LS vs. Coyote Revisited: Using Comp Cams’ DynoSim5 Engine Simulator

Ever since home-based personal computers have become a thing, companies have looked for ways to harness that computing power for more than games like Minesweeper and Solitaire. The company ProRacing Sim, LLC—now a division of the Comp Performance Group—saw the need for an accessible engine simulation software. Fifteen years ago, ProRacing Sim released the first version of its software, the original DynoSim.

At present, we are up to version 5 of the engine modelling software, with version 6 in the works. With a decade and a half of refinement, DynoSim5 has become a valuable tool in the arsenal of both high-end performance enthusiasts and engine builders alike.

By allowing testing of a wide range of components and specifications, without buying a single component or turning a single wrench, the engine simulation software allows you to save money and dyno time by being able to concentrate only on the areas which have shown promise in the simulation.

“I think it is perfect for small engine builders, or more professional guys that need to get in the ballpark quick on engines they do not build often,” says Billy Godbold, Comp Cams’ Valvetrain Engineering Group Manager.

To that end, in this article, we are going to take an in-depth look at the software and its features, then model two real-world engines and compare their performances on the dyno to their simulated performances in the program.

For the final act, we’re going to model a 5.0-liter Coyote engine, and compare it to a hypothetical 5.0-liter LS engine, using the rules established in our previous LS-Coyote Shootout series. Many of the critics of that comparison have pointed out that the LS and Coyote were not on an even playing field, displacement-wise. So, we are curious to see what would happen in an even-displacement comparison, theoretical though it may be.

The DynoSim5 Software

First, let’s get the technical stuff out of the way. The DynoSim5 program is a 32-bit program designed for Windows operating systems. The system requirements might have been considered exceedingly robust many years ago, but these days, almost any machine capable of running a modern version of Windows, will be more than capable of running the software.

Obviously, the faster your processor is, the better the program will perform when running the complex calculations required. For reference, we ran our copy on a two-year-old Windows 10 machine with a 6th Gen Core i7 processor, with 12 GB of RAM, and found zero lag in the program. Also needed is a CD-ROM drive, which is important to mention, as today’s line of netbook computers meet the system requirements otherwise, but don’t have an onboard CD-ROM.

The software utilizes the “Filling and Emptying” mathematical models for both their accuracy and fast processing times. By being “full-cycle” simulations, they calculate the complete fluid-dynamic, thermodynamic, and frictional conditions that exist inside each cylinder throughout the entire 720 degrees of the four-cycle process.

Conversely, other software relies strictly on calculations to determine volumetric efficiency alone, and then estimates horsepower and torque from that. Per the DynoSim5 manual, the DynoSim software performs several million calculations at each 500-rpm test point, with 41 test points calculated in a full power curve simulation.

“There are some assumptions that have to be made about thermal efficiency, windage, friction and dozens of other small factors, such that I prefer to use a dyno for horsepower comparisons,” says Godbold. “But you can learn a ton of useful information from these engine software packages.”

The most critical part of setting up a simulation is, well, the set-up. After all, as the computer science term says, “Garbage In, Garbage Out.” What that means, is that if you haphazardly enter parameters, you’re going to get a poor result. The key to a solid, accurate simulation is careful entry of the available data into the program. DynoSim is broken down into sections of data input to make things easier to understand, and a workflow to be established.

The Configuration Menus

In the Short Block menu, there are a number of predefined short-blocks to make your life easier, where the bore, stroke, rod ratio, and number of cylinders are all preselected. Alternatively, you can choose to enter your own bore and stroke dimensions, and when it comes to the connecting rod, you can either select a rod ratio, or a specific center-to-center rod length, and the ratio will be calculated for you.

Up next is the Cylinder Head menu. Here you have a number of options to let you be as general or as specific as you care to be. First, from a drop-down menu, you select the cylinder head type, chamber configuration, and general level of modification. You can also further model the cylinder head by entering the actual flowbench data of a specific cylinder head design into the program. You can manually select the valve sizes to use in your engine’s simulation, or additionally, there is an “Auto Calculate Valve Size” function in which the program determines the ideal intake and exhaust valve sizes. Interestingly enough, DynoSim5 will allow you to create configurations with up to three valves per port, for a total of six valves per cylinder.

