might be $5 to $10 each, but they're asking $124 each! A bit more searching and I came across a Chinese site that is inconsistent with specs for one set of part numbers and prices for a totally separate set, but seemed to suggest the Peltiers would be no more than $20 each.

[skewbe]

It's an interesting idea and might have some practical applications, but I can see a couple issues at the econobox scale :

The 1kw needed it's own radiator. For comparison, A 40 amp alternator can make half that much power.

There is likely increase in exhaust back pressure.

And specifically, I just ran out on a several mile run with 3 stopNshops, doing as much CODFISHing as I could, the car never even warmed up (got like 55mpg though). And I had to start it many times AND run the lights. I'd probably be on a defecit unless I deliberately burned more gas.

I need a kick starter for my little one litre

[Ryland]

I can't belive it's been almost a year sence I did anything checking on this topic... one of the other threds can be found at: http://www.gassavers.org/showthread....thermoelectric

at the time my figures showed that it would cost about $2,500 to make a generator for a small car, less of course if you spend more money on useing less electricity.

[Ryland]

I have to wonder how much buissness hi-z is doing, if they are doing any at all, because most of there web page hasn't been changed in 5-7 years... I suspect they are just re-selling the same thermoelectric generator you can get else where if/when they ever do sell anything.

[Bruce]

Perhaps a Stirling engine running an alternator from the radiator water/ambient temperature differential would be a bit more cost effective. However, it still has the same shortcomings in stop-and-go driving.

Gasoline ICEs are about 24% efficient at their best, so there's quite a bit of room for efficiency improvement if the waste heat can effectively be utilized.

[Sludgy]

Possibly the best way to reduce generator drag is to use a conventional alternator charging a deep cycle battery, BUT using a relay that energizes the alternator field windings only when the brakes are applied. This setup uses the vehicle's kinetic energy to generate electricity when braking. During accelleration, constant speed or or coasting, the alternator would not draw energy from the engine.

The cost would be relatively low. All that is needed is a deep cycle battery, a bigger (200 amp?) alternator and a relay in the brake light circuit. A voltmeter and bypass switch to make sure that you have enough juice during
extended highway driving.

A crankshaft-mounted (i.e motorcycle) alternator would be even better than a belt driven alternator because it would not have belt drag, but the bigest crankshaft alternator I've found is only 35 amps.

[Silveredwings]

I like this idea. The one AlexK found a while back (http://www.gassavers.org/showthread....7687#post27687) claims 4.8V at 1.2A output (at load) with a temperature differential of 100* C. I'm not sure how the numbers compare between the two part sources but I still think there's promise here.


Does anyone know what a car alternator has for a duty cycle? To pick totally arbitrary numbers, if a 60A alternator is on 50% of the time, then the equivalent is a 30A source on 100% of the time (or 432 Watts at 14.4V). To get this same output with the Tellurex Peltier modules, you'e need 72 (24 parallel sets of 3-in-series), or 14.4V at 28.8A producing about 415W. The cost of the Tellurex units would be $3096 (if they're still $43/ea).

I'd love to find some factory fallouts to play with. Though I don't know where I would find the surface area to heat and cool 72 of them suckers.

[Silveredwings]

Originally Posted by Sludgy

Possibly the best way to reduce generator drag is to use a conventional alternator charging a deep cycle battery, BUT using a relay that energizes the alternator field windings only when the brakes are applied. This setup uses the vehicle's kinetic energy to generate electricity when braking. During accelleration, constant speed or or coasting, the alternator would not draw energy from the engine.

The cost would be relatively low. All that is needed is a deep cycle battery, a bigger (200 amp?) alternator and a relay in the brake light circuit. A voltmeter and bypass switch to make sure that you have enough juice during
extended highway driving.

A crankshaft-mounted (i.e motorcycle) alternator would be even better than a belt driven alternator because it would not have belt drag, but the bigest crankshaft alternator I've found is only 35 amps.

As seldom as many hypermilers even use their brakes, I fear that battery would never get to charge.

I like the idea though.

[Ryland]

I think the idea of placing a thermoelectric generator, like a peltier junction on the exaust would have better results then using the engine coolent, because exaust will reach a higher tempeture much faster then engine coolent.
altho cooling the exaust will most likely have some negative affects, I would think it would cause less drag then something like a exaust turbine run alternator.
as far as simple, and cheap go, a cog belt or chain drive (98%+ efficent) rather then a v-belt with a standard alternator I think would be the way to go.
I would be interested in seeing a comparison of running an alternator at high speeds compared to low speeds to get a set amperage out of it, as I would think that an alternator might run more efficently at higher speeds, and if it's output is controled by the field windings, if you get away from the v-belt, and maybe even get some higher quality bearings (cermamic?) and clean all the electrical connections (gold plated? soldered? larger wire for less line loss?), then I would be intersted in seeing what kind of efficentcy can be gotten out of a traditional style alternator.
a while back I saw a over sized generator/motor that replaced your alternator, and sensed engine load, giving any car electric assist, and regenerative braking thru engine braking.

[DaX]

Originally Posted by Ryland

a while back I saw a over sized generator/motor that replaced your alternator, and sensed engine load, giving any car electric assist, and regenerative braking thru engine braking.

Now that sounds awesome! Let a chain driven alternator do some regen!

