Max torque from a 3.0l rebuild
 

Apologize if this has been asked elsewhere but search didnt give me much. I have an 83 ROW and I like to auto-x. Recently made the switch to a megasquirt setup (Bitz) with the intent of a bigger, better engine in the future. Engine has no know issues with about 100K on it but underwhelming in an auto-x environment and ho hum in street driving. I have not had much seat/tuning time since installing the MS but do notice a lot more throttle response. Assuming I reuse my case, what can I put together to give me the most torque? I have a lead on a 3.6 crank and rods as well as all the existing 3.0 bits. Local 911 wrecker has older parts so newer stuff, if needed, will need to be sourced. Reasonable budget on this with the expectation that once done its my "forever car"

Hi, are you sure your engine is in fine health? Normally the ROW 3.0 SC engine is quite peppy...
If it's not leaking oil,and has good leakdown, I'd look into cams, good tune on the efi kit, and shorter gearing plus an LSD if you don't have one.

As I mentioned Im into the tune phase so am optimistic it will be more responsive. My CIS had several issues which I was aware of.

My build question is for a replacement engine that I would like to build. Its not an immediate need but I know what I like to feel in the seat of the pants so it will be sooner rather than later.

Here is what is generally available
bore stroke & displacement
95 70.4 2994
97 70.4 3121
98 70.4 3186
95 74.4 3164
98 74.4 3367
100 74.4 3506
102 76.4 3746
104 76.4 3894
100 76.4 3600
99 76.4 3529
100 80.4 3789
102.7 80.4 3996
104 80.4 4098

here are the stud spacings, which limit bore
964/993 90mm
3.0-3.2-3.3 86mm
2.0-2.7 83mm

I suppose that you could build a 98mm bore x 76.4mm stroke(993/or 964crank) 3458cc motor which would likely be pretty torquey w/ the right tuning

the issue w/ 100mm bore on a 3.0 block is cylinder stability, similar to the 3.8 conversions on 964/993 w/ 102mm bores and 107mm spigots as opposed to 109mm spigots

JMHO, but for torque on a budget start w/ a 964 or better a 993

Al- I also have an '83 ROW SC, and echo Arlo's comment: Properly set up, that car will GET UP AND GO!! Mine had been acting a bit lazy a couple years ago, and my mechanic realized the injectors were a bit off. He cleaned 4 and replaced 2, and this car hasn't run this well in 20 years!

Mind you, it starts coming on strong at 3800 and keeps building to 5500. If you are seeking 'torque' at lower revs for autocross, that powerband may be less than ideal- I don't know that style well. Just my 2c from a distance.

In general torque output is proportional to displacement for normally aspirated engines. So bigger engine will have more torque.

john

Its that 3800 power band Im trying to get down. Being a cab, track days are done for me but never miss an auto-x. My last 2 cars were a lotus 7 clone and a miata. 911 is no comparison for power to weight which I accept but that off the line push into the seat is what I want.

The get out of the hole TQ from a 3.0 can be very impressive .
Here is a link to a dyno sheet and drag slip from my 3.0
This drag time is with a traffic light start , just letting the clutch out smoothly,
full out at 1,200rpm. I blew the Civic Type R away at the start, 3 car lengths ahead
at the finish line , 3" ahead, enough for the win, @ 101 mph.
This engine is all stock except for the cams, and warm up regulator .
With the right parts a 3.0 can make 260 hp and 235 TQ

Have fun, http://forums.pelicanparts.com/porsche-911-technical-forum/1073886-3-0-rebuild-performance-advice-2.html#post11082047

Thanks. Thats the kind of info Im looking for. My engine isnt broken so Im not looking to fix it, just planning for a future build to get the most out of it to make this my "forever car". I will always be upgrading something or else the car becomes boring to me.

For an auto x car, I would concentrate on reducing the weight of the car as much a reasonably possible, at your comfort level.

I regularly AX my car. It has an SC motor with carbs, backdated exhaust, midrange cams, and JE pistons and weights in a 2200lbs. It starts to pull around 3K rpm, and with my low geared trans it is perfect for AX.

I have known several guys with SC motors running CIS that were quick at AX, btw.

You may find this thread interesting. http://forums.pelicanparts.com/911-engine-rebuilding-forum/1114059-preferred-cam-timing-adjustment.html?msclkid=e97be9e1cfd011ec895678afa4 a29d54

Already well down the path to "enlightenment". Floor & engine sound deadening removed, bumper shocks for euro tubes, rear seats, backdated heat etc.. We run classes and my RWD 3.0l puts me with other 3.0l and its quite a spread in HP.

