Duke, thanks for weighing in on the DCR issue.

I have been playing with EA3.2 to verify some of the car rag dyno tests. The first one I chose was an article in Chevy High Performance called "The Ultimate Stick" (July 2006). In this article they took a 383 mule and swapped in six different cams to see the effects. I had modeled this on DD2000, and although DD2000 hit the torque and horsepower peak numbers very close, the low RPM torque was way off. When I plugged the motor into EA3.2, it did a very nice job of predicting the entire torque and HP curves. The values were within 5% and might even be tighter had I known their dyno air temps and pressures. EA3.2 even predicted a small inflection in the curves at 3000RPM that the dyno test showed.

The next motor I tried was in the latest Popular Hot Rodding (November 2006) in an article called "Budget Sledgehammer". Here EA3.2 missed badly, not just in the values but in predicting the peak torque and HP. I then plugged the motor into DD2000 and it agreed with EA3.2, at least for the torque and HP peaks. PHR is reporting 447 peak HP at 6200 RPM and 445 ft-lbs at 4500 RPM for peak torque. EA3.2 and DD2000 have the HP peak around 5500. This is a 350 with some pretty crappy heads and a smallish cam. (Heads are 60/54 @.1, 117,102 @.2, 173/134 @ .3, 216/152 @.4, 225/162 @.5, 232/169 @.6. Valves are 1.94" intake and 1.5" exhaust, port volume is 180cc. Cam is 224 duration, 108 LSA, 4 degrees advanced. Intake is a Pro Products dual plane. Carb is 750 CFM. Unspecified headers, but look to be 1 3/4". Compression is 10:1.)

I'm having a hard time understanding how these small heads push the HP peak to 6200 in a 350, but that's what their dyno says. I'm still playing with EA3.2 to see if there is something that can be tweaked so the results more closely resemble the PHR's dyno numbers.

 

My experience is that magazine dyno tests tend to show more power than lab dynos I'm familiar with, but lab dyno power is meaningless.

What count's is SAE net installed power including the vehicle exhaust system and front end accessories, and these can be correlated to chassis dyno HP using 0.85 driveline efficiency. DD2000 can only predict SAE gross on a lab dyno.

These data sets correlate well, except for low end torque, which usually tests higher than EA's prediction. In most cases the torque curve is flatter than predicted.

Duke

 

Duke:

I agree that the lab dyno could be way off depending on air temps, air pressures, and accessories but I would expect the location of the torque and horsepower peaks would be unaffected by the dyno conditions.

EA3.2 allows you to tweak the dyno conditions, but I can't get EA3.2 to match the "Budget Sledgehammer" dyno results for HP and torque peaks.

 

SAE gross laba dyno test data is corrected to standard sea level conditions - 29.92"..., SAE net is corrected to 990 mb, 25C dry air. These are the two default conditions you can choose, or you can input actual conditions as I do for Bonneville engines. The difference in standard sea level air density correction and SAE air density correction is about 4.5 percent

And like I said, dynos used for magazines tend to get higher power that EA's SAE gross prediction and also greater than I've seen on actual lab dyno tests that are corrected to standard sea level conditions. Go figure!

Don't waste your time trying to duplicate magazine dyno test reports. Other than less predicted low end torque, EA yields fairly accurate SAE net results as long as your models are accurate, and, of course, that's a critical issue, which takes some experience to get right. In particular, accurate exhaust system flow models are critical to achieving accurate predictions.

Duke

 

Thanks guys, this has been an interesting read. I am planning to buy one of these simulation software to plan my L98/stroker buildup.

I would like to test different combos and try out some ideas. I have no access to other data than that I can find in the net (part manuf. websites or airflow data comparisons etc.) so it would be nice if the sw has good database of go fast parts available on the market.

I looked at Desktop Dyno, Dynosim and EA but cannot figure out what features would be important. Can you please recommend what sw and version would be the right one for my use? Best place to buy it on-line?

Thank you!

 

Isn't the answer obvious?

I have both DD2000 and Dynosim and have not used them for years. They are the same program. Only EA allows you to model exhaust system flow and front end accessories and use different correction conditions, so only EA allows you to simulate SAE net power in an actual road vehicle, which is why it's all I now use.

Just google on "Engine Analyzer" to find the lowest price online source.

Duke

 

I had DD as well. I went with EA+ only because I wanted to be able to input better flow data for heads.

I like it, it is easy to master, has a good manual, buy or print it,. I think it is easier to model late model LSx motors from GM.

