Dave McDufford

This was dyno testing week. The purpose of the dyno runs was to satisfy my curiosity of how much power the rebuilt engine actually makes, make certain it would stay together, and to experience something I had never done before. The reason I am posting the results is I assume that some people are probably curious also.

The engine is the original block 327/350hp that I have rebuilt with Speed-Pro 2166 pistons, Crower Sportmans rods and the Speed Pro version of the LT-1 cam. The heads were pocket (street) ported by BES Racing Engines in West Harrison. Indiana and have 2.02/1.60 valves and stock valve springs. The intake is the stock aluminum 461 and it has a correctly dated Holley 2818. For the runs shown below the points were replaced with a Pertronix Ignitor II ignition module. Initially the coil was a Pertronix Flame Thrower II, but it was later replaced with an MSD box. The exhaust was either the 2.5 inch rams horns with the front sections of the stock Corvette 2.5 inch exhaust or a set of rusty old 4-1 headers Tony dug out of the closet.

The dyno testing was also done at BES Racing Engines. (The following is testimonial and not a commercial). I was very impressed Tony Bishoff (the owner) and his operation. He has eight employees who were doing some pretty serious work on some big engines and they were very busy. Due to ignition and carb problems (that were my responsibility) we turned what should have been a half day test into a two day trouble shooting session. While we did not fix everything because the parts were not available, I have a good idea of what I need to do. Tony only charged me for the half day session we agreed to. He runs a class operation.

I would also like to thank Rich who ran the dyno cell and Brian who helped scrounge parts.

To show some pictures, the following is the test cell prior to the first engine start.



The following is the engine running during the cam/engine break-in period (20 minutes).



The following is Tony connecting the 12 volt lead to the coil (in our jerry rigged system) prior to one of the runs.



The results a not straight forward because we could not get the secondaries on the carb to fully open until almost 6500 rpm. With the Holley carb the engine did not show a torque peak as expected around 4400 rpm but was flat at 310 +/- from 3000 rpm to about 6500 rpm. My best guess of the power potential of the properly tuned engine is as follows and is a compilation of several of the runs. The graph also shows the torque for run 20 (the strongest – demon carb/headers) and run 25 (showing the secondary problem). It is interesting that the peak power with the Demon carb/headers was about the same as the Holley/rams horn. The header gain was in the mid-range torque. The runs used for the blue and red lines in the chart are (click on the chart to see it better):

1600/2000 - used #23 - The only run at this low an RPM – unstable at high rpms.
2500/3000 - used #24 - Dyno more stable than #23 - not set to run as low.
3600/5000 - used #18 - Demon carb, but should represent a properly functioning Holley at this rpm.
5500/6400 - used #25 - Holley secondaries should be mostly open at this point.



Below are pdf images of the dyno sheets from the last eight runs. The changes to the configuration are noted in the title. The ignition was initially the Pertronix module and coil. After run 19 we substituted an MSD box for the Pertronix coil. The Demon carb was an 850cfm unit. All runs were made with racing fuel.

Run #18 Demon Carb/Rams Horns

Run #19 Demon Carb/4-1 Headers

Substituted MSD box for Pertronix coil.
Run #20 Demon Carb/4-1 Headers

Changed to rejetted and reworked Holley

Run #21 Holley Carb/4-1 Headers – 38 degree total timing

Run #22 Holley Carb/4-1 Headers – 41 degree total timing

Run #23 Holley Carb/Rams Horns – low range pull (from 1600 rpm)

Run #24 Holley Carb/Rams Horns – low range pull (from 2500 rpm)

Run #25 Holley Carb/Rams Horns – 45 degree total timing

Sorry for such a long post – it is probably more than most people want to see, but I thought some people might be interested.

I would also like to thank Duke Williams for his help in deciding what components in use in the engine and for taking the time for numerous conversation to explain why. I learned a great deal from him and from others on this and other boards.

For those who may be curious the BES Racing Engine site is http://www.besracing.com/

Thanks for looking.

Dave




[Modified by Dave McDufford, 12:02 PM 6/15/2004]

00fxd

Thats great info Dave, Thanks. Looks like the headers worked better than the 2 1/2's. Do you intend to normally run race gas? Are the Numbers where you expected them? Look Pretty good to me!

SWCDuke

To put this data in context, I have lab dyno sheets for a stock 40 over 327/340 rebuild- Duntov cam, stock 461X heads. As with Dave's engine, this data is essentially "SAE gross", which is the basis for the advertised ratings back in the sixties.

This engine essentially achieved the advertised torque rating of 344 lb-ft @4000, but power was way below advertised 340@6000. Actual measured peak power was 294 at 5000-5500, which was the limit of the test, but it would have fallen off at higher revs. The prediction from Engine Analyzer for the stock 327/340 configuration was about 300@5500.

