Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5

Flash's SC setup - bigger better faster stronger
#1

Now that I finally finished off the Stage 2 SC Kit, I can finally concentrate on my own car. This has been an incredibly long journey, that began 9 years ago when I first got the car. I knew then that I had a long row to hoe, but I was confident that I could end up with what I wanted.



I got there with the suspension and brakes. Interior came together quickly. Creature comforts and features came along well too. What was lacking was power. None of the bolt on components got me near where I wanted to be. I wanted that magic 10lb per horsepower. So, the SC kit was the obvious project. That did very well, but was still short of my mark. Then came Stage 2. Again, very nice, and gobs of torque, but the bragging rights still weren't there. I had surpassed the power of the Turbo S that porsche spent millions on, only to build little more than a dozen. I made the 10lb/hp goal at that point, but felt there was still more on the table.



As a part of developing the kit, I found the limits of the stock engine. The rods were incapable of handling 7lbs of boost, and the intake system was very limited. I ended up bending rods and really buggering things up. Then came the rebuild. Deciding how far to go was a real issue. I could just put it back together, and live with the power levels I had reached, or I could go deeper, and build it up to handle more boost, and thereby get more power.



I still wanted to run on 91 octane fuel, and pass California emissions tests. I still wanted the reliability and drivability of a stock car. This is a tall order, when factoring in that I wanted to hit about 325hp. So, the building began.



In the bottom end, coated bearings were a must. They can't handle the pounding of high revs, even in normally aspirated mode. If you are building up a motor for high revs (over 6k) then this is a must. The rods are also a problem, whether you have the early 1R or the later 2R. The reinforcing they did is not in the area of failure. Higher levels of boost, and higher rpms dictated a better rod. I went with Carrillo. Larger pistons were also chosen to resolve the damage done when the rods bent. The block got a kiss to clean things up, as the higher boost would find any anomalies and blow a head gasket. The rest of the bottom end got a complete balance and blueprinting procedure that ensured that every component was exactly the same.



For the top end, everything needed to open up. The OEM layout cannot flow enough air to make safe boost. Yes, you can cram it in there, but the temps really start to climb, and the efficiency goes way down really fast. So, the head was ported and polished, the valvetrain was changed, the intake was cleaned up, and a few other changes were made, to open up the flow and allow for more boost. I had to be careful though, because if I opened it up too much, the bottom end torque would go away, just like it would normally aspirated. Porsche did a heck of a job on the intake design. It resulted in a very flexible ability to produce torque across a wide band. They gave up some upper end ability, but not that much. It didn't take much clean up to make it flow well enough to make a lot more power, without giving up the low end. The head was also kissed for the same reason as the block.



The results were a disturbingly smooth engine, and a very thirsty one. The larger pistons and slightly higher compression needed more fuel. The opening up of the intake really made it thirsty, but opened the door for more boost too. I had to add almost 10% more fuel than the Stage 2 kit, just to get things even. This resulted in more torque, and much better throttle response, but I did not have nearly enough fuel to feed it with the stock injectors.



Then it came time to start opening things up and seeing what I could get. I was out of injector, so I had to start figuring that out. I looked at the 4 pintle misting type, and sourced one that flowed slightly better than stock, but they actually made things leaner up top, and by a wide margin. Between the valve pattern of the head, and the dead time and spray pattern of the injectors, the motor just wasn't happy. I tried some bigger ones of the same pattern, and got the same results. So, I had some custom injectors made. Bingo.



As soon as I opened up the SC a bit, and let it make a bit more boost, I immediately went past my target of 325 and hit 340, in the first pass. I had not even begun yet to tune things, or even to seek more boost. After a few passes I got things looking a lot better, and ended up with over 350hp! That's more than Porsche managed to do, with their millions of dollars of budget, on the one and only Le Mans prepped Turbo RS.



I went back a couple of days later to clean up the torque curve, and got things a lot better. The car is ridiculously fast now. I picked up a couple of foot pounds, and got a better mixture across the board. Boost is now up to about 7lbs. There is still no intercooler, and no need for one. As a result of modifying the supercharger itself specifically for this setup, intake temps are now DOWN from the Stage 2 kit.



I'm still not done, and am going back on Monday to fiddle with a few things. I expect to pick up another 4 or 5 horsepower, and a few foot pounds of torque. I'm pretty sure it won't be a lot of an increase, but more importantly, what I have planned will lower intake temps further, making it even safer to run. I still have quite a bit of mid-throttle tuning to do, but it's getting pretty close. Really it drives very well, but I can see by the air/fuel meter that I can get a whole lot more part-throttle torque by cleaning it up. Assuming I don't end up having to go to an even larger injector, and I really hope to not have to do that, as the very bottom end of running will be very hard to tune, I expect to be done with all of it by mid next week.



Here are a couple of charts, just to give you an idea of where I am right now. I'll post more charts after I get back on Monday.



Readings are taken at the hubs!!!!



   

   
Partial Post: Please Login or Register to read the full post.

