Tesla Semi

[mongo]

I don't think that it makes sense to have each motor/inverter have it's own battery pack. I do expect motors to have dedicated inverters (unless they put pairs together for packaging efficiency somehow, but they would still be separate logically as they need to drive wheels separately).

With this scheme, if you have 4 perfectly good motors and inverters and a pack fails you, you use a motor and inverter unnecessarily. If you lose an inverter you lose the corresponding motor, but no reason to lose the corresponding pack and thus range. What good is redundancy in motors if you're essentially stranded anyways (because you lost 1/4 the range)? Granted if you lose 1/4 of the pack, there's no engineering around that, but that's no reason to let inverter or motor problems result in the same problem.

By tying all the battery packs together into one larger pack using a bus bar, you achieve better reliability. As long as one pack and inverter/motor pair work, you can move, and the loss of a sub-pack is only going to lose range and power output (i.e., impact to drag racing and hill climbing but you still have 4 working motors), nothing else, and losing an inverter/motor pair has only a small impact on range.

Tesla already has the necessary technology with battery contactors and "smart fuses" to do all this, and the bus bar doesn't change the size of the wiring - max intended power draw determines max wiring size for both batteries and inverters (and motors being connected to inverters doesn't change so wiring is the same regardless). Smart fuses let you protect against problems without blowing pyros, or you can just use pyros if you want production cost to be cheap as a tradeoff with increasing the cost of maintenance for refurbishing/repair. The only thing that must be sized for total current capacity of all the packs is the bus bar itself, which is just a hunk of metal.

I agree that the joined pack helps in the event of an electrically failed DU. However, I do not think handling that mode is worth the system implications:

Fault current is now just under 4 times the single pack output. That means all the fuses need to be higher breaking rated than currently. A wring fault can pull the max, requiring all the packs to either open their contactors (if they can interrupt that current level) or blow their pyros.

Each drive unit needs at least one pyro fuses to handle internal faults, along with 2 contactors to isolate a failed unit (pack to chassis faults).

In the event of one pack performing better than another, it will have more current drawn from it, so the max allowable draw must be derated by the worst case pack to pack imbalance, otherwise the pack's pryo fuse will trip.


If the truck has onboard chargers, those could potentially be configured to transfer power from one pack to another in the case of a failed DU. That would solve the range problem without adding in the additional fault possibles. That may require another set of contactors.

 

[jerry33]

Ugh, Canada! The cold weather will be a real range killer for electric semis. Just parking overnight will cost many miles. Windshield deforsting, wipers, wet roads. Probably will take 50% more KWh/mile

We know Tesla's range claims are optimistic for Model S&X even in mild and sunny weather of Southern California.

Also geometric design standards in Canada (before metric-ization) were 666: 6% maximum grade (often exceeded in mountainous areas due to cost), 600 foot curve radius, and 60 mph speed limit. I foresee a "Canadian special edition" in the future to cope with this.

 

[KarenRei]

Right, if you zoom out far enough everything is a blip. ... So, we can agree that sharp inclines do exist but are not significant in the integral over all highways in US.... Tesla is not going to take over 100% of US trucking for years to come.

You're talking about something entirely different from me. I am - and I can't believe I have to repeat this yet again - talking about the article posted which claimed "diesel is dead" based on a claim of an average semi truck speed on US interstates of under 60mph. Nothing more. If you want to discuss that article and it's claims - specifically that the average speed of a semi on a long haul in the US will be under 60mph - feel free to quote and reply to me. If you want to talk about something else, please leave me out of it.

Now, do you or do you not want to argue that the average speed of a semi on a long haul in the US is under 60mph, and that - as the article does - you can multiply that 60mph times 11 hours driving to get the maximum range that US truckers drive in a day?

 

[kbM3]

They are also going to have to figure out a way to get max regen braking at all times, regardless of battery charge level. you can't have a loaded semi with half regen going down the grapevine. Trucks use a combination of gearing, engine braking, and service brakes to control loads, with no engine or transmission brakes on an electric they can't rely on service brakes during times of high battery charge.

With judicious route planning, this should very rarely occur. I’m sure the Semi will have the software to plan the optimal charge as long as the route is specified in advance.

Edit: Example starting at top of mountain where 200 kWh regen is expected on route, recommended charge would be 250 kWh below max.

 

[kbM3]

Trying to estimate how the Semi Megachargers would function / be designed.

