Electric Aircraft - The Thread

[mdrejhon 8]

12 minutes ago, nwt said:

Performance of today's Otters at the weights anticipated would be useful information.

On this point, I agree, more information from the industry will be welcome to execute better mathematics. 

Are there any pilots here that would be willing to help me run the mathematics on various types of battery packs

(Tesla scenario, LiFePo4 scenario, current 250kWh/kg scenario, future 300kWh/kg scenario, future 400 kWh/kg scenario)

I will get partial data from Magnix but because battery technology is an amazingly rapidly moving target (Thanks to Tesla lighting a fire under everybody who makes lithium batteries), we will have to do calculations separately on the hypothetical battery packs.

12 minutes ago, nwt said:

This is an inconsequential semantic difference that doesn't change anything about the discussion we've been having.

On this point, I disagree it is inconsequential because the mass difference will become much more major as 2020 progresses towards 2030, as explained in three previous posts --

In a "Designbattery for one skydive + 30min reserve" scenario (between-jump fast-recharging), the amount of battery for one skydive will become smaller/lighter, as an increasingly tinier percentage of a fully fueled load.  

Yes, it will still be heavier than the equivalent fuel for one skydive (partially fueled plane), but regardless, a full or near-full flight with full load of jumpers should be achievable, in this "design for one skydive" scenario.

The sufficiently powerful motor is already here, and there likely should be advocacy / trials even before the expected battery improvements that dramatically changes the mass savings.

This isn't nuclear fusion perpetual "won't happen for another 20 years" fantasyland stuff.

 

Edited November 12, 2020 by mdrejhon

[mdrejhon 8]

2 hours ago, olofscience said:

Back to the topic of electric aircraft then.

My main problem at the moment is MagniX hasn't released the weight or battery details for the electric caravan that flew earlier this year. If I had those numbers we'd be able to calculate many other  metrics such as climb performance, payload, etc.

I might be able to squeeze the numbers out of this sentence, but it's pretty annoying not having the actual numbers.

I'm not surprised.  Weight is a super-fast moving target.   

It's not a problem to have a light battery but you want a long-lasting battery, so you need a great cooling system.   A 750kW-output capable battery needs great longevity.  The technology for such lonevity has finally arrived, but the devil is in the details. 

The point being -- supercar electric cars are already racing at sustained 750kW on the racetrack using a battery lighter than the fuel load of a fully loaded Caravan!!!

The first megawatt supercars came out six years ago. (One megawatt = 1340 horsepower).  Since then, they've finally found ways to sustain-output more than half a megawatt for many minutes (>10 mins) -- long enough for a jump lift!

The "C" problem a solvable problem; the problem is making the battery longevity good through creative cooling systems and heat monitoring systems  (Tesla has succeeded in durable 1C-3C batteries on the racetrack, but we need twice that for sustained Magnix operations).  

The Tesla battery has more advanced temperature monitoring and charge monitoring than the Airbus A380 battery does. 

It's quite dismal how Airbus designed that first version compared to Tesla did to their batteries -- several thousand temperature sensors & several thousand charge controllers -- in those older 18650-based Teslas. The safety industry standard is 1 temperature sensor per cylindrical battery AND 1 temperature sensor per battery pack, as a double-redundancy, and always leaving a healthy SoC at the top/bottom ends of the range.  And a Tesla (original) consists of thousands of 18650 laptop battery cells -- so that's thousands of temperature sensors.  The Tesla charge controller quickly isolates bad cells in a best-effort.  Also, there were some major charge management mistakes done in the A380 batteries.

Now Tesla's moved to 2170's and soon 4680's.  Other battery manufacturers are finally copying the incredible battery monitoring breakthroughs that Tesla paved, and will be important for ePlane safety.

It's amazing how beat-up Tesla batteries are; no wonder they catch fires (but less often than gas fires of gas cars). 

Airplanes are treated more gently than cars, not having to deal with shocks of potholes and curb-bumps as often -- so just like electric cars catching fire less often than gasoline cars -- but, yes, they need to design electric airplanes to catch fire less often than avgas planes -- for trust of the public.

Given length of aviation approvals, it will take time, and one will have to migrate to safer lithium chemistries, and build enough safety margin.  However, experimental trials should proceed sooner than later in preparation for the incredible economics of an electric jump planes so advocacy should begin sooner than later. 

