Safe boost for internally stock 3.0?

[KobaltBlau]

Hi guys, I have definitely been searching and reading but would like to address this question directly:

What is a safe boost level for an internally stock early US 3.0, using an OK intercooler, say a stock 930 IC?

I realize this question has far too many variables hidden. In an effort to reduce those:

* 92 octane, 5% ethanol - Washington state premium gas
* 40k original miles, bone stock 78 SC engine (single plug, 8.5:1 CIS pistons)
* oil temperature kept under control - say not more than 230
* Programmable EFI, tune however you like
* Programmable ignition, can program boost to influence timing
* Wideband O2 with driver-visible gauge
* 1 or 2 EGT sensors/gauge, perhaps. Also perhaps CHT
* Possibly J&S knock monitoring, though I am not sure about sensor mounting
* only driven by a driver who knows how to pay attention to EGT/WBO2/CHT/etc
* Sea level
* Ambient temp generally 45-60F (Seattle)
* no water injection, methanol, nitrous, etc.

tsuter mentions that he ran 9-11psi for years in his BAE, non-intercooled early SC. Any other data points are very welcome.

I would likely be using a GT30 or even GT28 series garrett, so 76 or 77% peak compressor efficiency is possible. The goal with this engine would be to produce 10psi as close to 2000rpm as possible. I would expect to lose out on the high end, but would choose the compressor map to go as far as I could. Max power is not really the goal, but 300 would be more than enough.

I'm not sure how much this lower-rpm targeting really affects the safe max boost, but thought it was worth mentioning. Could reduce max boost at higher rpm if compressor gets less efficient if necessary.

Thanks in advance for your feedback!

[wwest]

Not worth the effort and/or cost.

[K24madness]

I think 9-11 PSI is safe for the motor but maybe not the fuel. I would use water/meth injection to help with the poor fuel or run race gas.

[mark houghton]

I've heard of similar builds running safely .5 to .6 bar...call that 9psi max. With your higher compression and our fine Washington State fuel even that might be pushing it. But I am far from an expert, just what I've heard from others.
I would agree with meth injection. Lots and lots of discussion on that (meth) topic, started recently by David Cole (Cole930).

[Dr J]

As far as the engine itself, the 3.0 is pretty tough. The weak spot is the Dilivar head studs that fracture. If you can replace them, it would be great; however, that can turn into its own project.

[KobaltBlau]

Thanks for the feedback, guys.

wwest, I am sure it is not worth the effort/cost if the goal is max horsepower, but my goals are different - it would be an interesting project with the goal of creating a 911 with more modern boost characteristics.

K24, water/meth injection is a good idea - I'd prefer not to add anything consumable to the car, but it might be necessary to meet my goals. Do you have experience with this? Do you have any guess how much more boost am I likely to be able to run?

Thanks Mark, I'll do some reading on Cole930's posts.

Dr J, I think the Dilavar studs are primarily compromised by corrosion rather than cylinder pressure. while mine are old, the car has not had much use and been in dry states its whole life until a few years ago. Still, something to watch out for.

Of course boost is not really a goal in itself, but I need to have an idea what my target would be looking at compressor maps to see what kind of boost curve I could achieve with available turbos.

Since tsuter ran 9-11psi for years with no intercooler, I would hope that I could run a couple of pounds more with a decent intercooler and really good engine monitoring.

Does anyone have an internally stock 3.2 that they are running more than 6 or 7 pounds of boost with? Compression ratio is higher there...

Andy

[sjf911]

I looked at a lot of the compressor efficiency curves and finally realized that the KKK turbos really were optimized for <1 bar so that is the way I went. It makes it simple as it is OEM and you can use a lot of used 930 parts for the install. Also, water/meth injection is very simple to install and adds a significant safety margin to your build. However, control of timing is probably the most important factor to avoid detonation short of running race fuel. You can run into diminishing returns as boost goes up and timing has to be retarded. Meth injection helps you maintain more timing and therefore more power.

[KobaltBlau]

Originally Posted by sjf911

I looked at a lot of the compressor efficiency curves and finally realized that the KKK turbos really were optimized for <1 bar so that is the way I went. It makes it simple as it is OEM and you can use a lot of used 930 parts for the install. Also, water/meth injection is very simple to install and adds a significant safety margin to your build. However, control of timing is probably the most important factor to avoid detonation short of running race fuel. You can run into diminishing returns as boost goes up and timing has to be retarded. Meth injection helps you maintain more timing and therefore more power.

Thanks again for your ideas, Steve. I am thinking about starting with 930 parts, I realize that the flange is a T3 flange too, so you could use a 930 J-pipe with a Garrett turbo to get started on a budget, of course your muffler is going to change then but easy to have one fabbed.

The point you make about optimized for <1bar is a good one, that's actually what got me started thinking about max boost target, because many of the GT series turbos have efficiency islands >1bar. It does not seem easy to fit a KKK smaller than a 3LDZ though, from what I have discovered so far - K26 and K24 using a different flange.