Looks like the loads will be huge (probably >1MW, and that's just for one connection!) and for short duration (~30 min charge, "peaky") -- which means very expensive to get that power from a utility hook up. Both from just capacity of hookup (1MW connection to grid will be expensive, and peak charges for that amount of electricity will be crazy)

So the obvious solution, as many have noted, is pull from large banks of PowerPacks instead of the grid. The PowerPacks could have a much smaller, near constant "trickle charge" from the grid (plus extra from solar) which would help control costs.

How these could differ from existing Superchargers however, can be interesting. Superchargers are built of stacks of many car charging (AC->DC) modules, and charge essentially directly from the grid (AC from Grid -> DC Car Battery).

If Megachargers will be essentially powered solely from PowerPacks, then they don't need the large amounts of AC-DC converters ($expensive$) that Superchargers use. PowerPack (DC) -> Semi Battery (DC). Not only is this more efficient (less conversion losses) but it would mean the cost of building these charging stations is pretty close to just the cost of the PowerPack + their (smaller) hookup to the grid.

Does this make sense? Am I missing something important? Anyones thoughts on the architecture of the Megachargers?


EDIT: Probably won't be more efficient in terms of conversion losses -- just changes where the conversion step occurs.

Sounds good to me except the part about trickle charge from the grid. Elon stated it would be pure solar. I realize he made the same promise for superchargers years ago and they are not there yet.

 

[kbM3]

This might be old news to y'all by now, but I did some research last night after the reveal that shocked me:



Full article: The Tesla Semi: Diesel Is Dead - Tesla Motors (NASDAQ:TSLA) | Seeking Alpha (I make no money when you use this link)

Did you account for the fact that the Tesla Semi is going to be much more expensive, therefore capturing an even greater portion of trucking revenue?

 

[generalenthu]

Link to the conversation on Tesla semi on a truckers forum.

Interesting to see initial negative reactions give way to more thoughtful responses.

That tesla truck is ugly!

Warning: quality of discussion is not great, though it's not like there's a high bar here at TMC in the general discussion threads.

 

[neroden]

You're talking about something entirely different from me. I am - and I can't believe I have to repeat this yet again - talking about the article posted which claimed "diesel is dead" based on a claim of an average semi truck speed on US interstates of under 60mph. Nothing more. If you want to discuss that article and it's claims - specifically that the average speed of a semi on a long haul in the US will be under 60mph - feel free to quote and reply to me. If you want to talk about something else, please leave me out of it.

Now, do you or do you not want to argue that the average speed of a semi on a long haul in the US is under 60mph, and that - as the article does - you can multiply that 60mph times 11 hours driving to get the maximum range that US truckers drive in a day?

I will simply point out that I've rarely, if ever, seen a loaded tractor-trailer exceeding 65 mph. Diesel semis basically can't do it except on dead flat roads, which excludes three-quarters of the country. I think we can state with confidence that the average speed of a semi on a long haul in the US is under 60 mph.

I see that you're in Iceland, so maybe you don't know this. Trucks don't go faster than 65 mph, and are usually around 60-65 mph even on the "open road".

So yes, you can multiply 60 mph by 11 hours to get an *overestimate* of the maximum range that a typical OTR US trucker drives in a day. There may be some exceptions in the flatter parts of West Texas or Nebraska, but most of the country? Yeah, that's the max.

You want to really make sure you overestimate? Multiply 65 mph by 11 hours. Either way, the Tesla Semi is sufficient for all single-driver applications, if there is an appropriately located Megacharger. (It's clear they're not even trying for the two-driver market.)

 

[TexasEV]

You obviously haven’t driven on any of the interstates in Texas or anywhere else between the coasts if you think the average truck speed is under 60 mph.

 

[LCR1]

I will simply point out that I've rarely, if ever, seen a loaded tractor-trailer exceeding 65 mph. Diesel semis basically can't do it except on dead flat roads, which excludes three-quarters of the country. I think we can state with confidence that the average speed of a semi on a long haul in the US is under 60 mph.

I see that you're in Iceland, so maybe you don't know this. Trucks don't go faster than 65 mph, and are usually around 60-65 mph even on the "open road".

So yes, you can multiply 60 mph by 11 hours to get an *overestimate* of the maximum range that a typical OTR US trucker drives in a day. There may be some exceptions in the flatter parts of West Texas or Nebraska, but most of the country? Yeah, that's the max.

