Engine Output Numbers

[Clark K.]

With all the horsepower numbers thrown about, I have been researching the facts. My '65 L76 A/C coupe has been recently dynoed at 203 hp @ 5,200 rpm. The dyno tech told me that it was running way too lean and the air temp. was 103 at the time. Both conspired to lower output. I have since had two sizes bigger (#78) secondary jets installed.

The most recent issue (Oct. 09) of "Corvette Magazine" had a tech question/answer that you may find informative.

According to "Ask a Tech Nerd":
Prior to 1972 models, GM used SAE (Society of Automotive Engineers) standards J245 & J1995 for testing engine output at the flywheel; This is commonly referred to as a "SAE Gross" rating; For example: the L76 Engine Option was rated 365 hp SAE Gross;

Beginning with 1972 models, GM began using SAE standard J1349. This standard was the output measured at the flywheel with the engine running in an (almost) "as-installed-in-car" condition (w/A/C, fan belt w/fan, water pump, alternator, air cleaner, etc. installed); This is commonly referred to as a "SAE Net" rating; This process gives the exact same engine an understandably lower rating. For the stock 1965 L76 engine using this standard, the output would drop 80-100 horsepower, to 265-285 SAE Net. Alarming as this seems, put it into perspective by comparing this output to the 160 SAE Net HP of the bigger '75 base-engined Corvette.

Beginning in 2005, GM began using a new testing standard that is arguably more "real world". The tech article did not quote the SAE standard, elaborate about the procedure, or give a conversion number for this standard.

The author did, however, state that a chassis dynomometer (dyno) test would measure (rear) wheel output, not flywheel output. The drivetrain would eat up about 10-20% of the engine output using this test. So, I assume gthat a new stock 505hp rated Z06 (LS7 engine) would probably "dyno" 404-455 horsepower at the rear wheels. By doing my own calculations, the stock 1965 L76 engine should produce 212-257 horsepower at the rear wheels.

I am impressed with the acceleration of my L76 Coupe, but I cannot begin to imagine the feeling of a C2 with a LS7 engine. Originality is special but I can begin to understand the motivation of those non-NCRS types that do engine swaps. -Clark

[John C.]

Re: Engine Output Numbers

Just for additional info, Super Chevy Sept. '09 issue engine dynoed three classic Chevy high performance engines:Z28 302, L76, and LT1.
302: 356HP @ 6500 332 lb-ft @ 4300
327: 352 @ 6300 363 lb-ft @ 4300
350: 353 @ 5500 391 lb-ft @ 3800

Interestingly, they also bolted 487 heads to the 350 to come up with a lower compression '71 era LT-1. This engine made 362 HP @ 5800 and 382 lb-ft @ 4000.

I believe these engines would perform better than factory engines as they adjusted the timing on the dyno vs. using the factory settings, used headers, and used a modern Holley 750 Street HP carb.

It's nice to see a magazine do this and illustrate that the original factory ratings were generous but that even so, they are still strong street engines.

[Clark K.]

Re: Engine Output Numbers

I would love for Duke Williams to weigh in on this one. By that, I mean, the Super Chevy test vs the tech information from Corvette Magazine.

BTW Duke, my '75 Cosworth 2.0 liter engine with stock EFI recently dynoed at 125 hp @ 6600rpm (100 degrees F ambient air temp). That output should be equal to a stock '75 Corvette 5.7 liter. -Clark

[Joe R.]

Re: Engine Output Numbers

I'm sure that Duke Williams will weigh in here with a more detailed response on the various issues that have been raised, but I would like to highlight one key point that I learned from Duke.

Many "engine build" magazine articles test the engines with open headers. This is the best possible configuration for a camshaft with a lot of overlap, because the headers and the low back pressure help to exploit the camshaft's overlap to achieve good exhaust scavenging. Many of the power curves published by cam manufacturers also use open headers.

However, the minute you add some back pressure (mufflers) to the system, the high overlap works against you, and even the beneficial effects of the headers are greatly reduced. Even a modest amount of back pressure can make a real dog out of some of those magazine builds.

