82 mm throttle body
 

Hey everyone, I wanted to share some pics of a project I'm finishing up. But first, a little back story...

2 years ago I was in the market for a sports car and stumbled upon a real garage queen - 00 base with 27k miles on it. I snatched it up and had Pat Williams do an IMS replacement for me and have loved the car ever since. Recently I moved to a rural area with wide open back roads and found myself wishing the car had a bit more oomph (mostly in terms of acceleration). So, the quest for more power began.

I know what everyone is thinking at this point "oh boy, not another one of these guys". To that end I will only say that I'm not doing this for racing purposes and I'm not looking to make the car spit flames out the back. I'm a tinkerer by nature and a bit of a scientist by trade so when really studying these cars it seemed to me that there are some obvious points that can changed and possibly improved. Plus, I'm curious to see what will happen lol. So, here we are, with the first part of that mad idea - bigger intake.

Here are the parts, it's really quite straightforward truthfully. Just a bit radical in execution
Early 996 plenum (for cable driven TB)
82 mm throttle body (cayenne turbo)
2 x 45 degree 3.5" silicone intake elbows
A short segment of 3.5" stainless
3.25" MAF housing (identical size to 996 but part is aftermarket Mercedes)
3.5" T shape short ram intake for GM (mostly found on C5 vettes)

Here are the porsche "speciality tools" used
3# mason hammer
4" hole saw and reciprocating saw
Factory jack (thanks porsche)

On to the execution
Gut the intake (drop the motor to do this, you can get the stock box out intact)
Strip the trunk and relocate the ECU a bit to gain slack on the wiring harness
Use mason hammer and jack to deform the cross member downward by 3/4 inch to gain enough clearance
Use hole saw to cut opening into firewall (gasp?!?!) between trunk and engine bay and use reciprocating saw to open up the too of the engine bay enclosure a bit.
Assemble and mount components

There are a few minor things from that point - opening up the wiring harness to get slack on the MAF sensor wiring to reach the new housing, extend the AOS hose to reach new attachment position, trim out the rough opening with rubber u moulding, and knock out the vents in the trunk to provide adequate ventilation.

And now, in the name of "pics or it didn't happen"...

Now look, I'm sure there are gonna be guys who are gonna gripe about not having dyno numbers (they will come but not until later in the project, lots more to come eventually) and criticize for loss of low end torque (which I realize is a very really possibility). But like I said, this is driven by a love of tinkering and a very itchy "what if".

The next step in this madness will be a 2.5" exhaust setup, and then plenty more after that. It's gonna take me a while to work through everything so bear with me, but I promise to keep pics rolling as the progress happens. https://uploads.tapatalk-cdn.com/202...6c279e7a4f.jpghttps://uploads.tapatalk-cdn.com/202...64c472d88d.jpg

Sent from my POCOPHONE F1 using Tapatalk

Looking Great so far!
Please keep us posted

Cheers!

Could you elaborate on this a bit? Where is your cold / fresh air coming from?

For now I've punch out the two vents in the rear of the trunk on the sides. These were there to vent the trunk when it heated up and communicates with the area where the stock muffler was located. If this isn't enough I can remove those popouts on the trunk floor to achieve the same effect. What I will ultimately wind up doing is using ABS to fabricate a dedicated "trunk space" and wall off all the intake components, which I am going to route down and out of those knockouts to twin turbos mounted below. With two young kids and a third on the way though, that part of the project is still quite a ways away lol. https://uploads.tapatalk-cdn.com/202...141b7f9679.jpg

Sent from my POCOPHONE F1 using Tapatalk

Thanks Muskoka! Btw I chimed in on your motor situation. My heart goes out to you.

Sent from my POCOPHONE F1 using Tapatalk

I stuck a 3.4 intake/throttle body and 3.8 997 exhaust manifolds on my 2.7 rallycross boxster. I can feel that i lost some low end torque but gained higher rpm HP. I'm rarely below 4000 rpm so it works for me. I moved the air filter inside the trunk and I'm surprised at how quickly the filter gets dirty. I've raised the trunk lid in hopes of drawing in more fresh air but I still need to find a way to get more fresh air inside the trunk. The air coming from the rear wheel well is very dirty. Granted I'm racing in dirt, so you should have a lot less than me.
http://986forum.com/forums/uploads02...1611627453.jpg
Here's my build thread

http://986forum.com/forums/show-tell-gallery/73877-6-months-live-build-thread-2.html

Very cool dude! I'm a big fan of your work, I nearly cried when I got to the end of your original build thread :(.