We don't have to use Tesla batteries but we have to compliment the brilliance Tesla brought to the table for other lithium battery manufacturers.

You can already cram the weight of 2 Tesla batteries including cooling weight equivalent (~200 kWh for 1.5 megawatt surge output, easily 750 kilowatt sustained) for the same weight as a fully tanked Caravan load. 

Obviously, we would use something else other than Tesla, as not as much cooling weight is needed due to cold altitudes -- but it all mathematically checks out already.   Supercar batteries are often lighter for higher power output, since their longevity doesn't matter as much as a Tesla, I simply use Tesla as a gold standard since they're pretty long-longevity batteries.

Now, power-wise, if we assume full max throttle 750kW for 8 minutes for a hurry-to-altitude, that will consume ~50% of a 200 kWh battery that fits within the weight of fully fueled caravan assuming we go by the latest weight data (lighter 4680 based architecture, under 2200lbs for 200kWh including water cooling -- unlike 2760lbs for the 18650 based architecture at 2x1380lbs for a Tesla P100D battery doubled-up).  That leaves 100kWh for emergency reserve. 

This is only napkin math at this stage -- but it shows this is all really within a stone's throw already, with even the lower capacities.

That's today's technology, only 200kWh per kilogram, including cooling. 

We haven't begun talking about tomorrow's 300-400kWh per kilogram implementations, yet.

Consequently, I'm pretty confident a long-lasting jump battery is a very solvable jump lift problem before 2030 -- the writing is on the wall.

Yes, the battery weight is a fast moving target due to many variables (chemistry, cooling, etc)

Edited November 12, 2020 by mdrejhon

[mdrejhon 8]

1 hour ago, RobertMBlevins said:

That said, my only thought is that these planes might need to have a reserve battery system installed, not a large one, but enough to provide just enough power to get down, in case for some reason the main batteries fail in-flight. A few extra minutes, say. 

Lithium batteries are already highly modular and this is probably a shoo-in feature.  Left wing battery, right wing battery.  (Ooops, forgive the political puns -- I know this ain't Speaker's Corner). 

Keep in mind that a Tesla car has more than 1,000 separate laptop batteries in it, and a charge controll activates/deactivates each and every one of them based on imminent failures.

An airplane will need more partitioning/firewalling of this, but a large lithium battery is already automatically modular -- they just will have to tree-subdivide it appropriate for aviation:

- Cell level & individual cell sensors 
- Pack level & individual pack sensors
- Cluster level (cluster of packs) & their emergency cutoff switches

It's only a simple architectural detail for modern large lithium batteries nowadays, as a simple subdivision matter.  Three clusters is probably ideal -- left wing, right wing, and underfloor.  The underfloor could be a reserve battery.

Also, the Magnix motor can theoretically be used as generator for regen descents, recovering between 5%-10% of the energy.  The electric trainer already can capture up to 13%, although student pilots find it scary to dive that deep.  But jump pilots don't mind doing steep descenbts, so wouldn't mind a bit of regen (if the location's electricity rates are expensive).

1 hour ago, RobertMBlevins said:

I was surprised to learn how FAR they've come with these electric aircraft. The recent article about the maiden flight of the Grand Caravan was pretty solid.

The kicker:

The Magni500 consumed about $6 worth of electricity during the Grand Caravan’s 30min test flight, he adds.

At those prices, we could essentially forget about regen.  Even a 6-skydiver Caravan load can already be profitable -- since motor longevity is stellar, assuming good battery longevity.   But by the end of this decade, I don't think we even need any capacity reductions.

Edited November 12, 2020 by mdrejhon

[nwt 131]

53 minutes ago, mdrejhon said:

On this point, I disagree it is inconsequential because the mass difference will become much more major as 2020 progresses towards 2030, as explained in three previous posts --

Right, but we were already both in agreement that the batteries will continue to lighten over time, which will bring us closer to feasibility. I don't think a single person in this thread would refute that.

What I'm refuting is the significance you ascribe to this arbitrary threshold of batteries required for jump ops being equal to or less than the weight of a full load of fuel. It's equally irrelevant whether you say "equal to" or "less than" or "equal to or less than".

A relevant statement would be "batteries required for a single jump flight weigh X, and if you replace the fuel in a typically configured jump plane with X, turn time will change from Y to Z."

