Hacker News new | comments | show | ask | jobs | submit login
New lithium-ion battery design – 2,000x more powerful, recharges 1,000x faster (extremetech.com)
148 points by derpenxyne 1644 days ago | hide | past | web | 86 comments | favorite

For people having trouble interpreting the graph, the legend from the actual paper says:

"The energy and power density of our microbattery cells (A–H) at low to high C rates, along with previous microbattery cells having 3D electrodes (MB1 through MB3). The plot also includes the performance range of conventional power technologies and commercial batteries from A123 (high power) and Sony (high energy)."

Where C rates stand for charge rates, with 1C being able to charge the full capacity of the battery in 1 hour.

The key take away for the new design's performance isn't the one quoted in the title (of the submission or the article), but rather "Compared with conventional supercapacitors, our microbattery delivers 10 × the power of a supercapacitor at comparable energy density, delivers 10 × the energy of a supercapacitor at comparable power density or has 10 × smaller volume than a supercapacitor at comparable performance".

Just by glancing at the graph, you can see that the design is orders of magnitudes less energy dense than today's lithium-ion batteries. This advancement will make high-power applications more feasible, but will probably not do very much to increase your phone/laptop's battery life.

Electric cars?

Well... it's tricky. If you assume that current electric cars have the performance of that A123 battery from the graph, and then pick something in the middle of the spread of the colourful dots, you get a roughly 1 order decrease in energy density, and a 1 order increase in power density. So, taking the Model S as an example, you have a 300 mile range, and a charge time of like 45 minutes at a super charger.

So... now you end up with like a 30-60 mile range range, and can charge up fully in like 3-5 minutes. So we've brought charge times into the realm of filly up gas, but you're doing it waaay more often.

If this is a useful advancement or not depends completely on what type of infrastructure we want to envision. I would certainly love a world where this was possible/feasible, but it's certainly not a grand slam.

Minor nitpick is charge time at a supercharger would remain 45 minutes. You'd need some kind of miracle superduper charger to deliver that kind of current. A cable that can handle that kind of current would be kind of unimaginable for the average civilian. It would resemble a telephone pole. Not the wires on the pole, but the pole itself.

A stereotypical thermodynamics class assignment was something along the lines of assume your 10 gallon gas tank is filled in a minute at the gas pump through a little half inch gas nozzle. Now calculate the thermal energy of that nozzle assuming perfect combustion in megawatts. Then given that the gas nozzle is about half an inch, look up what diameter copper wire would be required to transfer that kind of energy if it was electricity. The answer was pretty stunning.

For those of us that didn't study thermodynamics, what is the answer that's so stunning?

10 gal = 37.85L

37.85L x 36MJ/L = 1,362 MJ in the tank.

1,362 MJ / 60s = 22MJ/s from the hose.

J/s is Watts, so that's 22 megawatts. A cable that carried 200A at 110KV would do it. I think 000 gauge would handle 200A, it's just under 11mm in diameter.

Agreed, stunning it is that so common and reasonably safe a thing as a self-serve gas pump is a 22 MW chemical energy delivery system!

Some reflections:

* Perhaps an EV might be 10x as efficient with its input energy as a gasoline powered car. Perhaps 220 KW would do.

* Perhaps a robotically mobile, rather than a counterweighted, manually lifted and connected cable is in order. The present day robotic car wash design where the car stays still while the motile, rotating washing machinery moves around it comes to mind.

* Electrical connectors give me more trouble than wires and cables. A replaceable, quick-disconnect connector capable of 20KA seems more of a design challenge than the cables that would go with it.

* High voltage insulation and wiring is "easier" than high current conductance (low resistance) wiring, for example: Common engineering practice with 30-volt solar panels is to wire them in series so that the total string voltage is 300 to 600 VDC at 8 Amps, rather than in parallel so that the current would be 80 to 160 Amps at 30 Volts. Air is a lousy and unsafe insulator, but there are good liquid insulators which could easily, temporarily fill a two-pole, ground-surround high-voltage connector (displacing the air) before it's allowed to be electrically energized and ramped up to a high voltage. Precise ground fault current detection could make it safe from fault currents (i.e. shocks & shorts) in the milliamp range even as the cable carries hundreds or thousands of Amps. But any HV supply requires up-conversion at the station and down-conversion in the EV; whilst these can be 95% efficient, the effect of each conversion inefficiency is multiplicative, and they add weight.

