Photography Credit: Paris Van Gorder
Dynamometers are great tuning and research tools, but they do have their limitations. The primary limiting factor of a dyno is that it’s static. It tests power at the wheels while all other parts of the car are static, which is a condition you’ll never experience on any track we’re aware of.
So you end up getting a highly repeatable picture of power, but one that’s not always applicable to the real world, where air flows differently than it does coming out of a fan, the hood is closed, and the loads involved reflect the actual physics of the situation.
Luckily we have the closest thing to an accessible real-world dyno just two exits away in the form of Daytona International Speedway. This world-famous track at the World’s Most Famous Beach® hosts the SCCA’s Track Night in America program a few times a year, where normal shlubs get a chance to run the Rolex 24 course flat out for just $250 (before discounts). It’s one of the great values in track time in the country right now, especially for an event with such a fun, community vibe.
Photography Credit: David S. Wallens
But the long, full-throttle sections of Daytona’s banking also make for excellent real-world dyno pulls, which is what we were concentrating on during last Thursday’s TNiA that we participated in with our BMW 435i project. We’ve been using our MHD tuning suite heavily to log and monitor engine data during our sorting process–it was instrumental in showing us the dire need for better oil control, for example–but to this point we haven’t used it to do much actual tuning.
[How did our BMW 435i’s new oil-control measures work at Daytona?]
The MHD package includes several “off the shelf” tunes, which tweak performance parameters like boost, fuel and spark curves, and cooling strategies. To this point, we’ve been running its Stage 0 setting, which retains all the stock performance parameters but allows data logging of the ECU and real-time monitoring of OBD functions. For this Daytona event, we’d run through Stage 0, Stage 1 and Stage 2 tunes for our three sessions, each one hinting at unlocking slightly more sauce from the N55 engine.
If you recall, we’d dyno tested our engine after our first visit to BimmerWorld and found 301 horsepower at the wheels–awful close to the 303 horsepower at the crank BMW claims. Of course, BMWs are notoriously underrated for power, so we weren’t surprised to see our engine producing a likely 330-plus horsepower at the crank after measuring.
[Do we have an engine oiling problem with our BMW 435i?]
But we also knew that the N55 has a reputation for being fairly maxed out on power when using the OEM turbo. The N55’s twin-entry single turbo is fairly small. It’s designed to spool fast and function transparently, producing torque and throttle response but not necessarily big top-end power. And it does just that. In fact, we find our best performance when exploiting the 3500-5000 rpm range, frequently using a gear higher than we feel like we should in corners, and shifting well below redline. It feels a bit counterintuitive at first, but the VBox speed traces show that’s the driving style that produces the best thrust.
Still, more boost has to produce more power, right?
The results–as they say on the internet–may shock you.
We’re looking at a couple key data points here, boost being one of the most important. We pulled a snapshot of the ECU data from the long run from the exit of the Le Mans chicane, through the east banking, and down toward the start/finish line and Turn 1. Peak boost was around 10 psi, and the intake air temperature–which ran at ambient idle around 115° degrees on this 89° day—was in the 135-145° range during the pull. Trap speed at the start/finish line, according to the VBox, was 150.1 mph.
Nice. Let’s throw a tune on it and let it cook.
Wait, no. Not like that.
“Cook” is the operative word here, because with the Stage 1 tune in place, producing around 14.5 psi of boost in the same spot, the IAT had skyrocketed to 167°. Finish line trap speed was 148.7 mph, implying–if not outright showing–that we were making less power with more indicated boost.
For our final session, we installed the Stage 2 tune (all of these tunes rely on good-quality, 93-octane fuel, which we were using) and boost numbers increased even further. Peak boost was now 15.7 psi, IATs had now jumped up to the sous-vide-pork-chop temperature of 176°, and finish line trap speed dropped again to 146.8 mph.
Although our lap times were better during each subsequent session, a quick peak at Circuit Tools’ delta graph shows the truth. We were simply faster in the corners and more aggressive under braking on subsequent sessions. In the long acceleration zones, the tuned car would at best stay even and usually lose time to the stock flash, but it would make it up in the corners as the track gained grip and traffic thinned out toward the end of the day.
Most notable was that long pull from the former Bus Stop to the start/finish line. The spiciest tune fell off dramatically in acceleration above 100 mph as the IAT climbed due to the increased atmospheric pressure in the intake tract.
The green trace represents the Stage 2 tune, and the red trace represents the stock tune.
So what’s going on here? Why is more boost making less speed?
Well, the IATs give us the most likely path of exploration. Intake temperatures like that are almost certainly going to make the ECU pull timing to reduce the chance of detonation, thus taking away any of the power gains we might have enjoyed from increased boost.
The logical answer here might be increased charge cooler capacity, but it’s not really as simple as that and there are a few things to consider here.
For one, the stock intercooler is light. Like, crazy light. It weighs barely 7 pounds, and when a piece of hardware is located well in front of the front axle, lighter is better. Aftermarket intercoolers with their huge heat sinks weigh in at 28 to 30 pounds–not because they’re bad, but because they leverage the additional surface area of lots and lots of cooling fins to cool the intake charge more efficiently. Their mass helps them do their job, but it’s also a lot of additional mass in a fairly non-favorable location.
[Can we get faster laps in our Mk7 GTI by upgrading the intercooler?]
The other thing we’re dealing with here is a LOT of additional heat. Even in the Stage 1 trim, we’re producing 15% more intake heat the first time we go full sustained boost. That’s before we even get into any heat soak issues as the rest of the engine bay heats up. At Stage 2 we have 20% more heat to deal with to get back to our stock IAT temps.