One idea I had was to utilize the rear wing (I stole the idea from lambo) drill holes in the front face of the wing so that when air hits it it will be directed down and into the trunk. With enough perforations there will probably be plenty of airflow there. The problem obviously is water there, but if it's a dedicated track car I would think it might work well.

Sent from my POCOPHONE F1 using Tapatalk

Who or what is lambo? I've been thinking some kind of scoop too. Just enough scoop to pressurize the trunk to keep dirty air from entering elsewhere inside the trunk. Any water that would enter would leak out the bottom of the trunk and shouldn't be able to enter the motor.

http://986forum.com/forums/uploads02...1612017481.jpg
Ideally something like this is needed for the environment I race in. I'm not sure I want to take on building something that big and complex though

Wow! You guys have taken this to a whole other level! Very cool! :cool:

The lamborghini aventador has an active air management system that is ridiculously complex and equally impressive. One of it's components is the rear wing, the pillars of which are hollow and when appropriate will open up vents to channel air downward through the pillars. We obviously don't have pillars on outlet wings, but the body of the wing is hollow right? And the bottom of it communicates with the trunk. So you could theoretically turn the front of the wing into swiss cheese and it would (hopefully) pressurize enough to force air into the trunk.

Btw, those vents on the side should have a flapper on them to keep dirt out. I don't know if you still have them but it may help putting them back in to keep the dirt from the wheels out.

Sent from my POCOPHONE F1 using Tapatalk

Thanks man! I'm a huge fan of your videos! The series you did on your lowering springs gave me the info I needed to install my coilovers, and your intake videos were super cool. I was sad to hear you toasted the motor but I'll bet your loving the 987. How does is compare to the 986? Ever find yourself wanting to go back?

Sent from my POCOPHONE F1 using Tapatalk

Yeah, I was reviewing one of my videos and, you know, I do have rather a soft spot for that `04... I'm just glad that Vanessa, a racer chick, decided to buy it. More track days to come! :dance:

I probably did take those off as they're right next to my exhaust now. I'm going to seal them off with some sheet aluminum and raise the front of my trunk a little higher. If you ever look at your car after it's driven in the rain, you'll see the air coming off the side of the B pillar swings back towards the center of the trunk creating a slight higher air pressure right where the raised portion of the trunk would be. In theory, making a great location for a scoop.
http://986forum.com/forums/uploads02...1612076507.jpg

Did you do anything to seal the ecu from dirt and water?

Sent from my POCOPHONE F1 using Tapatalk

The dme is tucked up under and I've never seen the need to. I've had boxsters at my yard with no trunk lids and the dme stayed dry even during the worst rain storms. The immobilizer I do have inside a bag and disconnected from the floor though. I have it sitting upright between the rocker panel and seat. I also have holes drilled in the floor so water cannot collect inside the floor pans. So far everything is staying dry. Before I reinstalled the windshield, the fuel gauge would stop working when it would get rained on. Immobilizer aside, these cars are pretty solid electrically. I've power washed the whole car inside and out. I wouldn't recommend cleaning the interior of your street car with a power washer. lol

That's good to know! I'm a huge fan of rhino lining the interior of off-road vehicles, but I'm going for a bit more a subtle look lol.

If you get on Amazon there are a ton of aftermarket hood and side scoops. They're primarily used by wannabe ricers but can function as a real scoop and they don't cost much. Instead of raising the trunk lid you could just cut a giant hole in it and mount a scoop over the hole. Kinda the poor man's version of your pic earlier. The aero images of our cars that I have seen show that to be a relatively low pressure area but I'm sure it can't hurt.