[larryg]

I've been thinking about this lately. I know this sounds crazy but I think upstream restriction in the intake of a fuel injected car wouldn't affect FE. My thoughts are that the biggest restriction in the system is the throttle body and thats why you have manifold vacuum at all. So What if you put a turbine behind the air filter in front of the TB. certainly you would have to increase the throttle opening somewhat but the computer would inject the correct amount of fuel for the mixture anyway. The engines power out put is a function of cylinder filling which is basically a function of manifold vacuum anyway so- I think you would end up with the same vacuum reading at the same speed. no doubt the absolute power of the engine would be reduced due to the limited airflow but I cant find a reason for it to hurt the FE. the turbine could turn a small alternator. Why won't this work?

[skewbe]

Originally Posted by larryg

... Why won't this work?

You called?

Here's a real rough guesstimate:

I think my old chevy v8 had a 300 cfm carb on it, that's like what a 150 watt fan can produce. My metro has like 1/5 the displacement, so thats the equivent of a 60 watt fan moving air, now take that fan and use it as a turbine and give up another 50% efficiency.

you are left with a 30 watt generator, and that is at max rpm with the throttle wide open. I don't even know if that's enough for the fuel pump.

[skewbe]

Does anyone make efficient LED headlights?

I've been trying to think of a way to obtain a sustainable alternatorless configuration away from the grid and I always come back to dramatically reducing the electrical load. With several starts per trip and DRL's and electronic fuel injection, I'm stuck. Maybe I can pull my alternator and get a 100 amp alternator (or two) that I can rub up against a rear wheel with a lever so I can recharge sufficiently during the few times that I have to mechanically slow down, or something?

[rh77]

Originally Posted by Silveredwings

As seldom as many hypermilers even use their brakes, I fear that battery would never get to charge.

I like the idea though.

I've been doing some free-thinking on the topic as well. With my automatic, I don't have the luxury of bump-starting and rely on the juice available to fire up the ICE (the old battery continues to hold-out with lots of COFISHing, cold weather and a self-implemented daytime headlight policy -- knock on wood).

Perhaps whenever you coast -- meaning not necessarily when the brakes are applied, but whenever you are at complete throttle-lift a charging system would help (and would happen more often than braking for H-Milers).

Skewbe -- I've also heard of a crazy "fifth wheel" that can be deployed on decel (engine on or off) that turns a generator. During engagement, it would probably be like a plane's tire on landing -- complete with a puff of smoke (and would need a rudimentary suspension to handle road undulations), but if an appropriately-weighted flywheel is attached, the spinning momentum could continue to charge the even after the "landing gear" is retracted. The downside: hitting a pothole and losing the whole works. Maybe less dangerous (but still risky) would be your tire-on-tire idea that rides with the suspension movement of the wheel. I'd love to experiment with such an idea.

RH77

[Sludgy]

Originally Posted by larryg

I've been thinking about this lately. I know this sounds crazy but I think upstream restriction in the intake of a fuel injected car wouldn't affect FE. My thoughts are that the biggest restriction in the system is the throttle body and thats why you have manifold vacuum at all. So What if you put a turbine behind the air filter in front of the TB. certainly you would have to increase the throttle opening somewhat but the computer would inject the correct amount of fuel for the mixture anyway. The engines power out put is a function of cylinder filling which is basically a function of manifold vacuum anyway so- I think you would end up with the same vacuum reading at the same speed. no doubt the absolute power of the engine would be reduced due to the limited airflow but I cant find a reason for it to hurt the FE. the turbine could turn a small alternator. Why won't this work?

Larry, you're absolutely correct about using engine vacuum as an energy source. I'm old enough to remember when windshield wipers ran off vacuum. It was a terrible system, actually, since vacuum disappears when you press the throttle. So the wipers stopped when you were accelerating hard!

But using vacuum for generating electricity, it doesn't matter if the alternator stops generating for a few seconds, since the battery still provides juice.

[Silveredwings]

Originally Posted by rh77

Perhaps whenever you coast -- meaning not necessarily when the brakes are applied, but whenever you are at complete throttle-lift a charging system would help (and would happen more often than braking for H-Milers).

That would definately be true for me.

Skewbe -- I've also heard of a crazy "fifth wheel" that can be deployed on decel (engine on or off) that turns a generator. During engagement, it would probably be like a plane's tire on landing -- complete with a puff of smoke (and would need a rudimentary suspension to handle road undulations), but if an appropriately-weighted flywheel is attached, the spinning momentum could continue to charge the even after the "landing gear" is retracted. The downside: hitting a pothole and losing the whole works. Maybe less dangerous (but still risky) would be your tire-on-tire idea that rides with the suspension movement of the wheel. I'd love to experiment with such an idea.

RH77

I think the point is to reclaim otherwise wasted energy whether kenetic or thermal. As with hitting a pothole, I can imagine how suboptimal it would be to have a mere speed bump destroy a $3000 peltier-junction 'alternator' attached to the bottom of the exhaust system. For that reason, thermocouples on cars don't seem to be ready for prime time yet.

[Sludgy]

Thermoelectric generators were put into a pickup truck too:

http://www.nyserda.org/programs/tran..._Generator.pdf

$1500 buck is a lot of money to put into a truck for a 2% efficiency improvement.