94 Midnight Metallic Blue Cab Porsche 968 w/deviating cashmere/black interior and WAY too many mods to list - thanks to eric for creating www.968forums.com



"It isn't nearly as expensive to do it right as it is to do it wrong."
Reply
#2

Congrats! Awesome results! <img src="/forum/images/smilies/968/rock.gif" class="smilie" alt="" />
Partial Post: Please Login or Register to read the full post.
Reply
#3

thanks. really looking forward to monday. finally being finished is the primary goal now, not seeing how much more i can get. i wouldn't be going back, but i need to clean up the mix a bit way up top, and i want to see if i can fill in the small dips in the torque curve just a bit more.



pretty sure i'll end up at about 320hp and 270ft/lb at the wheels.
Partial Post: Please Login or Register to read the full post.

94 Midnight Metallic Blue Cab Porsche 968 w/deviating cashmere/black interior and WAY too many mods to list - thanks to eric for creating www.968forums.com



"It isn't nearly as expensive to do it right as it is to do it wrong."
Reply
#4

No really when can i order ?LOL



Nice curve enjoy .
Partial Post: Please Login or Register to read the full post.
Reply
#5

lol - funny - if only it were that simple. it has been a pleasant surprise to get this much power, and i am not afraid of anything going wrong, but you have to wonder about how the stresses of that much power will present themselves in the form of failures of under-engineered parts. i can already tell that the driveline, clutch and transaxle are at their limits, and i have had those done. even with new and beefed up parts, i can feel the limits.
Partial Post: Please Login or Register to read the full post.

94 Midnight Metallic Blue Cab Porsche 968 w/deviating cashmere/black interior and WAY too many mods to list - thanks to eric for creating www.968forums.com



"It isn't nearly as expensive to do it right as it is to do it wrong."
Reply
#6

That's a really unusual looking torque curve with all the localized peaks. Do you understand whats making it look like that? I'm struggling to understand which of the factors in the equation would create such localized maximums.
Partial Post: Please Login or Register to read the full post.
Reply
#7

keep in mind that the fluctuations are only 5-8ft/lb. that's a whole lot less than the OEM curve has with its peak point. it's just the way the graph is lain out. if i stretched the bottom axis, it would look almost flat. that being said, there are very few data points in the WOT fuel map. each one is hundreds of rpms apart from the next.



better?

   
Partial Post: Please Login or Register to read the full post.

94 Midnight Metallic Blue Cab Porsche 968 w/deviating cashmere/black interior and WAY too many mods to list - thanks to eric for creating www.968forums.com



"It isn't nearly as expensive to do it right as it is to do it wrong."
Reply
#8

Yeah, it's not the axis scaling that I was looking at, but the size of the torque fluctuations over such a small change in rpm. My Mazda has barely 1-3 ft-lb on a raw data unsmoothed torque curve. Just looks odd to me; never seen one that had so much variation over such a small rpm change unless it was due to VVT, or a variable resonance induction system, or some other factor that was coming into play. And then, it is usually only at one or two points on the torque curve. Yours has at least 6 localized peak values?? Just curious what could be making for such a variation. I assume your VE, fuel, ignition, etc. maps have few peaks?
Partial Post: Please Login or Register to read the full post.
Reply
#9

i would be extremely surprised if ANY car with this amount of power had only 1-3 ft/lb variance. if the car only has 150ft/lb at the crank, then 1-3 would be right. most dynos and all manufacturers use a wide scale smoothing tool to hide that stuff. i didn't do any of that. a 2% variation is the average variation. these are 2%-3% variation in those areas



again, the WOT fuel map has a total of 16 data points for the entire rpm range, and only 10 of them between 2k and 6k. the gaps between the data points are 300 to 500rpm each. you have to use those points and hope that you get an average curve that is better by manipulating them than it is without. modern cars have 10 times the amount of data points, and the ability to tune each injector separately. that makes things a whole lot easier.



this is actually a very smooth curve for this engine, and smoother than the factory curve if you turn off the smoothing tool.
Partial Post: Please Login or Register to read the full post.

94 Midnight Metallic Blue Cab Porsche 968 w/deviating cashmere/black interior and WAY too many mods to list - thanks to eric for creating www.968forums.com



"It isn't nearly as expensive to do it right as it is to do it wrong."
Reply
#10

Wow! That is some bad ass power!!
Partial Post: Please Login or Register to read the full post.
Reply
#11

So if you have to get new injectors after the next experiment will they (your first set of customs) work on the base SC with stage 2 any better than stock? Hint, Hint, dibs. I'm going for bigger, faster, better, stupider.
Partial Post: Please Login or Register to read the full post.
Reply
#12

nope. while injector increase would pave the road for more power, the boost limit is your big limiting factor. if you want more power, you have to build the engine to handle it. then you can uncork the boost.



then you get to tune that engine.
Partial Post: Please Login or Register to read the full post.

