and the intakes for the engine and the air pump.
In contrast, the intake for the M42 is on the left side , behind the headlamp and kidney. The scoop leads directly into the front of the air box through the smaller end of a tapered plastic cone. This is a more direct route than the M44, yet they both make for a somewhat restricted intake, eliciting comments like, "sewing machine sound " and prompting many people to upgrade to K&N Cone filters and the like. The effects of this type of modification on M44 engine output is open for debate and testing, but most people report a more aesthetically pleasing sound and apparent benefits in throttle response.
I've modified the airbox in my '97 ti sport (M44) to utilize both M44 and M42 intake pathways, simply by cutting a hole in the front of the airbox, removing it's innards, and adding the M42 outer plumbing. To complement the 'free-flow' condition sought, I added a drop-in K&N filter. The modification I made results in less of a 'growl' than a cone filter, but does impart a deeper smoother sound; I think I feel an improvement in throttle response but without at least a G-Tech, I hesitate to make any claims. It has the added advantage of appearing like an almost stock installation, and maintaining all the post-filter instrumentation and dynamics of the stock airbox.
Parts Needed: 1 Mr. Gasket Air Cleaner Adapter Ring, P/N 6404 or 6406 (Pep Boys, $5) 1 M42 Intake Scoop P/N (forthcoming from FLBEARCUB) (BMW, $14. I got mine from Bob Collins, Thanks Bob!) 1 K&N Air Filter P/N 33-2733 (mail order, ~$30) Tools needed: straight bladed screw driver 10 mm open end wrench 4" diameter hole saw, and drill with proper sized chuck ruler file (or Moto-Tool with grinder if available) utility knife chalk or sharp-tipped utensil for marking/scoring clear or black silicon sealant single edged razor blade
1. Disconnect the sensor wire, the intake hose, and the air pump hose from the top of the air box, then unlatch the 4 clips and remove the airbox cover. 2. Remove plastic cage/cone /foam assembly from airbox as follows: - on side of airbox closest to left fender, push down and in on second cage bar from the top, to disengage it from 2 'teeth' in the airbox (teeth can be felt through the foam). -pull up on cage assembly while pulling in on small latch above cone. Lift out cage and foam. 3. Remove 2 nuts holding lower airbox to fender (you may have to work around brackets and cables of the cruise control and ASC+T systems); work bolt/bushings out of the holes in the fender mounts, then on other side of airbox, disconnect the plastic duct by squeezing the ratcheted lever and pushing it down. Airbox should be free to remove now, but leave it in place. 4. Cut flange of Mr. Gasket Adapter Ring to 4 3/4 " in diameter, using utility knife or other appropriate tool. 5. Fit M42 air intake scoop into place between airbox and intake holes in car frame. The side of the scoop with the 'lip' should be towards the radiator. The flexible back end of the scoop should rest flush and snugly against the front side of the airbox. Score or mark at least a semi-circle around the scoop end onto the airbox. This is the hole you will be cutting, and you'll need to find the approximate center of it. Lift out the airbox. 6. Having located the center (the flange on the Mr. Gasket will give you about 1/4" error margin in any direction except down; watch the lower edges of the airbox!), drill a small pilot hole, then CAREFULLY!! and slowly start the pilot bit of the hole saw. As the saw snugs down to the surface of the airbox, check your positioning again; make sure no part of the saw is going off the lower edge of the box. Slowly cut the hole, controlling the drill so that you don't damage the other side of the box when the saw breaks through. 7. Clean up edges of hole with the razor blade and file, making sure to clean all plastic fragments out of the airbox. Center the Mr. Gasket over the hole in the airbox and mark the sections of raised 'ribs' on the box that need to be removed to allow the adapter to sit flush. File or grind off the minimum amount of the ribs necessary. Smooth the edges of the punched out plastic disc, and retain it, in case you ever want to 'revert' to stock by duct-taping it in on both sides! 8. Re-install the lower airbox, re-attaching it to the fender, and to the duct work on the other side. 9. Assemble scoop to Mr. Gasket adapter and reposition in front of the airbox. It should be a snug fit, and just the compression of the scoop will hold it in position for normal driving. 10. At this point you may reassemble the airbox, leaving out the lower innards, and installing the K&N in place of the stock filter. 11. I drove around a while to decide if I liked it, then removed the scoop and siliconed the Mr. Gasket to the front of the airbox. I let it cure about 14 hours, then carefully reattached the scoop.
The first 4-5 times I revved the engine past 4300 rpm, I got a scary, clacking noise from the area of the intake plenum. This went away quickly: my guess is that OBD-II was trying to make some adjustment for the increased air flow. My impressions after 1 day of driving it: - slight, pleasing growl when throttle is blipped.. - pleasing, deeper overall sound, very smooth, not over-bearing. - pretty sure she has a bit more "scat!" when you goose her. - refined, BMW-like sound, as compared to a cone. - looks stock!, or darn close!
I think this dual intake setup provides many of the benefits of a cone filter installation, without the attendant "hot air under the hood" and "heat shield " issues. It was relatively cheap and easy. Bob Collins donated the scoop from his airbox and the Mr. Gasket. The most expensive part was having to buy the 4" hole saw and mandrel (~$30 at Home Depot), and the hardest part was finding a drill that accepted the 7/16" shank of the mandrel. I finally went to a friend's wheel shop, and we ran the saw with a 11mm socket attached to his air wrench! So far I'm very pleased with the results; I'm gonna run the setup over 3000 miles in the next 2 weeks (trips to Mississippi and New York, from Florida) and see how it holds up.