On a side note, when you flashed to 996 row, did your 986 guages still work appropriately? Whenever I get an hour off of daddy day care this is next on my list to do. I've read some say everything was fine, I've seen other say not so much. Reading wayne's article on cluster swaps it seems like after flashing to 996 I can go back through piwis and actually change the cluster back to 986 while keeping the 996 tune, but I haven't seen anyone actually detail doing this.

Sent from my POCOPHONE F1 using Tapatalk

Well lookie lookie - I was reading into oil cooler swaps/upgrades and found this pic. They're using it to vent an external oil cooler in a 987. That would do the trick though.

https://uploads.tapatalk-cdn.com/202...2cb6c8b45f.jpg

Sent from my POCOPHONE F1 using Tapatalk

I have an automatic cluster in my manual car with a 996 ROW tune. Everything works but the check engine light is on and it's not on when I had the 3.2 ROW tune. Honestly the temp gauge is the only one I care about though.
I might do something like that. The more I think about it, the more I like the dragster scoop sticking way high. Instead of the intake drawing air from the trunk, id point it straight up. I'd build another engine cover that the dragster scoop would then attach to. Not only would it get a slight tunnel ram effect, the extra air not getting ingested by the motor will help with cooling the motor. I have an overheating issue when racing in 90+ temps. It's in the 90's or higher for half the year down here. My only issue with the dragster scoop would be that I would probably have to modify the roof. I really don't want to do that because when time comes to part the car out, that roof would become worthless.

edit; Another benefit of the dragster scoop would be that hot air would escape easier from the engine bay when the car is not in use in between runs. I thinking I'm about to ruin a perfectly good roof. lol

Roof scoops are awesome, like the lotus exige. Definitely a complicated build but I think that your logic is spot on. Keep us posted!

Sent from my POCOPHONE F1 using Tapatalk

http://986forum.com/forums/uploads02/aero1613188530.jpg
This "computer generated" aerodynamic picture somewhat shows what I'm talking about. That one line that starts on the hood and flows along the top of the door is close to what I see water spots do on my street boxster after driving in the rain. On my car it takes a sharper turn after the B pillar and flows right above the front corner of the trunk.

That's been my experience too. I actually dont mind driving with my top down in light rain so long as I'm above 35 mph because the aerodynamics keep the cockpit dry lol.

I wonder if it would work well to put a louvered panel or scoop along the top edge of the quarter panel to catch that line along the side. It shouldn't be difficult to route that air into the trunk and based on your racing pic it would probably be pretty clean and slightly pressurized at speed.

One thing I'm getting ready to do is to add a gt style spoiler (purely for look, guilty). With that in place though I think I will remove the stock spoiler assembly and cover that opening with a gutter guard that has a diamond mesh pattern. There won't be any ram air effect, just one more way for the trunk to "breathe" as I continue to add more stuff back there.

Sent from my POCOPHONE F1 using Tapatalk

http://986forum.com/forums/uploads02...1613312585.jpg
If I stay with the M96 motor, then I was going to use this scoop. By what I could tell from a rough measurement it should fit without too much modification to the roof. I believe it would fit through the rear window and the intake would be just above the top of the roof. About right where the red line turns to blue in the aero pic

That should do it. I would definitely try to get the horizontal aspect of that as far forward on the roof and possible so that it's not right at the break line where air separates from the roof line. It may be worth it to place some vortex generators in front of it too to keep as much air in their line as possible. How are you gonna duct the air into the inlet from that point though? Are you gonna plumb it to the TB or just use it as a breather for your filters?

Sent from my POCOPHONE F1 using Tapatalk

I was going to turn the intake tube with the TB as far forward as possible. I'm hoping the air filter will fit within the area after the neck.

That would be sweet, just plastiweld it to the top of the engine cover and route everything up. If it's the right size and shape that will be a hell of an end result! It will be tricky to get it through the back window though, maybe cut it in two, pass it through the window, and then plastiweld back together? Or just fab up a 2 piece window that assembles around it.

