OK, wiseass, so you need some space to vent. Here it is. Direct letters to the editor to email@example.com and we'll do our best to come up with a snide response. Or completely ignore you. But hey, at least you've got a voice.
To help Justin make it through another 25 hours, we're giving away $500 worth of goodies from Samco Sport. As one of the world leaders in performance silicone, Samco Sport has supplied Formula One, NASCAR and World Super Bike teams. Their product range includes a massive selection of replacement hose kits, custom parts for race applications, hose accessories and their newly introduced blow-off valves. To follow in Justin's footsteps, send us your feedback to firstname.lastname@example.org, or to Sport Compact Car - Inbox, 2400 E. Katella Ave. - 11th Floor, Anaheim, CA, 92806.
Letter Of The Month
I read, and enjoyed, Andy Hope's piece on Thunderhill's 25-hour race last year (May '08 issue). In the article, Andy mentioned the damage that his team's car, as well as the Daytona prototype car, incurred. I was a member of the pit crew for Team Midnight Performance at that event. We also had some damage to our car during the race, so I thought you'd find this picture interesting. This was taken just after we finished the race (I believe we took seventh in the E2 class). Most of the front-end damage was from 'contact' with a Mazda Proteg. Note the large piece of steel stuck in the windshield on the passenger side. That was road debris kicked up by one of the Car and Driver cars. It flew clean through the window and hit the rear of the cabin at around 2 a.m. The driver didn't know what it was. He just heard a big bang and then some rattling around in the car. We just stuck it back in the window for the picture.
Return To Glory
While researching the wonders of Group B Rally, I came across something interesting. A little section said that FISA (then the name of the FIA's motorsport regulatory division) canceled the proposed replacement series Group S, which would have limited horsepower to 300. This would have been safer (as already argued when this news hit years ago). This is a travesty and I believe someone should start a petition, probably online (like everything else today), to convince the FIA to revive the Group S idea in all its glory.
Do you have any DVDs of the USCC for purchase? Please let me know. I have the first three and then it seems they were no longer available, but you still continue to hold the USCC. I'd appreciate your feedback. Thanks.
Yes, we continue to hold the Ultimate Street Car Challenge. The next event is gearing up and information on it will soon appear in an upcoming issue. You can catch video from some past events at www.sportcompactcarweb.com or at www.gtchannel.com. -JL
I really enjoy your magazine and my opinion is often swayed for the better by what I read in it. So I hate to say this, but I'm calling out Jay Chen. In the July 2008 issue, Chen writes... 10 ways to save gas: #5) Engine brake. An engine uses less fuel under full vacuum, when engine braking, for example, as opposed to idling. The deeper the vacuum, the less gas is used. I currently drive an '04 SRT-4 equipped with a ScanGauge. The ScanGauge can monitor multiple engine parameters and one happens to be gallons per hour (GPH) of fuel used. At idle, the engine is sipping 0.4gph. Following the rule above, I should be able to leave the transmission in gear when slowing for a stop sign and save fuel. The real truth is that coasting down from 60mph in fifth gear only cuts fuel flow to 0.6gph. Vacuum reads 20mmHg, which is the same as idle. So Chen says that more vacuum should save fuel. Negative, Ghostrider. Downshifting to third gear at 60mph increases vacuum to 27mmHg. Damn near perfect vacuum as far as engines go. The downside is that engine revolutions increase to 4000rpm and, in turn, fuel consumption increases to 1.6gph.
I'm glad someone has written back on this. It's a complicated matter that could fill more pages than this magazine has. Without looking at what the SRT-4's computer is doing, I can't give an exact answer, but the concept still holds true. The ScanGauge might be deceiving you. Here's the skinny based on what I know from when my Master's thesis and when I was doing emissions and CARB tuning. I'm assuming you understand about A/F and closed-loop operation.
When OEMs tune their ECUs, everything on the fuel injection side is based around a base fuel map or model that tells the injectors how long to fire for given the load and engine speed. Other modifiers all work off this map, like long- and short-term closed-loop corrections, acceleration or cold operation enrichment. These pulse-width modifiers are added or multiplied onto the base fuel values to make a car run if the cylinders are still cold (and the mixture is harder to light), air is less dense, or to squirt in more gas for acceleration.
The base map, however, is tuned for near-stoichimetric (A/F of 14.7:1) operation. Every engine and injector is different, so that's where the oxygen sensor and closed-loop function comes in to add or subtract from the base fuel map, compensating for any discrepancies. Of course, this is ignoring all the other sub-routines, emissions, driveability and idle air control interactions the ECU has to juggle.
