The Ultimate Sleeper Engine Build-Off

So, this ought to be right up Avanti's alley:

I was playing around on Desktop Dyno the other day, and ended up producing quite a powerful sleeper. I thought it would be cool to see what you could come up with. Here's the rules:

You must start with a production engine (not a crate, race, or limited-production motor), you can bore it and stroke it, but you must use realistic bores. (No .120 overbores on weak, thin-walled blocks.) Furthermore, you must use currently available parts - no one-offs or customs. Also, you must list the price of the spec'd parts and any extra machine work. For argument's sake, we'll say that all the motors cost the same if found in a junk yard and all boring costs $200 (this is not sleeving). Don't price things like bearings, balancing, etc. Don't worry about ECUs, tuners, etc., just mechanical bits that Desktop Dyno uses. After a little while, I'll post a poll with everyone's sleeper engine, the cost to build it, and the power figures and we can all vote on who's is the best combination of power, value, and shock-appeal. Also, let's keep it to pump-gas and natural aspiration, with the idea that the engine could be street driven.

So, here's my sleeper: A common Mopar 318 - technically the later-model Magnum version.

.090" overbore from 3.910" to to 4.000". (Perfectly safe on the Magnum blocks.) [Disp. = 333ci] - $200.

4" Eagle 4340 Stroker Crank, Forged [Disp. = 402ci] - $690

Indy MA-X heads with the optional increased-size intake valves and porting and polsihing on the chambers, bowls, and openings - $1600.

Comp Cams Xtreme Energy 20-744-9 Cam, timing set to -8* - $333

TTI Exhuast 1 7/8" headers, w/ ceramic coat - $800

Stock intake, BBK Twin-52mm throttle body, flowing about 650cfm - $300

Using a mild 10.0:1 compression ratio, here are the results:



420 HP @ 5000 RPM
510 LB-Ft @ 2000 RPM
Cost: $3923

Yeah, buddy! Look at that torque curve - or should I say torque LINE! That's a damn goo cam grind, right there.

Not a bad little engine if I do say so myself. That would be awesome in my old truck! Yeah, that'd be a sleeper for sure. So, what's yours???

Desktop Dyno is all good in theory but honestly, the power numbers that you can get out of combinations you build is very inflated.

I replicated the old 5.0L that I had in my 90 LX on desktop dyno, and even manually inputed the cylinder head flow data, full cam specs, and everything. On a real dyno, my car put 284 to the wheels, for roughly 320 under the hood with a 5 speed. Desk Top dyno showed my engine making something like 435HP. Not even close to realistic.

From what I have seen, it can be very accurate. But for this, its fine. We're using hypothetical builds, and the same program. Engine vs engine specs should be accurate.

Sim, you could bump compression and still run on pump gas. It would just have to be premium. Shit, stock Hemis use 10.5:1 and you can get away with 87.

StalkerStang wrote:Desktop Dyno is all good in theory but honestly, the power numbers that you can get out of combinations you build is very inflated. …

I have heard that about Desktop Dyno. I can say they have gotten better with each successive version. I already had Desktop Dyno 2000 and Desktop Dyno 2003, and this summer I bought the trio pack of their most recent versions. Each software program does a very good job for what they are. Their quality control needs to be better at making sure errors don't ship though that's not a deal breaker.

This looks interesting Sim, I'll make a mental bookmark and take part at a later time.

Not sure it's that far off, Stalker. Indy is using their MA-X heads and their own cam with carbs to acheive similar HP and Torque ratings out of similar displacement engines. It's entirely possible that my combo could acheive the power figures listed, as I am using a different cam, the stock intake (which actually flows quite nicely), and MPI.

http://indyheads.com/newla.html

I'm not convinced Sim. Nothing on you.

Like I said I've built real combos on there and the power is so inflated. I just built a combo that I would like to do in the future and its in some 635+ HP range and I didn't even spec the cam, I just put a .600 lift general spec cam from the list in it.

Plus I don't like the limitations on certain aspects of the building. Those factors do affect power output and you can't control them.