You want to really make sure you overestimate? Multiply 65 mph by 11 hours. Either way, the Tesla Semi is sufficient for all single-driver applications, if there is an appropriately located Megacharger. (It's clear they're not even trying for the two-driver market.)

I drove 40 or 50K miles last year and none of that is accurate. I drive about 2-3,000 miles a month of personal us, especially during the racing season. I tow my RV around 65mph and get passed by plenty of semis in the hill country. I recently did a trip all the way to Florida and kept the cruise on 63 I only passed 3 or 4 semis, the rest passed me. A lot is also dependant on localities you'll see semis doing 75+ in Texas you won't see a truck dare go over 60 in California. It can get even more local than that, on the east side of houston trucks drive slower than the west side. It's really all dependant on the drivers just like the people that drive 75 in their car or those that drive 60, some people care about MPG and some dont.

 

[TNEVol]

The only exception is sleeper cabs with two drivers, which unfortunately the Tesla semi will not replace. They handle the government regulations by swapping drivers whenever one driver needs to take a break, and can therefore go until they need fuel.

Thankfully it's a tiny fraction of the market, because it costs a lot more to run with two drivers. I'm not sure what part of the market uses dual drivers, but I'm guessing it's extremely time-sensitive stuff which is too heavy to be reasonably sent by air. Anyone know?

Yeah for example Fresh Point. Produce from CA to Nashville every day

 

[BioSehnsucht]

Here in Dallas, TX it's common to see semi's doing 70mph on 635 and the various toll roads, and not uncommon to see them doing 75 either. Probably no more than half stick to 60-65, the rest are doing above 65 easily. And 635 is neither a steady incline nor flat - it's lots of small hills for all the overpasses of street crossings.

 

[Lessmog]

You're talking about something entirely different from me. I am - and I can't believe I have to repeat this yet again - talking about the article posted which claimed "diesel is dead" based on a claim of an average semi truck speed on US interstates of under 60mph. Nothing more. If you want to discuss that article and it's claims - specifically that the average speed of a semi on a long haul in the US will be under 60mph - feel free to quote and reply to me. If you want to talk about something else, please leave me out of it.

Now, do you or do you not want to argue that the average speed of a semi on a long haul in the US is under 60mph, and that - as the article does - you can multiply that 60mph times 11 hours driving to get the maximum range that US truckers drive in a day?

Thanks, but I pass. Two notes, as this has gone downhill already

1. You seem to take the article as read for part of your argument, but some of us will not go to SA or some other sites.
   
2. It's a mistake to conflate averages with particular cases, as you do in the purple quoted bits above.

With that, I leave myself out.

 

[Lessmog]

They are also going to have to figure out a way to get max regen braking at all times, regardless of battery charge level. you can't have a loaded semi with half regen going down the grapevine. Trucks use a combination of gearing, engine braking, and service brakes to control loads, with no engine or transmission brakes on an electric they can't rely on service brakes during times of high battery charge.

With judicious route planning, this should very rarely occur. I’m sure the Semi will have the software to plan the optimal charge as long as the route is specified in advance.

Edit: Example starting at top of mountain where 200 kWh regen is expected on route, recommended charge would be 250 kWh below max.

Folks, I give you the biggest electric dumper truck.

It will bring 65 tons of material from the quarry, downhill to a cement factory in Switzerland, 22 trips per day. Regen easily charges the battery, and also saves an important bundle on friction brakes. Much lower maintenance overall, plus negative fuel cost! Excess power will be exported to the grid at night.

From our Swedish forum.

 

[avoigt]

Folks, I give you the biggest electric dumper truck.

It will bring 65 tons of material from the quarry, downhill to a cement factory in Switzerland, 22 trips per day. Regen easily charges the battery, and also saves an important bundle on friction brakes. Much lower maintenance overall, plus negative fuel cost! Excess power will be exported to the grid at night.

From our Swedish forum.

No offense but different market segment. It does not help the discussion to compare apples to oranges......

 

[Lessmog]

No offense but different market segment. It does not help the discussion to compare apples to oranges......

Absolutely a totally different market, yes. But still kind of relevant to a discussion about hills, SOC and regen braking of heavy vehicles, I think.