If you are unsure of your cylinder heads’ specific flow numbers, DynoSim has a library of general chamber designs and levels of modification that can approximate the airflow capabilities of your combination.

The next area is the Combustion category. This area covers things like compression ratio—via an intensive compression ratio calculation worksheet—fuel type, Air/Fuel ratio, combustion chamber geometry, and ignition timing. Once again, the program is able to find the optimal timing settings for best torque at each testing point in the simulation.

In the Combustion category you are able to select fuel type – Gasoline, Methanol, Ethanol, Liquid Natural Gas, and Propane. In addition, you can select nitrous injection for Gasoline and Methanol in 25-, 50-, 100-, 200-, 300-, or 400-horsepower amounts (calculated at 25 horsepower per one pound per minute of nitrous flow). There is also a manual horsepower boost adjustment in the Nitrous Augmentation field.

When it comes to the Combustion Chamber geometry selection, there are nine different options to choose from, with a handy picture menu to help you choose the correct chamber shape for your cylinder head. With all the information entered, the program can then estimate the optimum ignition timing for the combination, or conversely, you can enter your own timing curve and see how it varies.

There are nine different combustion chamber styles to choose from, with an easy visual component in order to help you determine which best applies to your combination.

The next category on the list is the Induction category, which includes everything in front of the intake valves. 32 different intake manifold designs are available to be modeled, from standard-flow dual-plane manifolds to individual throttle body setups. The Total Induction Airflow menu has 15 different pre-sized carburetor models, or you can enter your own airflow. If you’re unsure, there is also a handy airflow calculator that will compute flow based on airflow per throttle bore, or known throttle-bore diameter.

Next, and potentially more complex, is the Forced Induction section. There are options for positive-displacement superchargers, centrifugal superchargers, or turbochargers, with a large selection of pre-modeled options of each. You can customize those models, or create your own, with setting your wastegate actuation pressures, pulley ratios, turbine size, and turbine A/R ratio. Additionally, you can add, remove, and adjust intercoolers in the system.

The second-to-last section covers the exhaust, which is everything downstream of the exhaust port. Options include: stock manifolds, high-performance manifolds, small-tube headers, large-tube headers, and large-tube stepped racing headers, with or without mufflers and/or catalytic converters attached.

The final, and probably most important and useful category is the Camshaft category. There are pre-defined generic grinds available, as well as actual Comp Cams part numbers in a library. In addition to the 799 camshaft files contained in the DynoSim5 software, an expansion called CamDisk8 with an additional 6,000 camshaft files is available. There is also the ability to enter the cam card data from any cam you can get the information for, or dream up. Additionally, there is a provision for Variable Valve Timing camshaft profiles.

“If you pick the cam you think you want, the software is excellent at predicting the port pressures and velocities, along with the shape of the power curve,” says Godbold.

The CamManager is probably one of the coolest and most powerful tools in the program. You are able to select camshafts from a library, enter specs from a cam card, or enter your own specs, and see how each spec change alters the profile.

One of the most useful features for figuring out how to refine a given combination is the introduction of Iterative Testing. Available in both the Quick Iterator and ProIterator versions, the former allows you to quickly click a “Best HP” or “Best TQ” button and it will make an analysis of the most efficient camshaft timing, or bore and stroke combination for your modeled engine.

The ProIterator tool analyzes bore/stroke combinations and camshaft timing, and also allows testing of different intake manifold types. Additionally, the ProIterator gives you far more control over the testing parameters than the Quick Iterator. It’s aimed at the more serious user and adds a level of complexity that not everyone needs or wants. Once either Iterator runs the tests, you can generate a new engine based on the changes it made. This will open a new engine rather than messing with your original file.

See the Pro Iterator in action:

The camshaft timing optimizer is where the processor power of your PC is really tested. It will generate camshaft specifications based on a given valve lift, which is optimized for one of four areas: Peak Torque, Area under Torque Curve, Peak Horsepower, and Area under Horsepower Curve.