I recently purchased the Engine Analyzer software and have been learning how to use it (with some assistance from Duke). The EA software lets you configure an engine and then simulate the effects of different cam timing and other parameters, such as exhaust back pressure.

I used EA to "build" a 400 HP small block engine based on one described in a magazine article. Then I added some exhaust back pressure corresponding to a typical high flow dual exhaust (500 CFM). Peak horsepower and torque dropped dramatically (about 50 each).

A related problem with engines that are optimized for peak power is that they typically produce very little power below 3000 RPM. Magazine articles often focus on only peak power, and often don't even provide data below 3000 RPM. That's okay for racing applications where the engine spends all its time in the power band, but it doesn't work well for typical street driving.

There are a lot of other considerations when comparing real world street performance and peak power ratings obtained on an engine dyno, but these differences were the most revealing to me. I like the fact that EA lets me experiment with various configurations and see the effects various changes with all other factors being equal.

[Duke W.]

Re: Engine Output Numbers

First of all "horsepower" numbers are worthless without full context, and a key context issue is whether the data is "observed", "corrected", and if corrected, WHAT ARE THE CORRECTION CONDITONS - SAE gross, SAE net, DIN, JIS...? Also, the type of chassis dyno is important. Mustang dynos typically read 5-15 percent lower than Dynojet dynos.

In addition to adding the exhaust system and front end accessories, SAE net air density alone reduces output from SAE gross air density by about 4.5 percent.

RWHP numbers should always be expressed as "SAE corrected", but as was the case with your post you didn't specify anything including the type of dyno. Usually the output printout will indicate the type of correction and actual correction factor.

If you test on a Dynojet, get the digital test files. Then you can view/print them with the WinPrep7 viewing software that is a free download from the Dynojet web site. The software allows viewing/printing of various data during each run and viewing/printing multiple runs together along with the ability to use various correction standards.

As a general rule for a front engine/rear wheel drive with a manual transmission, the driveline/tire efficiency is about 0.85 in direct drive, so a good estimate of SAE net at the flywheel is (fourth gear) SAE corrected RWHP divided by the above factor. Intermediate gear efficiency is about 0.82 and automatics are about 0.80.

Gross to net conversion varies with the details of the exhaust system and front end accessories, and as I said above, net-gross air density correction alone is about 4.5%. Typically, 0.80 is used as the net/gross ratio, but based on data I have for C2 small blocks with 2.5" OE exhaust my estimate is 0.89, so about 40 percent is air density correction and the rest is the exhaust system and accessories. The OE 2.5" exhaust is very efficient for a massaged SB - only about 3 psi backpressure, but gets restrictive with big blocks, and generates 5-6 psi backpressure at about 350 SAE corrected RWHP, so the net/gross ratio will be lower.

Most factory power ratings from the sixties era are grossly overstated - by 10-20 percent. If you were to take a typical "365 HP" Flint-built engine and test it to SAE gross standards on a lab dyno, my estimate is that it would show about 310-320 gross. Massaging the heads alone with no other changes will get it up to 350-360 and extend the useable power bandwidth 500-1000 revs.

If you search the archives you will find a couple of "327 LT-1" tests, one on a lab dyno and the other on a chassis dyno - about 360 SAE gross on the lab dyno and 278 SAE corrected RWHP on a Dynojet, however, since this engine was tested on a hot day with no extenal fan, the viscous clutch was "locked" which probaby cost about 10-15 HP.

A decent OE SHP/FI 327 should make 220-240 SAE corrected RWHP (if you can keep the fan clutch from engaging) in the 5000-6000 range, compared to the OE ratings of 340-375 gross.

So you can use the above data and ratios to compare these and your engine.

A bone stock CV engine makes about 90-95 SAE corrected RWHP. My stock long block with a little port blending/matching (probably doesn't do much with the OE cams since the head has so much flow capacity relative to displacement as machined by Chevrolet) reindexed cams, open exhaust, and no fan made 122 SAE corrected RWHP, so I expect there are some critical details you left out of your engine configuration or test details. There is no way in hell a "stock" CV will make anywhere close to 125 SAE corrected RWHP. The SAE net rating is 110 at the crank, so low nineties SAE corrected at the rear wheels is the expectation and fact.

