Hacker News new | comments | show | ask | jobs | submit login
700+ hp electric Honda S2000 built by high school senior (hackaday.com)
65 points by ek 1587 days ago | hide | past | web | 53 comments | favorite

There's much more information in the linked forum thread: http://www.s2ki.com/s2000/topic/975497-wait-that-isnt-an-f20...

For example, he also built a keyless entry system with an arduino board and an iPod. This kid is pretty amazing.

Does anyone else not see any of the pics submitted by the OP in the forum?

There's no way he's getting 700+ HP from those two warp 9 motors. It says each motor is capable of outputting 34HP continuously: http://electriccarinternational.com/media/img-prod/WarP_9_Sa...

That appears to be at 144 volts and 500 amps. His stated plan is to deliver 343 volts and 2000 amps. Assuming this doesn't melt the motors, he'll get more power than the datasheet claims. That said, his calculations might be a bit off.

144Vx500A = 72KW or 96.5HP. He's assuming 85% efficiency, which would give 82HP, not 34. Furthermore, if the motor were to deliver 100 ft-lbs of torque as stated at its stated maximum speed of 5800 RPM, that would be 110HP. This would obviously require more than 72KW of electrical power input.

Taking the ratio between the motor's stated performance and the electrical numbers given on the datasheet, and multiplying by the electrical power he intends to supply, we get 34/96x920 = 326HP. This too seems a bit off, as 35% efficiency is quite low for an electric motor. My guess is that the motor is actually incapable of producing its full torque at higher RPM.

My guess is he will actually produce something resembling the amount of power he's hoping for, but only briefly, after which they will deliver no power at all, ever.

Without the torque-speed curves for these motors, you just can't tell.

The actual graphs are at http://www.go-ev.com/images/003_09_01_WarP_9A_Graph.jpg and http://www.go-ev.com/images/003_09_02_WarP_9_SpreadSheet.jpg

My guess is that the motor is actually incapable of producing its full torque at higher RPM.

Correct. At 100 ft/lbs the speed drops to just over 2000 rpm.

I would guess these motors have a thermal fuse wound into the stator coil (source: spent a summer in a rewinding shop), and if he tries to drive them much outside the design limits, they'll simply cut out, perhaps permanently.

EDIT: the datasheet mentions a temperature snap switch, so it sounds like they have a non-permanent cut-out on thermal overload.

Nope, but it will drive a car, just not as fast as I'm sure he would like.

When he realizes his error, he'll be able to reason much better about electrical and mechanical engineering than many of his peers in school (if he decides to go to college) due to having actual experiences of failing in the real world (rather than just on paper). Good for him.

I also don´t see any kind of transmission. I don´t think it´s a good idea to mount the motors directly to the drive shaft.


http://electriccarinternational.com/media/img-prod/WarP_9_Sa... mentions 5.800 RPM (min. 2.000, max. 5.800 RPM). Thats too much to drive the wheels directly. Does anyone see a gearbox in the forum?

It still has a rear differential, stock ratio is 4.10.

Let´s see if he get´s this thing working. Building two e-motors and some batteries inside a car is a very "easy" task compared to building the "control unit" and cooling the batteries down. This two motors will draw a lot current, the electrics to control this huge amount of current will be a hard task to figure out. Another one is the battery cooling. 700hp will take a lot energy, the batteries will get very hot. If you ever had played a 3D game on your phone, you know what I mean. ;o)

If he get´s this thing working, it would be really great! I can´t wait to call a electric car my own. :D

If you interested in such kind of work, you should search on Youtube for "electric VW Golf II", "electric VW Polo" or "electric VW Käfer". ;o)

And if he want´s to build a "good" electric car, he should keep an eye on the EMF and EMI radiation.

Good point, although I can see the ricers using this to cause trouble instead of their boom boxes instead :(

Props to the kid for being such a whiz, but DAE shudder at the thought of a 17-year old getting behind the wheel of this thing? My mom's minivan had about 150hp and I still managed to do incredibly stupid shit with it. All the while believing I was an above-average driver...

