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.
The answer is that there's no replacement for displacement! Well, actually there is: the turbo. But once you got the turbo... there's no replacement for displacement! The 928 S4 has a 5.0 liter motor with about 622cc cylinder, and we've got two turbos on it. At ten psi, our motors are probably less stressed than some 951 motors from the factory. A 500 rwhp 16v 951 would be about as stressed as 1000 rwhp 32v turbo 928, and were not there yet.
The blue engine has stock thicker replacement head gasket, stock head bolts, stock open deck, and stock 4th casting revision head with some porting. We don't anticipate any problems with this setup as long as we control the knock and temperatures.
Sorry Tuomo. Edited.
What particular stresses do you refer to out of interest? Our next motor is 3.2ltr with pretty much everything modified not only for performance but for longevity. Our previous motors have been only 8v and we saw just over 550whp on a Dyno Dynamics dyno which tend to read lower than some others. This was at 25psi with 3.1ltr capacity. It had quite a big demise which we put down to ugly harmonics in the end. Having extra bits and pieces hanging off the nose of the crank to run dry sump etc may have contributed. Also having a knife edged crank could have been more susceptible to whip. Nevertheless we ran that motor for a year or so before it failed. The new 16v motor has had a few more precautions built into it to prevent this happening again however I am still not counting chickens yet in regards to how much boost we can put through this without seeing headlift again.
But can I assume that you've never had to deal with headlift and/or failure of the headgaskets? Looks like one of your 928 bretheren Todd has got close to the 1000whp mark. He hit 940whp @25psi 5ltr vs 675whp @ 22psi 3.1ltr in our 944 community. Not that comparing dyno results give a full indication. Can't see any updates to his thread since 2013 though. Perhaps he ran into further issues.
What particular stresses do you refer to out of interest? Our next motor is 3.2ltr with pretty much everything modified not only for performance but for longevity. Our previous motors have been only 8v and we saw just over 550whp on a Dyno Dynamics dyno which tend to read lower than some others. This was at 25psi with 3.1ltr capacity. It had quite a big demise which we put down to ugly harmonics in the end. Having extra bits and pieces hanging off the nose of the crank to run dry sump etc may have contributed. Also having a knife edged crank could have been more susceptible to whip. Nevertheless we ran that motor for a year or so before it failed. The new 16v motor has had a few more precautions built into it to prevent this happening again however I am still not counting chickens yet in regards to how much boost we can put through this without seeing headlift again.
But can I assume that you've never had to deal with headlift and/or failure of the headgaskets? Looks like one of your 928 bretheren Todd has got close to the 1000whp mark. He hit 940whp @25psi 5ltr vs 675whp @ 22psi 3.1ltr in our 944 community. Not that comparing dyno results give a full indication. Can't see any updates to his thread since 2013 though. Perhaps he ran into further issues.
I don't want to pretend to be an expert on engine stresses. However, this is how I think about it. First, we've got peak cylinder pressure as one measure. Not measuring it yet, but will do so down the road. Then there's the heat that needs to be moved away from the engine. And I guess the third thing I'd mention would be the various forces that change with the engine rpm.
The four valve head breathes easily, doesn't require high acceleration cams and stiff springs, and burns efficiently and quickly. The pipes in and out of the turbo in my car have low restriction, as does the intercooler. Exhaust back pressure is only slightly higher than boost, which also helps getting hot exhaust gas out of the cylinders. Crossplane V8 with fully counterweighted crankshaft also doesn't shake as much as four cylinder engine half the displacement. Those were my thoughts when I made the loose comment of my engine at 10psi boost being possibly less stressed than some stock 951s.
It's idling with the 80 pound injectors. Not tuned yet, but its running my with them. I don't think anyone has run 80 pound injectors before with the stock computer, but I might just not be aware of everything that everyone is doing.
What kind of injectors 80 lbs/h (840cc/min) do have? Duty cycle? Multihole? It would be interesting to see how well the engine can be tuned at idle and low rpm at low power conditions. Should be good for N/A appr. 100 hp per cylinder at 60% duty cycle.
Åke
What kind of injectors 80 lbs/h (840cc/min) do have? Duty cycle? Multihole? It would be interesting to see how well the engine can be tuned at idle and low rpm at low power conditions. Should be good for N/A appr. 100 hp per cylinder at 60% duty cycle.
