Twin Turbo 928 fixed and back out there terrorizing the streets!
06-04-2013, 10:35 AM
#631
Nordschleife Master
Thread Starter
John's notes:
I need to turn the boost down...it literally broke the tires loose on dry pavement in 3rd gear at like 70 mph during a roll on. ... It's to the point where it's just too risky to attempt any testing or tuning out on the street. ...It's very difficult to hold the throttle all the way down in 3rd gear on the highway, because the car accelerates very rapidly, even in the higher gears. ....when you come down on it, you literally feel your feet lift up off the floor due to the acceleration. It's pretty crazy and I'm sure the engine isn't loving being ridden that hard.
Let's put in those bigger turbos!
06-04-2013, 01:31 PM
#632
Racer
Join Date: Mar 2013
Location: San Diego, CA
Posts: 304
Likes: 0
Received 0 Likes
on
0 Posts
Yes, these compressors are pretty much maxed out now.
John's notes:I need to turn the boost down...it literally broke the tires loose on dry pavement in 3rd gear at like 70 mph during a roll on. ... It's to the point where it's just too risky to attempt any testing or tuning out on the street. ...It's very difficult to hold the throttle all the way down in 3rd gear on the highway, because the car accelerates very rapidly, even in the higher gears. ....when you come down on it, you literally feel your feet lift up off the floor due to the acceleration. It's pretty crazy and I'm sure the engine isn't loving being ridden that hard.
My conclusion:
Let's put in those bigger turbos!
John's notes:I need to turn the boost down...it literally broke the tires loose on dry pavement in 3rd gear at like 70 mph during a roll on. ... It's to the point where it's just too risky to attempt any testing or tuning out on the street. ...It's very difficult to hold the throttle all the way down in 3rd gear on the highway, because the car accelerates very rapidly, even in the higher gears. ....when you come down on it, you literally feel your feet lift up off the floor due to the acceleration. It's pretty crazy and I'm sure the engine isn't loving being ridden that hard.
My conclusion:
Let's put in those bigger turbos!
My conclusion: Add some weight in the back (say 100-200lbs in incremental weights in the spare tire area) give it a test drive if you like you can then add a full heavy sound system in the back (not near the center rear seat, I mean the back back near the hatch); then add bigger turbos
06-04-2013, 02:10 PM
#633
Nordschleife Master
Thread Starter
My conclusion: Add some weight in the back (say 100-200lbs in incremental weights in the spare tire area) give it a test drive if you like you can then add a full heavy sound system in the back (not near the center rear seat, I mean the back back near the hatch); then add bigger turbos
The above for testing this dyno mule. For the final car, the idea is to delete *everything* unnecessary from the front of the car and then move what can be moved of the rest to as far back as possible. I was thinking about creating some very simple .xls model of the weight distribution and then seeing where the moving gets you the most bang for the buck in terms of moving the center of gravity towards the center of the car and reducing the polar moment of inertia. Nothing has happened in practice here, it's been all engine all the time for a while now and it'll continue to be all engine for a little bit longer.
06-04-2013, 02:39 PM
#634
Addict
Lifetime Rennlist
Member
Lifetime Rennlist
Member
My conclusion: Add some weight in the back (say 100-200lbs in incremental weights in the spare tire area) give it a test drive if you like you can then add a full heavy sound system in the back (not near the center rear seat, I mean the back back near the hatch); then add bigger turbos
I know the answer in ptuomovs case.
My car was actually quicker when configured with bigger pulley on it during a few "testing" stages. It launched in a more predicable manner.
06-04-2013, 02:50 PM
#635
Racer
Join Date: Mar 2013
Location: San Diego, CA
Posts: 304
Likes: 0
Received 0 Likes
on
0 Posts
I know once its moving the weight shifts a little to the rear, my view point in general is center of the car for balance and handling and rear weight is for keeping the wheel on the ground.
For sudden acceleration you want to have more rear weight being you want the wheel on the ground while high speeds you want it balanced with the spoiler doing part of the rear weight job here.