That’s a lot of additional heat for a passive heat exchanger to deal with, even a very efficient one. So while we’re certainly going to have some serious discussions about an upgraded charge cooler and will likely do some testing, our expectations are tempered given the data we’ve gathered so far.
Eww canned tunes on track...yikes. I won't run them on the street. Typically just a throttle map and maybe an increase to load or boost targets. The worst of them removing the load based targets and just going boost.
A proper load based tune with more allowance for timing over boost should get better results with less heat. Add a gallon or two of E85 and then you may be able to do both.
If you do spike with some E85 you can bump up any of the pesky IAT/ignition modification tables. You'll get the power back.
The steady state part of tuning is easy- making power is incredibly straight forward. Can take time, depending on the number of variables that need to be dealt with, but it's not hard.
Most of the OEM calibration time is spent on starting and transients. Heck, all of the many thousands of miles driven during development is about transients.
E85 is the way!
So did these tunes not make more power on the dyno as well? Or are you saying heat soak on the track just negates the gains? You mention that the n55 is already pretty maxed out. I'm sure the MHD claimed numbers are on the highest octane- 95/102 ron and cold car, ideal day. Dipping down to 93, especially on the track if going to lose you some power. I know my s55 was pretty sensitive to heat even on a stock tune. I could see the vmax difference on long straights at Road America and when I had to fill up with 91 instead of 93 the difference was noticeable, even on a stock tune for 91, let alone pushing the poor stock turbo so hard.
Looking at the MHD numbers stage 1 vs stage 2 is 30hp and assumes an upgrade intercooler. That might be a reasonable expectation and mostly due to the intercooler. I think the biggest take away here is that the car has inadequate cooling to support more power.
Andy Hollis said:E85 is the way!
Not for Tuner it isn't :(
theruleslawyer said:So did these tunes not make more power on the dyno as well? Or are you saying heat soak on the track just negates the gains? You mention that the n55 is already pretty maxed out. I'm sure the MHD claimed numbers are on the highest octane- 95/102 ron and cold car, ideal day. Dipping down to 93, especially on the track if going to lose you some power. I know my s55 was pretty sensitive to heat even on a stock tune. I could see the vmax difference on long straights at Road America and when I had to fill up with 91 instead of 93 the difference was noticeable, even on a stock tune for 91, let alone pushing the poor stock turbo so hard.
Looking at the MHD numbers stage 1 vs stage 2 is 30hp and assumes an upgrade intercooler. That might be a reasonable expectation and mostly due to the intercooler. I think the biggest take away here is that the car has inadequate cooling to support more power.
Yeah I think that's going to be our first target: getting IAT's down. I concur with zdnet's assessment of canned tunes, but this exercise ws all about finding hurdles before we put the time and expense into doing a custom tune, and those IATs are a big hurdle. I'd love to be able to fill the thing up with corn sauce, but the SCCA says no E85 in the TT Tuner division, so we're stuck with pump 93. Once we feel like the hard parts are squared away, though, BimmerWorld has a great tuner they work with and we'll get them logged in and working on our ones and zeros.
JG Pasterjak said:Yeah I think that's going to be our first target: getting IAT's down. I concur with zdnet's assessment of canned tunes, but this exercise ws all about finding hurdles before we put the time and expense into doing a custom tune, and those IATs are a big hurdle. I'd love to be able to fill the thing up with corn sauce, but the SCCA says no E85 in the TT Tuner division, so we're stuck with pump 93. Once we feel like the hard parts are squared away, though, BimmerWorld has a great tuner they work with and we'll get them logged in and working on our ones and zeros.
Is that Epic Motorsports?
FWIW, I had great experiences using the canned Epic tune on my E46 M3 for several years. I later had them do a custom remote tune which didn't change the peak power but did pick up about 10 ft-lbs of midrange torque.
...and then later I had them do another custom tune that took ~ 80 horsepower off the top end so that I could race it in NASA ST4. :)
You can easily test intercooler capacity on the dyno - we saw the stock Mazdaspeed Miata drop 15 hp (IIRC) on back-to-back-to-back dyno runs as that factory IC heatsoaked. In fact, the dyno is GREAT for testing heat capacity simply because there is little cooling airflow.
In the article, you talked about the heat sink capacity of the intercooler. I don't think that's really what you're after - more heat capacity just gives you more time until the IC is heatsoaked. What you want is improved heat transfer from the IC to the air. That could be a more efficient core design (not necessarily heavier) or improved airflow. Just throwing a heavy IC in there for sheer mass is not the right solution.
Intercooler or airflow upgrades done right will probably help, but isn't that just a bandaid for a turbo that's now operating further and further out of its efficiency range? Great to be able to cool the charge before it hits the intake manifold, but you'll still be running very high exhaust manifold pressures, turbine inlet temperatures, compressor out temperatures, etc. Probably not an issue on the street, but for significant track use that stuff would be in the back of my head.
The stock intercooler may be enough for a more appropriately sized turbo pushing a few more pounds of boost. At what point does it make sense to make that change first?
gearheadE30 said:Intercooler or airflow upgrades done right will probably help, but isn't that just a bandaid for a turbo that's now operating further and further out of its efficiency range?
Yes.
Unfortunately, replacing the turbo would propel us into a TT division requiring sawzalls, and we really want to preserve the streetiness of this car as much as possible. So we're trying to find the most efficient band-aids possible at this point because the tiny little turbo on this thing is super efficient, but has very little headspace. We know that we're going to hit a hardcap on power pretty soon, but we're not there yet, so I think it's worth looking for what little low-hanging fruit remains.
Displaying 1-10 of 35 commentsView all comments on the GRM forums
You'll need to log in to post.