I've met anyone who doesn't describe them self as a above-average driver

Introductory blog post here: http://electrichonda.blogspot.de/2011/05/introduction.html

Juan's project is impressive, no doubt, but his analysis of the superiority of electric cars is suspect. He writes:

The average car uses about 20% of the potential energy of the fuel in useful motion while wasting the other energy as heat, noise and pollution. Electric cars use upwards of 80% of their 'fuel', a percentage which can be increased relatively easily as most of its losses are electrical.

The biggest problem is that he's treating the electricity in the battery as raw fuel. He does acknowledge that most electricity in the US comes from sources like coal, but ignores various losses between that coal and his car flying down the road. Let's add those in.

The best we can do from a coal plant is about 49%[0]. It may be possible to use the waste heat from such a plant for other purposes; I am not including such uses in this calculation as they do not have anything to do with powering the electric car.

The US power grid is around 93-94% efficient[1].

Electric cars don't get all the power from the grid in to the batteries as chemical energy. Charging batteries generates quite a bit of heat. The EPA estimates that the best electric cars are 62% efficient at converting power from the grid in to kinetic energy[2].

So we have 49% * 94% * 62% = 28.6% efficient at converting the chemical energy in coal to propelling the car. This is not a dramatic improvement over internal combustion. There is, of course the advantage of using power sources other than coal and oil, but the claimed efficiency is not quite what electric vehicle proponents would like it to be.

[0] http://en.wikipedia.org/wiki/Aved%C3%B8re_Power_Station

[1] http://en.wikipedia.org/wiki/Electric_power_transmission#Los...

[2] http://www.fueleconomy.gov/feg/evtech.shtml

If you want to include the efficiency of the entire electrical pipeline, then you need to do the same for the gasoline/diesel engine.

How much energy is used for the extraction, refinery, transport, and distribution of gasoline/diesel fuel? I suspect 26% efficiency for gasoline would also be extremely high compared to reality if those costs were factored in.

Last time I checked on average when you include extraction, refining, and transportation a higher % of coals energy ends up in the battery's than does Oil's in the gas tank. Initially, oil took little energy to refine etc, but we have started doing things like tar sands and cracking which take a lot more energy than earlier methods. While there is still some easy oil out there, Oil is worth 10x as much as coal vary energy expensive Oil starts to be worth extracting.

PS: If you compare solar cells with Bio fuels it's vary one sided with electric cars far out in the lead.

Splitting hairs now, but coal for the grid also has extraction and delivery costs - probably not equal to that of oil, but if you start from the earth for one you need to start from the earth for the other.

I'm talking about the efficiency of the pipeline starting with a fossil fuel and ending with a car moving down the road. It's also fair to start the comparison with a fossil fuel in the ground, but in that case you also have to start the electric pipeline at a coal mine.

Electric might well win by a significant margin, but it isn't nearly the 85% to 20% claimed.

Well, your car comparison started at electricity production. At the very least, your gasoline example should start at the refinery.

Does anyone know the efficiency of the average refinery? Or distribution costs? My suspicion is that they are actually just as bad as the overhead of burning coal.

I'm not sure what the 'efficiency' of a refinery is. In terms of energy consumed per energy produced? How about equipment/installation costs, costs to the environment, man-hours etc. All different units, hard to avoid apples/oranges issues.

28% is absolutely a dramatic improvement, over 20%. It translates to a dramatic decrease in coal consumed to drive the vehicle - close to a third less.

The EPA says the best internal combustion cars are 26%: http://www.fueleconomy.gov/feg/atv.shtml

And coal plants are about the worst yield of various energy plants.

Anyway, the whole thing is moot. The article may as well say "High School Student crams stuff under hood of car, claims it will kick butt if he can get it to work". I predict we'll never know.

Coal plants may be the least efficient, but they're the most common source of electricity in the US. To be fair, I picked the most optimistic efficiency I could find for a coal plant. I was trying to present a best-case scenario.

It looks like the best case with natural gas is 60%, or 35% total system efficiency. Yes, that's a significant gain over 26%, but it's a best-case scenario. The real-world efficiency of the most efficient internal combustion cars is fairly similar to that of electric cars.

It may be harder to measure 'efficiency' of nuclear (which we use here), solar and wind. Certainly they are more sustainable. Maybe we want the 'sustainability metric' instead.

