
Starting in first gear
Good morning. Somewhere in North York, Ontario, a 1973 BMW coupe is wearing an engine from an E46 M3 and a set of Motorsport stripes it did not earn at the factory. Bidding sits at $60k and the auction closes at 2:36 PM EDT.
Elsewhere, China's largest EV maker stood up at a truck show in Hanover and said it intends to become a European company, and the people who run Le Mans are working out how to refuel a race car with a liquid at minus 253 degrees Celsius in under two minutes. Neither problem is solved. Both have a schedule.
In today's newsletter, we'll get into:
Auction Desk: A widebody 3.0CS with an S54 heart, closing at 2:36 PM EDT
One Big Story: BYD wants a European address, and it's starting with trucks
Engineer's Corner: Liquid hydrogen at Le Mans, from the tank outward
Garage Economics: What a Le Mans program costs, and what hydrogen changes about the bill
Let’s roll.
THE AUCTION DESK
A 1973 BMW 3.0CS that borrowed its engine from an M3
The E9 coupe left Karmann's line with a carbureted 3.0-liter straight-six making about 180 horsepower. This one now carries the 3.2-liter S54 from an E46 M3, rated at 333 horsepower in the donor car, bolted to the M3's six-speed manual and a limited-slip differential through a custom driveshaft. It was offered on this platform in August 2018 as a lightly modified Polaris Metallic car and bid to $34,500 without meeting reserve. The current owner bought it in 2024, stripped it, and built what's on offer today: a Zender CSL Group 2-style widebody, Nardo Grey paint with Motorsport-style stripes, Air Lift 3P air suspension, carbon Sparco buckets, and 18-inch Alpina-look wheels on 225/45 front and 285/30 rear rubber. About 1,000 miles have been added since. The cluster reads in kilometers and shows under 61,000; total mileage is unknown.
The engine work is more thorough than the paint. The S54 was resealed in late 2025, with the VANOS unit and rod bearings replaced and Cat Cams camshafts fitted, and it breathes through a Turner carbon CSL-style airbox with a matching tune. Cooling was upsized with a Z3 M radiator plus oil and power-steering coolers, the brakes were rebuilt front and rear, and the seller lists adjustable camber plates and fabricated chassis reinforcements.
The line I read twice was the rod bearings. They are the S54's known weak point; the factory clearances run tight, and the E46 M3 community treats a bearing swap as scheduled maintenance. Seeing it done during the reseal separates a swap I'd bid on from one I'd admire from across the parking lot. The line I'd want more of is "fabricated chassis reinforcements." The E9 shell was never stiff and it rusts from the inside out, and a widebody on 285-section rears with air suspension loads it in ways Karmann never planned. I'd ask where the steel went before I asked about the stripes.
The Ontario registration is not the obstacle it looks like. At 25 years a car clears federal safety rules and at 21 it clears federal emissions, so the swap is invisible to Washington. It is not invisible to Sacramento. This coupe lives at the other end of Highway 401 from where I grew up, and I've done that border crossing with a car. The paperwork is a morning, not a month.
My call: Widebody plus air suspension narrows the buyer pool to people who wanted exactly this; the S54 and the documented engine work widen it back out. There is a reserve, and nobody outside the seller's house knows where it sits. This chassis has already been through one no-sale here, at $34,500 in 2018 before any of this work was done, and the current bid already sits $25,500 above that. Against $60k with hours to run, I'll say bidding ends between $72,000 and $85,000, and whether that clears the reserve is the seller's call, not the market's.
📈 Market signal: Retail listings for stock 3.0CS coupes run from the low $60,000s for driver-quality cars to roughly $100,000 to $155,000 for restored examples, and recorded sales average in the high $60,000s. Restomod asks stretch past $200,000, but heavily modified E9s tend to change hands well below their asks and well below their build cost.
ONE BIG STORY
China's biggest EV maker wants to be European. It's starting with trucks.

Source: BYD
On Monday, at the IAA Transportation show in Hanover, BYD executive vice president Stella Li said the company will launch its first heavy-duty truck in Europe next year and, over time, build in Europe everything it sells in Europe. "We become a European company," she said. The truck comes with a fleet package: financing, solar-powered charging, a workshop network, and mobile roadside assistance.