94 Midnight Metallic Blue Cab Porsche 968 w/deviating cashmere/black interior and WAY too many mods to list - thanks to eric for creating www.968forums.com



"It isn't nearly as expensive to do it right as it is to do it wrong."
Reply
#13

[quote name='flash' timestamp='1353712738' post='135186']

i would be extremely surprised if ANY car with this amount of power had only 1-3 ft/lb variance. if the car only has 150ft/lb at the crank, then 1-3 would be right. most dynos and all manufacturers use a wide scale smoothing tool to hide that stuff. i didn't do any of that. a 2% variation is the average variation. these are 2%-3% variation in those areas



again, the <acronym title='wide open throttle'>WOT</acronym> fuel map has a total of 16 data points for the entire rpm range, and only 10 of them between 2k and 6k. the gaps between the data points are 300 to 500rpm each. you have to use those points and hope that you get an average curve that is better by manipulating them than it is without. modern cars have 10 times the amount of data points, and the ability to tune each injector separately. that makes things a whole lot easier.



this is actually a very smooth curve for this engine, and smoother than the factory curve if you turn off the smoothing tool.

[/quote]



My Mazda was making about the same amount of power, boosted at 9 psi, so no real change there. I was also running a 16x16 matrix, so no change there. It was running an Electromotive TecGT, which if I remember extrapolates between the data points in the matrix to smooth the changes. So, that could be the difference. And, if you are still running the 968 factory ECU, the computer technology in it and the software it could support would be 15 years older than the TecGT. That too could explain the difference. The TecGT does have individual cylinder trim, but I never found a need to use it.
Partial Post: Please Login or Register to read the full post.
Reply
#14

the WOT matrix for this car is 16x1.



i am still using the motronic. that was a design factor required for CARB certification.



the dual resonant manifold, and the variocam on/off points also have an impact. these also could not be altered for CARB certification.



the fact that we are batch fire doesn't help either. same goes with the MAF instead of a MAP.



again though, 2% change in between any 2 data points is negligible, and i would question a chart that didn't have something along those lines. i'd love to see it flatter, and that is why i am going back on monday, but it's pretty tough to do with this ECU. you move one point and affect another. finding that happen medium is not easy. for example, i can get a flatter line, but at lower levels across the board. would that be better? at this point i blast through the dips so fast you can't feel them. this is the WOT map after all. the part throttle maps are entirely different, and much easier to make smooth.



as i said earlier, the STOCK curve is worse. it's all over the place if you turn off the smoothing. i fixed it out a lot for the Stage 1 kit.
Partial Post: Please Login or Register to read the full post.

94 Midnight Metallic Blue Cab Porsche 968 w/deviating cashmere/black interior and WAY too many mods to list - thanks to eric for creating www.968forums.com



"It isn't nearly as expensive to do it right as it is to do it wrong."
Reply
#15

Wow, given those restrictions, you should be pretty pleased with what you've accomplised. Nice results!



And, yeah I wouldn't expect to be able to feel those small changes on a moving car, given the amount of smoothing that would occur from the flywheel effect of moving all that inertia.
Partial Post: Please Login or Register to read the full post.
Reply
#16

i'm pleased, but being how i am, i'm also not giving up yet. one more shot and then i'll call it "done"
Partial Post: Please Login or Register to read the full post.

94 Midnight Metallic Blue Cab Porsche 968 w/deviating cashmere/black interior and WAY too many mods to list - thanks to eric for creating www.968forums.com



"It isn't nearly as expensive to do it right as it is to do it wrong."
Reply
#17

Have you played with adjusting the cam timing using the built in adjustment on the pulley? It would be interesting to see the effect of advancing or retarding the cam timing on power. I think there is about +/- 3 degrees flexibility in the slots.
Partial Post: Please Login or Register to read the full post.
Reply
#18

there is adjustment there. it can be used to fiddle with the curve. basically you can swing it one way for low end and the other for top end. been there done that. given that i was not looking for top end, other than the bragging rights of the numbers, because i could care less about anything over 6k rpm, as i will never be driving there anymore, now that i have all this torque, that direction was moot. swinging it to the other side made a certain amount of sense, but it messed with my emissions, and really cost me up top more than i was willing to give up. so i went in the middle, just like the factory did. go figure.
Partial Post: Please Login or Register to read the full post.

94 Midnight Metallic Blue Cab Porsche 968 w/deviating cashmere/black interior and WAY too many mods to list - thanks to eric for creating www.968forums.com



"It isn't nearly as expensive to do it right as it is to do it wrong."
Reply
#19

What dyno are you using?
Partial Post: Please Login or Register to read the full post.
Reply
#20

dynapak. far more accurate than a dynojet, and way better for tuning, though the numbers are a bit lower.
Partial Post: Please Login or Register to read the full post.

94 Midnight Metallic Blue Cab Porsche 968 w/deviating cashmere/black interior and WAY too many mods to list - thanks to eric for creating www.968forums.com



"It isn't nearly as expensive to do it right as it is to do it wrong."
Reply


Possibly Related Threads…
Thread / Author Replies Views Last Post
Last Post by ds968
06-16-2013, 10:48 AM

Forum Jump:


Users browsing this thread: 1 Guest(s)