Dave's 80 percent torque bandwidth extends from below 2000 to beyond the rev limit. The massaged 461X heads are what make the top end power and the LT-1 cam gets the most out of the heads without killing the low end, which is very important in a street engine. Also, the fact that power keeps climbing all the way to the 6500 redline is noteworthy!

I'm still analyzing the data, and the "hole" in the torque curve from 1500 to 3500 may well be somewhat due to a misbehaving carb secondary opening, but a relatively steep rise in torque at about 3500 is also somewhat characteristic of all the SHP cams. The carburetor was also way rich on the bottom end.

There is a decent chance that this "hole" can be at least partially filled in with additional carb tuning.

The actual measured CR of the engine is 10.35:1 +/-0.1 variation between cylinders. It should operate without significant detonation on pump premium with perhaps some juggling of the timing map. Race gas was used for the dyno runs to preclude any possibility of detonation and to allow flexibility in setting timing without getting into detonation.

As with any engine, some final tuning will be required once it's in the car to optimize performance across the rev range using pump premium.

Duke




[Modified by SWCDuke, 11:52 PM 6/14/2004]

66427-450

Looks like a nice balanced setup, I'm sure you'll like it; especially if you plan on using highway/economy gearing....... that wide power band would be just the ticket. I, personally, would use more cam (exchange some power below 4000, for more 4000-6500), but that’s just me/what I value in a C'vette engine.

Again, thanks for posting the data, looks like a well thought-out, durable combo......done with very good taste

Dave McDufford

Thanks for your comments.

I have a 3.36 diff and a wide ratio Muncie. The wide ratio gives me the equivalence of about a 3.90 diff with a close ratio for the first three gears (about 100mph) plus a better highway cruise.

I think I am going to like the performance once we get the ignition and carb issues sorted out. It sounded great with the open pipes both at idle and while running for power

Dave

kyosho

Thanks for posting. Looks to be a solid performing engine. Congrats.

SWCDuke

You're joking, right? The torque peak is about 4500 and power is still climbing at 6500 and you want "more cam"! "Highway/economy gearing"?

At 6500 the Mach index is 0.540. The heads are choked! And the flow numbers were excellent - better than Vizard's.

More cam will just kill the low end and peak torque without returning much of anything above 5000, and when you add mufflers, a bigger cam, i.e. more overlap, will kill the torque curve all the way across.

Eighty percent torque bandwidth from 1600 to beyond the 6500 rev limit, with power still climbing at the rev limit!

It doesn't get any better than that on a vintage carbureted engine!

Duke




[Modified by SWCDuke, 10:51 AM 6/15/2004]

tapio

Very nice report. Many thanks for posting it!

comp

66427-450

Why did I know you'd react that way :p: I agree, like I said, it's a good balanced setup.

My point is, when running thru the gears at redline he'll be about here.....
1st to 6500
2nd ~4800-6500
3rd ~5100-6500
4th ~4400 -

And for ME, remember I said for me/my use, I'd want to to position the power curve (and shape it's peak area) such that I applied the most power in the above RPM ranges......


[Modified by 66427-450, 2:34 PM 6/15/2004]

Dave McDufford

My decision was based on the idea that most of the time I will be driving the car in normal traffic and not be running it to redline from stop light to stop light. The low and mid-range torque for 99.9% of my planned driving combined with the ability to wind it up if I want, plus the sound is what I was looking for. I did not want a hot cam that would not be fun to drive on the street (including for the wife). Different strokes...

Dave

66427-450

oh, don't get me wrong, I think you have chosen very wisely for your intended use. plus, it still has that "stock" look, which i think is a real plus.

bcwaller

The numbers look "good", but I think the engine still should have a bit more in it. This is not too far from my old (blown) engine, but you have a ported intake which *should* give you better numbers.

My engine was a 0.030 over 327, 2166 pistons, CS112R Cam (I think it was a 300/300 gross, 224/224 @ 0.050, 108 llobe separation, .436/.436 lift cam), 462 heads (stock, 2.02/1.6 valves), 490 (L79?) intake, Holley 4776 double pumper, MSD ignition. Headers were take off's from a C4 (basically small tube shorty headers) into 2 1/2" side pipes, through a wide ratio Muncie and 3.36 gears.

I'm estimating more than 350 gross HP based on 1/4 mile numbers. My very first day at the track I ran 14.01 @ 97.8MPH, at 2900 ft. Correction factors distributed at the track (LACR Palmdale) brought this down to 13.75 @ 101 MPH. Take the weight of a C2 convertible and I estimate 350 gross HP from what I had.

