Edmunds.com Aptera Review Video!

[Prestadude]

Good video, some new stuff I did not know about such as the fact that the side crash reinforcement beam is carbon fiber or that the load limit for the cargo area is 250 pounds.

r_JV6QUIu5s&eurl
Link (http://www.youtube.com/watch?v=18eyA2CK_JU)

[futura]

Nice. Thanks Prestadude.
It seems the steering wheel is at a better angle in this vid.

[Eyelawdoc]

Nice video. Does a good job of showing the interior.

[Steve Martin]

Nice video.

[Apt3448]

Cool, yet another video! Didn't know about the weight limit (?) for the cargo. I like the acceleration shot at around 50 sec. This version of the 2e has the black side mirrors we saw somewhere (too much info!!) recently.I noticed that the car doesn't drag the rear wheel cover now - unlike at the track and road video. It is very low to the ground, but doesn't drag. Also this particular 2e doesn't have the black section on the rear wheel cover. The rear portion of the cover seems to slope up much more. I assume this is part of the reduced side wind plan. Nice to see several Apteras in the factory, and nice to see so much variation in the models. Steering wheel seems indeed also better positioned.

[Captain_Altair]

2 Comments:

1) The rear wheel does not appear to be on center line (see video 1:10). Hope they can move it a couple of inches to the left.

2) The range circles (see video 1:35) for one way or round trip should not be circles (unless you are at the Bonneville Salt Flats with no wind). The display should take into account the terrain(GPS maps), wind and temp(online weather reports), possible high speed routes (freeways) in all directions and calculate a more accurate range display.

[KarenRei]

2 Comments:

1) The rear wheel does not appear to be on center line (see video 1:10). Hope they can move it a couple of inches to the left.

I don't see that.

2) The range circles (see video 1:35) for one way or round trip should not be circles (unless you are at the Bonneville Salt Flats with no wind). The display should take into account the terrain(GPS maps), wind and temp(online weather reports), possible high speed routes (freeways) in all directions and calculate a more accurate range display.

Well, if Aptera wants a Google Maps-integrated range calculator that does take into account all of that, hopefully they know where to turn. ;)

[Apt3448]

[QUOTE=Captain_Altair]2 Comments:

1) The rear wheel does not appear to be on center line (see video 1:10). Hope they can move it a couple of inches to the left.

At exactly 1:10, for about half a second, there is a shot from the rear and the rear tire seems not centered in the wheel cover. Good catch! No idea what it means: wheel cover not yet tight in a vehicle being assembled? Never seen this before. One imagines the rear tire to be centered relative to the front tires.

[KarenRei]

Hmm, perhaps it's just that this monitor has poor resolution of dark colors, since I can hardly see the rear tire at all in contrast to the background.

[randyd]

2 Comments:

1) The rear wheel does not appear to be on center line (see video 1:10). Hope they can move it a couple of inches to the left.

No one noticed that before? I noticed it in one of the photos Josh took the night before TED. (http://www.apteraforum.com/showpost.php?p=26018&postcount=4) What's the big deal? It's reduces frontal area to put the suspension on only one side of the wheel. Symmetry is overrated. :)

[Captain_Altair]

Just had a thought - maybe they are still using the old rear wheel drive swingarm and matching rear fairing on this test mule. They might have just removed the drive belt and motor when they switched over to front wheel drive and just haven't gotten around to updating the rear suspension yet.

[jstdadd]

If in fact the door beam is carbon fiber, that doesn't impress me too much - the failure mode of carbon fiber under shock is 'catastrophic', in that if/when the beam fails it is a complete failure; with metal, you get deformation in a continuing fashion; eventual failure may occur, but a large amount of shock can be absorbed through deformation without complete structure failure. When you get a catastrophic failure in a composite beam, no further protection is available (I envision jagged shards of the beam protruding into the passenger cabin).

