small displacement turbos
 

This is not a project I'm attempting or anything, its just something I thought about today and realized Id never seen done before. Maybe there is no point. I dont know. Thats why Im asking here.

Has anyone here ever built or ever seen a small displacement turbocharged flat 6 porsche engine? Small displacement meaning an engine based on the early 2.0 aluminum case. I cant really see where it would be any cheaper than building a 3.0 based turbo, but Im curious about the possible engine configurations anyway. Anybody done this in a teener or a 911 that needed to use a 901 tranny?

If its pointless or if it has been covered before I apologize ahead of time.

TIA

I have ground cams for racing cars with under 2.5 litres. No boost until 5000, peak torque around 7000 and peak HP over 8800 RPMS.
Of course these were race only engines. A different camshaft could bring the boost and power band in earlier.
I think Porsche had a 2.1 litre turbo race engine too.

Well I see no problem with turbocharged 2L engine with low C/R pistons and smaller turbo. Actually, it can be done quite cheap as there are lot's of turbos in that range.

My suggestion would be older oil-cooled Garrett T3 that can be salvaged from SAAB- and Volvo cars. Slap on the intercooler and limit the boost to 0.5 bar and you shouldn't have problems with reliability either. Boost should come around 2700 RPM's with smaller A/R T3's....

How does the Saab t3 compare to the Ford T-Bird turbos?

Dunno what sits on T-bird's. T3 is generic Garrett turbo for smaller turbocharged vehicles in 80's and is easy to find cheap. We have plenty of SAAB's and Volvo's around so i know there are available at every junkyard.

There is even a "hot" version called "Trim 60" whick will flow more air.

Generally, they are worth 220-260hp at most and suitable for 2L motor.

There are better turbochargers available nowadays but most are watercooled.

Thanks for the response everyone. I appreciate the help.

Ok, this actually sounds pretty neat. 2700 rpm is not a undesirably long lag for a turbo is it?

Goran, PM on the way

TIA

I seem to recall that Martini was running a 1,7 litre turbo way back in the early days. Supposedly based on a 2,0 motor.

Of course, I am frequently wrong in my recollections.

I can't contribute must to the discussion except to say the a few years ago I acquired a 906 engine that came to me with a very German looking mechanical injection with a turbo enrichment pod on the pump. The system was for a mid engine car so it was either in a 906 or a 914. Displacement was 2.0 The engine was too valuable to leave like that so we took it off before it ran and I have no idea as to the out put. http://forums.pelicanparts.com/uploa...1085074343.jpg http://forums.pelicanparts.com/uploa...1085074352.jpg http://forums.pelicanparts.com/uploa...1085074384.jpg

does anyone sell a bolt on turbo setup for a carbd 1970's Flat 6 pork engine? id be very interested to find out!

thanks

With the 2.1l turbo motor you could run in PCA GT4. A gentleman back here in Maryland is just working the bugs out of his 2.1 turbo 914. I know for a fact that this motor makes considerably more power than the normally asperated 2.8 liters I have seen.

nein14-6, any details on the 2.1? Do you know if it was built on an old 2.0 aluminum case?

T-birds have T3s along with lots of other cars.

Definitely true about better turbos now (like mitsubishis or garrett GT), but most watercooled. I bet you could run a little cooling circuit for the turbo; might be too much trouble. Perhaps you could just run oil through the water jacket so you knew the cooling was sufficient.

I'd be very interested to see a smallish (2.0-2.4) displacement 911 motor with relatively high compression (8.5+) and 10-14 psi boost, and how it peformed (and cooled) compared to similar water cooled 4 cyl like saab/audi/modern volvo/etc "HPT" I guess beepbeep is doing this the cheaper (!!??) but probably not easier way by using that engine in the first place :D

nothing wrong with small displacement turbos. brabham had the 1.4 litrew bmw turbo in their bt55 F1 car and in qualifying trim it made 1400 horses then for the race they turned it down to 1100. they even used engine blocks from 2002's that had over 100000 miles because they stress was relieved from the case. cool stuff, i havent seen any small displacement porsche turbos on street cars yet...