For those who want to dive even deeper down the rabbit hole, there is also Dynomation 5, which is advanced computational flow dynamics software.

“Dynomation is much better for digging down into the pressure waves going on inside the system, and a big advantage for designing custom headers and intake manifolds,” Godbold says. “With Dynomation, I can set up an engine and then look closely at the wave tuning.  It could cost tens or hundreds of thousands of dollars to add pressure sensors to an intake and exhaust port to measure the pressures that Dynomation gives. The mass flow and velocity measurements are almost impossible to make without some sort of sophisticated engine modeling or CFD program, even after you have the engine wired up like Darth Vader’s bathroom.”

Simulation vs Reality – Coyote

In order to verify the accuracy of DynoSim5, as well as our ability to properly enter data, we decided to take two extremely well-documented engine builds and enter them into the program as baselines. You might remember the LS3 vs. Coyote Budget Engine Shootout we did about three years ago – those are the engines we’ll be using as baselines.

In order to validate both the software, and our use of the software, we’re going to be recreating these two engines in the software. With solid build sheets and dynos on file, they are the ideal candidates.

Using the build article for the Coyote as the guide, we “built” the engine in DynoSim5. Starting with the short-block, we entered the 3.649-inch factory stroke and a bore size of 3.640-inches, mirroring the Manley pistons used in the build. We then chose to directly enter the rod length of 5.933 inches, since we have the part number right there in the article. That brings our Coyote short-block to 303.8 cubic inches—or 4.978 liters.

For the cylinder heads, we chose the preset configuration type of “4-Valve, Pentroof, Stock Ports and Valves,” which automatically generated two intake valves and two exhaust valves per port. We converted the valve sizes from millimeters to inches, and came up with 1.456-inch intake valve diameters, and 1.220-inch exhaust valve diameters.

Moving to the “Combustion” section of the program, this is where having a well-documented build to act as a baseline comes in handy. The Compression Ratio Calculator needs six data points to accurately calculate total combustion volume, swept cylinder volume, and ultimately compression ratio.

Here you can see the data which goes into the compression ratio calculation for the validation Coyote engine. On the left, you can also see the intake manifold type (Boss 302) and the Total Induction Airflow rate which is calculated via the use of the 80mm stock throttle body.

After entering all of our variables, we came up with a compression ratio of 11.32:1. We set the fuel to gasoline, and the combustion chamber design to the “Pentroof” preset, which assumed a chamber timing requirement of 16-degrees.

Moving to induction, since the real-life Coyote engine used a Boss 302 intake for the test, we selected “Tunnel Ram, High-Flow” as the manifold type, and then calculated the Total Induction Airflow through the Flow Calc button. By using the stock throttle body size (since that’s what was used in the actual test) the program calculates the maximum airflow in CFM at 1.5 inches of Mercury, manifold pressure.

The exhaust is a simple selection of “Small-Tube Headers, Open Exhaust” which are the best way to describe the dyno headers used in the test.

Finally, we reached the camshaft section of the software. The easy part here is that we know the exact specs of the Comp Cams NSR Stage III camshafts used, thanks to the online cam card. It’s simply a matter of adding the specs into the CamManager and letting it generate a power simulation.

After the simulation, DynoSim5 shows an estimated peak power of 518 horsepower at 7,500 rpm (remember the program only calculates in 500-rpm increments) and 412 lb-ft of torque at 5,500 rpm. The actual numbers from the dyno cell were 510 horsepower at 7,600 rpm, and 398 lb-ft at 5,400 rpm. That is +3.5-percent on the torque and +1.5-percent on horsepower, with almost identical curve shapes and power peaks within 100 rpm of actual.

“’How would a longer or shorter runner on the intake or exhaust work with this cylinder head and camshaft?’ is much more reliable data to predict from a good simulation,” Godbold says. “A good simulation package lets you try 100-plus things in the virtual world to be able to predict a few things to test on the dyno, while giving you a very good indication of the relative power a longer of shorter header might give you over a standard combination. These programs are extremely valuable at helping you understand trends and make good choices throughout a development program.”

