U.S. semiconductor giant Intelannounced in March that it plans to spend $20 billion on two new chip plants in Arizona. Separately, TSMC (Taiwan Semiconductor Manufacturing Company) said it was going to build a $12 billion factory in Arizona, and chief executive C.C. Wei said Wednesday that construction had already begun.
The Grand Canyon State may not, however, seem like the most obvious place for a chip “foundry” or “fab” since the high-tech manufacturing plants guzzle millions of gallons of water every day.
At present, in the face of climate change, Arizona is facing a deepening water crisis and some of the state’s all-important aquifers have an uncertain future.
Arizona received just 13.6 inches of rainfall on average per year between 1970 and 2000, according to the NOAA National Climatic Data Center, making it the fourth driest state nationwide. Conversely, Hawaii and Louisiana recorded the highest levels of average yearly precipitation in the U.S. over the same time frame, reporting 63.7 inches and 60.1 inches, respectively.
“Water is a key element in semi manufacturing, but the infrastructure has been put in place [in Arizona] to ensure adequate supply to meet the industry’s current needs,” Alan Priestley, vice president analyst at tech research firm Gartner, told CNBC.
A key consideration of any new construction would most likely be contributions to enhancing the water supply infrastructure, he added.
Glenn O’Donnell, vice president and research director at analyst firm Forrester, told CNBC that chip fabrication plants “recycle water religiously,” adding that it’s a bit like a swimming pool in an enclosed building.
“You need a lot to fill it, but you don’t have to add much to keep it going,” he said. “Also, being in an enclosed space, a lot of the water that evaporates can be captured with a dehumidifier and returned to the pool. The fabs will do similar things with their own water usage.”
Intel notes on its website that it is striving to achieve “net positive water use” in Arizona and that it has funded 15 water restoration projects that aim to benefit the state. “Once fully implemented, these projects will restore an estimated 937 million gallons each year,” the company says.
Beyond water
TSMC and Intel, two of the biggest heavyweights in the chip industry, have chosen to expand in Arizona for several other reasons, according to the analysts.
Intel has had a presence in Arizona for over 40 years and the state is home to a well-established semiconductor ecosystem. Other major chip companies with a presence in Arizona include On Semiconductor, NXP and Microchip.
Intel now employs over 12,000 people in Arizona and the state is home to Intel’s newest manufacturing facility, Fab 42.
As Intel has increased its presence in Arizona, the local universities have “established a strong reputation for semiconductor design courses and research providing a highly-skilled work force for the local semi industry,” Priestley said. “This has helped create an ecosystem of companies to supply the products and services necessary to manufacture chips.”
TSMC will be “able to tap into these resources and [the] ecosystem of supply chain vendors,” Priestley said.
Local tax breaks and incentives “will have played a big part” in the initial site selection, he continued, noting that land availability, land costs, housing costs and the local economy will have also been considered.
Seismically stable
The case for Arizona doesn’t stop there. Its seismic stability and relatively low risk of other natural interference are appealing to chipmakers, O’Donnell said.
“A chip factory cannot shake, not even a microscopic amount,” he said, adding that they set such factories into the bedrock to keep them still. “Even a 0.5 Richter shake can ruin an entire crop of chips.”
That said, Intel does have some chip plants on the West Coast of the U.S., where the ground is more susceptible to earthquakes. The company has a huge presence in Hillsboro, Oregon, for example.
“The West Coast does have fabs but they need to take great measures to isolate the shaking,” said O’Donnell. “They don’t need such drastic measures in Arizona because it shakes a lot less.”
Arizona is also immune from most other natural disasters like hurricanes and wildfires, O’Donnell said.
With its bountiful sunshine, Arizona also boasts “dependable, plentiful and green electrical power,” O’Donnell said, calling out Salt River Project as a local power utility in the Phoenix area that caters to big consumers of power. A chip foundry needs power on the scale of a steel plant, according to O’Donnell.
Ultimately, it largely boils down to politics.
“The political machinery in Arizona is determined to make the state business friendly,” said O’Donnell. “More business equals more and better jobs equals more votes to the power brokers. The recent announcements by Intel and TSMC come via a lot of help from federal, state and local government entities.”
Porsche’s long-awaited Macan EV will finally hit US dealers on September 30th, and we’ve also just learned that it will have an EPA-certified range of 308 miles, or 288 for the “Turbo” version.
