Launch HN: Buoyant (YC S21) – Blimp drones for air freight
An airship (or blimp) is an aircraft that gets most of its lift from a lifting gas like helium. It’s the most efficient way to fly, which means it’s cheaper than any other aircraft for many missions. We’re starting by building an aircraft for middle-mile air freight in remote and rural areas—warehouse-to-warehouse or post office to post office. This is a $6B market in the US alone, and freight volumes are only increasing. By building autonomous blimps, we can lower shipping costs, increase quality and speed of service, and cut out millions of tons of CO2 emissions.
So far, we’ve built and flown four airships. The latest is 20 feet long and can fly up to 35 miles per hour. Here’s a video: https://www.youtube.com/watch?v=VEYOfVcwRhk.
We’re starting in areas where freight is already moved on small planes and helicopters. In these remote areas, availability of goods is lower, shipping takes longer, and it’s more expensive. For example, food in rural parts of Alaska is twice as expensive as in Anchorage. Small air freight is particularly expensive because of the cost of fuel, maintenance, pilots, and airport infrastructure. To make major cost reductions we need a new type of aircraft.
Ben is a mechanical engineer from Alaska and grew up in the outdoors, so is familiar with the challenges of remote supply operations. He has built aircraft tracking technology for the DoD, flat panel phased array antennas, GPS data loggers for motorsports, and helped start a company that made shirts with upside down pockets. I’m an aerospace engineer who’s built and flown spacecraft at two internet satellite startups, spending a lot of time on complicated flying machines. We met at MIT and have been friends for almost a decade since.
We spent years building satellites and antennas to provide internet connectivity to rural areas, and while doing so learned about the transportation challenges in remote places. Many drone delivery projects have focused on delivering small packages in suburbs, and are too short range or low payload to serve rural areas. We realized small airships were a technical approach that could work to move cargo in these areas, and decided to tackle the challenge.
An airship is the most efficient way to fly because it gets most of its lift from buoyancy, rather than spending energy on rotor lift or aerodynamic lift over a wing. This lets us fly further and carry more payload than other small aircraft. Other attempts at unmanned cargo aircraft have used quadcopters or quad-plane hybrid drones. These are useful for some missions but lack the flight efficiency to carry large payloads long distances. Airships have other benefits too: they are safer and quieter than quadcopters or multirotor-plane hybrids. If the motors fail, an airship floats to the ground, while a quadcopter comes crashing down. And it’s an easy way to build an aircraft that can takeoff and land vertically, like a helicopter.
Our airship is a fabric envelope filled with helium, with an attached payload bay, motors, and power system. It gets 2/3 of its lift from buoyancy, and the rest from aerodynamic lift. This combination is called a hybrid airship, and allows us to drop off a payload without needing to take on ballast. The aircraft flies autonomously and can take off and land in inclement weather, using centimeter accuracy GPS for approaches. The full scale version will load 650 lbs of cargo at one end, fly to the destination while we pilot it remotely, deposit the cargo, and return. Our first operational vehicle will be battery electric, with a range of 200-300 miles and a cruise speed of 60 mph. Future vehicles will have hydrogen powertrains for longer...
177 comments
[ 3.1 ms ] story [ 226 ms ] threadGood luck!
https://www.wolframalpha.com/input/?i=20ft+*+3ft+*+3ft+*+2+*...
= 28 lbs (13 kg)
How does it handle weather? You mention it can take off and land in inclement weather, but how does it compare with existing competitors? Intuitively (as someone with zero knowledge or experience in any of this), it seems like planes and helicopters could handle higher winds than a blimp - is that the case?
Our test site is by a wind farm so we have a lot of experience with flying blimps in wind, and we expect our max takeoff windspeed for the full size airship to be 25-30 kts.
What kind of inclement weather are we talking about?
Aircraft have to deal with some really inclement weather. METARs like 12026G41. You mention you are testing in Alaska, which is infamous for this.
How much of a problem will this be?
Have you done experiments into hydrogen safety?
The helium market's been pretty volatile since the U.S. government finished selling off the national helium reserve in 2019. But we think that's mostly due to short-term market dynamics because helium's a hard commodity to produce and there are only a few producers. There are lots of stories about helium running out, but we've dug in and haven't found much evidence that supply is actually drying up
We think we can build a hydrogen blimp safely. It's not on our tech roadmap currently, but in any case the best way to build a hydrogen blimp is to get lots of flight hours on a helium version first.
Interesting. Is this a case of people over-hyping the problem or just that prices are going to rise to a point that people will stop wasting it?
> the best way to build a hydrogen blimp is to get lots of flight hours on a helium version first
Can you just put hydrogen into an existing helium blimp or would you need to design a new one based on flight data?
I was under the impression they required radically different gasbag designs due to the explosion hazards (not to mention the increased lift)
The issue is more about the permeability of hydrogen gas. It can seep through all kinds of materials, including metals, and then embrittle them.