For all we know, from what has been presented in this thread so far, an aircraft with battery weight equal to or less than a full fuel load may not even be within max takeoff weight. So why is this information useful?

1 hour ago, mdrejhon said:

In a "Designbattery for one skydive + 30min reserve" scenario (between-jump fast-recharging), the amount of battery for one skydive will become smaller/lighter, as an increasingly tinier percentage of a fully fueled load.

Again, we all agree batteries will become smaller and lighter over time, improving feasibility. The fact that this will also mean they will become a smaller percentage of a full fuel load is tangential at best. They will also become a smaller percentage of any other thing that has mass.

 

1 hour ago, mdrejhon said:

Yes, it will still be heavier than the equivalent fuel for one skydive (partially fueled plane), but regardless, a full or near-full flight with full load of jumpers should be achievable, in this "design for one skydive" scenario.

You continue to hand-wave away the significance of weight on climb performance. This is so important that I think it is irresponsible to keep making such statements without any support.

 

48 minutes ago, RobertMBlevins said:

That said, my only thought is that these planes might need to have a reserve battery system installed, not a large one, but enough to provide just enough power to get down, in case for some reason the main batteries fail in-flight.

I think it goes without saying that there needs to be redundancy, but I imagine it would be similar to redundancy in fuel tanks--you have multiple of them but they all may see use in normal ops. No real need to have a completely separate module only for use in emergencies.

 

54 minutes ago, RobertMBlevins said:

just enough power to get down

Not to mention "getting down" won't really be the problem if your batteries go out lol

 

56 minutes ago, RobertMBlevins said:

Getting imaginative on cutting the gross weight of the aircraft might also help extend range. This is WAY long time ago...but the Germans did the same thing with their old zeppelins. Every single component on board was made with weight of that component in mind. Right down to almost the tiniest detail. 

I don't think this is any different from how aircraft are currently designed, anyway. Weight already matters a lot. If any new tech comes along to make airframes even lighter, electric airplanes will still have to compete with lighter-weight traditionally-powered airplanes using this tech. It's at least mostly orthogonal.

It's one thing that always bothered me about how hybrid cars are marketed: I want to see their fuel economy compared to the same vehicle but traditionally-powered.

57 minutes ago, mdrejhon said:

they need to design electric airplanes to catch fire less often than avgas planes -- for trust of the public.

This condition is necessary but insufficient. When fuel is on fire in an aircraft, there is almost always a firewall between the fire and the fuel tank. You can usually turn off fuel flow and extinguish the fire. A battery fire would be analogous to a fire in the fuel tank, a much more dangerous and rare scenario.

[mdrejhon 8]

4 minutes ago, nwt said:

This condition is necessary but insufficient. When fuel is on fire in an aircraft, there is almost always a firewall between the fire and the fuel tank. You can usually turn off fuel flow and extinguish the fire. A battery fire would be analogous to a fire in the fuel tank, a much more dangerous and rare scenario.

This is a good point. 

That said, keep in mind that battery fires start more slowly than fuel fires -- enough time for people to exit the car many minutes before exiting the airplane.   Regardless, the design of a battery fire should be sufficient to allow the plane to quickly glide to landing before the battery fire consumes the plane.

They will likely avoid lithium polymer (lipo) for an electric airplane because they can be more explosive.  Most other lithium batteries (the Li-On ones used by Tesla) slowly heat up and smoulder gradually, so it would take more than 10-20 minutes of smoldering before fires.  

Design of a battery fire safety will be important.
[replying to the other points out of sequence, keep tuned]

[nwt 131]

4 minutes ago, mdrejhon said:

They will likely avoid lithium polymer (lipo) for an electric airplane because they can be more explosive.  Most other lithium batteries (the Li-On ones used by Tesla) slowly heat up and smoulder gradually, so it would take more than 10-20 minutes of smoldering before fires.  

That doesn't sound so bad

[mdrejhon 8]

24 minutes ago, nwt said:

You continue to hand-wave away the significance of weight on climb performance. This is so important that I think it is irresponsible to keep making such statements without any support.

That's fair.  Let's get more data over the coming months/year to back up this better.   I agree that we need a lot more data.