Simulation vs Reality – LS3

Following the same process for the LS3, using the LS3 build article as a guide, we started off with the short-block category. The LS3 actually maintains stock displacement using an LSA forged crank with a 3.622-inch stroke, and Mahle pistons in the stock 4.065-inch bore, for 376 cubic inches or 6.136 liters (commonly rounded to 6.2-liter). The Lunati rods measure 6.125 inches, center to center.

The cylinder heads used in the build are factory-ported Chevrolet Performance LS3 units. Since we have actual flowbench data for the heads, we were able to plug that data into the program directly, along with the 2.165-inch intake and 1.590-inch exhaust valve sizes.

For the combustion data, we used the specs of the modified pistons, the 68.5cc chambers, along with the bore and compressed thickness of the 7-layer MLS head gasket, giving us a total combustion space of 75.53cc and a compression ratio of 11.2:1. We identified the LS3 chambers as “Wedge, Close, Fast Burn,” which the program calculated a timing requirement of 24.0 degrees.

For Induction, since the real engine uses a stock LS3 composite intake, and the program’s options are for a stock LS1 intake manifold, or aftermarket LS1 manifold, we chose the modified version to account for the LS3’s improvements over the original LS1 composite intake. Converting the Nick Williams 102mm drive-by-wire throttle body to an SAE bore size, generates a Total Induction Airflow of 1,084.8 cfm at 1.5 inHg.

For the LS-series engines, the software has stock and high-performance intake manifolds pre-modeled, which, combined with the throttle-body size selection, makes that part of modeling an LS-based engine easy.

Like the Coyote, we chose “small-tube headers, open exhaust” for the exhaust model for the LS3, as that is the most appropriate.

For the camshaft, we simply pulled up Comp Cams’ website, and used the specs for the 54-469-11 LSR, Rectangular Port camshaft, which has been designed for all-out power in the 2,000-7,000 rpm range.

Once the Simulation “Run!” button was hit, the numbers came up at 576 horsepower at 6,500 rpm, and 492 lb-ft of torque at 5,500 rpm. Compared to the actual dyno pull of the engine, the horsepower and torque peaks are dead-on the same RPM, while the horsepower is up 3.5-percent (576 vs. 556 actual) and torque is down 1.0-percent (492 vs 497 actual) over the real-world numbers, but with an identically shaped pair of curves.

“I’ve always been much more interested in predicting the torque curve shape with simulation rather than perfect absolute power numbers,” Godbold says.” There are so many combustion and frictional efficiencies needed for absolute power, where being able to optimize tuning is more important for component selection during a build.”

The two modeled dyno curves from the two validation engines (LS3 is the lighter red and lighter green lines), both of which match the shape of the real life dyno curves. Both hypothetical engines came in at 3.5-percent high on the power numbers, and within 1.5-percent on the torque numbers.

So we’re willing to call our validation testing a success, with peak-RPM being dead-on on both combinations, power and torque curve shapes being dead-on for both combinations. With horsepower numbers being within 3.5-percent of actual numbers on both combinations, and torque numbers being within 1.5-percent of both, we think we can pull off a fairly accurate 5.0 vs. 5.0 showdown.

5.0 vs. 5.0 Showdown – Coyote

For this fictional showdown, we are going to more or less use the rules from the original LS3-Coyote Shootout, but with a few additional restrictions. For displacement, we want both engines to be as close to 5.0-liters as reasonably possible. Both engines will run unmodified OEM cylinder heads, and even though this is all hypothetical, we want to keep these engines in the realm of real world capability.

For the most part, the Coyote engine from the actual shootout is a solid build, but we need to make a few theoretical improvements. First, to be addressed is the bore. By adding .020 inches over OEM, for a 3.649-inch bore size, we not only make for a perfectly square bore and stroke, but come out to 305.3 cubic inches, or 5.003 liters.

The next change is to hypothetically swap to Gen 2 Coyote cylinder heads. Their slightly larger valves and improved ports help our hypothetical engine breathe better, which is an advantage that can’t be overlooked in a non-exotic, OEM-only application.

