Engine Update and Exhaust Question!!

[jluetjen]

DTW;
I haven't seen any suggestion in my data that there will be a noticable difference in feel between a motor pulling 76 m/s at peak HP versus a motor pulling 96 m/s. Both are "good enough". But then I did all analysis because it was cheaper then building up 36 different configurations of motor so that I could try them all out. Once you get the motor into the car, it would be interesting to have you report the results or dyno data to validate my prediction. Specifically: when does the motor start to pull smoothly? When does it "come on cam". A dyno would be more precise, but a lot of us can't afford it!

BTW: For my records, does Dave know what the port size was before he started cutting?

[Wayne 962]

The ports are the least of your concerns at this point. You really need a copy of my new book (available in B&W form sometime next week or the week after).

Some concern areas:

- Time-serts are old news. They are not reliable, and you should use case-savers. If your mechanic wants to convince you otherwise, then I'll show you some photos of pulled time-serts.

- If using MFI, you do need to make sure that you have the appropriate t-body and pump matched to the camshaft. Not cheap.

- You need to have your cylinder spigots machined flat, otherwise you may tweak your engine case when you bolt it together. The mag cases bend and deform and create a condition where the center spigot can be several thousandths below the others...

$1000 for all this work sounds super-cheap, and my guess is that there will be corners cut. Are the rods going to be resized?

JE can make just about any piston - you might want to go up a half-point on the compression scale.

Remember, you get what you pay for. Walt (Competition Engineering) is considered to be one of the best (if not the best) in the US, and uncompromising on quality. Here's his price list:

http://www.competitioneng.com/Catalog.htm

-Wayne

[Early_S_Man]

John,

Your calculations don't seem to account for the 'taming' of the 'S' cams by increasing displacement ... this isn't a criticism, just an observation.

Your data on the cams and various hp and torque peak rpms seem to be based on the 2.0 engines. There is quite a drop in peak power rpms at the 3.0 engine size ... just using the 'S' cams and MFI as an example, and the progression downwards as engine size increased was incremental:

2.0-> 2.2-> 2.4-> 2.7-> 3.0

6800 6500 6500 6300 6200 rpm @ hp peak

5500 5200 5200 5100 5000 rpm @ TQ peak

Any comments?

One final data point ... the same, EXACT head part number was used for 2.2S, 2.4S, and RS 2.7 engines! Whether changes were made from summer 1969 to fall 1975 is unknown.

And, regarding Dave's expectation of increasing port speed with increased port size ... it is, in fact, just the opposite [perhaps not directly an inverse proportional relationship, but definely in that direction] when all else is held constamt, incresing port size REDUCES PORT VELOCITY!

[RSR.guy]

Dave,

Sounds like you are on the right track. I've had good luck with the GE-60 cams, although you probably don't want to go too much wilder for a motor that will see street use.

With regard to MFI, I prefer the fuel injection over webers. I agree with Doug, you may want to check in with Gus before you get everything assembled, cams timed, etc. You might also consider going straight to mechanical injection and avoid the webers all together. Once setup, the MFI is damn sweet. For me, it was a lot easier to bolt everything up on the bench rather than in the car. Gives the assembler a chance to look hard at all the brackets, belt tension/alignment and such.

I have not had a problem with backpressure (or lack thereof) and mechanical injection use. The only issues I've had in that regard are over-scavenging with some crazy header/flowmaster exhaust combos we were trying on weber carbed race motors. The MFI tends to dump a fair amount of fuel into the chamber in comparison, and the big key (as already pointed out by others) is to match injection pump space cam to engine camshafts.

Cheers, Chris

[jluetjen]

Your calculations don't seem to account for the 'taming' of the 'S' cams by increasing displacement ... this isn't a criticism, just an observation.

Hi Warren. You're right! The reason is that I ran all my data through a correlation analysis and regression analysis -- and -- engine size didn't seem to make a difference. It might also have been masked by adjusting the cam timing, which also wasn't in my analysis. But at the end of the day; it didn't show up. A classic case is the 906 cam used in the 2.0 906, 2.3 and 2.5 ST and the 2.8 RSR. It always generated peak HP at 8000 RPM and peak torque at 6200 RPM. So while everyone talks about the "taming" of the cam -- I don't see it in the data.