The passenger-car half of that plan is already under construction. BYD's factory in Szeged, Hungary, is due to start mass production next year. It exists because the EU put countervailing duties on Chinese-built electric cars in late 2024, at 17% for BYD on top of the standard 10% import duty. Build the car in Hungary and the duty disappears.
Trucks are different, for now. The countervailing duties cover passenger cars, and Chinese-built trucks currently face only the standard duty. Traton's MAN brand and other European makers have asked Brussels for an equivalent truck tariff. Li's answer on Monday was that BYD doesn't like the tariff, has to deal with it, and considers it short-term pressure until local production comes online. Read that as a company describing a plan it has already priced.
Why it matters:
The fleet buyer doesn't care about the badge. A logistics company orders on cost per kilometer over a million kilometers, and the person signing is not the person driving. With diesel up through 2026, one consultancy scenario puts an electric heavy truck's combined purchase, energy, and operating cost over a million kilometers at roughly half a diesel's. Treat the figure as a scenario; the direction is the point.
BYD is not alone. Six Chinese truck makers were counted as planning a 2026 European entry. Sinotruk and SuperPanther already assemble electric trucks at the former Steyr plant in Austria, and Skyworth, Sany, Dongfeng, and Foton are all on the Hanover floor this week.
The incumbents asked for a wall and are getting a neighbor. Brussels is weighing whether to tie electric-truck subsidies to European production, which would push every Chinese entrant toward exactly what BYD just announced. Demand isn't the problem: a Dutch electric-truck subsidy of about $95 million launched in January and was oversubscribed in one day. Price is the problem, and price is what Chinese makers are good at.
The takeaway: The word "European" in Li's sentence means a factory address, not a headquarters. A tariff is a line item that expires when the plant opens; a workshop network with roadside trucks is a fixed asset that doesn't. BYD is building the second thing while everyone argues about the first.
ENGINEER’S CORNER
How Le Mans plans to race on a liquid at minus 253 degrees
On June 11, during race week, Kazuki Nakajima drove Toyota's TR LH2 Racing Prototype around the full Circuit de la Sarthe. A demonstration lap, but the first by a liquid-hydrogen car, and it came the same month the ACO and FIA confirmed hydrogen cars in the top class from 2030: 1,200 kg minimum weight, dedicated tires, and an Equivalence of Technology system to keep a hydrogen car and a Hypercar within reach of each other. Manufacturers are expected to start designing before the end of this year.
The engineering starts with the tank. The FIA chose liquid over compressed gas because liquid hydrogen at minus 253°C is roughly 70 grams per liter, against about 40 for gas at 700 bar, and a race car has neither the volume nor the mass budget for the gas option. The cost of that density is a cryogenic vessel that has to survive a crash. Conventional cryo tanks insulate with a vacuum layer; the ACO's hydrogen consultant, Bernard Niclot, says new insulation materials may eliminate it. The tank rules are due to freeze by the end of this month, with full regulations next year.
Downstream of the tank, the rules don't pick a power unit. A hydrogen combustion engine is an adaptation of what teams already know and likely wins on weight and packaging, but it runs lean, so it wants more displacement, and lean mixtures make cool exhaust, which leaves a conventional turbo little to work with. A fuel cell has no hot exhaust, so nearly all of its heat goes through the radiators, meaning bigger radiators, more coolant, and more drag. What the fuel cell gets in return is a lower feed pressure, so it can run without a pump in the tank.
That pump is the detail I keep coming back to. Anything running in liquid hydrogen at minus 253°C cannot use conventional lubrication, and a pump with no oil in it is a bearing problem, a seal problem, and a thermal-cycling problem at once. The combustion route needs someone to solve that pump; the fuel-cell route needs a fuel cell engineered for racing from a blank sheet, which nobody has built.
Then there is refueling. The target is one to two minutes per stop. Toyota's liquid-hydrogen Corolla at Fuji showed why that is hard: between stops the transfer pipe warms up, so it has to be chilled with liquid hydrogen before any fuel goes into the car, with the boil-off returned to the supply truck through a second line. Gasoline cars will probably have to slow their refueling to keep the race fair. Niclot expects two to four cars when the rules land; Toyota and Ferrari have been inside the working group, and Hyundai is the name he floated next.