So my engine (with not exactly perfect cylinder walls and ring seal, but okay for an OE road engine) is about 145 gross. If I blueprinted it and raised the OE 8.2 (my actual measurement) CR to 10.5 it would be in the 160+ range, which is somewhat in line with Chevrolet's early reports of "170 gross" from development engines circa 1970, and I expect those engine had 10-11:1 CRs.

So how much horsepower do you want? Depending on test conditions and correction factors I can come up with just about any number you want! That's why numbers without full context are WORTHLESS!

There are several tests from the 2004 Roundup in So. Cal. with my comments on the CVOA Web site, and some testing done by Bob Maloy published back in the eighties in CVOA Magazine jibes with the recent results.

The data referred to in this recent article on various SB configurations does not apparently represent the actual "stock" engine configurations, such as inlet manifold, but no further details were provided, and it always seems that magazine tests show more power than tests from comparable configurations that I have in my files.

For an explanation of why two identical engines other than different strokes will make about the same power at about the same mean piston speed search the archives for "similar engines". This is why the 302 and 327 were about the same.

As is the case with most things in life, the devil is in the details!

Duke

[Duke W.]

Re: Engine Output Numbers

I'm sure that Duke Williams will weigh in here with a more detailed response on the various issues that have been raised, but I would like to highlight one key point that I learned from Duke.

Many "engine build" magazine articles test the engines with open headers. This is the best possible configuration for a camshaft with a lot of overlap, because the headers and the low back pressure help to exploit the camshaft's overlap to achieve good exhaust scavenging. Many of the power curves published by cam manufacturers also use open headers.

However, the minute you add some back pressure (mufflers) to the system, the high overlap works against you, and even the beneficial effects of the headers are greatly reduced. Even a modest amount of back pressure can make a real dog out of some of those magazine builds.

I recently purchased the Engine Analyzer software and have been learning how to use it (with some assistance from Duke). The EA software lets you configure an engine and then simulate the effects of different cam timing and other parameters, such as exhaust back pressure.

I used EA to "build" a 400 HP small block engine based on one described in a magazine article. Then I added some exhaust back pressure corresponding to a typical high flow dual exhaust (500 CFM). Peak horsepower and torque dropped dramatically (about 50 each).

A related problem with engines that are optimized for peak power is that they typically produce very little power below 3000 RPM. Magazine articles often focus on only peak power, and often don't even provide data below 3000 RPM. That's okay for racing applications where the engine spends all its time in the power band, but it doesn't work well for typical street driving.

There are a lot of other considerations when comparing real world street performance and peak power ratings obtained on an engine dyno, but these differences were the most revealing to me. I like the fact that EA lets me experiment with various configurations and see the effects various changes with all other factors being equal.

Very well said. I will add that the "327 LT-1" tested on a lab dyno was tested with both headers and manifolds. The headers made about 8 percent more peak torque, but only about 2 percent more power, and this was with open exhaust. With the OE exhaust system the results would likely be narrower.

Also, when you test on a chassis or lab dyno, start some pulls at 1000-1500, so you can find out where the 80 and 90 percent torque bandwidths start. But NOOOOOOOOOOOOO, they always start at 3000-3500, when the engine spends most of its running time below that.

A road engine that doesn't produce at least 80 percent peak torque at 2000 is going to feel real soggy in normal vintage car driving. I consider this to be a minimum spec for a high specific power output road engine, and 90 percent peak torque at 2000 will put a real smile on your face.

Duke

[Gene M.]

Re: Engine Output Numbers

Duke,

Do you have any results of installing the camshaft on 300 and 350 horse 327 a couple (2-4 degrees) degrees advanced? The only modifications I would consider is good clean up in the heads and port matching. I am assuming everything else being stock. I would be interested in any increase in power/torque coming off idle from a stop up to say 3500 rpm. Just street performance is my goal, no racing stuff.

[Duke W.]