Sent from my POCOPHONE F1 using Tapatalk

Hey woody, question for ya. I've been running now for a few hundred miles on this setup with 996 tune and big exhaust and I noticed that the car now likes to move! More specifically, around 4200 rpm at WOT she pushes me back in the seat, which I don't ever remember happening before. Weirdly though, it doesn't happen every time I pass 4200rom at wot. I can't quite pinpoint what is contributing to this happening (or not happening). I'm wondering if maybe the DME is still figuring stuff out.

I know most of your driving is dirt track, but I'm curious about what your experience has been so far? Big gains at high rpm? Are gains always there?

Btw, how's the shop build going?

Sent from my POCOPHONE F1 using Tapatalk

Every m96 I've driven seems to come alive between 3500-4500 and falls flat after 6000. There does seem to be a bigger jump in power at the 4000 mark with my 996 intake 2.7 rally car than with my 3.4 street car. I don't think it takes that long for the computer to learn and adjust. I have to disconnect the battery on my rally car at the end of the day. Every time I reconnect the battery, the car idles weird for about five minutes but seems to run fine after that. I don't know why you're feeling a jump sometimes and not the other. Aging fuel pump?

The building is going up pretty good for an old man working by himself. I start on installing the sheet metal tomorrow.
http://986forum.com/forums/uploads02...1619251786.jpg

Ike
I have experienced the same as you in regards to the torque curve around 4200 RPM under WOT or near WOT.
Every now and then it is like there is a little more power.
My car is a 2000S completely stock.
I thought that perhaps my MAF was going bad.
Or that my resonance flapper was hanging up.
But everything checks out ok.
The car runs great.
The only thing I could come up with is:
I can only get 91 Octane fuel.
So my thought is that the DME/ECU is holding back ignition timing to prevent pre-det or knock.
I think when I feel the little bit of extra power is when the ECU/DME is allowing a little more ignition advance.
Keeping my eyes open for some higher octane fuel to test my theory.
Thoughts????

Looking good so far! Doing metal by yourself is a ********************. I recently covered a barn in metal, it was hard. I hope you don't have any wind coming your way!

That's interesting about your idle. When my dme resets, I will idle a bit fast (around 1200 rpm) for a minute or two but it has always been smooth and then runs runs down to 650ish after that. I did notice though that for the first 100 miles or so I would get a bit of a wandering idle while driving in town, but that has stopped. That's what makes me think that it takes a while for the long term trims to really settle in.

Sent from my POCOPHONE F1 using Tapatalk

Woody is right that in general the torque curves for the 2.7 and 3.2 don't start to peak until past 4k. I had never noticed the "kick" at 4200 before I did my intake and exhaust mods though. That's definitely interesting that your stock s does the same thing.

At first I wondered if it could be variocam switching over at different points depending on throttle and load but from everything I have been able to find that it always a fixed transition at 5500rpm. I posted recently about this topic but I haven't gotten any responses yet.

I have read similar comments about the timing but I'm not sure about it. I think that the base tune has ignition advanced to as early as stuttgart deemed "safe" and the only adjustments the dme will make is to pull timing if knock is detected. I've not seen any documentation that the dme will advance timing further if the fuel is higher quality. I've also not ever read that the dme will advance timing beyond the base map under specific circumstances such as wot or rpm threshold.

With that being said, you could be right that at a high enough rpm and load the timing becomes such that 91 will knock and then the timing gets retarded. I run 93 but it's the same tank that's been sitting since before I started all this work last fall, so I'll be curious to see what happens with the next fill up.

The other thing I wondered was the fuel maps in regards to open vs closed loop. Woody and I are running the 996 tune on 2.7 and 3.2 motors, respectively, and we should be encountering rich conditions when the dme flips to open loop mode. I would think that I should be so rich in open loop mode that it would probably hurt performance, where woodys probably wouldn't be effected much. I don't know of any way to actively monitor what mode the dme is running in though, unfortunately.

Sent from my POCOPHONE F1 using Tapatalk

I agree that the DME is most likely set up with the most advance that is safe under the right conditions.
I wonder if timing advance in regards to Octane or rather knock is learned in a similar way as fueling maps are????
I have read that today"s knock sensors pick up pre knock that you or I would never detect.
On my OBII scanner it shows rough running feedback for each individual cyl. also ignition timing for each individual cyl. So the DME is pretty adaptable.

