Tesla Semi

[jhm]

I view battery leasing as a separate issue from the provision of charging and swapping infrastructure. A lease on a battery is just for the battery, not the electricity going into it.

So many battery leasees will want their own charging infrastructure that optimizes their own operations. So if you're Wal-Mart, you'd rather charge your trucks at your distribution hub using your own solar power or cheap industrial power from the utility.

Of course, there is still a role for Tesla to play in providing commercial Supercharging and swaps, but in both cases Tesla can charge a fee for the service or franchise this out to third parties like truck stops. I do not think the free Supercharging model will work in the commercial space as it has for private passenger cars. "Free" would induce bad logistics for companies that would be better served with having their own chargers. That is, you don't want to drive out of your way just to get "free" charging, but if you pay for your own power, you feel like a chump. Second, providing free charging also kills off any motive for third party providers to enter that market. So Tesla could find itself stranded in its own network, while commercial charging networks arise to serve all competitors. Finally, the amount of electricity per mile will vary enormously by how the trucker uses the vehicle. One may haul 20 tons uphill while another hauls 10 tons downhill. So Tesla could expose itself to adverse selection.

There is nothing wrong with truckers paying a fair price for charging and franchising the opportunity to be that service provider to other players. Moreover, suppose a truck stop wanted to offer commercial Supercharging at 350kW. They are going to need a few MWh of Powerpacks to back that up or be exposed to massive demand charges, like $7000/month per supercharger! For that kind of cash, they'd do better to get a small fleet of Powerpacks, and they may as well put solar panels wherever they've got the space. In short, franchising commercial Supercharging is an opportunity for Tesla Energy to build out power systems, which may ultimately serve all commercial EV fleets regardless of make.

One more little thing, did anyone notice that Morgan Stanley is assuming 1kWh/mile? I wonder how they got to that.

 

[SwTslaGrl]

This post Tesla Class 8 Semi Truck Thoughts shows there's interest in Tesla Semi.

 

[ohmman]

One more little thing, did anyone notice that Morgan Stanley is assuming 1kWh/mile? I wonder how they got to that.

I didn't notice that. I'll reiterate some disbelief of that number being realistic, based on my 575Wh/mi towing number. Another X owner is towing a heavier Airstream (out of specs) and reported over 700Wh/mi. But it's still nowhere close to the kind of load or frontal area of a Class 8 semi.

If they somehow achieve this, it'll be an engineering marvel and will instantly make it a competitive product.

 

[Electroman]

Nothing less then 3 kWh per mile. easily !

 

[larmor]

The main challenge then is to improve cd. Maybe the Semi will have some large gaping opening in the front with a weird battery configuration which needs air, the so called air battery. Think hyperloop pod meets semi tractor with air battery in the gaping intake...

 

[jhm]

Nothing less then 3 kWh per mile. easily !

Show us your math. Thanks.

 

[Ulmo]

I didn't notice that. I'll reiterate some disbelief of that number being realistic, based on my 575Wh/mi towing number. Another X owner is towing a heavier Airstream (out of specs) and reported over 700Wh/mi. But it's still nowhere close to the kind of load or frontal area of a Class 8 semi.

If they somehow achieve this, it'll be an engineering marvel and will instantly make it a competitive product.

Why does a huge vehicle have to have high energy efficiency mass density ratio compared to smaller vehicles? Huge vehicles have advantages, not the least of which is more room to swap battery packs. Some of that swapability is built in to trailer systems: just throw a hunk of battery in the trailer, and it can charge up at shore and the tractor can just flit about switching from load to load almost the same as now. When a load is too long distance for even that much battery, then the logistics of it can place it on train, swap trailers to a shore charge situation, or for time sensitive loads, swap battery some other way (moving interior load, or battery swap capability without moving load within a vehicle).

But, your idea of improving energy efficiency mass density via aerodynamic drag isn't all that unapproachable either: once we get self driving trucks, the usual vehicle unit size of cargo can go down (which logistics people say is more ideal), and that can enable considering more efficient drag shapes for certain loads. Other loads will still cause bad drag.

All of this will settle down into logistical patterns over the coming decades.

I really don't see it as a huge problem to tackle now. Anybody who can do current logistical analysis can probably do the type of logistical analysis necessary for electric trucks of the future designed with the types of thinking I used above. Any of those people who approach this issue in a timely manner in the near future will be well positioned to continue with their logistics.

 

[Ulmo]

I would think one of the great advantages of an electric truck would be increased torque. The bane of a truck drivers existence is the lack of torque necessary to maintain speed up a hill. Being able to maintain steady speeds up and down hills would be a benefit to every driver on the road. Add to that the huge amount of regen that a fully loaded semi-truck would gain going downhill and you have a compelling alternative to diesel. And, I wonder if the regen would reduce the likelihood of burning/failing brakes on long downhills?