My theory. The greater overall torque available from a bigger engine (all else being equal) tends to make the engine feel less weak at low rev's. For example let's assume a hypothetical cam which generates the following torque (X):

2000 RPM: 50% of peak torque
4000 RPM: 80% of peak torque
6000 RPM: 100% of peak torque
8000 RPM: 80% of peak torque

Most studies show that torque tracks engine size very closely. It is really hard to improve torque by much given a certain engine size. So you have a 2 liter engine that produces the following:

.5X at 2000 RPM
.8X at 4000 RPM
X at 6000 RPM
.8X at 8000 RPM

Everyone agrees that .5X is really anemic and the engine is only really useful in the car at 8X or above. The engine is really "off cam" until 4000 RPM and then it starts to "pull hard"

OK, now you take that cam and put it into a 3.0 liter engine. Now it puts out the following:

.75X at 2000 RPM
1X at 4000 RPM
1.5X at 6000 RPM
1X at 8000 RPM

Now everyone says that it doesn't feel so "cammy" because it most people can't detect a 5% points difference in torque between .75 and .8 at 2000 RPM. They just now that now it pulls at 2000 RPM while before it didn't.

If you also add in that lots of times people don't port the head when they change the cam, so now the 3.0 puts out the following:

.75X at 2000 RPM
1.2X at 4000 RPM (Smaller ports are improving swirl or tumble)
1.2X at 6000 RPM (Note the ports are choking flow)
.75X at 8000 RPM

Conclusion: Wow, the cam's a lot milder!

Now if you take this same 3.0 (or any normal American V8 which is choked off by the ports in the first case) and open up the ports resulting in the earlier results and suddenly you find people saying...

"Wow! Look at all of the HP that porting made for us!" Actually the HP was already there as a result of the cam, but the heads were the weak link and restricted the engine's performance.

'That's my theory. I'm open to rebuttal But please bring your data!

[Early_S_Man]

John,

As far as the 906-cammed engines of different displacement, I believe the heads changed progressively, with bigger ports and valves ... when the heads were almost the same, i.e., 906 & 911/21 & 22 with 38 mm ports, the power peak dropped to 7800 rpm, 230 hp with Webers & 250 hp with MFI.

I have always been amazed by the proponents of the 3.0 CIS engines (particularly the 3.0 Carrera and 930/10 Euro SC) engines and their claims of massive torque available and vast 'superiority' over the RS 2.7 engines, yet the data just doesn't support the claims!

Granted, the ROW SC engine does have more torque available at lower rpm, and peaks 900 rpm lower, but the numererical difference is ZILCH! RS 2.7 -- 188.4 lb-ft @ 5100 rpm ... 930/10 SC ROW -- 189 lb-ft @4200 rpm ... Carrera 3.0 188 lb-ft @ 4200 rpm ... and for comparison: RS 3.0 -- 203 lb-ft @5000 rpm ... '84 930/18 3.0 MFI SC/RS 184 lb-ft @ 6500 rpm ... USA 3.2 Motronic 192 lb-ft @ 4800 rpm

Conclusion ... build me an MFI 2.7 or 3.0 any day, and gear it LOW!!! Funny how some people claim superiority over something they have never driven! Those 'peaky' RS 3.0's sure must have been NICE with a 915/08 transaxle and 7:31 gearing!!!

[jluetjen]

As far as the 906-cammed engines of different displacement, I believe the heads changed progressively, with bigger ports and valves ... when the heads were almost the same, i.e., 906 & 911/21 & 22 with 38 mm ports, the power peak dropped to 7800 rpm, 230 hp with Webers & 250 hp with MFI.

I agree Warren. Here is some data that supports that what you are saying. All of the following engines are using the 906 cam.

Engine.....Peak HP RPM....Intake Port.... Gas Speed at peak HP
'65 906 2.0.....8000...........38mm.....78.0m/s
'71 911 2.3.....7800*..........38mm.....85.9m/s
'71 911 2.5......8000...........41mm.....83.9m/s
'72 911 RSR......8000...........43mm.....85.9m/s

Here's an interesting case that I got from this BBS:

Kurts 3.2.....6950...........38mm**...108.5m/s

Using the modified 906 known as RSR cam:
'74 911 RSR......8000...........43mm.....91.6m/s

* The 2.3 ST that you mentioned seems curious in this context given that they all supposedly have the same cam timing. The ports are no more restrictive then RSR at 8000. It might suggest that for all out race engines, you don't really want to go over 85 m/s. But then there is the '74 RSR.

** I'm pretty sure this is what Kurt is running, but I don't have footnoted where I got this from.

Hmmm, back to the drawing board...

[dtw]

Originally posted by Wayne at Pelican Parts
The ports are the least of your concerns at this point. You really need a copy of my new book (available in B&W form sometime next week or the week after).

Well perhaps I'll sign up- what's the going rate right now? 100 bucks for ring binder b/w with an IOU for the press copy? Sounds fine to me -

Some concern areas:

- Time-serts are old news. They are not reliable, and you should use case-savers. If your mechanic wants to convince you otherwise, then I'll show you some photos of pulled time-serts.