As said in a prior post, the designers of the simulations are getting better. These are screen captures of two features added since the 2K and 2K3 versions of the software.


Above are two screen captures from DynoSim5 with ProTools. Looking through the features, they've added variable valve timing and rod ratio which are big though not as big as the top cell, environment. Now you may enter the temperature and humidity at which the engine will be tested. I know DD2K and DD2K3 didn't take that into account, at least, it wasn't user specified. Prior versions may have been using a fixed variable, and if they were, who knows how far out from real numbers they were?

I have a feeling Motion Software is honing areas like Boyle's law, Charles' law, and other areas within thermodynamics, fluid dynamics, etc. As with anything, it's difficult to completely replicate real life. The recurring theme is, "they're getting better." I trust the current software to produce 90% accuracy of gross power output. If I put together a combination with 450 peak horsepower, I multiply by .9 and am now down to 405 horsepower. Then I factor expected real-world powertrain loss multiplying by .85 and I'm now down to ≈345 horsepower. I trusted the older versions to 80% accuracy of gross power output: [DD2K(3) power × .8 × .85 ≈ expected net power].

Powertrain loss is a funky thing. People (so far) have figured it to be a percentage, meaning more gross horsepower equates to a larger chunk removed. Removing 15% of 450 is ≈68 which drops 450 to ≈382. Removing 15% of 300 is 45 which drops 300 to 255.

Desktop Dyno is a lot like personal accelerometers from G-Tech, Escort, and other companies. They're fun to tinker with but don't take their results as gospel as these devices are off by a little. Is it possible to find the exact (factor / range / formula) these devices are out? Yes.

Fun thing to ponder: I wonder when we'll start to use magnetism or electromagnetism to transfer power from engine to wheels or for piston movement?

magnetism and electric magnetism is an electric motor. Now if you wanted to use them or some other technology to control the valves instead of a cam then you might be on to something. Individual valve control and variation independent of everything else in the motor could be useful (displacement on demand, VVT, etc...).

Just give me a 90 degree V-8, pushrods, two massive valves per cylinder, and F.I. I'm a simple guy. Thanks Chevy and Mopar! And if that sounds stupid to you, I have seven words for you:

National Association of Stock Car Automobile Racing.

Here's an interesting read for engine nerds.

I think your 85% powertrain loss it off. You realize that most modern automatics are more than 95% efficient, right? Some are 99%.

However, wheel power is funny. It can be affected by gearing, tire size, etc. In fact, through gear manipulation, you can produce more or less power at the wheels than at the crank. Gear Vendors has some good info on that.

If I build that engine, it will go in my 97 Ram, with a GearVendors behind the trans...

Simcik wrote:I think your 85% powertrain loss it off. You realize that most modern automatics are more than 95% efficient, right? Some are 99%.

However, wheel power is funny. It can be affected by gearing, tire size, etc. In fact, through gear manipulation, you can produce more or less power at the wheels than at the crank. Gear Vendors has some good info on that.

If I build that engine, it will go in my 97 Ram, with a GearVendors behind the trans...

If you find a drive train that puts 99% of the power to the ground I want to know what black magic makes that happen...

Yeah Sorta, I agree. There is a lot of wasted power rolling through the torque converter (or clutch), transmission, driveshaft, differential, axle, and out through the hubs to the wheels. So far I've found 85% to be the best guesstimate in difference between what the engine produces and what the whole car puts to the pavement.

Manual tranny with clutch fully locked in, low friction bearings, properly aligned shafts and gears... Viola.

And some of the new auto-shift manual transmissions are nearly perfectly efficient.

http://www.geartechnology.com/issues/0912x/minimize-power-losses.pdf

Granted, I didn't take the accessories into account, but you could find ways to minimize them... Like I said, 85% is too high.

Still got wheels and tires, diff gears, driveshafts.

Newer trans are more efficient than older ones for sure, but they still use a 13-18% loss for most drivetrain losses.

Simcik wrote:… Like I said, 85% is too high.

I'm not saying "85% loss." I agree, that would be too high. I'm saying "85% of" which slots right in to the 13% - 18% loss that Stalker mentioned.