Regarding downhill:

All dangerous vehicles interacting with other vehicles (such as on roadways) need backup systems for critical operations, in this case most specifically, brakes. That means the highest common capability of any unique set of braking systems on a vehicle combination, even in the worst possible working situation, needs to be able to safely operate. That means, that the backup brake system needs to be able to stop that truck. That means, about 35MPH in 6º grades downhill, even if the truck has a super duper electric regeneration system. What if the computer has a bug, or some corrosion gets on a contact? Suddenly, you've lost the regen braking. Now, you're depending entirely on the backup brake system. Getting 40 tons (or even 20 or 10 tons) to stop when it's already going 80MPH, 65MPH, or 55MPH down a 6º grade is generally too expensive in the backup braking systems (think fluid heated metal, generally as an analog to reliable energy transfer cost, traditionally heat dissipation), so that's why they're all built with 35MPH in a downhill 6º grade in mind. I don't have the specific numbers in front of me, so all those numbers above are generalizations. But my point is that it will be a while in economic cargo vehicle engineering before 35MPH can be substantially exceeded as a safe downhill speed.

Having said all that, by all means, going uphill faster is a good idea. Also, the safe downhill speed for each vehicle design and load combination would probably be somewhat different, and some might exceed 35MPH. For the meantime, they'd still have to line up in the same lane as other 35MPH and even slower loads (just watch the downhill trucks on the North side of the Grapevine for an example).

Engineers could make backup backup systems: one-time use stopping systems that are reliable enough that even during a regen failure the vehicle could come to a full stop from 80MPH-100MPH down a 8º grade overloaded with 50 or 75 tons and a few wheels out of commission (always over-design safety systems). If that is inexpensive enough, they can install that on the truck, and then make certain the operation of that backup backup system is good enough that it will actually stop the truck in all situations (including driver capability), and then the truck will just be parked there on the mountain, waiting for a tow, since the one-time backup system just went bust, shutting down the vehicle operability. It could even just computer-limit the speed to something like 15MPH for the regular backup braking system, so the truck could limp to a safer pull-out location before proceeding after fix. The backup-backup single use braking system could even be easily swap-able. Of course, the regen failure would also need to be fixed, meaning basically some type of drive train overhaul to fix it. So, there are some paths for faster driving downhill, but they'd have to be engineered and tested.

Whether or not and to what extent the regen effect on total system design and production reliability goes into the actuarial tables for the engineers to consider is kind of just part of the stuff the engineers have to consider, and the end product will be some total system that meets spec. I imagine that can vary from design to design and from vehicle to vehicle. Interoperability standards will be made up in the process, and sometimes governance will (can*) be used to streamline (in the theory that it makes it more reliable) as well as (theoretically always) verify the reliability of those standards.

* Theoretically, interoperability wouldn't be needed for single complete unchangeable vehicles, but I think for a while, that will be a niche product line for local use loads. Having said that, that might become what most citizens see in end point use, since they could probably find enough land to have fleets of discharged vehicles charging up for their next small short distance loads.

 

[Reciprocity]

Why does a huge vehicle have to have high energy efficiency mass density ratio compared to smaller vehicles? Huge vehicles have advantages, not the least of which is more room to swap battery packs. Some of that swapability is built in to trailer systems: just throw a hunk of battery in the trailer

The problem with the trailer idea is compatibility with current trailers and I assume there are a lot more trailers then trucks, because they already load them while detached from a semi. The trailer could be charged while loading but that would require the docs to be wired. Unless you are saying these trailers could work with existing ICE Semis? Which brings me to an idea that I had..

I had an idea of two different products. A semi with 500-1000kWh with 4 distinct and self contained packs that could be swapped or charged independently on a 350kWh+ child's toy as Elon refers to it. This would allow for up to 10-20x the speed of the current faster superchargers, depending on the capacity of the non-child's toy.

The second product would be a trailer that has its own motors and 2x 100kWh packs that could be independently charged like above. These trailers could be used with existing semi's to give the owner better fuel efficiency and safer control over the load. These trailers could also be autonomous to move around a shipping yard, but not drive on roads.

The long haul trucker would have to use the 1000kWh + powered trailer. Short haul could use the smaller pack and no powered trailer. Any truck could use the trailer to enhance fuel economy.

 

[Ulmo]

One other potential downside will be the additional weight. Semis are already blamed for the majority of road damage. A semi loaded with batteries and freight could potentially be even more damaging.

While I'm not convinced electric drive trains are usually heavier (they might be, in many situations, at least for the near term future), they could just reduce cargo to compensate. If cargo is reduced, then vehicle weight to carry that cargo weight can also be reduced. There is probably some sweet spot for this for each type of load. (That's one reason I advocate multiple swappable batteries per trailer and tractor -- about the size to put 3 on a tractor and about a dozen on a full size trailer -- so variable vehicle sizes and cargo sizes could be supported by number of batteries loaded.)

 

[Ulmo]

The problem with the trailer idea is compatibility with current trailers and I assume there are a lot more trailers then trucks, because they already load them while detached from a semi. The trailer could be charged while loading but that would require the docs to be wired. Unless you are saying these trailers could work with existing ICE Semis? Which brings me to an idea that I had..

I had an idea of two different products. A semi with 500-1000kWh with 4 distinct and self contained packs that could be swapped or charged independently on a 350kWh+ child's toy as Elon refers to it. This would allow for up to 10-20x the speed of the current faster superchargers, depending on the capacity of the non-child's toy.