I hope that makes sense. I prefer some deformation and continued protection to complete failure. The stress of the impact has to go somewhere, I just don't want complete failure of the side member because now that Hummer that hit me is coming into my lap. I'd like to think that the Hummer hits me, I bounce off, he hits me again with decreasing energy, etc. The passenger compartment might be becoming marginally smaller with each impact, but some protection remains.

I realize it is a tradeoff of weight vs. structural strength, with structural strength by weight being superior for composites, but I doubt you will find vehicle manufacturers using composites for door beams - I could be wrong...

[Captain_Altair]

Karen,
Just did a quick scan of your range calculator. Nice! I hope they incorporate something similar.

Scenario: You are in the mountains with a low state of charge facing a steep climb full of switchbacks that will completely drain the battery in 20 miles. Your hotel (and charging station) for the night are 25 miles up that road, but it is only 15 miles as the crow flies (straight line).

Question: The nav display shown in the videos apparently would place your destination inside the one-way range circle (20 mile radius) -- setting you up for a REAL problem. Is your range calculator capable of placing a big red "X" 20 miles up that road at the point where you would grind to a halt, 5 miles short of your destination? As well as 45 miles DOWN the mountain and into the flats where you would run out of charge in THAT direction?

[NmGfan]

Just had a thought - maybe they are still using the old rear wheel drive swingarm and matching rear fairing on this test mule. They might have just removed the drive belt and motor when they switched over to front wheel drive and just haven't gotten around to updating the rear suspension yet.

You could be right on the belt and pulley removal. There will still need to be a single sided swing arm though, so the asymmetry between the body skirt and rear wheel might remain.
:happy0025:

[gkra]

If in fact the door beam is carbon fiber, that doesn't impress me too much - the failure mode of carbon fiber under shock is 'catastrophic', in that if/when the beam fails it is a complete failure; with metal, you get deformation in a continuing fashion; eventual failure may occur, but a large amount of shock can be absorbed through deformation without complete structure failure.

To my knowledge, based on what's come out of Aptera in the past, is that there's a full steel safety cage, including steel side-impact beams in the doors, underneath all that composite bodywork.

[jstdadd]

To my knowledge, based on what's come out of Aptera in the past, is that there's a full steel safety cage, including steel side-impact beams in the doors, underneath all that composite bodywork.

The Edmunds video has the speaker holding a composite door 'beam' and talking about it.

[slk23]

When you get a catastrophic failure in a composite beam, no further protection is available (I envision jagged shards of the beam protruding into the passenger cabin)

Perhaps the door has both carbon fiber and aramid (e.g. Kevlar) components.

.

[jstdadd]

Perhaps the door has both carbon fiber and aramid (e.g. Kevlar) components.

.

Multilayer Kevlar might stop a bullet (useful in certain neighborhoods, certainly) but it fails catastrophically when embedded in resin. Bullet-proof vests use loose Kevlar mats, with the run of the weave at angles between layers; you still get bruises (and maybe broken ribs) but usually a handgun bullet will not penetrate. For military style vests, they use hard (usually ceramic) materials as well.

[Telchar]

Perhaps the carbon fiber bar is designed so that if it gets hit hard enough to fail, you're already dead anyhow. That would make the issue of carbon vs. metal moot.

[jstdadd]

:rolleye0003: Yes, if that could be shown to be true, then indeed it would be a moot issue.

[Matthijs]

To my knowledge, based on what's come out of Aptera in the past, is that there's a full steel safety cage, including steel side-impact beams in the doors, underneath all that composite bodywork.
This is from the Aptera website:

Composite body structures provide an impact-resistant exterior that is lighter than steel but three times as strong. Front crash zones incorporate race car technology that absorbs and deflects energy to keep the cabin from harm. Integrated high-strength aluminum door beams transfer crash loads into the body and away from the cabin. And a complement of driver and passenger air bags is incorporated to ensure that each passenger is always secure.

Edit: Here it states on the Aptera website:

Passenger Safety Cell/Cabin

We broke the rules again by creating a super-strong composite and aluminum safety cell to protect you and your passenger. We then added airbags, aluminum door reinforcements and more just because we like overdoing it on safety.