 

[SwTslaGrl]

It looks like a sleeper cab version is coming,

"It is initially designed for short- and regional-haul routes, such as those that run goods from ports to distribution centers, but Musk said a long-haul sleeper cab model is in the works."

https://www.trucks.com/2017/11/17/elon-musk-unveils-tesla-electric-semi-truck/

 

[ValueAnalyst]

It looks like a sleeper cab version is coming,

"It is initially designed for short- and regional-haul routes, such as those that run goods from ports to distribution centers, but Musk said a long-haul sleeper cab model is in the works."

https://www.trucks.com/2017/11/17/elon-musk-unveils-tesla-electric-semi-truck/

Here's to hoping for the day Tesla realizes an absolute minimum of ONE variant is required for max ramp speed.

Any other strategy, not only reduces shareholder value, but also slows down the mission.

 

[ccorda]

Bloomberg's Dana Hall turned me on to the idea that CA has a ZEV program coming for Trucks:

Dana Hull on Twitter

Which led me to some CARB documents

• https://www.arb.ca.gov/msprog/actruck/mtg/170425workshoppresentation.pdf
• https://www.arb.ca.gov/msprog/actruck/mtg/170830arbpresentation.pdf
  
• https://www.arb.ca.gov/msprog/actruck/docs/HDBEVefficiency.pdf

My takeaway is that at least in CA, Elon is right when he says his estimates are conservative.

Here is the most compelling chart IMO, and this just covers costs without any potential ZEV credits:

• The low MPG is based on the specific activity of the truck being typically used at low speeds around the port of LA, and then in traffic around LA. A longer distance truck would have more miles at a higher MPG, but also would do more than 25k miles.
• Note their fuel cost of 2.92 per gallon. Just searching in CA right now, prices are more like 3.25 - 3.50. On November 1, a 0.20 per gallon and a 4% sales tax increase kicked in, so it's hard to see these prices getting below $3.00 without a major overall drop in oil.
• They project a 30% maintenance savings for battery.
• This includes a one time $50,000 cost for setting up on site charging infrastructure.
• CARB's projected efficiency for the battery truck is lower than Tesla's claimed < 2.0 kWh/mi, but this would be route dependent as well -- though you'd think city miles would help.
• The assumed 300k purchase price is high, though I suspect the full loaded Tesla semi will be even higher.
• Their baseline assumption right now is for .13/kWh overnight charging from So Cal Edison, offset by a 0.06/kWh credit for selling back Low Carbon Fuel Standard (LCFS) credits. Without any further changes, CARB is saying net 0.07 is feasible for on site overnight grid charging right now.

All that is what gets them to the middle chart where even today they see battery as more cost effective.

Going forward, there are upcoming policies which tilt things further (last column of chart):

• SB 350, where utilities are spending about a billion dollars over 5 years to help enable transportation electrification. In this model, that decreases the infrastructure needs to 10k per truck. (don't fully understand the details on this).
• They want to increase the LCFS value so that it's 0.14/kWh by 2023. This essentially makes the electricity free.

Not included in the charts is the draft ZEV program. A class 8 semi with 500mi range would be worth 5.2 ZEV credits. While there is no class 8 ZEV requirements, Tesla would be able to sell their credits to the makers of all smaller trucks. Credits can start accumulating early, with requirements for manufacturers starting in 2023.

Included as attachments as some of the slides I found most relevant. I don't fully understand the LCFS program and how it works, so if anyone has any more info or can confirm it's saying what I think it is that'd be great.

 

[KarenRei]

I will simply point out that I've rarely, if ever, seen a loaded tractor-trailer exceeding 65 mph

Yes, that's because you're in New York, which has a 65mph truck speed limit. Meanwhile, other states have speed limits for trucks as high as 85mph.

Diesel semis basically can't do it except on dead flat roads

Yes, they can and do.

which excludes three-quarters of the country

You've apparently never looked at a topo map of the US. The vast majority of the US is flat.

I think we can state with confidence that the average speed of a semi on a long haul in the US is under 60 mph.

You would be wrong. And even if the average on interstates was 60 (it's not), that would hardly be the speed you could use to determine "diesel is dead". You need something like the 95th percentile for that, at least.

I see that you're in Iceland, so maybe you don't know this

I grew up and lived for many years in the US. Try again.

One of those states was Texas. 85mph truck speed limit. Some do indeed approach it. Almost none drive 60, let alone less.

Trucks don't go faster than 65 mph, and are usually around 60-65 mph even on the "open road".

"I see that you're in New York, so maybe you don't know this."