Live data from Hacker News

The Tesla Semi from an Insider's View After One Year: "Hot Mess"

bradmunchen.substack.com

171–180 of 279 posts

Re: The Tesla Semi from an Insider's View After One Year: "Hot Mess"

#171

This report is great. It certainly rules out the, “everything is definitely on track“ scenario. But itoesn’t tell us anything beyond that. It simply comes down to what Tesla is really trying to do. This could either be a “pump”, or it could be the normal shortcuts and growing pains that are part of an ambitious beta product. A couple of years ago I would’ve leaned much further towards beta product scenario, but I’m s…

I have worked with prototype computers and the process is WASTEFUL. To the outside it looks like the systems are a mess.

For example, proto systems usually don't have power management written, or fan control at the beginning. This means the machines run hot and noisy, consuming extra power and (safely) spinning the fans at full speed. maybe next proto will have a fan controller/firmware. You get prototype memory sticks. software is buggy.

given that, I can imagine that batteries are probably ignored by the chassis and software guys. I'm sure the battery guys are in a lab somewhere writing firmware and working on charging/discharging and managing the batteries.

also, I remember that tesla is spending a lot of work on the factory that builds the product, not just the product. These trucks are one-offs and probably assembled by hand, or poorly assembled by a prototype factory line.

Now, that said I may be giving them too much credit. I remember talking to a guy who had worked at microsoft in the early dos/windows days and I said "the engineers at microsoft probably don't know the new release broke this other software, etc" and he told me I was naive, that ms had meetings where they said "how can we OWN this?" these were the days when starting windows you typed "WIN" at the C: prompt.

Re: The Tesla Semi from an Insider's View After One Year: "Hot Mess"

#172
post #134

Earlier quoted context omitted.

> Ideally, yes - but the US is lacking in rail infrastructure, so the US needs zero-emission, road-based, logistics solutions. Are you aware that the US has the largest rail network of any country in the world, almost the size of the next two (both of whom are geographically larger than the US) combined?

That can be true, and we can still be lacking. Passenger rail usage rates are abysmal.

The US is too big for long-distance passenger rail. Where density is high enough, local rail mass transit exists.

Re: The Tesla Semi from an Insider's View After One Year: "Hot Mess"

#173

So at first glance this is based on an anonymous second-hand source, and written on the substack of a self-described Tesla short seller. How seriously should we take this? I'll note that in another article[1], the author claimed that the delay in GigaMexico would push the next-gen model to 2027 and possibly bankrupt the company. According to Isaacson's Musk bio, Tesla moved initial production of the vehicle to Austin…

Just because it's from a self-described short seller doesn't mean it's not factually true.

Short sellers obviously are incentivized to be uncharitable and focus on the negatives, but that also serves to bring the negatives to light where they might otherwise stay buried.

We benefit from hearing both sides, especially if we're holding TSLA or considering trading.

Re: The Tesla Semi from an Insider's View After One Year: "Hot Mess"

#174

Earlier quoted context omitted.

> Ideally, yes - but the US is lacking in rail infrastructure, so the US needs zero-emission, road-based, logistics solutions. Are you aware that the US has the largest rail network of any country in the world, almost the size of the next two (both of whom are geographically larger than the US) combined?

> Are you aware that the US has the largest rail network of any country in the world, almost the size of the next two (both of whom are geographically larger than the US) combined? Well of course, the US is a big country - but the overall density of rail (both freight and passenger rail) is lacking compared to Europe (and China too, I think?). I live in the Seattle metro area where the major cities (Everett, Seattle,…

The US rail freight volume was 2.2 trillion tonne-km in 2022, compared to 400 billion tonne-km shipped in Europe.

Meanwhile, US road freight was about 3.6 trillion tkm, and EU road freight 2 trillion tkm.

So not only does the US do about 5x the rail freight of the EU, US rail is about 38% of freight, compared to 17% in the EU.

Re: The Tesla Semi from an Insider's View After One Year: "Hot Mess"

#175

Earlier quoted context omitted.

Priuses manage to get 50 mpg highway without depleting charge or plugging in but I guess the same optimizations can't work on a big diesel as on a little gas engine?

Prius is just a little gas engine in a little and aerodynamic car. They can have a tiny gas engine because the electric helps acceleration from a standstill. The batteries are just weight on the highway

> The batteries are just weight on the highway

Weight is only relevant to acceleration (which is partially recovered), hills (partially recovered), and rolling resistance which is only ~5% of energy usage. Even with a consistent load on the highway the batteries can help by removing parasitic loads on the engine like alternators, a/c compressors, and power steering pumps. Depending on torque demand and specifics of the engine/transmission, thrashing the battery could also be more efficient than ICE alone.