Well, perhaps I'll be another data point for you, because at the beginning of the week half the case was done. Wayne, I'm definitely no expert on 911 engine building (this will be my first), so I'm not going to refute your stance on timecerts vs. case-savers. I do know a few guys that have been building 911 engines for years; two are pros with their own shops (like my local wrench that does mostly bone-stock rebuilds of tired engines, and there's David who builds some pretty outlandish racing engines), I've consulted all of them since the first time you made that comment. None of them have heard of problems with timecerts pulling, though they all recoiled at the mention of Helicoils (sorry about that pun).

- If using MFI, you do need to make sure that you have the appropriate t-body and pump matched to the camshaft. Not cheap.

Agreed. I've got mag t-bodies which can be bored if necessary, and I'm aware of the costs involved in re-camming the pump.

- You need to have your cylinder spigots machined flat, otherwise you may tweak your engine case when you bolt it together. The mag cases bend and deform and create a condition where the center spigot can be several thousandths below the others...

Sounds good, I'll doublecheck that David has taken these measurements and we'll take any needed action.

$1000 for all this work sounds super-cheap, and my guess is that there will be corners cut. Are the rods going to be resized?

It is super cheap. As I said, he gets a kick out of DIY guys and likes to help out. I've seen his engines at work and have confidence in his craftsmanship. I do not believe that corners will be cut. I hate being told "Told you so", so at this point I suppose I'm going to get another pair of experienced eyes on the work before I bolt it together. You guys sometimes make it sound that an engine is bound for a short, smoky, 10,000 mile life if Walt didn't do the machine work for the rebuild. I simply don't believe this to be the case. There are other guys out there quietly doing great work at reasonable prices, and then there are Motor Shuysters spooking everyone.

JE can make just about any piston - you might want to go up a half-point on the compression scale.

Note that the plan right now is new JE 9.5:1 CR pistons, 9.5:1 being suggested by many camps to be the max recommended CR for a single plug, street gas engine. Clarify?


Warren- yes, zuffenhausen cleared me up on relationship of gas speed to port size.

John- 32mm. I'll ask David about dyno testing, I know he doesn't have one but he's down near all the NASCAR race shops outside Charlotte and probably knows someone.

Chris- Yes, I think you're right, I owe it to myself to talk the operation over with Gus before sinking a grand into Webers. I'll give him a call soon.

Everyone- thanks for the great dialogue- this is definitely educational.

[jluetjen]

... And when your done your going to drive around the country and take us all for a ride with your new motor -- right?

[Wayne 962]

Agreed - Walt is not the only source in town. Obviously Walt cannot do everyone's engine.

However, Walt has a lot more experience than most, and has also learned a bunch of things that a lot of other people *don't* know. This is documented in my book. I'm not trying to sound like I'm pushing the book, it's just that I learned a *tremendous* amount from writing it, and I am shocked at how many potential mistakes are made out there! I'm not saying that you have to use Walt, but reading the book will certainly give you a long, long, long list of questions for your rebuilder. If he can't answer them, then I suggest that you move on...

Warren, the Porsche parts books are so screwy sometimes. Although the 2.2/2.4/2.7RS heads may have the same part number, they don't have the same cylinder spigot spacing. You can't use a 2.7L head on a 2.4, however, you can modify a 2.2/2.4 head for use on a 2.7L. (84mm versus 90mm). THe 2.7L head exposes teh wier CD ring on the 2.4 cylinder - it will burn out.

I'm not saying that you said it could be done, it's just that it's (incorrectly) implied by Porsche by them having the same part number...

-Wayne

[Early_S_Man]

Wayne ... the stud spacing was changed for the 3.0 engines to 86 mm, not the 2.7s!

And, the 2.8 RSR engine used stud spacing of 83 mm, exactly halfway between the 2.0/2.2/2.4/2.7 value of 80 mm and the 3.0 spacing!

[A Quiet Boom]

Warren, thank you for pointing out the "taming" of a cam via increased displacement. Unless the flow rate of the head changes in purportion with the displacement increase cams most certainly are tamed and quite often not a good choice on the larger motor. Excellent examples of this would be 260-351 Fords and 305-400 Chevys the heads on these engine families are very similar within each family, what would be a very peaky high rpm only cam in a 260 Ford is more of a midrange cam in a 351w. I have to wonder why no one on this board knows or talks about flowbench numbers for the various 911 motors since they are very helpful when selecting supporting parts for an engine build up. I mean flowbench numbers typically at .600 valve lift are commonly used when comparing V8 heads etc. why not for P-cars? I wonder if these numbers aren't closly guarded secrets by race engine builders.