The second product would be a trailer that has its own motors and 2x 100kWh packs that could be independently charged like above. These trailers could be used with existing semi's to give the owner better fuel efficiency and safer control over the load. These trailers could also be autonomous to move around a shipping yard, but not drive on roads.

The long haul trucker would have to use the 1000kWh + powered trailer. Short haul could use the smaller pack and no powered trailer. Any truck could use the trailer to enhance fuel economy.

Yes to all the above. I thought of this myself decades ago, and posted similar ideas here in years past. So have hundreds (at least, and at this point probably hundreds of millions) of other people. All of these ideas are pretty obvious to anybody who has done enough studying of these problems. My main takeaway is flexibility in mind for the engineers, and probably flexibility of product, as well, at least for the near to medium forseeable future, much as you described.

To deal with a few points:

Trailers and/or tractors can charge in a number of ways:

• Parked in yard (need to trench and wire yards -- medium expense)
• Parked at dock (would need faster charging capability -- higher expense (not the least of which are robotic connectors that wouldn't require death-inducing pinch points for operators), although some docks are sized such that they could charge slower)
  
• Solar roofs on trailers (would require sitting longer -- pretty high expense for most, lower expense for others)
• Swappable batteries (depends on design engineering initially done -- low expense overall, most flexible, but due to problems with battery swap, would need a lot of good engineering (don't want to drop that on your foot, hand, or head))
• In-road charging (very expensive, as far as we know)
• In-tractor batteries (high expense because of many limitations; essentially would use similar methods to above, with exception of a trailer roof full of solar panels)

All this points to the concept of doing all the above, except the in-road charging (for now).

Compatibility with current trailers is just an engineering problem from my point of view. Probably, electrified trailers would be able to travel on ICE tractors, and electrified tractors will be able to pull unelectrified trailers but with different distance limitations. States could allow longer combination vehicles or add-on parts to legacy vehicles to handle compatibility resolution, such as battery pack trailers on unelectrified trailers on electric tractors, but due to overall costs, this might not become one useful option, but it's all in the numbers. I think when the total system is added up, the numbers will work, one way or another.

The biggest danger is having a non-engineer make decisions about the system at any level. (Of course, an engineer CEO, engineer administrator or engineer politician would require input from non-engineer professionals.) (At least for anyone first to or near to first to market; those who wish to get less early advantage could just copy the lead of engineering leaders even though they themselves aren't engineers.)

 

[GoTslaGo]

Does anyone know how long it takes to fuel up a Semi? Can't find any info. Seems that would be the time gauge that we would need to compare whether a supercharger or battery swap makes sense.

Remember a Semi carries 250-300 gallons of diesel. If it takes an average 15 gallon car 5 minutes to refuel, an average Semi at 300 gallons could be 20x that number at 100 minutes. So you're looking at 1-2 hours. When I drove cross country many times (not in a truck) I do recall seeing many a semi "parked" and fueling for a long time (I would stop at these stops for a quick cat-nap, and often find the same truck still fueling as I woke up, or waking me up as they pulled out).

IMHO, if Tesla can optimize Supercharger speed then it actually may be faster or the same as current fueling speeds. So no battery swapping (and tedious logistics) would be needed.

 

[MP3Mike]

Does anyone know how long it takes to fuel up a Semi? Can't find any info. Seems that would be the time gauge that we would need to compare whether a supercharger or battery swap makes sense.

They make high speed fueling pumps for diesel, which I have seen up to 63 gpm. So 300 gallons could take as little as 5 minutes.

 

[GoTslaGo]

They make high speed fueling pumps for diesel, which I have seen up to 63 gpm. So 300 gallons could take as little as 5 minutes.

Is that common?

 

[MP3Mike]

Is that common?

Having high speed pumps is common, but I don't know that they are all as fast as 63 gpm.

 

[GoTslaGo]

Having high speed pumps is common, but I don't know that they are all as fast as 63 gpm.

Thanks!

 

[ggr]

Is that common?

Common at interstate truck stops, yes.

 

[GoTslaGo]

Common at interstate truck stops, yes.

Ugh, I'm getting old. Guess it has been almost two decades since I've driven cross-country like that.

 

[ohmman]

On the charging front, a Tesla patent application has recently been published (but appears to have been originally submitted in 2014) which allows for large contacts under the vehicle. Possibly more importantly, it moves thermal management from the vehicle's system over to the Supercharger. This really opens the door to fast semi charging.

 

[adiggs]

Patent applications filed with the US patent office publish 18 months after application - that's when the rest of the world becomes aware of their existence. That's when the process runs normally - I'm not a patent lawyer, but I do know there are numerous corner cases and it wouldn't surprise me if there are ones that can further delay the publishing of an application beyond the 18 month window.

One corner case I know of is if you file only in the US, then you have the option of keeping the application unpublished until it finally grants. In this case, then the first time anybody becomes aware of the application is when it publishes as a granted patent.


Looking at the picture more closely, the original patent was filed in 2014. This however is a continuing application to that original application - this one filed in Dec 2016. So in this case, more like 4 or 5 months from submission of application to publishing.