Good points about timing and diminishing returns as well. Stock NA dizzy advance curve isn't going to cut it!

[wwest]

Originally Posted by sjf911

I looked at a lot of the compressor efficiency curves and finally realized that the KKK turbos really were optimized for <1 bar so that is the way I went. It makes it simple as it is OEM and you can use a lot of used 930 parts for the install. Also, water/meth injection is very simple to install and adds a significant safety margin to your build. However, control of timing is probably the most important factor to avoid detonation short of running race fuel. You can run into diminishing returns as boost goes up and timing has to be retarded. Meth injection helps you maintain more timing and therefore more power.

It is my understanding that "detonation" has nothing to do with ignition timing but rather is basically the engine "dieseling" due to the heat of compression. These days with electronic fuel injection spark knock is often handled via enriching the mixture slightly, which is why our modern high compression Porsche engines can be run safely and efficiently on regular fuel.

[Dr J]

You may find this calculator handy:

SuperchargersOnline.com :: Supercharger Horsepower Calculator

[David]

Not for our engines, but it gives you some reference (and 98RON is about what our super unleaded is):

[cole930]

WWEST:

Too much timing advance creates excessive combustion chamber temps. Increased
combustion chamber temps causes detonation.

Cole

[RarlyL8]

I use the K27S for Carrera 3.2L turbo conversions. They are designed for our engines (low boost application) and the S spools ~300rpm quicker than the 7200.

[spuggy]

Most engines can tolerate detonation to different extents, it is often pre-ignition that causes the major damage.

All reciprocating motors have much the same sweet spot for peak pressure, 14 degrees ATDC - if your combustion chamber design causes a slow charge burn so you have to light it off early to achieve full combustion for peak power (too much advance), you can end up with excess pressure elsewhere in the cycle. (Which is why twin plugging helps, because you can run less advance for the same LPP.)

High combustion chamber temperatures in conjunction with excess pressure (too much advance, too much boost) causes detonation.

Pre-ignition can occur entirely separately from detonation (e.g. from badly prepared or internally damaged parts), as they are separate and distinct phenomena - but one cause of pre-ignition is induced by detonation - the detonation pressure spike breaks down the gas boundary layer in the combustion chamber, at which point things go downhill very fast.

The aviation engine guys know a lot about this stuff:

Engine Basics: Detonation and Pre-Ignition: Streetrod Stuff

[spuggy]

Originally Posted by KobaltBlau

* oil temperature kept under control - say not more than 230

Front Mount Oil Cooler. I've seen 230 idling in traffic, but I've never seen more than 210-220 under boost, even in ambients over 100F.

Originally Posted by KobaltBlau

* Possibly J&S knock monitoring, though I am not sure about sensor mounting

Bolted directly to the center of the block (under the fan shroud) works fine for me. The J&S is a real win as a bolt-on with CIS - as it also gives you programmable boost retard - but any decent EFI controller will include knock sensing and full spark/mixture control...

[KobaltBlau]

Thanks for all of the great feedback, guys. I have been doing a lot of reading and I have a lot more left to do. While there is of course an absolute limit to the cylinder pressure that this engine can take with a healthy combustion process, I do feel that detonation and preignition are the big limiters here. Not big news. To that end, I am increasingly convinced that I should apply a J&S Safeguard knock sensing system to control ignition timing (350HP930 had some success with a similar system, and spuggy I'm glad to hear you have one too, I was unaware of that).

In addition, despite my initial resistance to Water/Methanol injection, I think that WMI is a great idea here. I would plan to set up a WMI system with a failsafe (lack of water/methanol flow) that ran a boost controller. So, if there was a system failure or you ran out of water/methanol, the boost controller shuts down going to wastegate boost level. The wastegate could be set at 5psi or so, with the boost controller at max allowable boost with functioning water injection. The aquamist hfs-3 can do this but I think some of the snow systems can do this too.

I think the WMI could be pretty effective even without an intercooler, as run with by patkeefe. This raises issues on the track with having enough water/methanol to run a track session, but being able to use the washer fluid reservoir for this gives pretty good capacity, also as run by patkeefe. A 964RS reservoir could be used for the windshield washer system if I want to retain it.

Twin plugs would be very nice for this application but that requires major surgery I don't want to do.

