Matt Armstrong Used Honda Radiators in a Bugatti… And It Worked

Added: 2026-05-13

[00:00:00]Oh, it's so smooth. It's scary as well at the same time.

[00:00:04]>> Today, we're going to drive the Bugatti Shiron for the first time [music] since the accident.

[00:00:11]>> We come so close in the last video to being able to drive it. Honda Civic radiators just cooled a Bugatti Chiron W16 engine, not as a temporary measure, not as a rough approximation, as a fully functional replacement for components that Bugatti refused to supply tested at operating temperature and confirmed working. 982° C, fans on, temperature held. Honda Civic in a 3 million pound Bugatti. And here's the thing that most people watching Matt's video felt but did not fully understand. That temperature reading 92 degrees and holding was not just confirmation that the radiators worked.

[00:00:53]It was the answer to the question that has been sitting unanswered since Alex drove this car without coolant 14 months ago. By the end of this video, I am going to tell you exactly what that temperature gauge was measuring, why the Honda Civic cores were actually the right choice, and the specific moment in the test where everything could have gone wrong and did not. But first, the part nobody explained. Why did Bugatti's radiators cost so much and why did Honda Civic cores replace them? The price gap Bugatti refused to supply replacement radiators for this car. 10 separate assemblies destroyed in the crash.

[00:01:30]essential for the W16 to run at operating temperature. If Bugatti had supplied them, the price would have been consistent with their general parts pricing model. The same model that quoted £300,000 to fix a gearbox fault that turned out to be a 5 lb fuse. Bob sourced Honda Civic radiator cores instead. Not Honda Civic Type R, not a performance variant. A base model Honda Civic, the kind of car that costs £12,000 new. That gets 40 miles to the gallon. That is bought by people who need reliable transport at a sensible price. And here is what most people watching the video did not realize. The cores were not a compromise. Bob visited PMA Automotive Cooling, a radiator manufacturer in the UK. He went through the factory, understood how radiator cores are built, and found cores that matched the Chiron specification. Not identical, matched. The Chiron's original cores had 75 fins per unit. The Honda Civic cores Bob sourced had 73 fins, but the fins were slightly larger.

[00:02:33]The same total surface area, the same heat transfer capacity, a different physical configuration producing the same functional result. That distinction, same function, different configuration is the entire engineering story of this rebuild compressed into one comparison. Bugatti's parts are not superior because of exotic materials or secret engineering. They are expensive because Bugatti controls who can buy them. Bob found the same engineering in a Honda Civic catalog. And now I am going to tell you what happened when those cores were tested at full operating temperature because the story of the temperature test is not as simple as it passed. What the temperature test actually involved here is what was actually happening during that temperature test. The W16 engine was started. Coolant began circulating through the system. The temperature gauge displayed in Fahrenheit because there was no Celsius option on the Chiron's dashboard began rising. 48 F.

[00:03:31]Cold start, normal, then 90, 100, 150. At this point, the thermostat was still closed. Coolant was circulating in a small internal loop. The radiators were not yet in the circuit. The temperature Matt was watching was the engine side coolant only. Then the thermostat opened. This is the moment everything Bob built was tested. When the thermostat opens, the full cooling circuit activates. All 10 radiators come online. The hot engine coolant reaches the radiator cores for the first time.

[00:04:06]The aluminium and tanks Bob fabricated from sheet metal are exposed to full thermal load. Temperature continued rising. 194 Fahrenheit 90 C. the target.

[00:04:17]Then 197 199 201 Matt's voice in the video at this point the fans are on.

[00:04:24]Come on, cool down. He was watching the thermostat work, watching the temperature try to stabilize, watching Bob's radiators respond to their first real thermal load. 201 F is 94 C. Just above operating temperature, the fans running at full speed, pulling air through the 10 radiator cores. Then the temperature dropped 199 198 192 Celsius stabilist. And here is why that specific behavior the temperature rising slightly above the target and then dropping back is actually the correct response. Why overshoot then stabilize is correct. What was really being confirmed every engine cooling system overshoots slightly on the first full warm-up cycle. This is normal expected engineered. When the thermostat opens and the full radiator circuit activates the cold coolant in the radiators mixes suddenly with the hot engine coolant. The system has not yet found its equilibrium. The fans activate at maximum speed. The temperature drops.

[00:05:28]Then the system finds its operating point. Thermostat modulates. Fan cycle.

[00:05:34]Temperature stabilises. For Bob's radiators, the overshoot to 984 C before dropping back to 982 is textbook correct behavior. If the radiators were underperforming, if the fin density was insufficient or the end tank volume was wrong, the temperature would not have dropped back after the overshoot. It would have continued climbing. The fans would not have been able to remove heat fast enough. the temperature dropped back, which means Bob's specific choice 73 slightly larger fins versus the original 75 smaller fins produced equivalent or better heat transfer than the Bugatti original. But here is what the temperature test was actually confirming beyond radiator performance.

[00:06:15]The thing Matt said I am hoping everything works because Alex drove this car without coolant. The thing that had been sitting unanswered for months, the water pump seals. Water pump seals confirmed. The Bugatti Chyron has multiple water pumps. Each pump has a shaft seal, a component designed to run permanently wet coolant, lubricating the seal material, keeping it within its operating temperature range. When Alex drove the car without coolant, those seals ran dry. The heat built, the seal material degraded. The question was how much? A compromised water pump seal does not always fail immediately. It can hold under cold conditions. Under brief warm-up cycles, the failure reveals itself under sustained operation at full temperature, which is exactly what the temperature test was. When the temperature stabilized at 92° and held there every water pump seal in the system was running under its full design load for the first time since the crash, the dipstick came out clean. No milky discoloration, no coolant in the oil, no sign of seal failure. the seals held.

[00:07:22]And here is what that means beyond this specific test on this specific car.

[00:07:27]Matt's Hurricane comparison, the one he made on the second channel, the one that revealed how much this possibility had been on his mind, the comparison was valid. The Huracan's water pump seals were destroyed by a no coolant drive.

[00:07:40]The Chirons were not. Possibly because Alex's no coolant drive was brief.

[00:07:45]Possibly because the Chiron's pump seals are more robust. possibly because the specific circumstances of that early test drive did not generate enough heat to destroy the seal material. Whatever the reason the seals held and the temperature test confirmed it. But there is one more thing the temperature test confirmed that nobody is talking about.

[00:08:06]The vacuum fill. Why it all matters. The vacuum fill. At 3:00 in the morning, after the coolant had leaked from the newly modified bracket, after Matt's TIG welding attempt had failed, after Bob had come back and fixed it, the coolant went into the system by vacuum fill. 39 L sucked in by negative pressure through a system with 10 separate radiators, multiple circuits, and numerous high points where air could be trapped. Air in a cooling system is dangerous. An air pocket at a high point creates a section of circuit where heat cannot be transferred. Metal in that area heats unchecked. In a system as complex as a W16 with 10 radiators and air pocket in the wrong location can cause a local temperature spike before the warning system responds. The temperature test at operating temperature sustained with the thermostat open with all 10 radiators in circuit was the first real confirmation that the 3 in the morning vacuum fill had been complete. No air pockets, no local hot spots, no unexpected temperature spikes in any part of the circuit. The fill worked. So, here's the complete picture of what that temperature gauge reading 92° C. Fans on, temperature held actually confirmed.

[00:09:22]Bob's radiator cores are thermally equivalent to the Bugatti originals. The water pump seals survived Alex's no coolant drive. The vacuum fill at 3:00 in the morning was complete. The cooling circuit Bob and Matt assembled with Honda Civic cores, Home Depot aluminium, Harbor Freight tools, and a vacuum filler they nicknamed Mrs. Suck functions correctly as a system. All of that in one number on a diagnostic screen. 982°. Here's what the Honda Civic radiators passing the Bugatti temperature test actually means beyond this specific car. Bugatti's parts monopoly depends on one thing. The belief that their components cannot be replicated by anyone outside their supply chain. That belief held by every Bugatti owner who has ever paid a service invoice just took a significant hit. Not because Matt Armstrong claimed the Honda Civic cores would work, because the temperature gauge confirmed it. at operating temperature underload in the actual car. That confirmation is permanent. It is on YouTube. It is timestamped. It is searchable by any Bugatti owner who wants to know whether the parts they were told are irreplaceable actually are. Bob did not just build radiators. He created a documented proof of concept that sits permanently in the public record. For every Bugatti owner who needs radiators in the future, there is now evidence that Honda Civic cores correctly specified and correctly fabricated are thermally equivalent to the manufacturer's original. And the manufacturer cannot delete that evidence because 982° C does not lie. Honda Civic radiators cooled a Bugatti Chiron W16.

[00:11:05]982° C. Fans on. Temperature held. Three things confirmed in one reading. Water pump seals intact. Radiators thermally equivalent. Vacuum fill complete. Bob sourced the cores from a Honda Civic catalog because the fins matched the specification. He fabricated the end tanks. He machined the fittings. He rebuilt the radiator after transit damage. He came back when it leaked again. And the temperature gauge confirmed every decision he made was correct. Matt said it on camera. I think the Honda Civic rads have worked. He was right. Tell me in the comments when Bugatti watches that temperature gauge stably at 92° on a car they refuse to supply parts for. What do you think they feel? Because the Honda Civic that cooled a Bugatti is now part of the permanent record and it is going to be there long after the invoice for 300,000 gearbox repair has been forgotten. Here is the final thought on what Bob actually built. He did not build replacement radiators. He built evidence. Every future Bugatti owner who needs radiators, every Bugatti technician who quotes for them, every journalist who writes about manufacturer parts monopolies has a reference point now. Honda Civic cores correctly specified, correctly fabricated, 92° C.

[00:12:25]The evidence is permanent, searchable, free to access on YouTube, and the manufacturer who refused to supply the parts has no mechanism to remove it.

[00:12:36]Subscribe because the radiators passed.

[00:12:39]The first drive happened. The finish line is close and 92° C is the most expensive temperature reading Bugatti has ever witnessed.