Shuie,
I’m sorry, the gentleman spent a great deal of money on the motor and if very reluctant to give details. I know it is a sandcast case and it incorporates the 3.2 carrera style manifold with the intercooler sitting in the rear trunk.
You can find pictures of the car on www.clewett.com under “photo gallery” then “914” and scroll to the bottom of the page, the pink car and the motor associated with it.

also look into the turbo dodge setups.....i have a couple garrett T4's, they also will have the vacuum wastegate built into it and can accomodate 2 1/2"-3" exhaust outlets.
i have a vast collection of turbo dodge links/info if anyone is interested, these engines are either 2.2ltr or 2.5 ltr depending on the crank.

problems with 911 engines with high boost are twofold:

1. They have bad combustion chamer geometries, flame-front has long way to traverse so pressure-raise during combustion is somewhat slower. Can be band-aided with dual plugs though. Also, chamber-shape is more knock-prone.

2. They have individual heads which is worse when it comes to keeping "lid" shut when combustion occurs. Watercooled cars have one head per cylinder bank which means that there are many studs keeping it in the place. When combustion occurs (and it's almost always only one cylinder at a time per cylinder bank), head will be kept in place not just by head-studs surrounding that particular cylinder but by the adjacent ones too. Being single-piece cast object, that head will tolerate more boost before it starts leaking. 930 head, for example, is kept there by only four studs and will start "walking" when subjected to high boost. Even if studs doesn't crack, it will eventually lead to "uneven surface" problem and oil-leaks at cylinder/head mating surface. There is also continious heating/cooling problem related to engine load and RPM which isn't there on watercooled cars due to moderating effect and huge specific thermal coeficient of water.

Porsche knew this and used either disposable welded head/cylinder assys or watercooling and one-piece-per-row heads on it's race engines. So as much as we like aircooled engines (and i do as well), they are inferior design from beggining, kept alive overtime by huge R&D and customer demand. They are like tube amplifiers... inferior but lovely nevertheless.

Back to your initial question, I believe 8.5 C/R and 1 bar of boost on smallish 2.2L boxer six would be fully possible as long as other inherent problems are solved (mag cases are prone to stud-pulling, rods are weak etc.) Dual plugs would raise the limit some more.

Something similar could probably be made with early turbo/Carrera alu case, short stroke and small bore pistons. It would be one helluva nice-sounding engine...

What i would really like to do is to take a 2.7 Boxter engine, change rods/pistons and turbocharge it...now that would be a screamer. It has all good stuff already there: watercooled & hydraulically adjusted 4 valve heads with centrally placed plug.

That being said, what's the problem with using bigger displacement Porsche engines? They are stronger, will rev almost as much and are likely to be less stressed then 2L ones?

First of all, Goran, you and I have similar interests, although you are taking more action. Since I first drove a car with ~1 bar at 2000 rpm, I have been very impressed with this type of engine.

due to piston shape? chamber in head should be relatively knock-resistant with twin plug. (I have no data)

Agreed. However, is this really a problem at 1 bar and 8.5-9:1, if preignition is avoided, with quality steel studs? I'm not sure if anyone can give a definitive answer, but I tend to think it wouldn't be. As you probably know, combustion pressures increase radically when preignition occurs, and many owners of tuned turbocharged cars are not very careful about this. More care can ensure a longer lasting engine in many ways.

True, but the engines they were using those techniques on were running very high boost.

My main concern here is that I like the (original) 911, and I like the engine it has. I don't like the kind of crap most people who convert their cars to water cooled engines have to do for cooling plumbing and radiators. If you are able to complete your conversion without running front radiators, I will be very impressed (I'm not doubting, I just think that is a much nicer solution, like the guy with the geo metro engine did)

In practice, it would probably be easier to just do this on an SC motor. No longer small displacement, but strong and very nice. I would not consider single plug, and would try for 9:1. (heck, my engine has 9.3:1, and it probably isn't quite that high in practice, hmmm.) I'm not sure if a turbo small enough to provide 12psi say by 2500rpm could flow out to 7000 rpm, but that would be very nice. But you could build the hardware on this type of motor to go 8000.

interesting idea. But if you try to put that engine in an air-cooled body, you probably have more of a cooling challenge than you guys have with the SAAB engine.

I just read this now, but I agree. this goes with my SC comment.

Thanks for the interesting feedback Goran,

Andy

Im surprised youz guyz missed this

for my part, I didn't miss it, I'm just not as interested in torque peaks at 5700 rpm.

Neat race engine though.

BTW, all my posts in this tread assume a decent intercooler on the 9:1 1 bar setup, and twin plug as I mentioned.

369 lb-ft @5700 rpm is nothing to sneeze at....

at 2000 rpm its probably got a higher torque output than a NA 2.0...at 3000 rpm its probably got more torque than a 2.0 911S...at 4000 its probably got.... etc..

