2007 Exige S Intercooler Air Flow Study

[lotusforsale]

Originally Posted by qball

Not sure I buy your argument...

the flow from the roof scoop will also be under pressure from airflow being forced in at speed.

Total effective outlet area of the IC (taking out all the area blocked by the bars and fins, etc.) is still going to be greater than the area of the roof scoop.

This could further be assisted by some fans pulling through the IC

Not sure how this would work in THIS setup, but I do know Robains question applies to the airflow in the N/A Exige engine compartment.

The first time I drove my 06 Exige with the top off at speed, I reached up to the 'mail slot' opening...which indeed is open on the 06 even though the roof scoop is not open, and at 60 MPH air was actually being forced out of the roof scoop AGAINST the oncoming air, presumably from airflow through the side scoops, from underneath through the NACA ducts, and however the air was actually moving through the rear mesh on the deck lid.

I actually pointed this out when the S's first came out, and folks posted that because the scoop system on the S is closed as air is directed through the IC, it wouldn't be an issue. But deliberately directing air from the side scoops into the IC shroud may reintroduce that problem.

Or not. Anxious to see how this works out.

[apk919]

I just got in from a 3 hour drive and a very hot track day, so it will probably be tomorrow before I post detailed info... but here's the condensed version:

In the middle of the hot (95°F) afternoon, I ran three sessions on track; side scoops open, side scoops taped shut, and side scoops open. The histogram of Intake Air Temperature (IAT) showed that with the scoops open, the average IAT was 30° cooler (157°F vs 187°F). Maximum IAT was also around 30° cooler.

More analysis (and cool graphs) tomorrow.

[shay2nak]

Been away for a week. Glad to see this! Nice work Andy!

IAT was still 157 with the additional piping?! I wonder what normal street driving would yeild, but 95 is damn hot to be running the car out even out on the street.

Canyon driving will yield high 90s IAT and with accel bursts IAT will be around 100-110. It would be nice to have IAT be around <90. Coolant temp is around 182.

I'm pretty sure putting a heatshield between the engine and intercooler will help also.

[apk919]

Originally Posted by shay2nak

Been away for a week. Glad to see this! Nice work Andy!

IAT was still 157 with the additional piping?! I wonder what normal street driving would yeild, but 95 is damn hot to be running the car out even out on the street.

Canyon driving will yield high 90s IAT and with accel bursts IAT will be around 100-110. It would be nice to have IAT be around <90. Coolant temp is around 182.

I'm pretty sure putting a heatshield between the engine and intercooler will help also.

Thanks Ara! I have some info from some street driving too.. temps are nowhere near that high. That's why the real test is a track day... it's really impossible to have the engine under 100% throttle on the street for such a large duty cycle. There's probably room for improvement... I just installed catch cans last week, so my intercooler is coated with an oily film from running 20k miles without them. If I was able to clean the I/C out, it might cool the charge even more...

[ZJChaser]

Nicely done Andy, it was good seeing you again today and the setup on the car. You gathered some good data.

[robains]

Andy,

Did you collect "baseline" data before any changes to the roof and side scoops?

Also interested in exactly how you "taped" the side scoops off.

Specific data I would like to see is (if you have it):

1. Side scoops functioning as normal stock side scopes (only larger) with stock roof IC duct and your inlet wholes taped off from inside the IC duct.
2. Side scoops routed to roof IC duct (open).
3. Baseline with stock side scoops and unmodified roof IC duct.

Ambients temps provided in all cases. Also, are you running the stock IC? And do you have rear bumper removed? Cat or no cat? How much boost were you running?


Rob

[Thomasio]

[usurbrain]

You can't get anything past Rob...

Originally Posted by robains

Andy,

Did you collect "baseline" data before any changes to the roof and side scoops?

Also interested in exactly how you "taped" the side scoops off.

Specific data I would like to see is (if you have it):

1. Side scoops functioning as normal stock side scopes (only larger) with stock roof IC duct and your inlet wholes taped off from inside the IC duct.
2. Side scoops routed to roof IC duct (open).
3. Baseline with stock side scoops and unmodified roof IC duct.

Ambients temps provided in all cases. Also, are you running the stock IC? And do you have rear bumper removed? Cat or no cat? How much boost were you running?


Rob

[apk919]

Originally Posted by robains

Andy,

Did you collect "baseline" data before any changes to the roof and side scoops?

Also interested in exactly how you "taped" the side scoops off.

Specific data I would like to see is (if you have it):

1. Side scoops functioning as normal stock side scopes (only larger) with stock roof IC duct and your inlet wholes taped off from inside the IC duct.
2. Side scoops routed to roof IC duct (open).
3. Baseline with stock side scoops and unmodified roof IC duct.

Ambients temps provided in all cases. Also, are you running the stock IC? And do you have rear bumper removed? Cat or no cat? How much boost were you running?


Rob

OK, El Exigente...

I would have like to have gathered OBD-II data before modification (I did gather airflow data, however), but I was having trouble getting the OBD-II scanner to not interfere with the gauge cluster, among other things.

Here are yesterday's test conditions...

Car: 2007 Exige S, with Lotus Cup roof, and fettled mailslot. No modifications to the rear panel. TWRD CF side scoops. Nothing else should affect engine/intake temps.
Engine: Factory stock except for following: Gotham265 tune, Moroso oil pan, catch cans.
Intercooler: Factory stock.
Exhaust: Factory stock.

Track: Streets of Willow Springs. Comparison tests were run CCW, ambient temp around 95 degrees, between 1:30 and 4:00 PM (I'm sure that the hourly temps at Edwards AFB were similar).