I corresponded with Brian Brown for a few weeks after doing this, and with his consent, I've included portions of our dialog about the efficacy of the modification.
RC: I'm sending you this message because I respect the technical expertise and insights you give us on the ti list, and because I know that like me, you have an M44, and have been pondering effective modifications. I thought you might be interested in this mod that I've done, based on comparing M42 and M44 airboxes at the Florida ti-fest. I'd really appreciate any comments you might have about the mod, especially any possible negative outcomes from increasing air intake via this method (and any thoughts about the clacking sound, now gone). Does this mod actually, effectively, increase air availability to the engine?
BB: This seems to be a reasonable approach. It uses two of the factory's high pressure areas that are effectively shielded from picking up debris and water. While I haven't personally experimented with the air intake myself yet, I don't believe the stock M44 intake or exhaust are a significant restriction to the engine. I think this is the main reason the cone filters (and rear exhaust sections) don't show meaningful dyno gains. OBD-II isn't 'preventing' gains. I'm suspicious that the ends of the runners in the intake manifold are the main restriction in the entire air passage (this will need testing and verification). Changes in other areas of the air tract probably won't be very useful until the main breathing limitation is addressed. I think that because of European noise emissions laws, BMW tried to make the engine as quiet as possible. It appears that they did a real good job at cutting noise without hurting the performance. They did fall a bit short at producing an aesthetically pleasing sound. The lack of 'sound power' probably contributes more than anything to the ti's often perceived lack of quickness. (General Motors has an entire department with the mission of making their cars sound powerful while keeping them quiet.) If you like the way your mod sounds, then that's a bonus. It would be interesting to get a 'numbers' comparison on your mod. Third gear acceleration times would probably be the best way to make a before and after comparison. Stationary dyno tests aren't good for this type of comparison because they lack the wind pressure on the front of the car (unless a wind tunnel is also used). (I think that this factor is often overlooked on other cars where a cone style filter does show a dyno gain). I think your approach could be more of value on an engine with internal mods that increase its air requirements.
RC: After 3 and a half weeks on the road and over 5000 miles, I finally got a chance to run and time the car with and without the dual intake mod. I ran the tests as follows: - 10 timed acceleration runs from 2500-4000 rpm on ~2 mile stretch of flat, remote divided roadway. 5 runs in each direction; car was held steady at initial rpm for ~5 seconds then launched at WOT (wide open throttle) and timed until final rpm was reached. - 10 timed runs from 4000-6000 rpm on ~7 mile stretch of highway. 5 runs in each direction, same protocol.
After completing the "dual intake" runs I covered the secondary hole in the airbox with the cut out piece and duct tape from the inside, and reinstalled the plastic cage and baffling. I drove around for 20-25 minutes, varying my speed and doing several trial acceleration runs as outlined above to allow the ECU to recalibrate. I then repeated the above test protocol for the "stock" intake runs. Possible confounding variables (aside from timing error) may have been light wind gusts, and a possible 2-3 degree ambient temperature drop between the dual and single intake runs. The results? Figures never lie and liars never figure! According to the stats I ran on the data, there was no significant difference in acceleration time in either rpm range between the stock and modified intake. Don't you HATE being right ALL the time???!!! : )
BB: I wish I was. (Author's note: This guy is TOO modest!!!)
RC: The mean values for the runs with the stock setup in both ranges were actually a little lower, but as I said, not statistically significant. Now, despite the numbers, in just driving around immediately after "undoing" the mod, my ears and my butt told me the car was not as quick. Funny thing, tha' !
BB: I continue to be impressed at how sensitive the 'butt dyno' is to sonic horsepower. Even when I know for sure that it's not the case, a louder car often seems quicker to me. I'm starting to think that it might be worthwhile to make comparisons while wearing some high attenuation hearing protection type ear muffs.
RC: I did all the test runs at WOT, as it was the only consistent throttle input I could come up with; I wonder if the mod may not perform differently at partial throttle positions, and how I could test this.
BB: I think that part throttle operation would be affected even less by air intake mods. Don Wentz has run comparisons between stock and a cone filter. He has told me that his ti unquestionably runs slower times with the cone filter.
I still think that your approach is one of the most interesting ideas I've heard for increasing the area of the air intake, especially since it retains the stock inlet points and a somewhat stock appearance. While it doesn't appear to be useful for an otherwise stock engine, it's possible that it might still be a useful approach for a modified engine.
Epilogue: After taking the car back to a "stock" intake configuration and finding that the mod gave no tangible performance enhancement I was a little disappointed, and left it stock. While washing the car two days later I suddenly got the urge to revert to dual intake. I went for a drive afterwards and my grin returned! Moral: Dammit Jim! I'm a wildlife biologist, not an engineer - but I know what I like! And on the track, where so much is mental, feeling faster probably makes you a little faster. In theory, it's possible to add all of the M44 intake plumbing and shrouding to an M42 to get the same effect, but until additional engine mods that can utilize greater air flow are available, it probably would not be a cost-effective route. As Brian's comments and those on recent list traffic indicate, most of our intake and exhaust mods will continue to be aesthetic and cosmetic until other restrictions in the engine's breathing are addressed. Thanks to Brian Brown, Bob Collins, Ken Weingold, Stephen Wolff, and Albert and Vincent of The Driver's Side, Gainesville Florida, for help, parts, ideas and advice!
http://www.318ti.org/notebook/M44_dual_intake/
September 17, 1999