Since fueling maps can be learned my guess is that fueling even in open loop conditions is learned. With the narrow band O2 sensors used on these cars they are constantly going back and forth from closed to open loop. WOT, and deceleration are two examples of when they will go from closed to open loop. So it makes sense to me that fueling would be learned even in open loop conditions.
So you may or may not be running rich in open loop.
A way you can monitor this is to hook up an OBDII scanner that shows open-vs- closed loop and watch it change back and forth as you drive.
I have done it several times with my scanner.

So today I got to test my AFR guage. Its actually a solenoid controller from innovate motorsports that integrates a Bosch wide band afr sensor. What I saw was rather interesting (and a bit concerning).

On startup at idle, AFR was running 15.1-15.3! Too lean! Under light loads this would fluctuate back down 14.5-14.9. At higher loads this would go lean again, 14.5-15.3. At WOT it dropped way rich, down to 10.5-11.3.

Interestingly, with foot off gas, it would peg lean (over 22).

Numbers in the drive home were a little less lean but Instill saw 15+ a fair bit of time.

I can explain the WOT richness - switching into open loop, the computer is fueling a 3.4L engine. Even with the biggest air intake in the world that ****************s just not gonna match up on a 2.7

The closed loop operation afr bothers me though. I would assume that with as sophisticated as our computers are (combined with what I perceive to be a general cover-your-ass approach from Porsche in regards to performance aspects of this car) that the computer would shoot for 14.7 or less at all times. I was really quite shocked to see 15+ so much.

With that being said, if the computer is gearing the fueling to hit a target above 14.7, I think that may be just one more piece of the puzzle as to why we see issues like bore scoring, overheating, etc.

I did have my battery disconnected this weekend so maybe it just needs a few days to relearn short term trims again.

Anyway, I'll keep things updated over the next few days to see if things change much.

Sent from my POCOPHONE F1 using Tapatalk

Ike
That all seems like pretty normal numbers to me.
A couple of things to keep in mind:
14.7 fuel air ratio is not what the engine needs under all conditions to perform best.
An example is WOT.. it should go rich. Heavy engine load requires a richer condition.

Foot off the gas your getting 22: exactly you don't need fuel in the air to coast. your getting 22 because you don't need fuel under that condition but the engine still pulls in air.
Hell you don't need spark. Some modern day engines cut off fuel and spark all together under off throttle conditions. An AFR sensor will see all air (or nearly so) under off throttle conditions.

Another thing: 14.7 fuel to air is the ideal ratio for Catalytic Converter performance.
Not engine performance under all conditions.
On a modern day computer controlled engine the primary function of the DME/ECU system is "Catalytic Converter performance, engine performance is secondary.

The DME/ECU under closed loop conditions constantly fluctuates the fuel air mixture above and below the 14.7 fuel air ratio. That is why O2 sensor signals constantly fluctuate in a sine wave. The DME/ECU sort of looks for an average under closed loop conditions.

Hey blue,

Thanks for the response. That's interesting about the sine wave pattern, which is exactly what I'm seeing. That seems a bit less efficient than a "hold steady" approach at control but maybe it's more effective to continually correct.

That's also interesting about the cutting off of spark while coasting. It makes sense to do so for fuel economy, as long as the alternator keeps spinning to keep up with current load demand.

The reason the lean tendency was surprising is just the excess heat and likelihood of detonation under those conditions. I see your reasoning though that if the motto is "emissions first, engine second" then it makes sense to do so.

With that being said though, the numbers on my drive this morning were much more centered around 14.7, especially while cruising under moderate load. I really wonder if the short term trims just needed 20 miles to adjust the mixture. I suppose 20 miles of lean isn't too bad considering the detonation sensors/ignition retard built in to the dme.

I think it's fascinating to learn more about how these computers actually do their job. Unfortunately, Bosch me 7.2 is kinda like this black box with very little info about the logic and processes involved. I don't think I've seen a single thread on this forum which is technically detailed regarding this (beyond basic principles of open/closed loop, short/long term fuel trims). I can't even find a good one on nefmoto, which is probably the best site on the web for that type of info.