I'm sure there's stuff I've missed, but I think those are the main reasons why a number of hybrids have highway EPA ratings 10-15 mpg higher than the highest MPG ICE car ever produced.

Re: The Tesla Semi from an Insider's View After One Year: "Hot Mess"

#176

I'm really hoping the (not Tesla) USPS electric NGDV doesn't experience this same over-promise-under-deliver rollout. The ICE variants of those new vehicles get only 9 mpg (when the AC is running, 14 when it's not).

With the USPS involved, I fully expect a complete and total fiasco.

Re: The Tesla Semi from an Insider's View After One Year: "Hot Mess"

#177
post #19

Earlier quoted context omitted.

We need to go to net zero somehow. With carbon capture or carbon offset technologies near non existing, we have to make the switch sooner or later

But the current manufacturing of EVs and batteries isn’t carbon neutral or net zero until well into a vehicle’s lifecycle. It’s still worth doing but we should understand the true cost.

I mean, the true cost is that EVs will eventually get to net zero during their lifetime.

ICE never will.

Re: The Tesla Semi from an Insider's View After One Year: "Hot Mess"

#178
post #3

Why does anybody in the entire world want a semi with a 5 second 0-60?

Possibly not the buyers, but as someone who often gets stuck behind them struggling to overtake or accelerate on a hill, I wouldn’t mind.

Semi trucks should be required to stay out of the leftmost lane. Most of them are governed at 70mph or so and take forever to overtake each other, creating a logjam of traffic and frustrated drivers.

Re: The Tesla Semi from an Insider's View After One Year: "Hot Mess"

#179

> The constant use of the battery pack is unprecedented and is causing reliability problems for Tesla and other EV truck makers I don't understand why this is a surprise and why they think they can muscle through it. I didn't think our current battery technology is built for this scenario. > During the installation process, the Megachargers took down the power grid for a good chunk of the city This is my fear with ev…

> This is my fear with everyone eventually plugging in their electric vehicles at 5pm. If I plug my car in at 5pm, it won't start charging for 7 hours, when it's economically rational for it to do so. Time-of-use pricing seem to solve this. EV load is among the least time-sensitive of all power consumers. Set up your pricing right, and the EV charging will follow.

What if you need to go out again at 9pm?

Re: The Tesla Semi from an Insider's View After One Year: "Hot Mess"

#180
post #12

Batteries on semis dont make much sense to me. Hybrid systems on heavy inner city vehicles make a lot of sense. As do hybrids in mountainous areas. But long distance hauling? BS. Semi engines are large with very low power to displacement ratio -> they are very fuel efficient. Truck builders' customers are very sensitive to fuel consumption. I find it very hard to believe that marginal grid power (ie thermal plant), t…

Priuses manage to get 50 mpg highway without depleting charge or plugging in but I guess the same optimizations can't work on a big diesel as on a little gas engine?

There's a few things going on that come to mind.

Big trucks often have other systems that historically can be dependent on having a good consistent power draw source.

Also, air brake systems; The brakes on a power unit are designed to 'fail safe', so rather than using hydraulic fluid to close the calipers, uses air pressure to keep the calipers open. There's a bunch of efficiency implications here but in general you either have to make a work around to that or take efficiency losses of a PTO or electric compressor. Also, not sure whether a different approach requires DOT approval or not. 'What about all those wheels and regen braking' one might ask? Then you'd be talking replacing/retrofitting trailers which has a bunch of implications for an owner/operator.

Oh. Transmission losses, and the realities of a diesel; One thing that really helps the Prius (and vehicles that use similar arch) is that it (and many/most of the more efficient hybrids) use a Fixed-Ratio transmission; The engine uses various techniques to maximize efficiency at a wide band. Diesels, on the other hand, are often most efficient within a (relatively) much more narrow power band... Also in general, rig diesel engines aren't exactly designed for 'stop-start' patterns.

Then, you've got the delta-V problem; The more weight you add the more energy it takes to move stuff. My SWAG (i.e. completely uninformed and friday night math) is that a power unit would have to have a battery with the capacity of a plug-in hybrid to provide meaningful value.

Then, drag coefficient. A truck has a much larger profile from every angle which has a sizable impact on efficiency.

Oh, and make it all with expectations for long service intervals.

Post reply on HN