We’ve been waiting what seems like forever for the Porsche Macan to come out – Seth even got to go see it in Germany last October – and now the car is finally (almost) here, arriving later this month in US dealers.
Porsche told us that the ships carrying the cars are en route, and depending on which coast you’re on, they should arrive in the last week or two of this month. But Porsche and its dealers have been communicating the Sept 30th date for Macan availability – so if you’re looking forward to this car, you’ve only got a couple weeks to get your affairs in order (you can use our affiliate link to contact local dealers and get in line).
And today we’ve learned one of the final steps before getting these cars on the road has been submitted, as the Macan EV has been officially rated at 308 miles EPA range, or 288 miles for the Turbo. These numbers are lower than the European 381-mile WLTP range, but WLTP ranges are always higher due to different testing protocols.
So we expected a range of around 300 miles for the Macan EV, and that’s what we got. Though Porsche also told us that range will be “10-15% higher in real world.”
These range numbers translate to an MPGe rating of 98, or 91 MPGe for the Turbo version. Both of these numbers are higher than any Taycan efficiency numbers, which is somewhat incongruous given the Macan is a larger vehicle.
However, other preliminary US reviews we’ve seen showed the Macan having high-200s mile range. We haven’t had a chance to do a range test on the Macan ourselves, yet, so we can’t confirm those numbers.
So, as usual, “your mileage may vary,” but it looks like the car will have more than enough range for buyers.
It’s also capable of 270kW charging, which Porsche says will allow it to charge from 10-80% in 21 minutes. This is plenty quick enough to fill up at a lunch stop, long bathroom + stretch break, or whatever else, and get you back on the road without significant delay.
In this day and age, quick charging speeds is really the more important thing to focus on anyway, and there are big changes on the horizon in that respect, with Porsche committing to NACS connectors in 2025.
However, despite the Macan EV being a 2025 model, it will retain the previous SAE CCS port, and will not use the NACS part for the foreseeable future. So you’ll have to stay tuned for more updates in that respect, including potential adapter availability (Porsche is currently not on Tesla’s NACS “coming soon” page, and the NACS rollout has been slowed by Supercharging chaos caused by Tesla CEO Elon Musk’s impromptu firing of the entire Supercharger team).
If our coverage of EVs has been helpful to you, you can use our affiliate link to contact your local dealers about the 2025 Porsche Macan, and ask them to put you in line for the Macan EV when it shows up at the end of this month.
Hydrogen-electric plane developer ZeroAvia has completed another successful financing round, led by some of its previous investors and some new ones. The sustainable aviation specialist plans to use the fresh funds to expedite the certification of its first powertrain and support selling its in-house components to other electrified aviation OEMs.
ZeroAvia has tasked itself with delivering 40—to 80-seat aircraft with up to 700 miles of range by 2027. So far, its sustainable technology has amassed some heavy hitters’ interest (and funding) in the segment to help push development forward.
In 2022, ZeroAvia secured over $30 million in funding, including investments from American Airlines, which joined Alaska Airlines and United in the hydrogen-electric plane venture.
2023 included several new partnerships and a fresh round of funding led by Airbus, Barclays, and Saudi Arabia’s “living laboratory,” NEOM. In late November, ZeroAvia announced a deal to provide up to 70 zero-emission planes to sustainable startup airline EcoJet, which looks to become the world’s first all-electric airline.
This past July, American Airlines committed to a large purchase of zero-emissions engines alongside a fresh investment in the aviation startup’s technology as part of a Series C fundraising round. Today, ZeroAvia announced it has extended upon that Series C round, which has now been completed for a total of $150 million.
ZeroAvia adds more names to its investment rolodex
The sustainable aviation company shared details of its extended financing round today. This included a 20 million euro £20m (23.7M euros) investment from the Scottish National Investment Bank, aka “The Bank,” which joins other investors like American Airlines, International Airlines Group (IAG), and ITOCHU Corporation.
ZeroAvia shared that the round was co-led by Airbus, Barclays Sustainable Impact Capital, and the NEOM Investment Fund (NIF). UK Infrastructure Bank joined as a cornerstone-level investor, and existing shareholders like Breakthrough Energy Ventures, Horizons Ventures, Ecosystem Integrity Fund, Summa Equity, Alaska Airlines, Amazon’s Climate Pledge Fund, and AP Ventures also participated.