Other than those two changes, the file for this version of the Coyote engine is identical to the actual Coyote engine used in the Shootout, including the stock 80mm throttle body.

Our hypothetical Coyote engine is very similar to the one we actually built for the Coyote-LS3 Shootout, however we added .020-inch to the stock bore to bring it to 5.003 liters, and “swapped” to Gen-2 Coyote heads.

5.0 vs. 5.0 Showdown – LS

For the LS side of things, there are quite a few ways to go about getting 5.0 liters of displacement. Unfortunately, most of them involve custom destroked cranks and would wind up with something with a monster bore and super short stroke. So we decided to start with an iron-blocked 4.8-liter LY2 truck engine. Coming from the factory with a 3.780-inch bore and 3.267-inch stroke, a simple .078-inch overbore would bring the engine’s displacement to 305.5 cubic-inches or 5.006 liters of displacement.

While that might sound like a fairly large overbore, there are quite a few instances of the 4.8-liter blocks being bored to the factory LS1 bore size of 99mm or 3.898 inches and living. Having a custom-diameter LS piston made isn’t a huge undertaking these days, and we’d need a custom piston anyway to bring the LY2’s OEM compression up from its meager 9.1:1 factory compression ratio for this comparison.

For the cylinder heads, there’s really not much imagining that needs to be done, as the LY2 comes with the cathedral-port 799 head, which offers decent flow from the factory. With their 209cc intake ports and 74cc exhaust ports, and a chamber diameter designed to work with the 3.78-inch bore engines, the 3.858-inch bore of our hypothetical engine will work well with the 799 heads.

Since this hypothetical test is meant to fall on the more factory side of things with unported OEM castings for both engines, and the 799 castings are generally accepted as the best factory option for the under-4.00-inch bore size, we don’t feel like we are putting the LS engine at any disadvantage. On the flip side, as they are the original head found on the LY2, which is a Gen IV LS design, it’s only fair that we use them when comparing to second-generation Coyote head–so no, we aren’t giving the LS an unfair advantage, either.

 

We were able to locate flowbench data for the 799 cylinder head, so coupled with that data, and the 2.00-inch intake and 1.575-inch exhaust valve sizes, the program has quite an accurate representation of the LY2 head.

For the Combustion section, we went with a piston with a 12cc dome, which, when coupled with the 799 heads’ 64cc combustion chamber and a .041 compressed-thickness and 4.100-inch bore head gasket, give the combination an 11.28:1 compression ratio. Like our previous LS validation engine, we selected “Wedge, Close, Fast Burn” as the combustion chamber type, gasoline at 12.5:1 AFR as the fuel, and 24.0 degrees of chamber-required timing.

For the induction section, it was a relatively easy choice. Since, in real life, we would run a F.A.S.T. LSX 92mm intake manifold and matching throttle body, and that option exists in the dropdown menus as “LS1/LS6 HP Runners/Mods”, it’s a no-brainer. We then spec’d a 92mm throttle for 882.4 cfm at 1.5 inHg, Total Induction Airflow. For the exhaust, like all the rest, simple dyno headers were selected with the “Small-Tube Headers, Open Exhaust” selection in the program.

The real fun came in the camshaft selection area for this particular engine. Previously, all of the other engines we’ve modeled have had camshafts picked for them. For this engine, we are going to take a two-step process. First, we used Comp Cams’ CamQuest selection software to narrow the choices down. By selecting “High-Performance Street, Choppy To Mild-Rough Idle” we’ll keep the camshaft profile in the spirit of the original rules.

After looking through the recommended camshaft grinds, we decided on the Comp 54-444-11, which comes with a “Great Fit” recommendation from the software, and is advertised as a “High-RPM street/strip camshaft, for use with FAST LSX intake.” Since its advertised powerband is 2,000-7,000 rpm and we’re trying to keep up with the high-winding Coyote in this comparison, it seems like a great fit.