When the engine is under negative load, such as when the wheels are driving the engine and the throttle is shut, the engine is pumping even less air than it needs when idling. More vacuum means less air, and at the same A/F, less air means less fuel. The base fuel map tuning values will always reflect this. Injector pulse widths are less at 27mmHg than they would be at idle at, say, 20mmHg.
On top of that, OEMs employ a gas-saving strategy called deceleration fuel cut-off (DFCO). Above, say, 2000rpm-with the throttle closed and vacuum below idle-the ECU will override the base fuel map injection values and shut the injectors off almost immediately. So, say you're engine braking to a stop or riding Fifth gear down a long hill, look at the injector duty on the ScanGuage (assuming it updates fast enough). It should say zero when you are in DFCO. Where and when this happens differs from car to car, as OEM tuners also have to deal with emissions implications as the fuel is cut off and the cylinder goes lean, plus the herky-jerky motion of negative torque. When you tip in the throttle, the injectors will begin fire.
Depending on how long the injectors have been off, sometimes additional enrichment will be added to the base pulse width to replenish the boundary layer of unatomized fuel on the intake port walls (this isn't a waste, it's just re-filling the bucket from which the cylinders are fed) to avoid lean misfires, again for emissions reasons. This stragety will also vary from car to car and between ECU algorithms. But just about all cars do this regardless of whether it uses an MAF- or MAP-based load measurement. The electronic BOV on the SRT's speed density-based Delphi ECU might add variations to the management method (most likely for emissions reasons), but it should still have fuel cut regardless. The throttle base BOV might be inhibiting DFCO.
My guess for your ScanGauge reading is two-fold. First, it displays fuel consumption in gallons per hour. That's a rough number (necessary for low load, to avoid confusing a driver with 200mpg with a little coast, or 0mpg at idle) and only reflects a fuel flow rate with little resolution. You should be looking at injection pulsewidth in milliseconds or miles per gallon, which are both better estimates of fuel economy.
The fact that you were decelerating on flat ground has obvious implications. Try repeating the same experiment down a two-to-three-mile-long hill in gear, then coasting in neutral (good luck finding that kind of elevation change in your neck of the woods). When the car (in neutral) hits terminal coasting velocity (rolling and aero resistance is equal to the force pulling it down the hill), look at the mpg and injector pulsewidth. Compare that to engine-braking down the hill in Fifth and wheel speed isn't changing. In gear, the injector duty cycle should be zero-meaning you are in DFCO and saving gas. If not, then the people at Dodge decided to open the idle air bypass to smooth out the ride and thus increase the vacuum to above the DFCO threshold. In that case, try the same hill in fourth gear. Regardless, DFCO can't happen at idle, which is why you save gas by engine braking. Automatic drivers are just out of luck.
The second issue I see is that the ScanGuage communicates through the OBDII port. There are five different OBDII protocols, each with different communication speeds, but on the SRT-4, the best-case scenario is one data read per second. That's an eternity in car speak when injectors fire over 10 times a second at idle. Add more channels, like wheel speed, throttle position, plus whatever else, and the system will slow down significantly. Even if the ScanGuage logs data at its optimal one scan per second and averaging fuel consumption over the 15 seconds you were decelerating to a stop, it's missing the thousands of corrections the ECU itself is processing between each one-second sample.
If you still think the pattern is full Ghostrider, get something that can communicate and log in real time, like an AEM EMS, and examine the log file after you're done driving. If you want to get really fancy, import the log file into Excel, numerically integrate the area under the injector pulse width curve, and multiply by injector size in cc/millisecond, times the number of injectors to get actual fuel consumption over a known time. Then take the distance traveled (by numerically integrating the wheel speed over a known time), divide it by fuel consumed over the same period and convert it from radian/cc to mpg. Then you have real-world mpg. Send me the log and Excel files, and I'll trade you for a sticker. Or you can trust me because I've done it before. -JC
Letters To The Editor
Our e-mail address is email@example.com
Our fax number is (714) 978-6390.
Sport Compact Car
2400 E Katella Ave., 11th Floor, Anaheim, CA 92806
Letters must contain the author's full name,address and home telephone number.
Letters may be edited for content and/or length.
Joey Leh, Editor-in-Chief
Jay Chen, Engineering Editor
Andy Hope, Senior Editor
Jong Cadelina, Art Director
Edward Loh, Editor-at-Large