Re: Engine Output Numbers

I would not recommend installing either of those cams advanced from OE indexing, but might recommend retarding them depending on the owner's objectives and how well the heads flow.

If you want maximum low end torque the 300 HP cam is excellent even if retarded four degrees, and the head work will yield about 10 percent more top end power and at least 500 more useable revs.

I also designed a custom cam that swaps the inlet and exhaust lobes of the 300 HP cam with the same LSA and is installed with fairly late indexing. This design appears to provide the best torque bandwidth from off idle to about 6000 revs. I originally designed it for Powerglide, but it should work just as well with a manual when high low end torque is an objective, but it will still make equal or better than OE L-79 top end power.

The "Special 300 HP" cam I designed (L-79 lobe on the inlet side 300 HP cam exhaust lobe on the exhaust side) ,failed to produce the expected top end power. It only makes about the same power as an OE L-79, but made better than expected low end torque - about 90 percent at 2000.

Both of the above cams are designed for the typical flow characteristics of pocket ported heads and idle just like an OE 300 HP engine.

Unfortunately, with the demise of Crane I don't know how to get them ground. Despite an extensive search I have not found any other grinder that has the required OE lobe masters.

The above two projects along with some isentropic flow analysis during the overlap period provided me with additional insight on valve timing. If you want to maintain the smooth idle of the 300 HP engine while maximizing top end power, the modest overlap period must be phased fairly late or there will be excess exhaust pumping losses at high revs which will eat up the additional power produced by a late closing inlet valve.

Duke

[Joe R.]

Re: Engine Output Numbers

[ATTACH]Ported 327-300 EA plot.pdf

IMG_3817_S.jpg[/ATTACH]

Duke,

Do you have any results of installing the camshaft on 300 and 350 horse 327 a couple (2-4 degrees) degrees advanced? The only modifications I would consider is good clean up in the heads and port matching. I am assuming everything else being stock. I would be interested in any increase in power/torque coming off idle from a stop up to say 3500 rpm. Just street performance is my goal, no racing stuff.

Hi Gene:

The "pocket porting" that Duke advocates will have a much bigger influence on power than port matching. Pocket porting focuses on the "pocket" or "bowl" directly under the valve, and is not as extensive (or expensive) as traditional full-race porting.

I have attached a photo of a pocket ported area on the heads I have on my 327-300.

Also attached is a pdf scan of a plot from Engine Analyzer that compares a stock 327-300 to the exact same engine with pocket ported heads. Output power is SAE net through a typical Corvette dual exhaust.

The flow numbers used for the pocket ported heads were actual measurements on my heads, while the flow numbers for the stock 462 heads were in the EA library and look correct compared to other flow data I have seen for the stock 462 head.

If you look at the plot it is easy to see why Duke likes pocket porting. The 327-300 cam responds very well to the improved flow. You still get most of the low end torque of the 327-300 cam, but useable torque and power are greatly improved at higher rpm.

If you are uneasy about pocket porting your original heads (I got over it, but I understand the concern), just buy a used pair of similar heads and get them pocket ported instead. As long as they have the correct external double-hump casting mark, no one but you will know they aren't the original heads!

[Duke W.]

Re: Engine Output Numbers

I might add that Joe's head flow data is about the best I've seen on these vintage OE heads - equal or better than any I've seen with the larger 2.02/1.60" valves, which is why I don't recommend increasing the valve sizes on a 327/300.

As you can see, the head work boosts the top end considerably without any meaningfull loss on the low end and no effect on idle characteristics. So it's like a 300 HP engine up to about 4000 and like an OE L-79 above, which is the best of both worlds.

Retarding the OE cam four degrees will improve the top end more with only a minor effect on the low end and no effect on idle.

Retarding the cam results in less exhaust gas dilution of the subsequent inlet charge. During this period as the piston approaches TDC on the exhaust stroke and both valves are open, but only slightly, cylinder pressure builds up and exhaust flows through both the exhaust and inlet valves, and by delaying the overlap event more exhaust is pumped through the exhaust valve and less through the inlet valve, so subsequent charge exhaust dilution is reduced, which increases high rev volumetric efficiency.