[RoninLB]

Walt at Competition Engineering did my machine work..sure there are other good 911 machinests in the U.S.. not too many, but you can find them..I stumbled on CE, and Walt was the Man...after the job was ordered he would continue to answer my phone calls/so what..well, I would have a written list of specific questions and talk fast to get off the phone ASAP. but the time spent on the next phone call would grow for the same "time alloted" next group of questions...I was a little confuised till I figured out what was going on. that oversized outdoor forest person actually respected what I was doing to a 2.7 engine..he held my hand thru my wrench act of the rebuilding/big time..like I really became friends with him, no $hit...and this whole thing went down years ago, so any infactution has worn off/I have a clear head on this one..

Anyway, IMHO, if you decide on CE, have a specific purpose of what your engine will be used for before you call him..don't waste too much brain fuel of what has to be done..think ahead about your wrench act and burn the fuel getting the rebuild done correctly/no problem/bullet proof/easy. Walt's a friend..........Ron

[dtw]

Originally posted by Wayne at Pelican Parts
I'm not trying to sound like I'm pushing the book, it's just that I learned a *tremendous* amount from writing it, and I am shocked at how many potential mistakes are made out there! I'm not saying that you have to use Walt, but reading the book will certainly give you a long, long, long list of questions for your rebuilder. If he can't answer them, then I suggest that you move on...

Dude it's your forum, push the book

The upshot of being out of town all the time is that this engine is not having bolt the first turned on it for several months. That'll give me time to pick up your book and make sure I've addressed all applicable questions.

Funny you should say that John- this car has been so controversial among family and friends that zuffenhausen and I think I should just install a driver seat and nothing else. That way if someone wants to ride shotgun, they must prove their worthiness by providing their own seat. This is not a problem for most of my friends, as they all have 911 seats lying around their garages. Ha!!

[CamB]

Just to add to the cam/engine size/hp debate.

My engine is a 3.2RS - S cams and short stroke (98mm p&c) 3.2. I'm cheating with extra-high compression and EFI.

I have:

- 1978 SC case and crank, rods
- 98mm Mahle P&C with 10.3:1 c/r
- twin plug (964 dist)
- S cams
- ported and polished heads - 42mm inlet, 40mm exhaust
- standard (1978 - same year as case I assume) SC valves
- early MFI throttle bodies bored out to 40mm, converted to EFI
- MoTeC 3D controlled EFI
- SSIs and standard 2 in 1 out exhaust

Peak power is 270hp est flywheel (245hp at hubs) at 6220rpm and 231 ft/lb at 5120rpm. Correction factor is times 1.1.

Click here for dyno graph (100kb)

I'm probably giving away a little bit with the single out exhaust, but in the words of Rolls Royce, the power level is now "adequate". Even with EFI, the engine is still a little fussy at low rpm...

By the way, I now need 98RON and it is only semi-available (easy in Auckland, hard in the country). I have now realised this is a bit of a pain in the arse for long distance/remote touring. I am going to investigate if I am ok with 96RON...

[Wayne 962]

Warren, sorry if I wasn't 100% clear - I wasn't referring to the stud spacing - yes, as you pointed out they were the same. The cutout in the heads for the 2.7L pistons are too large for the 2.2/2.4 pistons, hence you can't use the 2.7L heads for the 2.2/2.4 engines. In the same manner, you can't use a 2.7L case for a 2.2 or 2.4 engine.

Hope this clears it up...

-Wayne

[Wayne 962]

I echo the sentiment about Walt. Not only is he good, insist on quality, has a waiting list, but he's also a really nice guy.

-Wayne

[jluetjen]

Cam;
You've been holding out on me! I was looking forward to your dyno data to test my spreadsheet.

Here's what I predicted for your motor:
HP: 253 @ 6300 (BMEP of 163.9 psi and intake gas speed of 80.5 m/s.)
Torque: 213 @ 5100 (BMEP of 162.6 psi)

Your Actual:
HP: 270 @ 6220 (BMEP of 177.1 psi and intake gas speed of 79.5 m/s)
Torque: 231 @ 5120 (BMEP of 176.3 psi)

Not too bad if I can say so myself , but a little conservative in regards to the delivered HP. I've been thinking about adding a term for CR which I have not had up to now and see if it get's me closer. Right now my calculations assume that the engine's CR will be the same as an S's, up around 9.5 +/-. You've got 10.5:1 and that obviously makes a difference.

Generally S's ran BMEP's of about 162 psi at max HP and 165 psi at max Torque. You're doing quite a bit better then that. The engines with 10.3:1 CR's like yours (and usually 906 cams) generally turned about 172 psi at peak HP and 185 psi at peak Torque.

Hmmmmm......

I'd say that you did pretty well for yourself.

Thanks for sharing!

[CamB]

Sorry John! I thought you had seen the numbers scattered through my other random posts.

Either way, I am extremely pleased - I've got heaps of power and it is wicked the way it gets up on cam like a smaller motor but with bags of torque below 5000 as well.