I taped off the side ducts with blue painter's tape. Tape was completely intact and still sealed after the test run.

EDIT: Note that the ducting was taped off, not the side scoops... see the previous photos to see the position of the ducts in the side scoops.

Here's a bit of the data from two consecutive sessions... the first is with the ducts open, the second with them taped off. The red trace is Intake Air Temperature, the green trace is Engine Coolant Temperature.

You can actually see the separate laps in the data... SOW has two (relatively) long straightaways, and you can see the IAT temps spike during the WOT periods...

[shay2nak]

damn that is a huge difference. I still think you should dump the stock IC. That thing is such a featherweight a lit match underneath it would cause it to heatsoak.

[apk919]

Originally Posted by ZJChaser

Nicely done Andy, it was good seeing you again today and the setup on the car. You gathered some good data.

Thanks AJ... good seeing you there too, and thanks for the helpful information on your setup as well as parts sources, etc. It's very cool to see the variety of solutions to forced induction/intercooling... everyone benefits from the shared information.

Oh yeah, that includes you too Frank!

[Thomasio]

FWIW, here are the raw #'s form my supplemental roof ducting airflow study:

Really Light Stuff - Carbon Fiber Exige Rear Window

None of this can be compared apples - to - apples, just too many variations... but it is interesting to see the benefits of adding an air pathway in addition to the "mailslot."

[apk919]

Originally Posted by Thomasio

FWIW, here are the raw #'s form my supplemental roof ducting airflow study:

Really Light Stuff - Carbon Fiber Exige Rear Window

None of this can be compared apples - to - apples, just too many variations... but it is interesting to see the benefits of adding an air pathway in addition to the "mailslot."

Cool, that's great data! It's interesting to see that at higher speeds the flow rates are not linear in your data... I haven't take any real data above 80mph

[Thomasio]

Originally Posted by apk919

Cool, that's great data!

While my setup pales in comparison to yours, I utilized a passenger to take readings while I carefully drove on private roads.

Re: non-linearity, I'm not certain why. We took about [edit: 100] data points, but there was about a 5 - 10 mph variable wind..

I need to think about the implications for a while. I'm blown away that the mailslot was that much of a restriction... blown away. With the supplemental roof ducting, I expected only a 50% gain in airspeed through the intercooler.

[apk919]

Originally Posted by Thomasio

While my setup pales in comparison to yours, I utilized a passenger to take readings while I carefully drove on private roads.

Re: non-linearity, I'm not certain why. We took about 200 data points, but there was about a 5 - 10 mph variable wind.. I need to think abou the implications for a while. I'm blown away that the mailslot was that much of a restriction... blown away. I expected a 50% gain in airspeed through the intercooler.

Yes, I think that the smallest cross section of the roof scoop flow path (in most cases, the mailslot) determines the flow through the ducting. When I opened the mailslot by 25%, I got a 25% increase in air flow.

Also, headwinds, tailwinds, and crosswinds add random noise to the data... that's why I used a square course to do my measurements, and then did a linear regression of the data to determine the most likely relationship between vehicle speed and I/C flow. I won't even get into the data integration and interpolation software that I had to write to put all the data together...

[Thomasio]

Originally Posted by apk919

Yes, I think that the smallest cross section of the roof scoop flow path (in most cases, the mailslot) determines the flow through the ducting. When I opened the mailslot by 25%, I got a 25% increase in air flow.

Also, headwinds, tailwinds, and crosswinds add random noise to the data... that's why I used a square course to do my measurements, and then did a linear regression of the data to determine the most likely relationship between vehicle speed and I/C flow. I won't even get into the data integration and interpolation software that I had to write to put all the data together...

Your data analysis is far better than mine, I just posted the raw data from a one hour drive at varying speeds. Luckily the wind was fairly low today. Each point in that graph is the average of about 10 data points. Excel drew its own lines. At some point I'll do the math and make a better graph.

Maybe the "octopuss" has merit!? Add both our systems together!

[apk919]

Originally Posted by Thomasio

Your data analysis is far better than mine, I just posted the raw data from a one hour drive at varying speeds. Luckily the wind was fairly low today. Each point in that graph is the average of about 10 data points. Excel drew its own lines. At some point I'll do the math and make a better graph.

Maybe the "octopuss" has merit!? Add both our systems together!

I think it does have merit... I like the "ductwork through the rear window" approach for purity; it makes more efficient use of the roof scoop without adding drag or stealing air from anywhere else. I've resisted it because it's a bit hardcore for a street car (plus I'd clearly have to ditch the glass rear window for one of your CF ones, or a Lexan/plexiglass one). For a pure track car, it's a great solution though.

It would be interesting to see the flow with both solutions combined...

[Thomasio]

Originally Posted by apk919

It would be interesting to see the flow with both solutions combined...

I'll contribute a window-delete!

[qball]

How about combining Thomasio's roof duct bypass, apk919's side ducts, plus two pull and two push fans on the I/C and two inline blowers from the side-ducts.

I bet we can get 3000 cfm through that IC.... LOL

[Thomasio]

Originally Posted by qball

How about combining Thomasio's roof duct bypass, apk919's side ducts, plus two pull and two push fans on the I/C and two inline blowers from the side-ducts.

I bet we can get 3000 cfm through that IC.... LOL

Every little bit helps... the thing is, with all of the ducts in the world, if you are sitting in traffic with no airflow your intercooler is heating up. Fans help there where no ducting can. Of course, the better the ducts flow, the faster you will cool your intercooler after you are stuck in traffic and get moving.