When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
I haven't owned a car yet that I didn't end up modifying and I've made topics in the past about my 981 and 982s with emperical data gathered first hand (in contrast with various manufacturer's claims) so now it's time for my 991.2.
Somehow I get easily bored without goals and my biggest goal of '23 was go sub-7 mins BTG on the Nurburgring, but that came a little too easy so I stepped it up trying to beat the 2017 OEM lap time and managed that as well on the final track day of the year. The car had minor track mods but it was on stock suspension and stock exhaust headers/center muffler since I didn't want to cheat too much. But to continue in 2024 without any changes would be, again, boring, and mods are fun to test and document so here we are.
Most fun bits:
- ECU tune by Ehresmann in Germany
- Dundon street headers
Not as "testable":
- MR coilovers
- MR rear wing and front splitter upgrade
I've only done step 1 now, the ECU tune, but there's plenty to talk about already. Let's start with a dyno graph.
540 DIN horsepower and 496 Nm (366 ft-lbs) at the crank. Stock was 508 DIN HP so in any case the "gain" is +32 max. There's a lot of reasons why this is all you need to know and the measured WHP is redundant and less informative than the crank numbers this dyno spits out, but I'll try and keep this initial post from becoming a wall of text.
Of note is that an ECU tune on a non-RS GT3 gains loads across the entire rpm range which is very nice. You don't see this kind of consistent gains when you change the exhaust side since the math on that side is a little different. The GT3 ECU is capped in terms of mass air flow that it allows, so IN PARTICULAR ATMOSPHERIC CONDITIONS the tuning gains are massive and huge value for money. Not that this was cheap, but I know Ehresmann since my 981 and they do a good job. The ECU is part RS mapping copy/pasted and part their own.
It's important to frame this data correctly. If I was selling ECU tunes I could now claim 32 HP gained from my proprietary tune, or even 40 if I ignored the stock reading and just decided that it made 500 stock which is Porsche's OEM number. But that would be misleading (shocker!). First of all, a big part of these gains is simply air that the car could have sucked in all along but was told not to by the ECU. So when you uncork it, big gains. However that "extra" air isn't always hanging around. On a hot day or at higher altitude that free headroom isn't there and the stock car may not even be able to reach its air flow limit. There are still some gains as we're adding more fuel (the stock ECU is constantly trying to be as lean as possible and suffers quickly in hot weather) and have a bit better mapping all around but we might be down to 10 or 15 over stock rather than 40. As a side effect of running richer: note how much smoother the tuned graph is vs the stock one where it is fighting itself trying to run lean while keeping the ignition timing.
The stock car would keep producing around 500 HP no matter how good the conditions and for as long as it could once they get worse to the point it's no longer hitting its cap. The tuned car's power will vary all the time from insanely powerful when cold and low to almost equally down on power in thin air. Unfortunately for me, the Nurburgring and Spa are at 1500-1800ft elevation so I won't be getting a massive amount of lap time from the tune.
The upcoming addition of the Dundon headers will be adding power that works "everywhere" though. Because the gains there are mostly reduced pumping losses + better exhaust gas scavenging they won't be affected much by the conditions. It will be interesting to see those on the dyno - and I will probably compare the power with the stock and tuned ECU there as well. ETA of those charts will be mid April.
One final small mod that I've made is the Soler Performance "86mm" throttle body and BMC filters. I don't expect it is doing a whole lot, but it's good value for money (in a delusional Porsche modding kind of way) and my engineering brain likes the elegance of it versus the more blunt stock throttle body. If there is something to be gained from stuff like an 8K Dundon plenum with big throttle (there almost certainly is NOT on a non-RS with stock ECU unless the densitity altitude is proper crap) then I'll catch most of that with this slightly better flowing throttle body. I'm thinking 2 or 3 HP if I'm lucky, but I'm unlikely to test the difference unless I get very bored.
Once I add the MR aero I can't easily compare straigtht line speeds to before anymore, but I plan to have full 100-200 km/h and 60-130 data before I mess with that. In terms of lap time goals for '24 I'm a bit stuck since everything I could realistically compare to is a bit too far up the road. A recent time I could chase is the stock GT4 RS, that's only a few seconds off my old time. Of course the really tasty one would be to make my car the fastest ever manual car around the Nordschleife. That would require beating the Viper ACR's 7:01. 10 seconds faster than I've ever gone. Seems impossible. Maybe if I start a fundraiser like they Viper crew I could afford to bolt a bunch of Verus aero to it just for the attempt
As you know, I have Dundon street headers with a standard rear muffler, so very interested to see the difference that makes.
As you are testing the headers on a different day, will you be running a baseline with just the stage 1 tune again, then once the headers are fitted to give more accurate results?
I definitely think you should be aiming for the fastest manual time on the Ring now. Extra power, plus MR suspension & aero......another 6 second lopped off your time, so somehow you have to find another 4 seconds or so!
Maybe also aim for fastest 991 version gt3 round Spa.
Are you going to fit new suspension arms to help give consistent geo settings? That may also shave some time?
It's not practical to dyno again before fitting the headers but I can do dragy times right before and after which will give an adequate comparison I think.
6-7 seconds from mods is possible in theory, but then I'm still 3-4 short and I got reasonably lucky with that run in October to lose zero time to traffic. The benefit of official lap attempts is that you're the only car on track and you know for a fact there's no fluid spill around a blind corner. Worth a couple of seconds on its own but that kind of luxury is unavailable to amateurs.