* Then there's the charging station itself. If it has eight charging stations at, say 250 KW each, that's 2 MW draw from the utility (or an underground group of batteries recharged at a lower rate from utility power?) when all stations are charging cars. Not trivial; just the 500 KVA (call it 500 KW for discussion purposes) utility distribution transformer outside building where I work is the size of four refrigerators, not including its switchgear, all of which is enclosed by a 20 foot tall fence surrounding about 200 sq feet. Multiply that by four in volume to get 2 MW supply required from the utility.

But 110kV into the car is totally out of the question. At a more reasonable battery voltage of 1kV, you'd need 20kA. That'd be a fat cable indeed.

But 110kV into the car is totally out of the question.

It would be pretty unreasonable. But burning gas at 100% efficiency is actually impossible, so it's sort of a silly scenario anyway.

...you'd need 20kA. That'd be a fat cable indeed.

Maybe not as fat as you'd think; this guy figures 50KA here, and you wouldn't want to work with an extension cord like that every day, but it's not like you'd need a crane to lift it, either:


Whoa am I posting late sorry.

There's a simpler way to do the analysis which doesn't require heat of combustion of typical gasoline analysis and flowrates and such. Assume it takes a KW for a horsepower (perfectly accurate for 1 sig fig) and it takes a couple horsepower (kilowatts) to go down the highway on a long road trip. To at least one sig fig this is correct. Also it takes about 3 times the chemical gasoline energy to generate electricity because IC engines are not terribly efficient. We'll call it 10 KW gas equivalent to drive on the interstate at a constant-ish high speed.

Now what fraction of the time of a road trip do you spend pumping gas vs driving? Well it can't take more than 4 minutes to fill a 10 or so gallon gas tank, and on cross country trips I can get a good 8 hours or so outta one tank, so to one sig fig I spend WAY less than 1/100th the time pumping gas vs driving.

So multiply 10 KW by 100 times the flow rate (conservation of energy and all that)

Another way to look at it, is think of the flow rate, lets say 5 gallons per minute. The kind of engine required to burn a couple gallons per minute resembles my uncle's hospital UPS backup engines, certainly in the megawatt class. This is also in the thousand horsepower (aka megawatt) class airplane engine and race car category.

The OP didn't give enough information to produce a specific answer. For a higher voltage and lower current (same power), the wire could be thinner with the same outcome. The reason electricity utilities string million-volt lines between cities is to minimize the heating losses in the conductors, not to impress the civilians.

It's the same reason electric car designs favor putting a lot of cells in series -- a high-voltage electric motor gets more of the power delivered by the batteries, less of it is wasted in the connecting wires. Also, in the car's charging circuit, the high voltage choice maximizes power transfer for a given connecting wire size.

You still need to convert this high supply voltage to a lower voltage to match the cell voltage for charging. The currents involved mean this converter is still likely to dissipate a lot of heat, even if it is extremely efficient.

> You still need to convert this high supply voltage to a lower voltage to match the cell voltage for charging.

No, instead you charge the cells in series. That delivers the same power while avoiding the problem of having to produce overly high currents.

> The currents involved mean this converter is still likely to dissipate a lot of heat, even if it is extremely efficient.

Yes, but there are ways to minimize this loss factor. Obviously a fast-charging scheme for a car is going to involve a lot of current, but the system designers do all they can to minimize that current. The primary way they do that is by both charging and discharging the cells arranged as much in series as is practical.

In the Tesla design, there are 11 modules connected in series and delivering 375 volts nominal (and requiring 375 volts to produce an optimal charging current). So there's no issue of changing individual cells at their normal cell voltage, instead the system charges modules consisting of 621 cells in an optimized series-parallel scheme:


Quote: "Tesla Motors refers to the Roadster's battery pack as the Energy Storage System or ESS. The ESS contains 6,831 lithium ion cells arranged into 11 "sheets" connected in series; each sheet contains 9 "bricks" connected in series; each "brick" contains 69 cells connected in parallel (11S 9S 69P)"

By 'cell' I meant the overall pack, not the individual unit within the pack.