The funding will enable the aviation startup to accelerate its progress toward certifying its first hydrogen-electric plane powertrain for commercial operations. Per ZeroAvia founder and CEO, Val Miftakhov:
We have closed an exceptionally strong financing round to help us deliver the clean future of flight for the entirety of aviation. As a purpose–driven impact investor, the Bank is an ideal partner for ZeroAvia. Scotland’s ambitious net zero targets, its strategic focus on hydrogen and its strong existing aerospace skills base make it an attractive place for ZeroAvia’s UK production operations as we scale into a major aerospace manufacturer.
In addition to locking in flight certification, ZeroAvia says the $150M in funding will help it begin sales of its in-house aviation technology, including electric motors and fuel cell power generation systems, to other companies.
ZeroAvia has already flight-tested a prototype of its first ZA600 engine, implemented aboard a Dornier 228 aircraft at its UK base, and its application for certification with the CAA is already underway. Additionally, the company has completed advanced ground tests in the US and UK of its ZA2000 system, which can someday help sustainably propel 80-seat regional turboprop aircraft.
That larger and more advanced propulsion system includes cryogenic tanks for LH2 and proprietary high-temperature PEM fuel cell and electric systems.
Tesla’s rollout of Full Self-Driving v12.5 has failed so far, and owners want to know what happens next?
In 2016, Elon Musk announced that all future Tesla vehicles would come equipped with the necessary hardware for self-driving capabilities, even specifying “level 5 self-driving,” which implies the ability to operate autonomously under any conditions. However, shortly after, Musk acknowledged that Tesla might require more onboard computing power than initially thought, leading to the introduction of Hardware 3 (HW3).
Musk assured that HW3 would enable full self-driving (FSD) capabilities, promising retrofits for earlier models that had purchased the FSD package. Following this, Tesla introduced Hardware 4 (HW4), a more advanced onboard computer system, but did not offer retrofits for older models with HW3, maintaining that HW3 was sufficient for achieving self-driving through software updates.
Initially, Musk claimed that FSD improvements would first be optimized for HW3, suggesting that HW4 might lag behind by at least six months. However, Tesla reversed this approach with the release of FSD version 12.5, which was first deployed to HW4 vehicles. Musk explained that optimizing the software for the less powerful HW3 would take additional time, hinting at the limitations of HW3 in handling the latest software advancements towards unsupervised self-driving, a capability Tesla promised to HW3 owners since 2016.
This rewrite aims to streamline the narrative, focusing on the evolution of Tesla’s self-driving hardware and software, and the strategic shifts in deployment and optimization of FSD capabilities between HW3 and HW4.
Musk said that it would take ten days to adapt v12.5 to HW3.
Not only was the update to HW3 late, but Tesla also confirmed that it was running a smaller model than on HW4.
On top of all that, now three weeks later, Tesla has yet to push v12.5 to the vast majority of FSD vehicles with HW3. Tesla appears to only have pushed v12.5.1.5 to some Tesla HW3 owners and it is now moving HW4 cars to v12.5.2.
Social media and Tesla forums are full of Tesla HW3 owners asking why they haven’t released a new update since v12.3.6 earlier this year despite Musk’s comments.
In its “AI roadmap” released last week, Tesla now claims that HW3 will get the same release as HW4 starting with v12.5.2 this month.
However, v12.5.2 is already in the consumer fleet for HW4 cars and v12.5.3 is already being tested in the beta fleet.
Electrek’s Take
This article is mainly to correct our article from last month that claimed Tesla was pushing v12.5 to HW3 since it turned out to be a very limited release.
Earlier this year, Elon said that Tesla was not compute-constrained for training FSD anymore. He also claimed that the training compute combined with v12’s full end-to-end neural nets would enable much faster software improvements.
And yet, the vast majority of HW3 owners have only received v12.3.6 this year.
That, combined with the fact that Tesla’s AI roadmap makes no mention of unsupervised self-driving whatsoever, and Tesla seemingly stopped promising it on new cars, has completely killed my hopes of Tesla delivering on its self-driving promises on HW3 cars and it has greatly limited by hopes of the same for HW4 cars.
I wouldn’t be shocked if Tesla fully shifts its self-driving strategy to the dedicated robotaxi, but I have no idea how they plan to make HW3 and possibly HW4 owners whole.