An excel sheet can do the high speeds balanced thing easy, its the high acceleration which you need some type of formula which likely you can nab from a Carrera race forum or even a HondaAccord/MazdaMiata forum being they have a bigger street racer base with turbos/supercharged/nitro stuff (not perfect but the basic formula is closer than a Carrera formula being Carrera is usually not super modded due to the cost vs Japanese Rice Rockets)
The problem is trying to find the balance of which is acceptable for you (too much weight = slow; too little weight = no rear grip; too much rear weight = bad car balance; too much center weight = not enough rear grip)
Speed vs power... I don't think his car has hit that divide yet 
For sudden acceleration you want to have more rear weight being you want the wheel on the ground while high speeds you want it balanced with the spoiler doing part of the rear weight job here.
An excel sheet can do the high speeds balanced thing easy, its the high acceleration which you need some type of formula which likely you can nab from a Carrera race forum or even a HondaAccord/MazdaMiata forum being they have a bigger street racer base with turbos/supercharged/nitro stuff (not perfect but the basic formula is closer than a Carrera formula being Carrera is usually not super modded due to the cost vs Japanese Rice Rockets)
The problem is trying to find the balance of which is acceptable for you (too much weight = slow; too little weight = no rear grip; too much rear weight = bad car balance; too much center weight = not enough rear grip)
Last edited by eijun; 06-04-2013 at 03:39 PM.
06-05-2013, 06:53 PM
#636
Nordschleife Master
Thread Starter
Right now it's all about power, and speed is a secondary consideration. Down the road, maybe we'll think about speed as an objective as well. So far we haven't.
That said, purely as an unintended byproduct, car is also getting faster...
In addition to finding the boost bottleneck, we've also come up with a new tuning hypothesis. The hypothesis is that we'll make the same peak power and wider power band at 650 ft-lb if we do the following:
(1) put in generous spacers to the intake manifold on both sides
(2) change the software such that the flappy is not opened before 5500 rpm, or maybe not at all.
We think this will be da bomb!
I am forecasting reader skepticism.
Now, don't get me wrong, I think that on a completely stock 928 S4, the intake manifold spacers marginally reduce the peak power. No +10hp gain there, more like -2hp loss! This about -2hp can be recovered with Sharktuning by keeping the flappy open at high rpms, but still even that gets you to about zero. Of course, nobody can reliably measure +/-2 hp on a chassis dyno, but you get the point. The plenum spacers don't give more peak power on the NA car, maybe a little unshourding benefit but that's it.
However, we're in the boost land. We are going to use the spacers to move the first torque peak from about 3150 rpm to about 2900 rpm by adding the largest available spacers. This was among the lessons learned in the intake manifold threads http://speedtalk.com/forum/viewtopic.php?f=1&t=35634 and https://rennlist.com/forums/928-foru...and-ideas.html. This "spacing out" the plenum will increase torque at low end where turbo is not yet fully spooled and thereby produce more exhaust gas to spool it earlier. We can easily fill the corresponding hole at 3400 rpm in the torque curve with boost. The result is -by my expectations- that we'll add about 100-150 rpm to peak torque region (>600 ft-lb) in the low end. We're not trying this in the next test, but the low compression engine will surely be get a completely "spaced out" intake manifold.
In the next test, we are planning not to open the flappy. On a normally aspirated car that would reduce the torque pretty dramatically at about 4200 rpm, which would be bad. On my turbo car, it's good! Why? Because we can detune the most detonation prone rpm range 3600-4800 rpm while still running the stiff wastegate spring we need to go into low twenties in terms of boost above 6000 rpm. Keeping the flappy closed will knock out about 10% of the torque at 4200 rpm, which in boost units is about the same as taking the boost down from 14 to 11.5 psi at 4200 rpm. In fact, it'll reduce knocks even more than just turning boost down from 14 to 11.5 psi because the intercooler sees hotter charge that is then expanded in the manifold to be colder (a la new Carrera Turbo expansion intake manifold).