 

I kept playing with EA's models to see if I couldn't get it to match up better with the "Budget Sledgehammer" from Popular Hot Rodding. EA was predicting the HP peak was lower than the dyno results.

It looks like EA has factored in a roll off at high RPMs for hydraulic lifters. The budget sledgehammer was using beehives for a lightened valve train and potentially better high RPM performance. In EA I swapped the hydraulic lifters for solids and set the lash at EA's minimum of .004" to mimic the hydraulic lifters profile. This pushed the RPM peak up closer to the dyno numbers.

 

Could you post a response for me, to clarify some issues. Actually, the responses from the members to each other are pretty accurate.

Int Vacuum in the tabular results is for the manifold vacuum at WOT during the run, at each RPM. The responder who explains that a 600 CFM carb should have 1.5" Int Vacuum when the engine is flowing 600 CFM of air is exactly correct. We have some other corrections for manifold type, dual vs single plane, individual runner, etc. If the engine is Super/Turbocharged, this number turns into Boost in PSI.

Idle Vacuum (in the Special Calculations section below the tabular results) is like it says, how much vacuum at idle. The higher this number, the smoother, more "streetable" the engine.

Yes, we are overly conservative on Piston Speed, and in other areas of the programs. We have no idea of the ability or background of our users. In the US, they have these things called "lawyers". We like to stay clear of them, and the kid who could hire them after he rev's Dad's station wagon to 8000 RPM because a computer program said he could.

Yes, the Pro can give different answers than the 2 smaller programs. These programs are quite complex inside the code, and the Pro is MUCH more complex than the others. That's why it takes minutes to come up with results compared to hundredths of a second as in the smaller ones. If we have a question here at Performance Trends, we run the engine through the Pro.

The Std Engine Analyzer does not do Piston-to-Valve clearance. The Plus and Pro versions do.

Ten years ago or so, we used to have a bulletin board. It had some good discussions, then it turned really nasty, lots of profanity, and took a lot of upkeep on our part. We started another bulletin board about 6 years ago, with a different board, but it got very little traffic. Therefore, we dropped it about 5 years ago. Now we have the User Community page, which is not that great, but is a place to volunteer yourself. You give your name and email address and act as a source. Others can then contact you to get any files you want to volunteer. Here's the link:

http://www.performancetrends.com/User%20Community.htm

Thanks.

Kevin Gertgen
Performance Trends
feedback@performancetrends.com
248-473-9230 (ext 2 for sales) fax 248-442-7750

 

Kevin:

Thanks for the reply. Obviously I didn't wear you out with my questions.

Are the PTV models quite accurate? The reason I ask I was using EA+ and with certain cams beings used with the LSx motors, some of the clearances being modeled were .050 and even worse. I know my numbers are good.

 

Kevin:

Thanks for taking the time to write in. Your technical help and support is the reason I switched to EA3.2.

 

In answer to Chalky:
Piston to Valve (PTV) is mathematically much simpler than estimating performance, and should therefore be more accurate. However, errors are always possible because of incorrect user inputs or approximations within the program. Because these errors can destroy an engine, we always display the disclaimer "This calculation is NOT a substitute for checking clearances with clay."

Some of the inputs for PTV are hard to come by and not published by the manufacturers. A good example is Deck/Valve Clearance, the distance the valve is set back from the deck surface of the head.

Second, the program assumes a flat top piston. SO, if you come up with a -.050 valve clearance (valve goes into piston .050"), but there is a .150 valve relief or dish, you still have .100 clearance.

Third, the exact cam profiles the program assumes from its simple inputs may or may NOT be the same as a particular cam. We are only matching the profile at the .050" points and the max lift point (centerline). Variations in ramp ratings, rocker arm ratio changes (from various geometries), etc will introduce additional (typically small) errors in valve lift.

As we always state, use the programs to find trends. Like if I switch from this cam to this cam, my PTV clearance is likely to get much tighter.

Thanks.

Kevin Gertgen
Performance Trends
feedback@performancetrends.com
248-473-9230 (ext 2 for sales) fax 248-442-7750

 

The the program have inputs for valve seat dimensions besides angle? (as in the length of the sealing surfaces on the valve and the seat)
Exhaust system length?
Header design and length?
Intake manifold runner length/ volume?
Valve spring tension and valve train weight?

I am preparing to write my own simulation based largely on the teachings of Gordon P. Blair and Charles Taylor. However I am looking for a more readily available program to recommend for friends interested in learning about engines and what makes them tick.

MAtt