Once we inserted the cam specs into the program, we then ran the ProIterator for both peak horsepower and horsepower under the curve, to see how close or far the chosen camshaft’s specs are from ideal. On the peak horsepower simulations, the Iterator only gained 1.6-percent—or 9.1 horsepower—over the shelf cam.

Here are the ProIterator results for camshaft optimization using the Comp 54-444-11 camshaft as a base. On the left is the optimization for peak horsepower and area under the curve on the right. As you can see, the off the shelf camshaft is almost dead-on in the peak horsepower section, and is more of a tradeoff in powerband as opposed to an increase in the area under the curve.

In the horsepower under the curve ProIterator simulations, it was a bit more significant, with  power under the curve (between 3,000 and 7,500 rpm) increasing by 3.5-percent. However, when you overlay the two curves, you notice all the increases are below 6,000 rpm, with significant loses between 6,500 and 7,500 – basically a tradeoff in where power is made.

Based on the rules and spirit of this competition, we decided to “run” the shelf grind camshaft for the simulation. Were this a real life engine, we would be placing a call to Comp’s tech line and asking for verification of our selection.

“I love it when a customer will play with something like DynoSim a while and then talk with one of our guys about what they would recommend,” Godbold says.

“Chances are that the exhaust opening will be the biggest difference, unless everything is very close. Generally the software will like the cam that our guy picks almost as much as the ‘iterated’ cam the program picks,” Godbold says. “Either cam would be pretty good, but I hope the one with some more experience thrown in is a bit better, as exhaust opening is extremely difficult to model.” Once we clicked “run” on our hypothetical LY2 engine simulation, everything was in the hands of the zeroes and ones inside the computer and program.

5.0 vs. 5.0 Showdown – The Results

After all this reading, you’re probably dying to know the results of the two modeled 5.0-liter engines. Were all the Blue Oval fans right, saying “cubic inch-for-cubic inch, the Coyote makes more power?” With peak numbers of 536 horsepower at 7,500 rpm, and 416 pound-feet of torque in a plateau spanning 5,500-6,000 rpm, the Coyote takes the horsepower victory. However, with 500 horsepower at 6,500 rpm and 419 pound-feet at 5,500 rpm, the LY2 edges out the Coyote in torque, especially when you factor in the discrepancies we found between the real world results and our modeled validation engines.

The only real way to call this is going to be an average power/torque number. One of the genuine surprises we came across in this test is the similarity of the curve shapes for these two engines. That shape similarity will make the average power numbers far more similar than one would expect from an LS and a Coyote, on the surface. So as to not give the Coyote an unfair advantage, as it makes power to 8,500-plus rpm, we’re going to average power and torque between 3,000 and 7,500 rpm.

The final 5.0-versus-5.0 comparison graph. The lighter colored lines are the LS, and the darker colored ones are the Coyote engine. I’m not sure anyone involved in this article expected the curves to be this similar, and the overall numbers to be so close to one another.

The average numbers created by the Coyote are 381.6 horsepower and 373.4 pound-feet of torque. The LY2’s average horsepower came in at 379.7 horsepower, and its average torque is 376.8 pound-feet. So, once again it’s a split decision. However, if we were to extend that average RPM range another 1,000 rpm, the Coyote would take a decisive lead, while if we dropped the RPM range 1,000 rpm, there would not be a huge change in the LY2’s favor.

So while absolute numbers aren’t really what DynoSim5 is aimed at, we are confident in the shape of the curves being very close to reality and are genuinely surprised at how closely the two engines’ curves resemble one another. For those who would challenge this test based on the use of stock cylinder heads, stay tuned, as we have a future installment planned in which we’ll “run” ported factory heads, or maybe even aftermarket heads, on the LY2 short-block, and compare it to a Coyote with ported cylinder heads. We’ll also be diving into the forced-induction component of DynoSim5 as we see if boost changes the outcome on the 5.0 vs. 5.0 showdown. After all, this software is just too much fun to play with.

Article Sources

About the author

Greg Acosta

Greg has spent nineteen years and counting in automotive publishing, with most of his work having a very technical focus. Always interested in how things work, he enjoys sharing his passion for automotive technology with the reader.
Read My Articles