So maybe this is an update in the Aptera design?

[jstdadd]

What do you think is an update? The aluminum or the composite door rail? I believe that the design is tubular steel or aluminum with composite members somewhere. I am just musing on the Edmunds guy holding a tubular composite piece and stating that it was a door rail. If that is the door rail right next to the driver or passenger's elbow it COULD (I am NOT saying that it DOES) have a catastrophic failure mode that could make it look like a broken tree branch with sharp shards of material at the break.

[evmavin]

Not to mention something so rigid does not absorb energy like steel and would most likely rip out the attachment points which are the weakest link. if the aptera shell is so strong how is the front a crumple zone if all the mechanics roll under yet the under shell is stiff. A stiff shell would not seem to "crumple" properly and absorb energy. Yes, you can hit it with a hammer but does that make it safer in a rear end collision or are you decapitated by a stiff plane of composite sheering off from the rear?

[Telchar]

The unfinished Aptera shells you see don't have a nose. That is the stiff, strong part. As I understand it, the front crumple zone is the nose itself, which gets smushed under the main shell. That is the crumpling action, which should leave the main shell unharmed, so they can make the main shell as stiff as they want. I'm not sure how it's supposed to work for the rear zone.

[KarenRei]

Indeed, you can make composites weaker through all sorts of means -- even just sloppy/lazy production ;) You can even choose different strengths along different axes by how the fabric is aligned.

[jstdadd]

Indeed, you can make composites weaker through all sorts of means -- even just sloppy/lazy production ;) You can even choose different strengths along different axes by how the fabric is aligned.

I could conceive that they could put composite 'tubes inside tubes' using minimal adhesive at particular places. It acts rigid in one direction, but if compressed along its long axis it could 'fail' in a predictable manner and provide a kind of 'shock-absorber' effect. For the front end, sacrificial weak points would let engineers design the failure modes of each individual part from impacts at various angles with velocity vectors at differing angles. You could do a heck of a lot of 'what-if?' analysis using finite-element-analysis and kind of integrate the compression of the front with regard to the force absorbed.

I just would kind of like to know what happens to a large composite door rail when it gets hit directly in the middle of its mounting points and fails completely. Not that aluminum or steel is necessarily superior, but I can imagine steel 'bending before it breaks' and absorbing energy along the way.

On the other hand (is that three hands?) you look at the layered composite 'feet' on artificial legs that have great flexure, and you could absorb huge forces by using such structures kind of like leaf-springs in the front end. You might fail the 'sandwich' but in a much less catastrophic way. You might be able to 'dial-in' the deceleration vectors with a controlled failure in the structure.

This article talks about composite leaf springs: http://tinyurl.com/bhvsdo

[johnvall]

Noticed, in the video, 100 mile charge, before it was 120.

[KarenRei]

Noticed, in the video, 100 mile charge, before it was 120.

100 miles = 2 passengers, climate control, normal freeway driving
120 miles = 1 passenger, no climate control, steady 55mph.

[johnvall]

120 miles= 1 passenger, no climate control, steady 55 mph. In another words get on the freeway at 2 am. Price keep going up, millage keep going down.

[jstdadd]

Johnvall: Would you prefer that they listed the maximum mileage at 35mph? What's your point?

[johnvall]

I rather see posted a driven range of x amount of miles under normal driving conditions. To achieve 55 steady, it is near impossible and to certain extend misleading.

[SEGsby]

*blinks* Um, it has Cruise Control... :sad0126:


I rather see posted a driven range of x amount of miles under normal driving conditions. To achieve 55 steady, it is near impossible and to certain extend misleading.

[KarenRei]

120 miles= 1 passenger, no climate control, steady 55 mph. In another words get on the freeway at 2 am. Price keep going up, millage keep going down.

But that's what was stated from the beginning... they pointed out that that range was at 55 mph.