Dave McDufford

bc,

Run 20 shown above included 4-1 headers, a Demon 850 carb and the MSD box. It is fairly close to what you describe.

It basically made 360hp from 5700rpm to 6500rpm with an extremely flat torque curve for the entire run from 3500 to 6500. I suspect this is likely the best I can get out of the current major components and I consider it to be very impressive, particularly the torque curve.

One point that Tony Bishoff made was that his dyno is "honest". It has not been turned up to show more power than is really made. He said when he bought it used it was setup to show 50 more lb/ft of torque than reality. The guy he bought it from said this was normal and everyone did it to keep the customer happy . Tony said it now calibrated correctly.

I think the 20th run is very consistent with the times with your older engine.

Thanks for the comments.

Dave


btw, do you know Duke from Redondo Beach?


[Modified by Dave McDufford, 8:13 PM 6/15/2004]

bcwaller

I never had a chance to dyno my old engine, so I never had a chance to see what the wheel horsepower numbers were, let alone let someone inflate them for me!

And yes, I've met Duke a few times and we talk on the phone as well. He even came over to help me do the aborted engine install. Maybe I will finally get that sucker in this summer...

SWCDuke

The optimum shift point for this power curve is about 7000- where it starts to seriously rolloff (according to the simulations). This would increase average power over the rev range that is established by gear spacing. Since the optimum shift point is already well above redline, it wouldn't make much sense to shift the torque curve farther up the rev scale, with the inevitable loss of low end performance.

Engines are not one-dimensional. If you set up an engine just for absolute peak power at the rev limit you will end up with an overall dog!

It's not peak power that maximizes acceleration, but average power, so the optimium shift point is where the power falls off to the point where it would be equal in the next higher gear based on gear ratio spacing, and this point is already well above the recommended rev limit.

The wide gap between third and fourth will cause acceleration to fall off, but by that time you're through the lights in the quarter mile, so it doesn' make any difference unless your are running a longer than quarter mile acceleration course.

If Dave ever want's to "see what she'll do", the engine will pull all the way to 6500 in top gear. With 3.36 gears and OE revs/mile tires, that will be an honest 150 MPH, and if he keeps his foot in it, it will rev higher.

Duke




[Modified by SWCDuke, 7:20 PM 6/15/2004]

66427-450

" The optimum shift point for this power curve is about 7000- where it starts to seriously rolloff (according to the simulations). This would increase average power over the rev range that is established by gear spacing. Since the optimum shift point is already well above redline, it wouldn't make much sense to shift the torque curve farther up the rev scale, with the inevitable loss of low end performance.................
It's not peak power that maximizes acceleration, but average power, so the optimium shift point is where the power falls............ "


>>> "Power"?. So, you're saying it's all about (engine setup/gearing) to try to maximize the average "HP" applied ? Not the "torque" ?


-----------------------------------------------------------------------------------------

"Engines are not one-dimensional. If you set up an engine just for absolute peak power at the rev limit you will end up with an overall dog! "


>>> Well yeah...... it's just that some guys are way more concerned with top end power (~4500-7000), than under 4000RPM power. It’s a use dependent thing........... what I mean is, we all define "dog" in a different way, and that's cool.

-----------------------------------------------------------------------------------------

".........If Dave ever want's to "see what she'll do", the engine will pull all the way to 6500 in top gear. With 3.36 gears and OE revs/mile tires, that will be an honest 150 MPH, and if he keeps his foot in it, it will rev higher."

>>> Do you think the 327, in a midyear, would have the torque to pull to 6500-7000 in 4th, with just 3.36's?

Dave McDufford

I made the calibration comment to state that I felt the numbers I had posted were accurate. It was not a shot at anyone and I hope you did not take it that way. You were estimating 350hp from your engine. My configuration which you expected to be higher due to porting was 360hp. I was trying to agree that things seem to fall in line.

Dave



[Modified by Dave McDufford, 4:00 PM 6/16/2004]

SWCDuke

It's power that accelerates a car and power that determines top speed. Power is a dynamic measurement - energy per unit time and is a product of torque and RPM, so you can achieve power with high torque and low revs or low torque and high revs.

Torque is just a static measurement. It is primarily a function of displacement and compression ratio, but is highly affected by the volumetric efficiency and internal engine friction curves. Once the first two parameters are fixed the remaining two issues are where system engineering is applied. The task it to develop flow paths and valve timing that maximize VE/torque throughout the operating regime and do what is reasonable to reduce internal engine friction without compromising durability. You design to achieve VE and torque curve characteristics and then compute the power from the formula.

If designing a racing engine you are only concerned with the VE/torque characteristics in the upper 20 to 30 percent of the allowable rev range to achieve maximum average power in this range. For a street engine the entire VE/torque curve from off idle to the rev limit must be optimized if you want an engine that will have flexiblilty and strong acceleration throughout the entire rev range, good throttle response, and driveability characteristics.