It's not hard to measure the efficiency of nuclear or solar; it's just not as useful as directly comparing two systems that run on fossil fuels.

Hm. How do you compare uranium fuel efficiency to coal? Chemical vs nuclear reactions - by weight? cost? heat?

You don't compare them, as such comparisons aren't useful.

You can measure the efficiency of a nuclear plant by comparing the total amount of energy released in the reaction to the amount of electric output. Such measurements are commonly used to compare different reactor designs to each other.

In the end, there's only human efficiency. How much effort in people-power does it take to get how much energy. Indirectly or directly. We often get confused about money and chemistry etc. They are not really comparable to human capital.

I just wondering where he's getting the cash to convert his s2k into an electric drag racer. I wish I had that kind of access to car and batteries as a kid.

It says he runs his own CAD business, and I imagine he's from a fairly cash-rich background.

He was 16 when he started the build and was using an AMG Mercedes ($100k car) to pick up parts.

I'm going to assume he comes from a fairly well off family.

He also mentioned he's getting some sponsorship in the forum.

Interesting that he bought the batteries "in china" (his scare quotes), I'm wondering if they are off market copies? Having built high current battle bot battery boxes I get really really twitchy around "good deal" battery packs. That said, its amazing how 'off the shelf' this tech is becoming. Hats off to this guy for taking the challenge.

His project inspired me to write a blog post: http://blog.octanenation.com/news/high-school-senior-builds-...

It's a complex undertaking and he will definitely learn more than his classes in high school!

Just read through the whole thread on the s2000 forum, I'm an auto enthusiast but I was more interested in how this kid knows how to do all of this more than I am interested in the car. 16 years old and you know how to retrofit an s2000 with a custom electric motor and build CAD models of it!?

He's 16 years and 300 months old.

What's the weight of the 104 batteries and the gain/loss after removing the gas engine and replacing with electric?

A123 cells are about 180wh per kg, hes using 20ah 4 at 104 cells in series. so thats about 104 3.7 = 396 volts at 120ah. multiply that together you get 47kwh of energy. These batteries would be around 500lb.

These battery pricing must've came down a lot over the past couple years because this would've cost something like $40k couple years ago, but a quick google for pricing looks like the battery pack he is using can be bought for around $13k pricing source: http://www.a123rc.com/goods-468-Excitingly+Powerful+A+123+20...

500lbs in batteries and a quick search shows each of his motors is 156lbs. Since the gas motor he took out weighed 326lbs, he's got a net gain of ≈ 480lbs. That's a lot to overcome.

That's a 16% increase in weight and a 312% increase in power. I used to be a frequent poster on s2ki (autocross an AP1), the weight isn't that much of a hinderance when you increase power that much. Braking and turning will suffer.

These cars offer excellent balance, handling and braking from the factory. I'm guessing all of those may suffer after gaining that much weight.

I had an '05 S2000. I wish I still had it, wonderful car.

Agreed. Best car I ever had (and totalled).

Add on (or take off I guess) the exhaust system, gearbox and fuel tank and he's probably shaved a bit more off as well.

s2000 is a rwd car and likes to spin. this is not a trivial question.

"The battery pack will consist of 416 cells with 4 cells in parallel and 104 in series...The batteries will be A123 20Ah prismatic cells. " Approx 460 pounds just for the cells.[1]


Notes: http://www.a123systems.com/prismatic-cell-amp20.htm

~500g each (1.1 lb) based on a quick google [tbc]

Only a 15% loss of power to the wheels is very low.

Just the drivetrain loss on muscle cars is around 30% approximately.

Honda has fantastic performance gearboxes and drivetrain loss is usually in the low 20%.

Also most muscle cars are slushboxes and the torque converter ruins efficiency.

looks like tesla's are about 88% efficient, so definitely in the range.

I'll be interested to see how he's going to deal with the torque problem that Tesla had to solve.

I'm interested in how he's dealing with even simpler problems like charge time and battery life.

Me, too. I don´t see any battery cooling system. This will be a problem, if he´s getting it to work. With these big motors, the batteries will get very hot ... and maybe, burn down ... or at least don´t work very long.

He's using a COTS charging module: you'll see it if you look through the linked thread.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact