The race at Phoenix was the first non-restrictor-plate race. A number of drivers experienced engine-related problems, leading some media outlets to start blowing the “EFI problems” horns as loudly as possible. Mark Martin, the pole sitter, was an unfortunate casualties of a “flipped circuit breaker”. One of the most interesting exchanges to me was a series of tweets and a radio interview with Mark Martin’s Crew Chief Rodney Childers (@rchilders55) in which Childers repeatedly said it not “an EFI problem”, the radio commentators persisted in saying that it was.
Here’s his verbatim tweets (and if you’re not following him on twitter, please do!)
Man I hate that!! We had a breaker POP on our ECU for the fuel injection about half way. Which makes it switch to a safety fuel map.
It popped about half way.. it didn’t affect the performance. Just the Mpg, which made us have to pit. But we are really happy..
That made our mileage go from 4.2 to 3.8… And no way we could have made it. Good job to mark and all the Aarons guys though.
The EFI deal isn’t really the issue.. probably a wiring issue that we have to figure out. We had the same deal happen in practice.
Each car has a relay box, which acts sort of like Mission Control. I’ve said before that the ECU (Engine Control Unit) is the brain of the EFI system. The ECU collects information from a number of sensors located in different places on the car that measure things like humidity, pressure, temperature, and air-fuel ratio. The ECU makes decisions on what to do next based on the information it gets from the sensors, and it acts on those plans by sending messages to the rest of the car through the relay box.
What happens if one of the sensors stops working (or a wire breaks and the sensor, even though it works, can’t send information)? If the ECU believed the data it was being given, it would have a rather warped view of reality and could start telling the car to do all types of goofy things, including some that could actually damage the engine beyond repair.
The EFI system is smart – smart enough to know when it is getting suspicious data. Instead of acting on that data, a relay is triggered and the ECU changes over to an alternate engine map (what Childers called a ‘safety fuel map’). The alternate map isn’t optimized for performance – it’s purpose is to allow you to keep running, but your performance won’t be as good as you would have with the information from the sensor and the optimized engine maps. As an analogy: You eat more efficiently when your eyes are open. If someone blindfolded you, you could still eat, but it wouldn’t be as efficient (or clean) as with your eyes open. This relay system protects the engine from the actions of a confused ECU. You might not run as well, but you also might not destroy a whole engine.
The relays work on the basis of a voltage threshold. They’re like an electron bouncer. A relay has a maximum current or voltage it will tolerate. Bob Pockrass reports
these relays have limits at 5V and 100 mA. If an electrical signal comes through with a higher voltage (or current, depending on the type of relay), the relay switches everything to an alternative circuit. Unlike the circuit breakers in a house (which are normally on/off, in contrast to these, which are circuit 1/circuit 2), it’s very difficult to re-set the relays. It’s not like there’s a switch that the driver or a crew member can reach down and flip back. When they flip, you’re pretty much stuck with them for the rest of the race.
Teams will be looking for anything that might cause the relay to flip. That could be a wire that vibrated loose, or an electrical spike (like, say, the spike you get when you turn a switch on or off perhaps…?) I understand some teams are switching to boxes without relays – but then you take a chance that something gets out of whack and you blow up your engine entirely. One of the real pains in the neck with a short or a transient spike is that they can be hard to reproduce. You can do the same thing 10 times and the problem may only happen one time out of those ten. Makes for some long nights for the engine shops.
Childers tweeted that the mileage dropped from 4.2 mpg to 3.8 mpg – that’s just about a 10% drop in efficiency. Phoenix is a one mile track, which means that the change in fuel mileage cost them nine laps per tank of fuel. When teams are scrapping for one-percent increases, a 10% decrease is just going to kill you.
Keep in mind that we’re in uncharted territory. NASCAR teams are subjecting these systems to environments they haven’t seen before. I expected a few problems like this at Phoenix and I expect a few more in Las Vegas. Teams will figure out how to make their system withstand the high-vibration/high-temperature world under the hood of a race car and drivers will gradually learn which of their techniques still work with the new system and which ones they’re going to have to change.
And if you’re wondering exactly what an engine map is, I’ve got a video going up on Friday to explain it.