But yes hydrogen is cheap enough that refillable bottles and more leakage is cheaper than helium in terms of just the features of lifting gas. Hydrogen reactivity makes for other materials, engineering, and maintenance challenges that can be ignored with helium.
As best I can tell, this is the case. The price of helium has gone up quite a bit, but non of the commentary links that to the actual production volumes of helium or size of remaining reserves.
> Can you just put hydrogen into an existing helium blimp
We'd need to modify the envelope design, but the rest of the design wouldn't need to be modified (aerodynamics, controls, propulsion, etc.)
What altitudes do you intend on operating in the "v1"?
Interesting to see it go the other way too. Hybrid Air Vehicles near me is the remnants of a cancelled US military autonomous blimp project, now aspiring to use the large helium blimps originally powered by diesel engines to pilot 100 civilians passengers as an alternative to turboprops (i.e. use the bits of the tech and the business model that's been around in some form since before Hindenberg)
I do think freight with electric propulsion (over distances battery-powered heavier than air aircraft can't compete even if they're also autonomous) is an easier environmentally friendly sell.
How delicate do you suspect this will be? Will a bored or malicious person be able to cause harm with a laser, an air rifle with pellet or BB, or gun? In the event of a puncture what is the failure mode like? Are you worried about birds? Are you concerned about banditry? I know the last might sound silly, but you'd only really need to go to our grand-parents and great-grand-parents to find some train robbers.
I could see Hollywood making use of this for filming in remote locations. Or other situations where you need to get something heavy and oddly shaped up or over a mountain. Relief work as well where the roads aren't operable. Wouldn't surprise me if this had applications in agriculture either, if anyone is still using planes might be nicer to have an autonomous blimp instead. Maybe stuff like introducing fish to alpine lakes, if you've got good control of vertical height you could probably design a safe exit system.
Most of that is supposing that getting the blimp somewhere to then be used is easy. How is transportation of those things when not in use?
Most of the time we'll be operating over areas without people, so if the airships do get shot at, they should be in an area where they can safely land and be recovered. The envelope will be tough enough to handle birds landing on it.
Regarding banditry, we hope that we are providing enough value to these communities that there will be peer pressure to prevent it.
It's relatively easy to transport them when not in use, either keeping them inflated and flying to the end location, or packing them up and assembling/inflating on location.
How does piloting work? Is the goal to make these things completely autonomous, or will there be a remote pilot at all times?
What's the business model? Can I buy one of your blimps, or are you going to become FedEx of the low skies?
Business model is something we're still figuring out. We have interest on both sides, but at first we'll definitely be operating themselves as the 'FedEx of the low skies.' Once the airship is certified (1-3 years after we build it), we can sell them.
If you build the infrastructure for the latter, one operator could likely manage multiple craft.
That said, what does either type of pilot do, if the vehicle misbehaves? What's the abort protocol, esp. over populated areas? Given a 600-lb payload, self-destruct is not an appealing way to exit.
Either way, seems like a cool project- I wish you luck!
Our goal isn't to maximize efficiency, but rather to build something that fits a market need and that we can deploy quickly. We actually think that the quest to build huge, efficient airships has lead past airship projects astray. There are several projects that have struggled to build 10 ton or 50 ton vehicles, but no past attempts to build something in this size class.
Lift to drag ratio describes the aerodynamic efficiency of the vehicle. Typical values, depending on flight conditions, are 4-5 for a helicopter, 8-13 for small planes, and 12-16 for our airships. (Large planes can get as high as 17, but a larger airship would be also be even more efficient). https://en.wikipedia.org/wiki/Lift-to-drag_ratio#Examples_of...
From the L/D ratio and the propulsion system efficiency you can calculate the total energy used, and divide that by the payload mass to get energy/kg/km.
Sure, I don't doubt a blimp is more efficient when the wind is blowing it to where it wants to go, but what happens when the wind doesn't cooperate?
EDIT: yes, the Wikipedia page states that L/D numbers are usually presented as graphs, because they are dependent on speed. This means that without knowing the wind speed the blimp is facing, especially given it travels at a lowly 35 MPH, you can't really calculate an L/D number. For the helicopter and airplane, the wind matters less as they are going much faster.
The figure of merit is maximum (speed * L/D), not L/D alone, and this is where the efficiency of airships fall apart. Due to their greater parasite drag compared to aircraft (necessitated by the lifting gas volume), airships generally can’t match the same per mile efficiency, and even if they can, not at the same von Karman efficiency.
L/D is a ratio of forces, and work = force times distance. Going faster at the same L/D decreases the travel time but increases the power used, for the same energy usage. Generally L/D depends on speed, but you can get an apples-to-apples comparison of the energy usage of two different aircraft by comparing only L/D.
I think we might be discussing two different figures of merit. The efficiency that I care about is "energy per lb of payload per mile." You haven't defined what you mean by "efficiency per mile," but from the reference to von Karman efficiency, I think you have a figure-of-merit in mind that includes speed - specific resistance or productivity or something like that.