The other point I want to point out is it's not exactly that irresponsible to say this, is that at $6 per load, one can scale back the skydvers quite a bit to restore a lot of climb performance and then still make a profit.  Still hugely more proftiable than a C182...

A 9-person slightly-slower climb (11-minute instead of 8-minute) Electric Caravan could perhaps work just fine on dropzone accounting if the fuel-up is only a few dollars (whether $6 or $30). Yes, slower turnover than an avgas carvan that way -- but sooo much cheaper operationally when also considering the lower maintenance. 

The goldilocks spot is simply calculated by the cost vs time, which might turn out to be a number similar to that -- less than max load, but quite usable.  Something that climbs fast enough with those Early Version 1 Jump Batteries. 

Which can be later be replaced in five years (in subsequent ePlanes, and in battery replacements in a decade) with a lighter Version 2 to increase jump capacity later.  The improved profits can be saved towards such inevitabilities, in theory.

By 2030 it looks highly probable that no skydiver-load-scaling-back is necessary.

Edited November 12, 2020 by mdrejhon

[nwt 131]

Just now, mdrejhon said:

That's fair.  Let's get more data over the coming months/year to back up this better.   I agree that we need a lot more data.

The other point I want to point out is it's not exactly that irresponsible to say this, is that at $6 per load, one can scale back the skydvers quite a bit to restore climb performance and then still make a profit.   A 9-person slightly-slower (11-minute instead of 8-minute) climbing Electric Caravan works just fine on dropzone accounting if the fuel-up is only a few dollars (whether $6 or $30).

By 2030 no skydiver-load-scaling-back should be necessary.

I agree that the apparent savings is quite significant. However, I don't think it's as simple as you present it: Yes, maybe you can take fewer jumpers up and still profit on the load. However, that ignores other factors I have already mentioned and simply being revenue-positive on the load is not enough. For example: how many additional planes would a DZ need to buy, house, and maintain in order to meet their current jump capacity requirements?

Maybe with such dramatic savings on fuel it will make sense to work around these factors. However, that doesn't go without saying--you need to show it.

[mdrejhon 8]

10 minutes ago, nwt said:

I agree that the apparent savings is quite significant. However, I don't think it's as simple as you present it: Yes, maybe you can take fewer jumpers up and still profit on the load. However, that ignores other factors I have already mentioned and simply being revenue-positive on the load is not enough. For example: how many additional planes would a DZ need to buy, house, and maintain in order to meet their current jump capacity requirements?

Maybe with such dramatic savings on fuel it will make sense to work around these factors. However, that doesn't go without saying--you need to show it.

That's going to be a big purpose of advocacy. 

We need more accountants, pilots, mechanics, etc, working on this -- testing out scenarios, crunching the numbers, visiting Magnix, monitoring the news, etc. 

Then that way, I can shut up, and let everybody's numbers speak for themselves.

P.S. (edit) One thing to keep in mind is that jump capacity scaleback, if any, is likely temporarily minor.  For a "trial" it is probably only 1 electric jump plane at a major dropzone such as Perris or Deland.  Preferably 2025 intead of 2030.  By the time a whole jump plane fleet goes electric, it would be long into the 2040s or beyond where there is no longer notable capacity compromises.

Edited November 12, 2020 by mdrejhon

[olofscience 420]

1 hour ago, mdrejhon said:

Since then, they've finally found ways to sustain-output more than half a megawatt for many minutes (>10 mins) -- long enough for a jump lift!

I'm skeptical about this - with 750kW output, the P100D's battery would only last a theoretical max of 8 minutes, probably less. Do you have a source?

[gowlerk 1,901]

29 minutes ago, mdrejhon said:

Regardless, the design of a battery fire should be sufficient to allow the plane to quickly glide to landing before the battery fire consumes the plane.

A battery fire in a wing would almost certainly result in a structural failure making glide, or any other controlled flight impossible in a very short time.

[mdrejhon 8]

On 11/12/2020 at 6:30 PM, gowlerk said:

A battery fire in a wing would almost certainly result in a structural failure making glide, or any other controlled flight impossible in a very short time.

The million dollar question is what “very short time” means.

There are some lithium batteries, where upon short circuiting it, it takes a long time before it becomes a conflagration — the chemistry, the construction, etc.  Whereas, others (some unprotected lipos) will just simply expand / explode almost right away.  