What did you decide upon?

I havent yet. Still looking at options. Id like to build this as a second engine so what I find as a core will dictate most of it. If no core, then Ill play with what I have over the winter. Currently a torquier cam will be the first step.

I had two broken head studs on my 3.0 so I was planning on rebuilding it into a 3.2SS. I did the same thing - bought megasquirt, installed ITBS, and worked out some bugs before I took the engine apart.

I was very familiar with the impact of cubic inches in dealing with VW engines (built a 2332cc with ports/valves/induction/cams geared toward torque). There is no substitute for cubic inches.

I started calling around to find shops in my area and stumbled upon William Knight. After a conversation with him I went down the cubic inch route. I ended up building a 3.55 with ITBs and MS3X and love it. The power from low RPM is just nuts and there is no need to ream at high RPM all day long - it just pulls, and to me, for street driving, this is where the fun is. Anyway, I'd recommend giving him a call. He's a good friend and a great resource for nearly everything in the air-cooled performance space.

Spoken with William. We talked back and forth and he threw me some ideas and then life got busy. Have to reconnect with him again.

What is this 3.55 you speak of?

If it were me, After weight (as some have noted), I would first look at "gearing"...a 915 with the 7:31 ratio (or similar) and 15 inch wheels is a winning combination at speeds under 70...

It starts with the 3.0 case, custom crank, rods, and pistons (98mm), open the '81 heads to 39mm intake (CNC), 964 oil pump, etc. William selected the cams based on my preferences (I like torque for street driving). It's a mild mannered street engine with loads of low/mid range torque.

Is that custom crank around 80mm stroke?

I think Joes crank was 78.4 mm. we made it from a 3.2 crank.. It gives better rod journal size and higher piston speed. WK

No Motec, you used a Megasquirt? More info please!

I didn't use Megasquirt,. He has made it work out though

Yes, I used MS3X. Aside from some injector driver issues, it's worked fairly well for me. It's full sequential, 2 O2 sensors, knock sensors, etc. I have no experience with other ECUs, but MS works great and is easy to understand/use.

FWIW, William calls it a Commodore 64. Hahahaha... and I get that, but until I can swing, and have the patience to install, Motec M130 or EcuMaster Black, Megasquirt will have to do.

Joe - I get the Commodore 64 joke, but I don't really get it. These ECUs all take inputs, compare to tables, and give outputs to manage fuel injection (and also spark), don't they? Don't they all interpolate between points on a matrix? How much better is a larger matrix? Are things happening in our engines (not F1 stuff) so fast that ECU speed matters that much? Especially if the tune is a special one, like for WOT operation, where you don't care about cruise/part throttle, MPG, even idle, etc?

Exactly, Walt. What's the baud rate of a factory Motronic ecu? 8000 or so? And they're still running fine today. And before that, Bosch used *gasp* analog computers and rotary switches for throttle position!

It's a joke. Megasquirt comes from a long history of DIY, build-your-own-board, etc, history that, to me, gives it a certain stigma. I also understand that not all ECUs are created equal in the math and the hardware. I've had some issues with the hardware having to replace injector drivers because the drivers that came in the MS3X were not reliable enough to drive a Toyota smart coil. I found a few posts where others had experienced the same thing and ended up soldering in new drivers. Problem solved. But, because of this, I too question the reliability of Megasquirt. Beyond this hardware issue, it's been fine.

The other aspect is simply the math/programming. Not all algorithms are created equal, either. Some are simply programmed better than other using PID vs PI and using different control schemes versus other schemes. I'm not an EE and I'm not a programmer, but I know enough about both to discern the difference.

For me, MS3X has been perfect. Tunerstudio is great to work with (intuitive and easy to use/understand). Megalog viewer is great to work with. The wiring, setup, amount of control, features, etc, are all great, to me. I am a DIYer and love the process of learning, doing, and redoing in an attempt to make things better/perfect so it's a match made in heaven. I have rewired my engine 3 times now improving the craftsmanship, adding features that I didn't think I would need/want, etc. And I am at it again - engine out, adding a second O2, adding a CHT, and a few other nick/nacks.

If I were building professional cars, racing professionally, or looking at this from any other lens than the one I use, I would heavily consider Motec, Haltech, ECUMaster, or some other, more reputable, ECU to install that has the backing and resources to develop them to a higher standard than my perception of Megasquirt.

All said, Megasquirt is hard to beat in terms of value, features, install base, forum support, tech support, etc. For a beginner, I would choose to use MS3X every time.