The earlier post has an injector test sheet that has the part numbers etc.
I don't want to pretend to be an expert on engine stresses. However, this is how I think about it. First, we've got peak cylinder pressure as one measure. Not measuring it yet, but will do so down the road. Then there's the heat that needs to be moved away from the engine. And I guess the third thing I'd mention would be the various forces that change with the engine rpm.
The four valve head breathes easily, doesn't require high acceleration cams and stiff springs, and burns efficiently and quickly. The pipes in and out of the turbo in my car have low restriction, as does the intercooler. Exhaust back pressure is only slightly higher than boost, which also helps getting hot exhaust gas out of the cylinders. Crossplane V8 with fully counterweighted crankshaft also doesn't shake as much as four cylinder engine half the displacement. Those were my thoughts when I made the loose comment of my engine at 10psi boost being possibly less stressed than some stock 951s.
Yes I have discussed measuring cylinder pressures with our US builder. Haven't quite got there yet. Without wanting to pollute your thread by listing all the 'stuff' we've done on our new motor I'll just say "a lot!!" It's very impressive how you've managed to retain many of the stock components in your build without over stressing it. My wallet is extremely jealous! Congrats on that.
Yes I have discussed measuring cylinder pressures with our US builder. Haven't quite got there yet. Without wanting to pollute your thread by listing all the 'stuff' we've done on our new motor I'll just say "a lot!!" It's very impressive how you've managed to retain many of the stock components in your build without over stressing it. My wallet is extremely jealous! Congrats on that.
One of the goals in this (science) project has been to figure out what is absolutely necessary to modify to make big power with turbos and what can be left stock. We were able to take the completely stock engine about 700 rwhp without any internal engine modifications and without any obvious reliability problems. The blue engine has some improvements that seek to improve reliability independent of the turbocharging, lower the compression to allow for higher manifold pressure, and improve breathing to allow the turbo system to function more efficiently.
The expensive and unreliable things all relate to high rpms in 928 engines and this blue engine isn't going to run any more rpms than stock GT engine.
One of the goals in this (science) project has been to figure out what is absolutely necessary to modify to make big power with turbos and what can be left stock. We were able to take the completely stock engine about 700 rwhp without any internal engine modifications and without any obvious reliability problems. The blue engine has some improvements that seek to improve reliability independent of the turbocharging, lower the compression to allow for higher manifold pressure, and improve breathing to allow the turbo system to function more efficiently.
The expensive and unreliable things all relate to high rpms in 928 engines and this blue engine isn't going to run any more rpms than stock GT engine.
You've achieved that comfortably. Well done. Looks like Todd revs his motor out to 8500rpm from what I read. Clearly much has been changed in those motors.
Originally Posted by V2Rocket
Patrick...maybe you should just put a 928 engine into your racer...
A 9248 is born! No, I'm not a fan of V8's in these bodies. If we can get a comfortable 600whp + then that will be sufficient for our needs. Reliability is what we really want Spencer.
First, the 80 lbs injectors seem to be working ok, which is a major win.
We're still trying to chase down the parameters for the injector opening time / offset. The time depends on the injector type, supply voltage, and fuel pressure.
The relationship between the injector offset by failed high-school physics logic and some spec sheets should be about as follows. The failed high-school physics guess for the injector opening time = a + b*(1/sqrt(c*supply voltage^2 - fuel pressure force), motivated by solenoid force being proportional to the square of voltage, fuel pressure force providing a constant offseting force, and the time to travel a distance at constant acceleration being inversely proportional to the square of the net force. (This logic doesn't really work that well, but don't worry, I'm not planning to quit my day job... ;-)) In practice, the Sharktuner interface takes the user-provided offset value at 13.1V and then applies a correction to the offset time based on a table inside LH.
Given that our fuel pressure regulators are boost referenced, and thus keeping the difference between the manifold pressure and fuel pressure in principle constant, we don't need any corrections to fuel pressure in the offset parameter or the flow tables. (Better have those fuel rails well dampened, though, to avoid resonances that could really throw the pressures and AFR's off!) It may make a difference for translating the spec-sheet offset parameter to the correct offset parameter for our application, though.