From your link: "A full recharge of the battery system requires 3½ hours using the High Power Connector which supplies 70 amp, 240 volt electricity"

Multiply that current by 10-100 to 'charge it faster' and you have a huge problem with thermal dissipation, not to mention using truly massive cables.

> Multiply that current by 10-100 to 'charge it faster' ...

No, that can't be done without damaging the batteries. A fast charge in this context means multiplying the default charging current by roughly three, not 10, and certainly not 100.

Once you start getting into the thousands of volts, it becomes very dangerous. That stuff will arc through the air and kill you. There is a reason why we use a hundred or two volts in consumer appliances.

Not necessarily. If the connection would be unmanaged, as having the power available for drainage all the time, then you're right! ...but we have computers and all kind of complex systems now. The cable may be "empty" of that high voltage current until the cable is safely plugged in and ready. After plugging, before unleashing energy, the car's system and the power source's system may first communicate and verify their sensors if everything's ready to go, like isolation and other things. The high voltage risk and its management it's an engineering solvable problem.

You have the wrong idea of how voltage works. High voltage electricity can and will kill you before the ground fault interruptor (or whatever) can kick in. It takes time to break a high voltage circuit, (on the order of milliseconds), and during that time, you will die. If the current is across your heart it's especially lethal.

If the voltage is high enough, there is no way to break the circuit anyway. The electricity will just arc across the (ionized) air, even if the computer physically disconnect it. Assuming the computer itself isn't killed by the spark, which is pretty likely. This is why linemen on high voltage lines don't even try to electrically insulate themselves when they're working with it live. Ask wikipedia:

For high-voltage and extra-high-voltage transmission lines, specially trained personnel use "live line" techniques to allow hands-on contact with energized equipment. In this case the worker is electrically connected to the high-voltage line but thoroughly insulated from the earth so that he is at the same electrical potential as that of the line. Since training for such operations is lengthy, and still presents a danger to personnel, only very important transmission lines are subject to maintenance while live. Outside these properly engineered situations, insulation from earth does not guarantee that no current flows to earth—as grounding or arcing to ground can occur in unexpected ways, and high-frequency currents can burn even an ungrounded person. Touching a transmitting antenna is dangerous for this reason, and a high-frequency Tesla Coil can sustain a spark with only one endpoint).

You understood wrong my statement. Assume that you have two independent connecting channels in the same connecting tube. One is a live low-voltage channel for communication protocol and the other one is high-power (but dead in idle) pumping channel. When you plug the tube you automatically connect both channels. After that, having the info-line established, having assured that the couple has been locked & safe (i.e. insulated), both systems might then agree on opening the second high-power channel. It won't matter if it will take more than a few milliseconds to open or cut the power with the tube being in a safe coupling state. By "safe coupling state" I mean that not even the contact with the possible ionized air is allowed, the line being covered/insulated all the way. You just design the systems to not allow power to drain from the source unless it's safe. I also think that it's already time to design all the systems like that, not only the high-power car-recharging ones. This would make a difference in lives saved.

Google "ground fault interruptor." What are you are talking about already exists. And it doesn't solve the problem.

Could electric car battery use a dual type battery pack to optimize charging and range?

You could do that, but energy density is the key metric here.

Energy density determines the size of the battery required to store a specific amount of energy. There's only so much space in the car, and you also have to factor in that any battery that will be moved by its own power loses efficiency with lower energy density. You've got a huge battery to move, and physics dictate that accelerating this battery is going to use more energy.

So, let's say you split this hybrid battery 50/50, your quick charge-discharge battery quickly becomes too small to deliver meaningful range, and you've given up space that could be used for the traditional battery.

The Tesla Model S already uses 7000 individual battery cells. This is how it's able to charge as quickly as it does with the supercharger. The main limitation of household charging is lack of power.

That isn't necessarily true. Them using small batteries to make a larger battery is what allows them to make their car sized batteries cheaper, since the smaller sized batteries are already mass produced for things like laptops. For charging time it depends on how those batteries are connected.

Don't you mean a charge in 24-30 seconds? Otherwise your example is just that of a 10x smaller A123 battery.

45 min is 2700 seconds, so 10x faster charging is 270 seconds or 4 1/2 minutes.