The overlap period is characterized by the concept of "split overlap" - the point where both valves are open the same amount. On most OE cams this occurs about 4 deg. BTDC, so retarding the cam 4 deg. moves this point to TDC.

Retarding the cam also delays the inlet valve event for a greater inertia effect that keeps filling the cylinder past BDC, which also increases high rev VE, but can reduce low rev VE (and torque) too much if over done.

The broadest road engine torque/power curve is produced by generous head flow and a relatively short cam that has late phased low overlap, and this is typical of modern Corvette engine configurations. Also, since the pocket porting process increases the E/I flow ratio, more inlet than exhaust duration is beneficial, which is what swapping the lobes on the base cam does.

Duke

[Gene M.]

Re: Engine Output Numbers

Duke and Joe,
Thank you for sharing your findings. Since I do not want to loose any bottom end torque, I'll keep the 300 hp cam straight up.

But you did mention the benefits of longer intake vs exhaust profile. This is cause for my next question, Will 1.6 rockers on the intakes help on a stock 300 hp cam? I assume the cylinder head push rod cast slots may need opening up on as needed basis. Maybe 1.7 even better if it doesn't screw up valve train acceleration up and down. I realize this increases the lift more than duration.

All this may require screw in studs and spring change?

What about the same 1.6/1.7 rockers on a 350 hp 327?
.477 lift and .507 lift, maybe a bit much? I physically tried the 1.6 with no coil binding 350 hp 327 stock everything.

The cam card from Speed Pro (Sealed Power) for my 350hp 327 says .447" lift at 291* on both lobes, 30*IO 33*EC 81*IC 78*EO at .006" lift and 1*IO -7*EC 41*IC 49*EO at .050" lift. This sound to me like it is 1.5* retarded, yes?

I don't know what the negative 7 is all about........

Sounds like you guys have done a lot of playing around with different possibilities. As I keep picking your brain, thank you again for sharing.

[Timothy B.]

Re: Engine Output Numbers

Gene,

I think the -7 would indicate 7*BTDC @ .050 lift. Usually the exhaust valve closing is ATDC @ .006 SAE specs.

[Duke W.]

Re: Engine Output Numbers

I DO NOT recommend any kind of aftermarket rocker arms - OE equivalent ONLY!

If you want more "power" retard the cams 4 degrees from OE indexing, and don't mess around with Mickey Mouse "solutions".

The loss of low end torque is minimal and may not be noticeable SOTP.

Duke

[Joe R.]

Re: Engine Output Numbers

Hi Gene:

Duke's reference to having the intake duration longer than the exhaust duration is only for pocked ported heads. Pocket porting increases the exhaust flow by a larger percentage than the intake flow, so the ratio of exhaust-to-intake flow (called the E/I ratio) changes considerably.

This ratio change simply reflects the fact that there is much more room for improvement from pocket porting the exhaust ports compared to the intake ports. So, if you do the best job you can on each port, the degree of improvement is greater on the exhaust side.

Stock 462 heads have an E/I ratio of about 65%, and factory cams are designed to account for this with (typically) an exhaust duration that is longer than the intake duration. Pocket porting changes this ratio to about 80%, so the requirements for an optimal cam become different. With pocket ported heads, an intake duration that is equal to or less than the exhaust duration will yield better overall performance than a typical factory cam that is biased the other way.

I think the factory cams are about right for the factory heads. I think that the potential gains from using 1.6 ratio rockers will be small, but I suppose you could do this if you want.

[John H.]

Re: Engine Output Numbers

Will 1.6 rockers on the intakes help on a stock 300 hp cam? What about the same 1.6/1.7 rockers on a 350 hp 327?

Gene -

For what it's worth, about four years ago, HOT ROD magazine ran a test to determine what changes resulted from using 1.6:1 rockers. They built a baseline engine, and dynoed it (lab dyno, not chassis dyno) with four different camshafts, using two different sets of rockers (1.5 and 1.6) with each camshaft. The data showed the max gain with 1.6 rockers was 4hp at peak power; hardly worth the effort.