The ideal goal is to have constant torque across the rev range, but this is virtually impossible. Modern techniques such as variable valve timing and variable inlet runner geometry go a long way to achieve the goal of constant torque, but with fixed valve timing and flow path geometry one must achieve the best compromise for the intended service of the engine, and the best measure of engine flexibility is torque bandwidth. For a good performing high performance street engine, establishing a requirement of 80 percent peak torque at no more than 2000 revs is a good criterion. On a racing engine the bottom end of the torque bandwidth is not important, assuming the gearing will keep operating revs between the torque peak and rev limit.

All naturally aspirated IC engines are characterized by torque curves that rise to a peak and then fall off, and the VE curve mimmicks the torque curve. The VE falloff at high revs is primarily due to increasing inlet system velocity, which increases frictional losses, and the less falloff at high revs, the more efficient the flow paths. The best measure of high rev output is the upper bound of the 80 or 90 percent torque bandwidth. This combines both high rev VE and engine friction, which can become dominant at high revs. (It increases with the cube of engine speed.) On Dave's engine the top end of the 80 percent torque bandwidth is beyond the rev limit and the top end of the 90 percent bandwidth is about 6000. This indicates minimal high rev volumetric efficiency/torque falloff and very strong top end power. The problem is how do you achieve this without killing the low end.

Valve timing, primarily the inlet valve closing point, is used to shift the torque curve up or down the rev scale, but shifting it up too far hurts low end torque and low rev performance, and once the inlet flow path is choked as indicated by an inlet Mach index approaching 0.55, delaying inlet valve closing has little effect on peak power, but kills low end torque. At this point the only option is to improve the flow paths or boost the engine.

Overlap is used to take advantage of exhaust system wave dynamics, but with conventional exhaust manifolds the wave dynamics cannot be harnessed, and this places an upper bound on effective overlap. On a racing engine, overlap is juggled along with inlet and exhaust system flow path lengths to maximize average torque in the upper rev range, and high overlap in conjuction with proper inlet tract and exhaust system pipe lengths can considerably enhance high rev VE/torque with VEs as high as th 120 percent range, however, such engines have very narrow power bands in the upper rev range and will often barely run below 3-4000 much less creating any useble torque/power in this range.

With conventional exhaust manifolds, too much overlap will hurt the torque curve, even in the upper rev range. This places a relatively conservative limit on overlap, and the design goal should be to achieve minimum average exhaust backpressure. The exhaust valve opening point has the least effect on the torque curve, but opening it too late can hurt top end power without adding much to the low end. Overall, it is better to open the exhaust valve too early rather than too late.

Pushrod engines have more restrictive flow paths than DOHC 4V engines because of having to snake the inlet ports around the pushrod passages. Traditional hotrodding techniques have been to apply ever more radical valve timing to achieve more top end power, but once the flow path is effecitively choked radical valve timing produces decreasing marginal returns at the top end while causing increasing marginal losses to torque/power in the lower rev band.

The purpose of this project was to extract high power with a flat torque curve without changing the visual appearance or basic operating characteristics of the engine. This requires retaining the OE inlet system and exhaust manifolds, and compression ratio was limited to allow opeation on unleaded premium without having to retard ignition timing exessively, which will hurt the torque curve. The remaining options to achieve this goal were improving the inlet flow paths to achieve strong top end power, primarily head work, and selecting valve timing to create as flat a torque curve as possible with 80 percent peak torque not later than 2000 revs.

The first criterion was certainly achieved as evidenced by the upper bound of the 80 and 90 percent torque bandwidths and the fact that power is still climbing and the 6500 rev limit. The bottom end torque requirement was exceeded with 80 percent at well below 2000.

A lot of vintage Corvette owners want to improve engine performance without altering appearance and without changing the overall "character" of the engine including low end torque, driveability, and idle characteristics. In some cases (30-30 cam engines) owners want more low end torque without loosing top end power. This is how to do it on a typical SHP engine and the results are impressive compared to a stock L-76 or L-79.

If you can build a stock appearing L-79 with more top end power and better torque bandwidth, I'd like to see the dyno sheets.

If all you want to talk about is peak power or highest average power in he upper end of the rev range, start another thread. That wasn't the objective of this engine project.

Duke

66427-450

[QUOTE]......If all you want to talk about is peak power or highest average power in he upper end of the rev range, start another thread. That wasn't the objective of this engine project.[QUOTE]

I agree, and didn't intend for my comment to divert attention from the threads topic. Like I said earlier, I think he met his goal and developed/built himself an superb combo.

In any case, thanks for the detailed explanation .........