Choose a proper cam, time the cam approprialtely, combine with some of the new efficient turbos and you could have a wicked small displacement turbo engine.

Ive been squireling away parts to buil a similar engine for years.. I have a set of iconel 935 header, TWM throttle bodies, GT3 oil pump etc...

I think ill stick it in my 914, instead of my 911

Hold your head
 

[QUOTE]Originally posted by beepbeep
problems with 911 engines with high boost are twofold:
They have individual heads which is worse when it comes to keeping "lid" shut when combustion occurs. Watercooled cars have one head per cylinder bank which means that there are many studs keeping it in the place. When combustion occurs (and it's almost always only one cylinder at a time per cylinder bank), head will be kept in place not just by head-studs surrounding that particular cylinder but by the adjacent ones too. Being single-piece cast object, that head will tolerate more boost before it starts leaking. 930 head, for example, is kept there by only four studs and will start "walking" when subjected to high boost. [QUOTE]

Supertec Head studs will reduce and under certain circumstance eliminate head movement. One customer is running our head studs at 45 ft/lb of torque. Although we don't recommend head torque in this range it seems to work with his 700+ hp.
Watch for the SUPERTEC HEAD STUD KIT on the pages of Pelican parts soon.
http://forums.pelicanparts.com/uploa...1085785571.jpg http://forums.pelicanparts.com/uploa...1085785621.jpg

According to my calculations, our engine will pass the boost treshold @ 2600 RPM and reach 1 bar @ 3200 RPM.
Full boost won't be achieved until 4400 RPM though :-) That's the price you have to pay when using big turbocharger. We'll see how good my estimate works in practice.

Flame-front has finite speed that is pretty constant regardless of RPM. With two plugs, you "lighten" the mixture from two sides so whole mixture is combusted faster, which leeds to more distinct pressure peak in cylinder. It means that more of that pressure can be used to push the piston downwards and you don't need to lighten it too early. Twin-plugs won't make the chamber less knock-prone directly but will allow you to achieve same power with less boost or more power with same boost (as more of cylinder pressure is converted to work)

Speaking about chamber design, domed pistons are usually big no-no when it comes to knock resistance. You want your combustion chamber free of sharp edges and slightly spherical. This is little easier to achieve when you have four small valves instead of two big ones.


Unfortunately, i believe that it happends even on stock 930's, given enough time/miles. Many are modified which only hastens the process. It's not immediate failure but slow cumulative process that progressively gets worse. It is also speeded up by the fact that head/cylinder temperature is fluctuating all the time on aircooled engines which means they will sustain many more thermal cycles (with moderate delta T though), compared to watercooled engine which effectivly only gets one per drive. Different thermal properties of heads and cylinder, however slight, will eventually lead to leaky seals after prolonged "grinding"...refered as "uneven mating surface".

Not so terribly high, actually. Around 1.2 to 1.4 bar, with C/R ranging from 6.5:1 to 7.0:1. Note that those boost figures are nowadays used on cars with C/R of 9.0:1 and higher, thanx to advances in combustion-shape geometry and knock-detection.



Yepp, that is going to be our biggest challenge. I'm against cutting out the front trunk and spoiler and spoiling (sic) the air-flow underneath. My initial idea is to use a radiator adjacent to engine, on the cold side. If that's not enough, we'll run additional small cooler in front lip. Third way is mounting radiator high up in the wing, but it will raise CG so it's last resort.


You can get a pretty decent idea of where your boost treshold will be by calculating mass flow at certain RPM's and plotting it on compressor map. My guess (without really calculating) is that Garrett GT32 would fit the bill quite nicely. It might blow hot air at 7000 RPM but it's not a showstopper, just a price you pay for low-end grunt.


Hehe. No, I was thinking of turbo-converting a existing Porsche Boxter (2.7 "the hairdresser edition"). It's light and has middle-engine. Very good platform. I understand that PAG doesn't want to develop it too much, it would start eating into bread-and-butter 911 segment ;-)

My pleasure Andy!
Cheers!

Definitely True! I'm just curious about a turbo 911 based engine that reaches peak boost (and first reaches high torque) at 2000-2500 rpm. More a street engine than a track engine, certainly not a race engine like that one. That engine is very interesting in its own right, I've just been thinking about something different :D

PS Goran I don't have time to respond now but I will later, your reponses are again very interesting.