The takeaway: The hydrogen rulebook is being written around the tank, and the manufacturers will write the engine. Whether a hydrogen car can lap Le Mans is settled; Toyota did it in June. The open question is the ninety seconds in the pit lane.
GARAGE ECONOMICS
What a Le Mans program costs, and what hydrogen does to the bill

When the Hypercar class was drawn up in 2018, the FIA and ACO put a target on it: about €20 million per season for a two-car factory team, €16 million for a privateer buying a car, and the number was cut on purpose to get IMSA and the manufacturers to sign. That was the brochure. Figures now commonly cited for a full factory program sit in the €30 million to €40 million range, with development budgets higher again. I've been close enough to a small manufacturer's racing budget to know the brochure number is the entry fee, not the bill.
The manufacturers keep coming anyway. Nine brands raced the top class this June, and Ford and McLaren arrive in 2027, which puts at least ten on the grid. In July the FIA extended the Hypercar homologation cycle, originally due to end in 2029, through the end of 2032. A ten-brand grid with six years of rules certainty is what a €30 million check buys.
The other side of the ledger is why the ACO can keep raising the price. Attendance has climbed every year since the pandemic, from 244,200 in 2022 to 350,105 this June, when Toyota beat BMW by 10.9 seconds after 5,190 kilometers. The centenary in 2023 drew a reported 113 million television viewers across 196 countries, and Rolex, TotalEnergies, Michelin, and Goodyear sit on the partner board. That is a ticketing product, a broadcast product, and a sponsorship product with a manufacturer roster that keeps growing.
Hydrogen changes four line items:
For manufacturers, it is a second program, not a cheaper one. A 2030 hydrogen car sits on top of a Hypercar budget, and only a company with a road-car hydrogen strategy can charge part of it to the road-car side. That is why Toyota and Hyundai are the names in the room.
For the ACO, the constraint is infrastructure, not demand. "More than six cars breaks the refueling system" is a capacity limit on liquid-hydrogen supply and pit equipment, and the organizer pays for that, not the entrants.
For the incumbents, the cost shows up in the pit lane. If gasoline refueling is slowed to keep hydrogen cars competitive, the brands that didn't build a hydrogen car pay for the ones that did, in seconds rather than euros.
For everyone, the 2032 extension is what makes 2030 rational. Nobody spends a hydrogen budget to enter a field that resets two years later.
The takeaway: Le Mans sells manufacturers a stage that 350,000 people attend and 113 million watch, and it sells fans more badges on one grid than any other series can manage. Hydrogen is a bet that the road-relevance story can be renewed for another decade without the grid shrinking. The ACO has paid for the fuel station. Which manufacturers pay for the tank is the open question.
🏁 THE COOL-DOWN LAP
The XM from Monday bid to $63,000 and didn't sell. Reserve not met. The bid cleared the top of my $55,000 to $60,000 call. A relist is the usual next chapter.
Brampton is four days out. Stellantis's September 20 deadline for Brampton Assembly is Sunday. The watch-number from #16 has run out of runway.
Goodwood Revival runs Friday through Sunday, with the Bonhams sale on Saturday. The results will show where E9-era BMW money sits in the UK.
IAA Transportation runs through Sunday in Hanover. Tesla has a European-spec Semi on the floor, Volvo claims up to 700 kilometers for the FH Aero Electric, and the Chinese contingent is the largest the show has had.
The gasoline side of the Le Mans grid is not waiting. McLaren and Ford both have 2027 Hypercars in testing, and Ford's runs a 5.4-liter Coyote V8.
Share the Downshift. Forward them this email. Every subscriber from your link gets you closer to the Daily Downshift starter-pack. Unlike the XM, it has no reserve.
That's a wrap. See you Friday— same inbox, lower gear.
Was this email forwarded to you? Subscribe free at Daily Downshift. Drop a gear. Read the good stuff.
Consider this the feedback loop: reply with what worked, what didn't, and what you want more of. Reader replies are the single biggest input into what gets written next.