If you have any experience watching some of the battery tests of the more slow-cascading lithium battery designs, you’d realize that it is possible to design it in a way that it would take long enough before temperatures/expansions raised to structure-failure levels.   There are designs that give enough time for landing before structural failure, at some cost of battery density. 

Lithium battery self-destruction rapidity varies all over the map.  Sometimes a Tesla screen have displayed a warning more than half an hour before the first flames occurs; so there is some precedent to the ability to warn in advance of a potential impending battery fire as well; those are often reported more often than fuel fires which also happens to airplanes as well; (also consider many vehicle lithium batteries — not just Tesla — apparently has had way, way more active monitoring of all the battery sensors than the lithium battery in the first flawed A380 design).

But yes, this will probably be a major item of FAA approval.  The fire incidence must be acceptably low with a wide margin of safety for the public to accept electric aviation — both from a frequency perspective and also a warn-in-advance perspective.

Edited November 17, 2020 by mdrejhon

[olofscience 420]

1 hour ago, mdrejhon said:

those are often reported more often than fuel fires which also happens to airplanes as well; (also consider many vehicle lithium batteries — not just Tesla — apparently has had way, way more active monitoring of all the battery sensors than the lithium battery in the first flawed A380 design).

The issue was with the 787, not the A380.

You can't just extrapolate this time from Tesla cars to aircraft - Teslas are made of steel, which withstands much higher temperatures than the aluminium used in aircraft.

Since TWA 800 aircraft have also had inerting systems and fire suppression systems, but these won't be as effective for a thermal runaway. The 787 issue was eventually worked around by putting a heavy steel container around the battery and a venting system to vent gases in case of thermal runaway, but the same solution would kill an electric aircraft's viability.

[lyosha 46]

Not sure if this has been mentioned here or not, but there have been recent developments in a hydrogen fuel cell plane.  In my opinion this shows more promise than batteries.

 

https://www.aopa.org/news-and-media/all-news/2020/september/30/hydrogen-powered-piper-circuits-pattern

[coticj 0]

@mdrejhon You keep mentioning massive fuel cost savings. A typical caravan has (the ones we operated) had a fuel burn of 60-70L of Jet A1 per load. Current fuel price my country is 0.41€ per L. This comes to 1.85€ of fuel per person. Sure, the electricity is 4 or more times cheaper. But saving 1.4€ per ticket in a ticket that costs 30€ doesn't seem much.

And for the fast charging that you are mentioning. The power consumption would be the same as for a wind tunnel, meaning huge infrastructure investments and massive electricity bills (I don't know if it is the same in US, but in my country the power company charges you for the power you use and the "available power" - for example household 7kw supply costs 5€ monthly and 21kw costs 25€).

I don't think conversions are the way to go. Most (non Tesla) car manufacturers tried with converting existing models, but all of them found that it is just better to start from scratch.

Lastly you keep mentioning the batteries that would be needed for a load + 30min reserve. What kind of battery do you think this would need?

 

 

[nwt 131]

10 hours ago, coticj said:

@mdrejhon You keep mentioning massive fuel cost savings. A typical caravan has (the ones we operated) had a fuel burn of 60-70L of Jet A1 per load. Current fuel price my country is 0.41€ per L. This comes to 1.85€ of fuel per person. Sure, the electricity is 4 or more times cheaper. But saving 1.4€ per ticket in a ticket that costs 30€ doesn't seem much.

This is the kind of math I've been wanting to see and I've been pretty skeptical that it would come out favorable. The cost savings in fuel looks much more compelling without this context, and that seems to be the main driver of enthusiasm.

Quote

And for the fast charging that you are mentioning. The power consumption would be the same as for a wind tunnel, meaning huge infrastructure investments and massive electricity bills (I don't know if it is the same in US, but in my country the power company charges you for the power you use and the "available power" - for example household 7kw supply costs 5€ monthly and 21kw costs 25€).

Another really important point that is obvious now but hadn't occurred to me.

Edited November 18, 2020 by nwt

[nwt 131]

Ironically, if aircraft electrification does take over, my idea for building cheap tunnels with unwanted turbines may actually become feasible...

[olofscience 420]

9 minutes ago, nwt said:

This is the kind of math I've been wanting to see and I've been pretty skeptical that it would come out favorable. The cost savings in fuel looks much more compelling without this context, and that seems to be the main driver of enthusiasm.

