for my next $500 question...

in trying to properly size a turbine for my application, I need help with engine delta P. I understand the theory but I have no real world experience with it.

my goal for this project is steady 5lbs of boost from as early as possible to redline, with most driving happening between 3k - 4.5k, with 4k - 6k being where i really want the power.

using BW matchbot, a 57 or 58mm turbine wheel will let me generate 5 psi around 2500 rpm, which is fantastic! except that past 5500 rpm my engine delta P starts to drop with my 6000 rpm value being -1 and at redline it is -2.

a larger turbine wheel (64, 67, 70) will keep that delta p to positive numbers to 5000k and then 0 to redline, but I am obviously sacrificing my early boost to about 3000 - 3500rpm.

so, my question is this: is -1/-2 delta p enough to start to hurt exhaust scavenging and cause power losses due to the recirculation of exhaust back into the cylinder during valve overlap, which on my engine advances significantly at 5500rpm (just past where the smaller turbine wheel pushes delta p negative).

as always, comments and criticisms welcome.

best,
ike

btw, i swear i will start contributing back to the forum when i have enough knowledge to help someone lol

 

So is there an engine that belongs with this turbo, or a vehicle, or an application?

Engine output effects turbo sizing more than anything else, so size and as much of the configuration as you know.

Vehicle: weight, gearing, tire size

Use case: A turbo for a towing app is different from a street car is different from an autocross car is different from a road race car is different from a drag car.

Let us know what your parameters are and the forum will start spitting suggestions at you, you can look further at those suggestions and ask specific questions about the recommendations.

Don't worry too much about the math. At some point, once you have spent some time here someone will tell you the specs I asked for and you will just spit out the answer. A lot of this boils down to experience and shoving a lot of examples into your brain library.

If your interested in the math portion for academic reason: @gruntguru and @Boost Engineer are incredibly smart in that department. Search their posts. I use a lot of simple math to explain basics, but these two are another level.

 

Sorry, I didn't mean to leave out important details, I figured that was a more universal question that an application specific one. The car is a 00 porsche boxster with a 2.7l flat 6 with CR of 11.4:1. Curb weight is around 2800 lbs, I'm running 255/40/17s michelin ps4s in the back. The tranny is a 5 speed manual, gear ratios are 3.5/2.117/1.428/1.090/0.837.

There really aren't any ways of getting much power out of this motor other than to swap it with the 3.6l of the same era 911, so that's why I've started looking into converting it to FI. There aren't any good ways to lower CR on the engine so I'm stuck with a low boost project. At stock setup, this motor really doesn't make any real torque until you hit 4k rpm. Hence, most daily driving is done between 3-4krom but real driving is 4k-6500k. I do not drag, the car will be for fun and occasional autox.

Between 2k and 6.8k (redline) rpm I need 6-16 lbs / min at PR 1.35 to meet my goal. This will give me nearly 250ftlbs of torque across the board which should be fantastic compared to the stock setup (and even destroy a 911). I'm doing a twin setup because of the small displacement and my desire to have early power and to keep my flows high while maintaining low PR, and the setup needs to be remote because of size constraints (making this choice even more important). I've looked at BW 5862 and GT 2860. The problem with both of these compressors is that I will be over the surge line at low rpms. Then I found the mitsubishi large 16G wheel - the compressor map looks perfect for this application.

So then the question comes to the turbine. I think mitsubishi's setups are really neat because the big 16g can be had with 3 different turbine setups - td04h, td05h, and td06h, each of which come with different ar housings. I forget the exact measurement of each of those turbine wheels (I don't have my notes in front of me right now) but comparing them to BWs wheel sizes using the charts in match it I think that the td04 is too small, the td05 will give me what I want but will lead to a negative deltaP at higher revs, and the td06 will cut my early boost by a bit but will keep the deltaP positive through the range.

So this is where my question regarding deltaP comes into play. Good performance stock NA setups run with exhaust pressures of at least 1-2 psi. So will a deltaP of -1 or -2 at high rpms actually hurt me in FI an application? I can very easily see why we would always want to build a system built with positive deltaP through the entire range, but what role does this play in the real world. I know that these setups always require give and take, I'm just trying to be as scientific as possible with my selection so the project doesn't end with "ahhhh shit"

On a more global level, how accurate is BW match it when determining these numbers? I would love to know how these numbers actually hold up in a real application.

I scoured the internet looking for more information on this topic, but detailed discussions about the topic refer to it's role in egr, not to it's role in performance. I've seen those guys names brought up a few times in other threads through so I will definitely read through their posts to learn more.

Thanks again for the input!

 

Matchbot is as accurate as the inputs you give it. So if you can get those dialed in you will get very good plots out of it.

For your goals I would avoid the mitsu turbos, they are old tech compared to the modern borg or garret units. If you don't have a tight budget one of the 52mm EFR series would meet your goal and have some packaging advantages. The s252 would likely do very well for you, either one would be happier at a higher boost level than what your looking for (most turbos like to make lotta boost).