Anyway, for me the enjoyment will have to be postponed. Looks like I'm leaking oil from the passenger side bank. Probably plug tubes, I should have replaced them when I changed the plugs and coils over the winter. Back on to the stands we go!

Sent from my POCOPHONE F1 using Tapatalk

IKE
Most people tend to think that the O2 sensor signal drives the DME/ECU.
But what really happens is the DME/ECU drives the O2 signal.
The O2 signal on these cars operates between a low of around .1 millivolt to a high of about .950 millivolts. Stoichiometric or 14.7-1 is at .450 millivolts.
So the DME/ECU goes a little rich to drive voltage down then goes a little lean to drive volt up. Constantly. The DME/ECU uses the response from the O2 sensor as feedback (closed loop) info to determine fuel requirements or fuel trim to drive the O2 sensor signal.. Which in the end is about Catalytic Converter performance.

Stoichiometeric or 14.7 lbs. of air to 1 lb. of fuel is the ratio at which there is exactly enough molecules of air to combust ever molecule of gas.
So 14.7-1 produces an exhaust that is most easily cleaned by the Catalytic converter.
To rich (excess gas molecules) Cats plug up. to lean(excess air molecules) Cats burn up.

But 14.7-1 it is not the ratio at which fuel air burns the coolest or produces the most power.
That is why you see a richer mix at WOT or high load and leaner mix at idle or very little load.
Engine performance needs vrs Cat converter needs.

Most of the time cars are operated in the lower end to middle of the cruise range as far as engine load goes. That's why you see the air fuel ratio closest to 14.7-1 in that range. 14.7-1 meets engine needs and Cat needs equally in that range.

The shutting off of fuel and or spark at off throttle conditions does save fuel but it is more about pollution.
With the constrains put on automobile manufactures by the EPA everything (as far is engine performance goes) is about pollution control. Performance (power) or fuel mileage is secondary.
EPA restrictions are the biggest reason you can get more power and sometimes fuel economy from an after market tune. It is also why after market tunes are illegal in some states.

Blue, thanks for the knowledge brother. Always more to learn about this stuff. I've been working through an automotive textbook to get a better understanding of different systems and this convo definitely put the fuel chapter next on my list.

I will say that the last drive I took (15 miles home) the afr stayed 14.5-14.9 the whole time during closed loop operation. I guess the stfts really did play into that. Which I think is interesting because it implies that the dme is programmed to run lean until it can dial in 14.7 more precisely, or that this setup provides more air than the stock 996 setup (less likely since the MAF should be able to dial this in precisely without the help of the fuel trims). Thinking this through though it may just represent a difference in MAF housings since I am using a 3.25" diameter honeycombed pipe from a BMW V8 intake - same diameter as the 996 but maybe with fewer curves in the piping I am flowing air more efficiently.

I'll have the battery unhooked while replacing the plug tubes so I'll be curious if it goes back lean for the first drive after that.

Sent from my POCOPHONE F1 using Tapatalk

82 mm throttle body
 

Edit. Just sent Ike PM on ROW tune)

Could you kindly advise where to get a 996 ROW tune to upload to a 01 3.2? I am trying to complete the process and have the new TB, but got a cayman intake plenum which doesn’t fit. I think I need a 996 C4S plenum. Pardon my ignorance.


Sent from my iPhone using Tapatalk

Hey man, I just saw this post. I pm'd at the end of the week, hope you got it. Why doesn't the cayman tb fit? Good luck!

Sent from my SM-T595 using Tapatalk

Cayman Plenum lists Part number as 987.110.416.01
Unsure. As I prepare to get to this project, the 996T looks like a completely different shape than this part I have here in my hands. This Cayman part has a 30 degree bend, and ~30 degree radius off center to said bend.

I have the TB part no 997.605.115.00

And I’m confrused- I know I need silicon elbows and longer screws but don’t have the parts list handy. That problem, plus the inability to access the ROW tune led me to post. I will go check your PM now. Thank you brother. -new986owner potentially former 986owner haha ;)


Sent from my iPhone using Tapatalk