The rate that the battery charges is independent from the capacity of the battery. Assuming the new tech is 10x less energy dense and has a 10x faster charge rate, if you compare one of these new batteries with capacity equal to that of the Model S battery, the new battery will charge 10x faster but occupy 10x the volume, since it is less energy dense. A new battery of this kind that occupied the same volume as a Model S would contain 10x less energy than the Model S battery, but it would charge to full capacity 100x faster than the Model S.

There should be two multipliers though right? If it charges 10x faster and you are charging to a 10x less total energy it should charge 100x faster right?

Imagine shorting this capacitor.

Weight, Energy density. and cost are what's most important for general Electric car energy storage. So if this cost's more or weights significantly more it's next to useless for that. However, they might have some use as part of a more efficient regenerative breaking system.

PS: As to faster charging, issues are going to Limit that. A 99% efficient battery charging 100x as fast as Li-Ion is going to dump a lot of heat vary quickly. Duping heat is not really an issue for small battery's but scale this up into the 50+kwh range for an electric car or even just a cellphone battery and you will have significant issues.

About charging speed, you're invoking another problem, that of (over)heating. That however, deserves a separate context of its own to be addressed in. I think it's safe to assume that a lot of efficiency can be safely salvaged from that 100x improvement.

One parameter not discussed so far is self discharge rates. Maybe a problem maybe not, either way that parameter is going to be important.

For simple regenerative braking its not so big of a deal but if you're hoping to use it as a primary energy source its a big deal.

This is a classic case of a submission title taking liberties with the meaning of the article it links.

This submission's title: "New lithium-ion battery 2000 times more powerful, recharges 1000 times faster"

The actual title: "New lithium-ion battery design that’s 2,000 times more powerful, recharges 1,000 times faster"

The missing word is "design", a word that is crucial to the meaning of the article. The article discusses a design for a battery, not a battery. How important is that? Here's the last sentence in the linked article:

"For this to occur, though, the University of Illinois will first have to prove that their technology scales to larger battery sizes, and that the production process isn’t prohibitively expensive for commercial production."

It's a design, not a battery.

The solution to this problem, as I have pointed out many times before, is to copy the article's title, do not invent your own.

If HN would not force you to reduce titles to < 80 characters then likely the original title would have been used.

The original is 95 characters. "design that’s " and two comma's in the numbers get dropped and we're down to exactly 78.

Dropping words from a title in order to fit the 80 char limit is a bit of an art if you don't want to change the meaning, and making a new title is frowned upon because then you care not using the original title. Effectively you can't post an article with a title > 80 characters and not break the rules.

> Effectively you can't post an article with a title > 80 characters and not break the rules.

I never knew that. Thanks for the heads-up. I find it annoying that people rewrite headlines, but I didn't realize there was an actual reason.

"New lithium-ion battery design 2000x more powerful, recharges 1000x faster" fits with out altering the meaning at all.

I didn't say you couldn't do better, I just pointed out the likely source of the problem.

GGP says

> copy the article's title, do not invent your own.

Unfortunately you can't always do that. Dropping words is one technique to shorten the title, rewriting it is another, both techniques are employed.

We agree on the topic. Indeed it is tricky to strike the right balance when trying to edit something down to fit and keep the meaning. I probably do this dozens of times a week trying to get the full intent of a thought down to 140 chars. :)

Could always opt for the scientific notation.

"New Li+ battery design - 2,000x more powerful, recharges 1,000x faster"

Now you just have to know that Li is Lithium and the + is for an ion.

> Could always opt for the scientific notation.

> "New Li+ battery design - 2,000x more powerful, recharges 1,000x faster"

In that spirit (and just for fun): "New Li+ battery design - 2k x more power, charges 1k x faster.

I originally tried scientific numeric notation, but that didn't actually shorten the title: 2x10^3 versus 2,000. :(

> So far, the researchers have used this tech to create a button-sized microbattery, and you can see in the graph below how well their battery compares to a conventional Sony CR1620 button cell.

That sounds like a battery to me. The new design has actually been built and tested. The quote you pulled out refers to scaling it up to things like lasers and F1 cars. A button-sized battery sounds fine for a lot of electronics today. Just not an F1 car.

Building one in the lab likely does not give you the ability to build a million of them at a decent price in a factory.

Yield probably is low, and the current process will likely be way too labour intensive.