An exhaust to intake pressure of 2:1 is not going to hurt you, and is going to be almost a necessity for you to get boost in early. As a word of caution: putting boost into an engine that is not designed for it very early in the RPM range is really hard on it. You end up making a lot of cylinder pressure. I don't know the reputation of the engine your working on for strength, but the car is fairly light, has relatively deep gearing and small tires, that should help keep it alive.

 

I looked at the s252 but had previously ruled it out since a PR of 1.35 is below the lowest wheel speed line. I didn't know how to extrapolate data below that line so I didn't want to chance it.

Is there a problem with the mitsubishi turbos? I know there are some design differences like the shaft diameter being much larger and no incorporation of water cooling for most of them, but I haven't found any real "negatives" about them. Sure, they don't come with billett wheels, anti surge housing, or ceramic BBs but I didn't really expect to benefit from any of those in my application. Truthfully the simpler and more reliable the design, the better off I am at this point.

Thanks for the heads up about boost at low rpms, I had not encountered that info. Why is that? In general this motor does not appreciate "lugging" at low rpms so maybe I should rethink about where I want my boost.

In general, when changing a NA to FI (which I know is generally not a good idea but just humor me for a minute) is it best to try to design the system to augment the stock output curves? I would love to get a flat torque curve starting at a lower rpm, but your comments make me see how that could be too hard on the internals.

 

General rule to keep an engine alive is bring peak boost in past the natural aspirated peak torque curve of the engine. Remember the engine will make peak cylinder pressure at peak torque, so by feeding boost in after that point you keep the pressure lower.

How you or the OEM design the engine to operate, and what components the bottom end has will ultimately decide what its going to take for boost early. More important is to look at the application. For instance towing inherently puts a lot of stress on an engine, where a light vehicle with lower gearing and limited traction will be much easier on an engine. But you could put a bunch of stress into it if you load in the turbo while the engine is already lugging on the highway.

As for the pressure ratio your right. A 1.35 is quite low, really no turbo is going to be super happy there. If your not above 1.5 a supercharger or engine swap would be a better option. Superchargers are great for low boost and low RPM, no matter the SC they are a rising boost curve. A turbo is making max boost as soon as it is spooled unless you have some fancy controls going on.

Remember to factor in altitude for your boost calcs. If your at 6000 feet atmospheric is approximately 3 psi lower. That changes your PR, it allows you to essentially have the first 3 psi free.

The deal with the mitsu turbos is just that the wheel designs are old, they are generally less efficient. Not bad turbos, just dated.

 

I appreciate all the advice. After talking with you I think that I will shoot for building boost later in the rpm range. Its really not feasible (or fun) to wait until after the NA torque peak as that doesnt happen until 4500 rpm on this motor, but I can definitely set it up to spool when the curve slope starts to straighten out a bit past 3k.

Speaking of altitude, I definitely factored all that in my decision-making. I'm near sea level so no freebies for me. My "vision" for the setup is actually to emulate running the car at 1.5 miles below sea level (hence the flat boost curve). All this was actually inspired by how much better the car pulled on a subzero temp day. I can't drop the temp, but I can certainly pack in a bit more air... I know it sounds silly, that's just how my brain works.

I thought about a SC for this application but there is zero room in the engine bay for a traditional setup. I considered a "remote"setup where I switch the transaxle to an audi quattro and run a pulley with two roots blowers off the rear drive flange of that housing, but that's above my capabilities of fabrication. Maybe after I blow this engine up I will try that approach though lol.

Seriously though, thank you for your time and advice. If you're ever in the lake cumberland area give me a about and I'll buy you a beer.

 

It's not that you won' t make more power before the tq peak, it's just that you generally don't want to bring in full boost before tq peak. If you target 16psi max boost, it would be best if that didn't happen until after 4500rpm. No reason you can't be at 10psi before that and enjoy the swell of torque.

 

I suppose that brings up the age old question of "how much is too much". I haven't been able to find a lot of detailed builds in conversion of NA motors to FI, but from what I have seen for a CR of 11.4:1 it probably best to keep boost at a max of 5. That's why I am shooting for a flat boost profile that builds early, as opposed to just about every other build that builds boost to a peak. My reasoning is not so much to create a "forced induction" as much of an "augmented" one, if that makes sense.

After talking this out though I'm not sure why I want boost before 3K rpm, other than just out of principle. I think there are several good reasons to do so, the most important one being to avoid anything all things that may increase the likelihood of this ending in catastrophic failure.

 

Search LS turbo or Coyote turbo for examples of NA engines turned into FI. The latest Coyotes are 12:1 compression.

Look... You've got a turd. There. You make what? 200hp on a good day? Go ahead and put a turbo in place. Let er rip. If it pops, so be it. What did you lose? A 200hp placeholder.

 

Well stated! I get roasted on the 986 forums for pointing this out but this motor is awful and I secretly hope to blow this one up to do a v8 swap. My wife won't let me just throw out a perfectly good engine though.

Thanks for the lead on the coyotes. I didn't realize they were such high compression.

As always, the help is much appreciated!