The inquiries and experiments to get the correct offset parameter for these Siemens 80 lbs/h injectors is ongoing. However, we already know its going to work out as the idle is stable an pulse widths are not very close to the non-linear and/or the inconsistent range of the injector.
As a second update, we're doing some computer work to figure out how to build a quiet exhaust for this car. The Silent Night exhaust currently on the car makes a lot of power but also a fair bit of noise. I need a quieter exhaust for this neighborhood of mine...
I've attached a useful (to me) pdf presentation about what kind of noises each kind of engine configuration produces. It conveniently has our firing order V8 as an example case.
Based on the attached presentation and some other thoughts specifically about turbo motors, we're now simulating an exhaust that has a cross-over in the form of Y-pipe. That y-pipe (or may not) may cause some turbulence and/or losses at high rpms, but should reduce noise and may do something positive to the torque at very low rpms. The simulations are mostly to determine whether there are any pulse tuning effects that need to be taken into account, or whether the only thing that matters is the steady state flow. The simulated exhaust configuration:
At this point, my pulse tuning thoughts are as follows. These are not verified by any sense, not even in a simulator let alone in physical reality. This is merely the current state of my evolving thinking:
- At very low engine rpms when the turbine hasn't spooled yet and can't be realistically expected to make any meaningful boost, the engine tunes somewhat like a normally aspirated engine and the turbine looks somewhat similar to an open but constraining pipe. In this regime, one wants to time the reflected low-pressure wave to the collector (defined as the point where the pulses from different cylinders first combine) at the time when high-pressure wave arrives from a cylinder to reduce the interference between cylidners. Call this normally aspirated regime.
- At slightly higher rpms when the turbine is in the process of spooling, there's still some normally aspirated tuning effects. There's a new effect, however, of timing the pulses on both sides of the turbine wheel such that the turbine is most effective in extracting energy from exhaust and producing boost. I don't understand enough about turbine efficiencies to know how these should be timed, just that they may make a difference. Call this transition regime.
- At mid-range and high rpms when the turbine is fully spooled and the exhaust manifold pressure is high, the turbine gas flow is close to sonic. Downpipe waves don't have much impact on what happens upstream of the turbo or even on the turbine speed. All that one can do is to make sure that the exhaust gas flows efficiently out of the turbine outlet. Call this the turbo regime.
The trick is to design an exhaust that works well in all three regimes while suppressing sound. Easier said than done.
I was actually thinking of X-pipe usage after twin turbos, it will give some scavenging boost and do not have Y-pipe drawbacks like possibly too small flow area and such.
Surely the exhaust pulse strength after turbos aren't that high, but velocity should still be relative high to create scavenge.
Have you already made simulation using X-pipe or is there something which makes you to skip the whole X-pipe idea?
I did something like this mainly for reducing the level of loudness, but also reducing drone. The cats are located in middle of those two X.
I was actually thinking of X-pipe usage after twin turbos, it will give some scavenging boost and do not have Y-pipe drawbacks like possibly too small flow area and such. Surely the exhaust pulse strength after turbos aren't that high, but velocity should still be relative high to create scavenge. Have you already made simulation using X-pipe or is there something which makes you to skip the whole X-pipe idea?
I don't think there's any meaningful difference between x-pipes and y-pipes for cross-plane V8s if the sizing is correct such that the mean flow velocities are similar. Different angles and constructions can make a bigger difference than the choice between x and y. (Kind of like with the original chromosomes of people vs. whatever is there after various surgeries... ;-) )
If you look at the exhaust sound slide deck pages 14 and 15, you'll see the x and y sound is very similar, too:
Yowza!! So you have made and tested all these configurations?? That's extreme. Sorry, to recap, are you running twin GT30 turbos? I would doubt you'd have any lack of driveability even at lower c/r piston. However I wonder if you had considered some of the Borg Warner EFR twin scroll range of turbos? I'm shifting from a GTX3582 to a twin scroll EFR9180 at the moment after good reports from someone with a similar motor. Not quite buttoned up yet but they do enjoy a good reputation among quite a lot of racers now. Very responsive.
Yowza!! So you have made and tested all these configurations?? That's extreme.
No, but I stayed in Holiday Inn last night! Those are generic simulations by someone else produced using the GT Power suite. See the whole presentation from the earlier post.
02-11-2017, 04:37 PM