But the economics of our vehicle and use case don't really change with speed. The middle mile legs that we're flying are not fully-utilized - the most important thing is to get the cost per flight down, so that we can dispatch the aircraft whenever there's a delivery ready.
What's the full-scale design going to look like? Will you continue to use the standard blimp shaped envelope? Or will you transition to a fat flying wing type design?
The full scale design is a fatter shape that approximates a flying wing. That's why the aircraft in the demo video is two lobes - the final version will have an aerodynamic cover between the lobes.
Having a bunch of them in the air in a constant cycle synced with the warehouse could compensate for the lost speed and could result in more distributed last mile centers. So package delivery is closer to ordering time, rather than sitting in queued bunches.
Say it takes 1000 m3 hydrogen to lift 1 m3 concrete. Image a pyramid shaped cone 4000 meter long with a bottom 4 m3. Now imagine a balloon with a diameter of 8000 meters.
Clearly much to big to build [for now]. We pump the hydrogen out of it. How much extra lift does that generate before the balloon implodes? What other materials can we use? Would it be possible to make it smaller than a km?
It was originally a napkin calculation from someone who did. The pun [if you like] is that a flying city is doable but one has to think much much bigger.
I’m sure your guys have done a ton of prior art, but I want to encourage you to make sure Amazon hasn’t patented this in any capacity yet.
A close friend that worked at Prime Air shared how years ago their division had a full time legal staff dedicated to patenting all engineers ideas. He had gotten several patents even being entry level FWIW. They had several concepts very similar to this and it wouldn’t surprise me if they were able to get that IP locked down. Like I said, I expect you all have checked and rechecked but thought it’s worth mentioning.
The search is very cheap and easy, and then you can decide if the risk is worth it.
1) You are now aware of the patent and continuing to willfully infringe
2) You're going to spend days obsessing on how your idea is different, or they didn't think of this or that, or this tiny edge case convincing yourself of non-infringement.
In both cases any patent litigation attorney worth their salt will absolutely murder you. The best part is so many broad patents exist that everything infringes on one.
I wish we would tie patent fees to the assets of a company, or person. Basically the wealthy boys would pay much more for a patent, and low income individuals woukd pay less, or nothing.
So a big company might think twice about an exponential government fee when they are on their 100 patent.
I think its hard to avoid the elephant in the room that blimp cargo has been a dream ever since the invention of the blimp, and if we put your technology claims to one side (because they are more an assistant, not an enabler) I'm not too sure what makes you any different ?
Look at what is perhaps your most recent competitor ... Flying Whales.
Going since 2012, burning through cash like there's no tomorrow ($30m Quebec, €225m France etc. etc.) , and so far maybe promising some sort of prototype by 2024 (already slipped from 2021).
What really makes you so different from prior dreamers and in particular your present competitors ?
Putting the French specific comment to one side because I don't have sufficient knowledge to comment, what I would say is that having "skin in the game" is no bad thing (it focuses your mind, and demonstrates your commitment to others).
That's one of the problems with the US funding scene.
On one side its great because people will throw obscene amounts of money at you if your marketing pitch sells well enough.
One the other hand, it does lead to the situation where your commitment to the business only goes so far because you've basically always got someone else picking up the tab.
Most founders aren’t taking a risk (they tend to come from wealthy families) they are executing a vision.
I’d bet the median return on startups is negative.
In theory, having that much skin in the game sounds like a pro with small cons. The reality is that is kills innovation, leaves the few brave enough who fail indebted and hampered for years, and makes the successful a fraction of the scale (and disproportionately founded by those already rich).
The profile of the person founding them would be different. You'd have istrionic risk takers, borderline gamblers carrying the company for the first say 5 years, then they'd be bought out and a more stable person would be brought in.
The VC model enables the economy at large to artificially skip the first step. It's artificial because Silicon Valley companies which basically start from zero and arrive in the S&P500 in 20 years or less, they have very low turnover in managment and executives.
It's like a Vietcong learning how to fly an F-35.
As long as the US side doesn't seek to brush over the failures by attempting to focus people's attention on the successes. And also as long as we have a fair definition of "success" .... an exit strategy that comprises being able to hype up a blatantly overvalued IPO is not "success", for example.
Let's face it there have been some quite spectacular failures in the US in recent years. As well as many more less spectacular that have burnt through millions of dollars in cash before calling it quits.
All that glitters is not gold. The US has a mentality of "throwing more money at the problem", but that does leave you in an environment where larger and larger burn rates become acceptable.
I'll reserve my comments about my own personal experience but I think the characterization of poor execution is a bit sweeping.
Where I think the French and American start up communities differ is that failure is not generally accepted in France, whereas in the US it's acknowledged and somewhat accepted to have a few "duds" on the resume.
Hopefully we've responded to practical constraints in other answers, but if not let me know questions.
Why still hundreds of thousands? Blimps are an incredibly simple, cheap technology, and drone hardware is completely commoditized at this point.