Some of the savings will be cheaper "fuel",  but the main savings will be in the MRO. 2 or 3 orders of magnitude fewer moving parts adds up to a lot of maintenance savings, as well as normal operating temperatures being quite a bit lower (not counting thermal runaway). Fewer parts to TSO will also save a bit on certification (which the manufacturers will pass on to users anyway).

For turbines, capital cost could also be a major source of savings - superalloys in the turbine section aren't cheap to make, electric motors would be a lot cheaper. However this benefit is wiped out by expensive lithium batteries, hence some attention on overcoming the current disadvantages of LFP.

[sfzombie13 307]

14 hours ago, olofscience said:

 Fewer parts to TSO will also save a bit on certification (which the manufacturers will pass on to users anyway).

 

you haven't seen how companies operate very often, have you?  i doubt anyone will pass on any (probably minuscule anyway) savings to the customer when the other option is more profit. 

[coticj 0]

I don't think that fewer parts will make certification cheaper. The highly flammable batteries will probably be much more expensive to certify...

[olofscience 420]

29 minutes ago, sfzombie13 said:

you haven't seen how companies operate very often, have you?  i doubt anyone will pass on any (probably minuscule anyway) savings to the customer when the other option is more profit. 

If their margins can afford it, and it means taking more market share from the incumbent turbine manufacturers then some will do it.

I actually work for a fairly successful company who uses this exact strategy.

[olofscience 420]

1 minute ago, coticj said:

I don't think that fewer parts will make certification cheaper. The highly flammable batteries will probably be much more expensive to certify...

Hence the search for less flammable battery chemistries, and the LFP research NASA is doing.

Certification will definitely be cheaper with fewer parts - each part needs its own pile of documentation. Salaries of people to make those documents isn't cheap.

[gowlerk 1,901]

26 minutes ago, olofscience said:

Hence the search for less flammable battery chemistries, and the LFP research NASA is doing.

It is unlikely that less flammable chemistry will be the answer in any time frame that we are discussing. The answer will have to be better quality control and a more robust design and manufacturing process that is able to achieve certification for aviation use. Since there are currently aircraft in testing I would assume someone has addressed the problem already. But there must be a cost and weight penalty involved. 

[riggerrob 558]

May I suggest that a wise DZO will invest in a half-dozen sets of batteries and re-charge them mid-week? 

By re-charging batteries during off-peak hours, the DZO can save money on electric grid rates.

With solar cells on the hangar roof, the DZO might be able to save even more money .. or sell surplus electricity to neighbours.

Finally, with those half-dozen sets of batteries fully-charged by Saturday morning, they can swap battereis and fly until noon before worrying about re-charging.

Also the time spent swapping batteries might be needed to allow packers to catch up.

I used to work at a DZ that had too much lift capacity with a King Air. After three or four loads, we would run out of tandem rigs. So we stopped jumping for a half-hour while packers caught up, TIs trained and dressed the next batch of students and the pilot refuelled the airplane. The other reason we flew three or four loads back-to-back is that it costs hundreds of dollars to shut down a turbine engine .... wait for it to cool ... then re-start. Every hot-cold cycle costs hundreds of dollars worth of life to a turbine engine. The number of hot-cold cycles becomes increasingly expensive as you approach the end of the over-haul cycle since turbines were originally designed to take-off-cruise-land only two or three times per day ... nothing like 3 to 5 cycles up-down per hour.

Edited December 2, 2020 by riggerrob
add a sentence

[riggerrob 558]

Within our lifetimes the petroleum industry is going to stop refining 100LL aviation gasoline. Then we need an alternative to keep flying.

Consider the Romanian Air Force's dilemma back in the early 1990s. After the fall of communism, Romanian refineries stopped making 100LL and it was prohibitively expensive to import 100LL gasoline from Greece, so the RAF grounded their fleet of I.A.R. 823 trainers. Those trainers were powered by the same Lycoming IO-540 engines as Cessna 182 and 206. Those airplanes sat idle for a decade or so, then were bought up an American dealer and sold on to the warbird crowd, who were wearing out their similar-sized Beechcraft Mentors.

Debate all you want, but DZOs are going to need new a fuel source within the next 20 years.