Furthermore, not opening the flappy when Bruce Banner is turning green from biting his lip makes tuning a whole lot easier without tripping knock retard and sending the torque curve to the EZK penalty box. This in turn will allow us to tweak the column rpm definitions in the software and move some columns from the mid range and to the above 6000 rpm where power is made. If the flappy trick works in this test, the next test will see whole different fuel and spark maps.
That said, purely as an unintended byproduct, car is also getting faster...
In addition to finding the boost bottleneck, we've also come up with a new tuning hypothesis. The hypothesis is that we'll make the same peak power and wider power band at 650 ft-lb if we do the following:
(1) put in generous spacers to the intake manifold on both sides
(2) change the software such that the flappy is not opened before 5500 rpm, or maybe not at all.
We think this will be da bomb!
I am forecasting reader skepticism.
Now, don't get me wrong, I think that on a completely stock 928 S4, the intake manifold spacers marginally reduce the peak power. No +10hp gain there, more like -2hp loss! This about -2hp can be recovered with Sharktuning by keeping the flappy open at high rpms, but still even that gets you to about zero. Of course, nobody can reliably measure +/-2 hp on a chassis dyno, but you get the point. The plenum spacers don't give more peak power on the NA car, maybe a little unshourding benefit but that's it.
However, we're in the boost land. We are going to use the spacers to move the first torque peak from about 3150 rpm to about 2900 rpm by adding the largest available spacers. This was among the lessons learned in the intake manifold threads http://speedtalk.com/forum/viewtopic.php?f=1&t=35634 and https://rennlist.com/forums/928-foru...and-ideas.html. This "spacing out" the plenum will increase torque at low end where turbo is not yet fully spooled and thereby produce more exhaust gas to spool it earlier. We can easily fill the corresponding hole at 3400 rpm in the torque curve with boost. The result is -by my expectations- that we'll add about 100-150 rpm to peak torque region (>600 ft-lb) in the low end. We're not trying this in the next test, but the low compression engine will surely be get a completely "spaced out" intake manifold.
In the next test, we are planning not to open the flappy. On a normally aspirated car that would reduce the torque pretty dramatically at about 4200 rpm, which would be bad. On my turbo car, it's good! Why? Because we can detune the most detonation prone rpm range 3600-4800 rpm while still running the stiff wastegate spring we need to go into low twenties in terms of boost above 6000 rpm. Keeping the flappy closed will knock out about 10% of the torque at 4200 rpm, which in boost units is about the same as taking the boost down from 14 to 11.5 psi at 4200 rpm. In fact, it'll reduce knocks even more than just turning boost down from 14 to 11.5 psi because the intercooler sees hotter charge that is then expanded in the manifold to be colder (a la new Carrera Turbo expansion intake manifold).
Furthermore, not opening the flappy when Bruce Banner is turning green from biting his lip makes tuning a whole lot easier without tripping knock retard and sending the torque curve to the EZK penalty box. This in turn will allow us to tweak the column rpm definitions in the software and move some columns from the mid range and to the above 6000 rpm where power is made. If the flappy trick works in this test, the next test will see whole different fuel and spark maps.
06-11-2013, 09:16 PM
#637
Nordschleife Master
Thread Starter
A quick update:
The stock engine won.
The compressors have hit the centrifugal compressor choke point, aka "stonewall". By our math, in the last dyno experiment the compressor internal air speed went sonic, which creates a sonic wall after which the mass flow rate doesn't really increase with turbine speed. This means that the gt3071r-wg turbochargers are completely maxed out on the compressor side. The stock engine held, without any problems, the maximum amount of air that we could throw at it with these turbos.
I am impressed by what the factory put together there in 1987.
The stock engine won.
The compressors have hit the centrifugal compressor choke point, aka "stonewall". By our math, in the last dyno experiment the compressor internal air speed went sonic, which creates a sonic wall after which the mass flow rate doesn't really increase with turbine speed. This means that the gt3071r-wg turbochargers are completely maxed out on the compressor side. The stock engine held, without any problems, the maximum amount of air that we could throw at it with these turbos.