I don't think I'll do anything more to the suspension after the MR kit, I suspect the returns will be tiny in terms of performance.
I'm looking forward to tuning mine once the CPO is up. I already have the dundon headers I wonder if it makes sense to get the Dundon specific pro tune for them?
I'm looking forward to tuning mine once the CPO is up. I already have the dundon headers I wonder if it makes sense to get the Dundon specific pro tune for them?
Hard to say, I remember people not having great results with their GT4 header pro tune but that was a while ago, and since IIRC they use the Cobb flasher you can try the built in tunes plus any other pro tuner so there's flexibility
Very interesting! Especially the dyno graph. Normally, when tuning, the torque curve "pivots" meaning that if you gain on top you loose in the bottom, and vice versa. Here you're gaining basically everywhere, like an offset. I'm intrigued to understand what you mean with that the ECU doesn't allow all the available air in. Only way that could happen is by not opening the throttle fully, or do something with the VVT:s, or how do you mean?
To run a bit richer is probably very effective, I imagine Porsche maps the engine to L=1 even at WOT, if they can taking exhaust temp limits into account, so L=0.88-ish will give performance, for sure.
It will be very interesting to see further progress!
I'm intrigued to understand what you mean with that the ECU doesn't allow all the available air in. Only way that could happen is by not opening the throttle fully, or do something with the VVT:s, or how do you mean?
Yes, it cuts the throttle plate angle to where mass air flow does not exceed the limit. The RS ECU allows for more air, hence the +20 HP more on paper. But even that is capped with a programmed limit, which is done to give more consistent performance in different conditions - of course it can only compensate up to the point where it starts dropping power due to high ambient temp or alt.
To run a bit richer is probably very effective, I imagine Porsche maps the engine to L=1 even at WOT, if they can taking exhaust temp limits into account, so L=0.88-ish will give performance, for sure.
Yep, I believe .88 is what it's going for now. Stock it will do lambda 1 for as long as it can, it does start to add fuel eventually while it rides the edge of knocking. This is why they are quite temperature sensitive as well, they dyno quite a lot lower in warm weather even after DIN/SAE correction.
Very interesting - thanks for the info! I understand. Makes sense that they left "a bit on the table" for the RS version, when developing the std GT3 engine, and doing this by keeping the throttle not fully open is simple, yet elegant! Side note: I worked many years ago at a Swedish OEM. We had a 2.4l 5-cyl engine N/A that produced 170PS. The car consumed around 9.0l/100km in the EU-testcycle. The Swedish government issued a new tax relief for cars that consumed 8.6l/100km, so definite business opportunity there. We solved the consumption easily with a higher final drive, but the car became slower obviously. To be able explain that to the customer we closed the throttle a little, only the last 1000-1500 rpm to the redline and Pmax became 140PS... So marketing could tell the customers, "yes it's a bit slower, because it has 30hp less because we pushed down fuel consumption..." Haha! This is more than 25 years ago - crazy how time flies...
Your tune - is it available on the market?
Last edited by erik_plus8; 01-25-2024 at 05:42 AM.
Side note: I worked many years ago at a Swedish OEM. We had a 2.4l 5-cyl engine N/A that produced 170PS.
Nice! I started modding with a 1995 olive green 850 T-5R ;-)
Eventually I had a gearbox in it from the 2.0L 10V engine which had a shorter final drive than all the others (as it was a gutless engine) and it worked pretty nice except it capped the top speed quite badly.
Your tune - is it available on the market?
Sure, Ehresmann automotive is a commercial enterprise
Nice work!
I have access to the gt3 ecu, and tables. The fuel is hitting 1.0 all the time until there is a rise in egt, then it hits .88 stock. I picked up aprox 20hp across the board on my .1 gt3, putting more fuel in at .84, cooler charge for high compression.
The timing is pretty agresive in these engines, and if knock is detected it will pull a few degrees of timing ,some tweaks to the cam timing can help power too.
Yes, mass air is limited, so raising the limit will help in good weather, raising headroom on the engine..
The fuel is hitting 1.0 all the time until there is a rise in egt, then it hits .88 stoc.
Thanks for the info!
In this case I couldn't be hands on unfortunately but I did a lot of logging and testing of tunes on my 982 GT4 which is probably running similar stratagies from the factory on its Bosch MG1 ECU. The 981's Siemens was more primitive. I've come to suspect that most fuel in the US is of at least somewhat lower quality even if we're comparing 93 vs 98 RON which should be the same in theory. On a dyno with loads of cooling I never saw any richer than .92 on the GT4 stock, but when M Engineering sent me a map they went straight to .82 and lost a bunch of power as the stock ECU had even been running more timing than what they put in. But put the same map on a car on US fuel and a typical dyno with poor fans (and in Florida for good measure) and apparently it picked up a bunch of power in those conditions. Eventually I got them to crank it up to .86 lambda and we found a few horsepower over stock but on that car it turned out to be not worth it as the engine is basically being run all out already. The GT3 with its "race derived" engine tuned down versus the RS is a lot more rewarding already.
The GT3 with its "race derived" engine tuned down versus the RS is a lot more rewarding already.
(I have been saying this for a long time, they didn't leave much on the table)
The gas is crap in the usa, and I have the best conditions possible being at sea level.
Going to look at building a map to run 100% e85 down the road.
01-24-2024, 04:42 AM
JCv's 991.2 GT3 manual modifications thread (dyno, dragy,...)