To try to answer the question originally posed directly, I've added some notes from real-world responses found in my search. I have added notes about ignition timing and other relevant details where available. All engines have stock internal configuration for their type as far as I can tell. No WMI used except where noted, nothing known about fuel except where noted, but nobody mentions race gas:

tsuter no IC 8.5:1 3.0
some distributor mods
9-11 psi
"over the years I have found 9-11lbs boost to be fine for 91-93 octane pump gas"

iamchappy no IC 8.5:1 3.0
MSD BTM boost retard ignition, auxiliary fueling
8-9 psi

friend of mb911 no IC 8.5:1 3.0 BAE T3/T4
7.25psi
dynoed high 280s hp

rfng no IC 9.3:1 3.0
7-9 psi
"With the right distributor you can run 7-9 psi no problem non-intercooled"

sammyg2 no IC 9.3:1 3.0
no ignition mods
7 psi

sjf911 IC, WMI, 9.3:1 3.0
8 psi
"the most boost I have tried is 8lbs ... I think the only way to really explore the limit is with an advanced knock detection system and a lot of nerve."

lr172 no IC 9.5:1 3.2 stock internals
megasquirt, controlled ignition timing "I taper from 32 down to 19 at 160 kPa"
8.7 psi
no noticeable knocking

pjv911 no IC 8.5:1 2.7
no ignition mods
7-8 psi
when first installed, 87 octane still in tank, no pinging at all with 8psi

copbait73 no IC 8.0:1 2.4 CIS
no ignition mods
6psi okay, some knocking at 8psi

I think sjf's comment is most interesting, I believe the variance in these numbers is more due to the tuner's nerve than the ultimate potential. In reading all this it does not seem like most of these folks have pushed their engines to the point where they could tell they had detonation, copbait73 excepted.

[copbait73]

Originally Posted by KobaltBlau

copbait73 no IC 8.0:1 2.4 CIS
no ignition mods
6psi okay, some knocking at 8psi

I think sjf's comment is most interesting, I believe the variance in these numbers is more due to the tuner's nerve than the ultimate potential. In reading all this it does not seem like most of these folks have pushed their engines to the point where they could tell they had detonation, copbait73 excepted.

I’ve not been active for a while, but I tend to join in when discussions move from the real world motor you have of air-cooled architecture designed nearly 40 years ago and modern engines. The 901 motor was a marvel for it’s time but frankly time has passed it by.

This said I love them and own 2 cars and 4 motors. The 901 motors ultimately have all of the limitations well known to it’s type despite Porsche throwing years of engineering and $$$$$$$ at it). Most air-cooled motors, 901 included, will still run when they lift the heads due to overtemp or too much boost. Water-cooled types must maintain this critical joint or they are toast. Big problem for them 40 years ago and a bonanza for Porsches. This feature was key to its TURBO racing success for years.

This engine’s limitations regarding boost vs. C.R. was well defined by the factory in the mid- 70s. I published the chart they generated back then. You can add turbos of higher efficiency and massive intercoolers but the limitations are basically the same. An air-cooled engine is ultimately limited by it’s ability to get rid of the serious increased waste heat that is the product of ALL power increases.

If you burn racing gas or E85 you can make more than the chart suggests. If you ceramic coat parts you can make marginally more power, large expensive intercooler, very high engine RPM or just plan old boost with no care for the odds of damage…...send the bill to the sponsor.

Just put on the turbo, set your boost at 6-7PSI. Drive it in the summer, let oil temp get up, drive up a crowded street in town for maximum heat soak of the engine structure, roll down your windows, then turn on to an empty side street with high curbing. Get 100% into the throttle from 3K to 6K. Pay special attention above 5K rpm and just after a shift. Bump the boost in 1PSI increments over a period of several days. You will eventually hear damaging knocking. I have it with stock timing and 6PSI (no water injection) and 8PSI (on water). BTW, I love water injection for the street. It works where you need it inside the combustion chamber.

[cole930]

Well put copbait !!!!

Sure makes a great case for adding WMI !!!!

Cole

[copbait73]

Originally Posted by cole930

Well put copbait !!!!

Sure makes a great case for adding WMI !!!!

Cole

Never broke a 911 engine but broke rings and holed a piston on a Type1, 6.25:1C.R. air-cooled VW motor running 24PSI. Rebuilt and added home-grown W/M and she ran at same boost till I sold her.

[cole930]

A short quote from an engineering article I read while doing research on WMI:

In a piston engine, the initial injection of water cools the fuel-air mixture significantly, which increases its density and hence the amount of mixture that enters the cylinder. An additional effect comes later during combustion when the water absorbs large amounts of heat as it vaporizes, reducing peak temperature and resultant NOx formation, and reducing the amount of heat energy absorbed into the cylinder walls. This also converts part of combustion energy from the form of heat to the form of pressure. As the water droplets vaporize by absorbing heat, it turns to high pressure steam (water vapor or steam mainly resulted from combustion chemical reaction), that would add engine output. The alcohol in the mixture burns, but is also much more resistant to detonation than gasoline. The net result is a higher octane charge that will support very high compression ratios or significant forced induction pressures before onset of detonation

Also in WWII the Army Air Corp experimented with WMI on supercharged aircraft engines. During the testing it was noted that they could ran as much as 25 times the compression ratio before they saw detonation. WMI was eventually used on many military aircraft such as the P51 Mustang, the B52, and numerous turbine aircraft. In the case of the P51 the WMI was noted as a significant factor contributing to the P51's reversal of air superiority in the European campaign.

Cole