I am impressed by what the factory put together there in 1987.
07-14-2013, 11:30 AM
#641
Nordschleife Master
Thread Starter
Update:
The gt3071r compressor hit the sonic choke which means that the intake become mass-flow limited. With this stock engine and this tune, what it means is that once the car hits 700 rwhp at some rpm it can hold about that same horsepower level up the rpm band but not produce more power. The boost becomes irrelevant, it's the mass flow that is now fixed.
This left us at a cross-roads: Either start optimizing the engine to be as efficient as possible in using that capped air flow level better or swap in larger compressors.
We've chosen the second path. The car is on a lift and the old turbos are off. Custom turbos are being built as no stock Garrett turbos fit in the space we have. The compressor cover starts life as a Garrett part number, but needs to be cut, welded, and remachined. The turbine cover starts life as a stainless steel aftermarket turbine housing, but needs to be cut, welded, and remachined. The exhaust manifolds need to be modified to fit these custom turbos. New downpipes are needed as well. The next step is to get the engine running with this new, proof of concept turbos.
Once the proof of concept turbos are ready and tested, the stock engine will come out and the low-compression hot-rodded engine will go in. The final version of the turbos will be built and TBC coated, speed sensors fitted, etc.
The gt3071r compressor hit the sonic choke which means that the intake become mass-flow limited. With this stock engine and this tune, what it means is that once the car hits 700 rwhp at some rpm it can hold about that same horsepower level up the rpm band but not produce more power. The boost becomes irrelevant, it's the mass flow that is now fixed.
This left us at a cross-roads: Either start optimizing the engine to be as efficient as possible in using that capped air flow level better or swap in larger compressors.
We've chosen the second path. The car is on a lift and the old turbos are off. Custom turbos are being built as no stock Garrett turbos fit in the space we have. The compressor cover starts life as a Garrett part number, but needs to be cut, welded, and remachined. The turbine cover starts life as a stainless steel aftermarket turbine housing, but needs to be cut, welded, and remachined. The exhaust manifolds need to be modified to fit these custom turbos. New downpipes are needed as well. The next step is to get the engine running with this new, proof of concept turbos.
Once the proof of concept turbos are ready and tested, the stock engine will come out and the low-compression hot-rodded engine will go in. The final version of the turbos will be built and TBC coated, speed sensors fitted, etc.
07-14-2013, 11:28 PM
#642
Addict
Rennlist Member
Rennlist Member
Join Date: Feb 2004
Location: Monterey Peninsula, CA
Posts: 2,374
Likes: 0
Received 16 Likes
on
12 Posts
Tuomo,
Would a GTX3076 or GT3582 cold side fit with the existing hot sides? Or is the space available the criteria for fitment...? Also, will you guys be using external wastegates or sticking with the internal..?
I am curious, as I have been debating something silly with my silly self...
Would a GTX3076 or GT3582 cold side fit with the existing hot sides? Or is the space available the criteria for fitment...? Also, will you guys be using external wastegates or sticking with the internal..?
I am curious, as I have been debating something silly with my silly self...
07-15-2013, 12:09 AM
#643
Nordschleife Master
Thread Starter
Tuomo, Would a GTX3076 or GT3582 cold side fit with the existing hot sides? Or is the space available the criteria for fitment...? Also, will you guys be using external wastegates or sticking with the internal..? I am curious, as I have been debating something silly with my silly self...
Everything would be really easy to size and select if it weren't for the real estate problem. Almost nothing fits under there. If it weren't for the space constraints, we'd probably stick a couple of GTX3582R's under there and be done with it. Or a couple of twin-scroll turbos, pairing 1+2, 3+4, 5+7, 6+8. But this is all worthless daydreaming as there is no room whatsoever for off-the-shelf turbos and ideal manifolds.
07-15-2013, 06:53 AM
#645
How much back pressure are you seeing now, and with what A/R size on the turbine housings?