It would be interesting to understand the true architectural and design differences between MS and Motec. My own experience, which is very limited, is that the likely failure cases of the design are probably more important than the differences in processing speed.

My friend's MS3 developed two very frustrating failures that were likely due to MS design or ECU hardware degradation. The first was consistent stumbles at various RPMs with no logged indications of the cause. The second failure caused a complete, random stall when driving at highway speeds, with no warning and again no logging of any indications of reason.

We spent a lot of effort and time to try to diagnose these 2 issues, even though one was repeatable and reproducible, and the other random and frankly pretty scary to reproduce: drive around at highway speed until the car stalled.

Both were (finally) fixed by applying firmware updates to Megasquirt.

This is just a single anecdote and I have no interest in maligning MS (I much prefer TunerStudio and MS tuning maps over say the older Motec M84 ECU Manager), but this was a disturbing failure case. Perhaps this was an edge case that very rarely occurs, or time will tell that it is a failing piece of hardware. I couldn't say.

I would love to know if this is a MS-specific design / hardware issue or not. Perhaps someone has insight.

I don't know if or how often other ECUs have similar failures. Personally, the ECU relying on firmware updates to continue to function safely left me with some concern. YMMV

Either more displacement or forced induction will provide more torque lower in the RPM band.

As others said, you should probably check ignition advance and/or fuel enrichment - an SC should be fairly peppy/responsive low/mid-range. I still run SC cams in my 930, even though they're all done by 5500 RPM - purely because I think the low/mid-range response is more useful than slightly more maximum power at the expense of mid-range. Lower rotating (clutch, flywheel) and/or reciprocating (rods/pistons) mass doesn't hurt.

You'd probably need quite a lot of extra displacement for it to help appreciably. Like a 3.6 swap. Not generally noted as a cheap approach.

On the other other hand,.a full-size turbo good for 400+ HP (K27/HF - fairly trick 30 years ago) makes 6 PSI around 2500 RPM, and opens a stock (0.8 bar) wastegate by 3000-3200 depending on gear (eg "load"). Makes ~80% of peak torque around 3000 RPM. Kind of a difficult trick for an N/A motor, without variable cam timing and/or induction lengths.

A smaller turbo (eg low-boost setup) will spool faster. More efficient/modern designs (ball bearing, better compressor wheels, split volute, variable impeller vanes) don't hurt spool either.

This could be done relatively cheaply/easily (J pipe, turbo, MAP sensor, BOV, wastegate, muffler). If you stayed down around 75-100HP, it wouldn't seem especially unrealistic/unreasonable (at least, not to me, YMMV) to ignore (or defer until convenient) clutch, transmission, brakes, suspension, oil pump, cooler etc upgrades. Not like you're full-chat down the Mulsanne straight.

You already have precise fuel/spark control; big help right there. If you use active knock detection/mitigation, you might be surprised as to how much boost you can actually run.

"Adding power makes you faster on the straights, subtracting weight makes you faster everywhere” - Colin Chapman

It's obviously impractical to get an SC down anywhere near a Lotus 7. Caterham 7 620 makes 310HP weighing in @ 1342 lbs. Don't drag one of those for pinks...

On the other hand, extra 100HP (with near-linear power delivery) would certainly help with soccer moms in minivans wanting a point-by. Adding 250HP gets you comparable power-to-weight to 20 year old Ducatis, Gen 1 Dodge Vipers, lesser Ferraris and the like.

Just sayin'.

Saw a pic of a supercharger on a newer 911 engine and the numbers from that impressed the hell out of me. Owned a couple of GM cars with 3.8's & superchargers in them and when you hit the go peddle, it went. Still more stuff to consider

Supercharger installs were quite popular back in the day apparently, the perceived advantages being linear power delivery, and boost from idle (eg no wait for turbo spool). Although against that last is overcoming parasitic drag to drive the compressor (meaning you might not break even until higher in the rev range). Supercharger tech has likely similarly advanced in the intervening 40 years or so...

I think many consider DIY install packaging pretty simple with a turbo. Mount the turbo on a bracket where the hot side won't melt everything, connect some pipes - job done. In comparison, hallmarks of a supercharger install tend to be a substantial mount and a honking belt drive - some of the older units are pretty bulky.

Even old turbo designs spool faster and are much more linear than mismatched early 70's units. Folk lore(?) says the 3DLZ was originally intended for a big diesel truck, being all that was available to the factory off-the-shelf in 1974 - and hence the infamous/atrocious turbo lag. It's (more or less) "free" to drive them, given that you had no other plans to do anything useful with the exhaust gas...