True. But lutusp was shitting all over the submitter for dropping the almighty "design" word from the title and then claimed this is not an article about a battery but only about a design for one. That claim is wrong. It is more than a design... it is an actual battery. Maybe there is only one unit that exists. But that is still one more than a design. And if the tech proves sound but too labor intensive, your next thing to tackle is to make it easier to build. Isn't that how bleeding-edge innovation works?

Bleeding edge innovation usually occurs when researchers and engineers make progress using known or easily-modifiable processes. Most "breakthroughs" are dead-ends because process or scaling weren't considered in the initial research.

True technological innovation needs not just engineering feasibility but also economic feasibility. I'm not trying to say that the research that "dead ends" is pointless. Progress is progress. All I am trying to say is that it doesn't mean economic viability is right around the corner.

Often, researchers are well aware of scaling or manufacturing hurdles when they embark on the experiments. For the sake of academic careers and exposure, the research will get hyped regardless of the viability of implementing the research.

> But lutusp was shitting all over the submitter for dropping the almighty "design" word from the title and then claimed this is not an article about a battery but only about a design for one.

Yes, but guess what? That's true -- it's not a marketable product, it's a laboratory curiosity. That difference mustn't be dismissed. There are any number of research developments that never become products for reasons other than their theoretical properties.

> Isn't that how bleeding-edge innovation works?

Not exactly. If if were, a solar panel that converted a higher percentage of the sun's energy into electricity would beat a panel that produced more power per dollar. But in reality, it's the other way around.

> That sounds like a battery to me.

Not when the public thinks a battery is something you can buy for a reasonable price. This reminds me of the solar panel business, where the driving market issue is not cell efficiency, it's cell cost per watt. A panel that produces more watts per dollar will usually beat a panel that's more efficient at capturing the sun's energy.

It sounds like they've developed a prototype, which is further than a design but also not a production model.

Agreed. But it is a functional battery more than it is "only a design"

Fair enough. Is a prototype flying car a design or a product? Can I get one -- not the design, the product?

Neither... it is a prototype. Somewhere between just a design and a product is a wide variety of other states. A single, physical item that actually functions as designed may not be a product yet (or ever) but it is leaps and bounds beyond "just a design." If "prototype" and "design" mean the same thing then a lot of people get that wrong everyday.

The original title wasn't the greatest either, probably engineered to be link-bait. Nobody cares how fast you can get power out of the thing, they care how much total energy it holds - and the article is clear that it is comparable or slightly worse than current technology.

Low internal resistance can make life very exciting in case of short circuit.

Maybe a decade ago there was a fad of sticking a bare, unbiased LED across a button cell and relying on the internal resistance of the cell to current limit the LED. Then you superglue a magnet on it and toss it and you've got a LED throwie. If you tried that with a super low internal resistance battery like this, the effect would be vaguely similar to sticking the LED across a power outlet (boom).

For some part I agree with you but it's difficult to make the cut down to 80 characters for titles.

To be clear, this is the ExtremeTech blog, all of their commentary leans heavily toward wishful thinking.

For context a number of surface area improvements in Lithium-Ion anodes have failed at scale due to the creation of lithium dendrites (metal 'wires' that grow during recharge) which have either rendered large sections of the anode worthless and reduced the capacity of the battery, or created rapid exothermic reactions (aka fires) in the battery.

Its on the BBC New site as well which is actually much less clear in explaining that this research is applicable to the supercapacitor space as opposed to 'normal' Li* batteries.

Where's that xkcd chart when I need it for every announcement like this...

Ah here we go: http://xkcd.com/678/

To be clear: high power density, but similar energy density to current batteries.

At some level, it doesn't make too much sense to go that much higher in the power (which I'm assuming is the same as discharge rate, since voltage is fixed). Current lipoly batteries are rated at 30C (Amps per watt*hour). Essentially this means that the battery can be fully discharged in a minimum of two minutes. It doesn't seem like there are that many applications that require higher power output.

Agreed. I would argue that many (most?) higher power applications would be better served by ultracapacitors anyway...

cars, for one, especially if high power output correlates with high power input for charging.

Exactly. It's very rare to find an application where a rapid discharge matters (battery powered guided missile maybe? A drone launcher that boosts a device to altitude and then comes back down on a parachute?), but a rapid charge speed is almost universally desired. At the numbers being quoted here, for example, it would become faster to simply charge your phone than swap the fiddly battery.