[jhm614]

*blinks* Um, it has Cruise Control... :sad0126:

Even with cruise control, traffic conditions can make it tough to maintain a steady 55 mph for an extended (2+ hour) time.

j.

[SEGsby]

Well, in LA, you'd be doing waaaay less at rushhour, so your range would be much higher? ;)

[NmGfan]

Yeah, range would be higher. There is no idling in an EV, so creeping along in commute traffic will increase range so long as there isn't severe accelerations and decelerations.

:happy0025:

[slk23]

Multilayer Kevlar might stop a bullet (useful in certain neighborhoods, certainly) but it fails catastrophically when embedded in resin. Bullet-proof vests use loose Kevlar mats, with the run of the weave at angles between layers; you still get bruises (and maybe broken ribs) but usually a handgun bullet will not penetrate. For military style vests, they use hard (usually ceramic) materials as well.

I've seen aircraft cowlings made with Kevlar layers (combined with fiberglass I think) and it resulted in a tough, somewhat flexible structure. I don't know how that would handle a large impact.
.

[n_dawg]

Well, if Aptera wants a Google Maps-integrated range calculator that does take into account all of that, hopefully they know where to turn. ;)

Unfortunately, neither you nor anyone else not-at-Google has the resources to make that work without bringing Google Maps' servers crashing to a halt. A simulator and a range calculator are very different. For the latter, fundamentally you need to cache all the map data to do it in a reasonable amount of time. Google doesn't like you doing that. :-\

Actually making a range plot would involve traversing the entire reachable graph from that location, storing some simulation data at each location. At the least you'd want to record the (simulation) time, state of charge, and total distance to that point.

[KarenRei]

Unfortunately, neither you nor anyone else not-at-Google has the resources to make that work without bringing Google Maps' servers crashing to a halt. A simulator and a range calculator are very different. For the latter, fundamentally you need to cache all the map data to do it in a reasonable amount of time. Google doesn't like you doing that. :-\

Not true, for several reasons.

1) I disagree about there being a fundamental distinction between a "simulator" and a "calculator".
2) The map data isn't invented by Google. You can get it for free from the USGS.
3) If your concern is range plots, there are hundreds of algorithms you could use that would give reasonable performance and accuracy. The entire space of routes is NP-complete, but you're only looking for "optimal" or "likely" routes in each direction, not *every* route possible. A naive implementation would be to make your shape be a regular polygon, with evenly spaced vertices, and then see how close to those vertices you can get on a charge. More sophisticated versions would involve loading road segments up with precomputed energy consumption figures under varying conditions and doing a lowest-energy tree search with interpolation between those cached figures.

[n_dawg]

Not true, for several reasons.

1) I disagree about there being a fundamental distinction between a "simulator" and a "calculator".

No, fundamentally no. However, without re-implementing Google Maps and just using the interface provided, a range plot would be incredibly inefficient and would end up doing a huge amount of duplicate queries.


2) The map data isn't invented by Google. You can get it for free from the USGS.

True. I mentioned OpenStreetMaps earlier as another great source of data (they scraped the USGS data as well).


3) If your concern is range plots, there are hundreds of algorithms you could use that would give reasonable performance and accuracy. The entire space of routes is NP-complete, but you're only looking for "optimal" or "likely" routes in each direction, not *every* route possible.

Right. Traveling salesman is NPC, but who cares? You only need to do shortest-path.

EDIT: While technically what you stated is not "traveling salesman" per-se, it's the closest optimization problem I could think of to "traverse every possible path through the graph," which of course is non-terminating.

A naive implementation would be to make your shape be a regular polygon, with evenly spaced vertices, and then see how close to those vertices you can get on a charge. More sophisticated versions would involve loading road segments up with precomputed energy consumption figures under varying conditions and doing a lowest-energy tree search with interpolation between those cached figures.

Yeah, I had the latter solution in mind. You wouldn't even really need to pre-compute the weights – just make a sufficiently simplified function to return the [edge] weight.

I maintain that a real-time "energy contour" map would be incredibly nifty.