The title is completely misleading.

Yeah, it's literally correct if you understand the difference between power and energy but most people don't.

Including the author considering he is claiming that this would leed to smaller cell phone batteries...

I was hoping for 2000x battery life for my laptop...

Uh, let's combine this with the other 3 articles claiming a 10x more long lasting battery, and we'll have a 1000x longer lasting, 2000x more powerful, 1000x faster recharging battery.

I really love progress, but I've seen too many of these "N times better" articles and no actual consumption products by now.

Same with these memristors, MRAM and feRAM, and 1000x faster internet.

> 1000x faster internet.

I don't know about you, but my first internet connection was 56 kbps, now I can easily buy 100 Mbps one (I don't cause 10 Mbps is good enough for me) and it would be much cheaper than what I had to pay 15 years ago for that 56 kbps, even including inflation.

Progress is here, we just don't notice (also media and inventors overhype). But in the end the progress is still happening unbeliveably fast. It's decimal order of magnitude every 5 years!

The one thing they did not mention was the battery weight and size. The biggest barrier for electric cars and other battery uses is the ratio of energy to weight and size.

Today, this ratio enables about 70-100 miles travel for an electric car with a similar weight and size as a gasoline car.

will be interesting to follow this initiative.

If a car battery can be made to charge to 50% capacity in about a minute with a total range greater than 60 miles or so, then a very novel solution to charging electric cars becomes possible -- induction coils at stop lights and along highways.

In the bigger cities and between them we could just connect electric cars to grid the whole time, and put energy meters in each car. Like trolleys. No need for big batteries inside the car, just supercapacitors for a few minutes worth of driving if there's some problem with the grid and you need to park safely.

I wonder if people could buy electric cars like they buy cellphones - you sign a contract with a provider, you pay X USD per month for the car and access to the grid, additionaly you have access to free regular car if you need to drive somehwere the grid doesn't exist yet(but you pay for the fuel by yourself).

The problem is infrastructure, I think.

Some busses have been running on this kind of wiring in cities for ages.

Yep, we have trolleys in my city and they work OK. Still less popular/more difficult routes are served by regular buses.

"2000 times more powerful, recharges 1000 times faster " - so it will still take twice as long to charge it fully then?

No, the energy density is not all that unusual. Assuming the battery is 2000AH and is 100% efficient, the "2000 times more powerful" means you can (theoretically!) discharge it 2000 times faster than the benchmark battery (e.g. 2000A over 1 hour vs. 1A over 2000 hours).

The 1000x recharge rate means you can put the energy back in 1000x faster than the benchmark battery (e.g. 1000A x 2 hours vs. 1A x 2000 hours).

Due to the energy density being comparable, I would expect the actual size of a 2000AH New Improved[tm] lithium-ion battery would be comparable to the size of a traditional 2000AH "benchmark" battery.

Disclaimer: All the numbers made up out of thin air using values that were easy to do in my head.

It discharges/charges much faster, that is all about the new battery.

Discharging fast make it more possible to catch fire.

Overall, it seems not very useful for smartphones.

Even if it doesn't catch on fire, discharging faster isn't what I want for my phones battery.

But nowhere does this article even talk about using this for smartphones or in any form of application.

"In real-world use, this tech will probably be used to equip consumer devices with batteries that are much smaller and lighter — imagine a smartphone with a battery the thickness of a credit card, which can be recharged in a few seconds."

Design != Build != Shipping

I love this research but have seen similar articles for the past decade. At scale production is hard, just ask A123.

The ideal: Design, build, sell.

The real: Sell, build, design.


I wonder if 2,000x more powerful means generates 2,000x more heat and the risk of exploding is increased by 2,000x as well?

The most exciting application they can imagine for a super-battery is smaller cell phone batteries?

That doesn't seem very forward thinking.

Excellent surface area to volume ratio for cooling.

If you want to shove tens of amps into something tiny its gotta be nearly perfectly efficient and/or excellent cooling and/or be tiny to get a great surface area/volume ratio.

I bet this would rock for bluetooth earpieces.

Also people expect some apps like cars to last for 15 years, but if you're "nervous" about long life, your stereotypical disposable cellphone is a great app.

Great, my Android will be alive 6 hours longer

Oh good, I can play Angry Birds longer.

And it will be available in 10 years.

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