How accurate have you found DD2000 to be?
12-06-2002, 12:07 PM
#21
Drifting
Member Since: Sep 1999
Location: Wylie TX
Posts: 1,411
Likes: 0
Received 0 Likes
on
0 Posts
Re: How accurate have you found DD2000 to be? (81vette)
When I ran my DD2000 vs Chevy High Performance dyno runs, I used advertised and found it to be real accurate with the actual dyno pretty much across the entire RPM range.
12-06-2002, 02:44 PM
#22
Race Director
Member Since: May 2000
Location: Redondo Beach USA
Posts: 12,487
Received 1,973 Likes
on
1,188 Posts
Re: How accurate have you found DD2000 to be? (Vetterodder)
Duke, in my simulation the difference is well under 2% (407hp @ 6,000 w/331ci, 400hp @ 5,500 w/383ci) but I still don't accept that peak power would be less if measured on an actual dyno.
Another quirk I found was with carb size. That 407hp with a 700 cfm carb jumps to 418hp at the same rpm with an 1100 cfm carb and no other changes. It also shows a power increase throughout the entire range, even at 2,000 rpm.
Another quirk I found was with carb size. That 407hp with a 700 cfm carb jumps to 418hp at the same rpm with an 1100 cfm carb and no other changes. It also shows a power increase throughout the entire range, even at 2,000 rpm.
You need to read the manual and understand DD2000's limitations. It assumes idea fuel-air ratio and mixture distribution, which is not the case with an oversized carb or 2x4 cross ram manifold. A bigger carb means less inlet restiction, so it will show more power. Also, keep in mind that DD2000 only simuates WOT operation with the above assumptions, so low rev power and part throttle behavior is a function of using common sense for carb sizing and your ability to properly tune the fuel and spark maps.
I've found that DD2000's weakest link is valve timing characterization, and it seems to provide most accurate results with the valve events defined as .006" actual VAVLE lift, which is usually what aftermarket cam grinders use to specify their "advertised duration", BUT they use a rocker ratio of 1.5:1 for SBs, but this is not the case. I measured mine and found they are 1.37:1 at low lift and 1.44:1 at max lift. I took an accurate lift crank angle diagram for the LT-1 cam and using my preferred clearance and actual rocker ratios, I was able to pick the valve events to within one degree accuracy, so I have good confidence in my results.
BTW my production '63 L-76 yielded only about 300 HP. The pocket porting and LT-1 cam got me to close to 350HP and more peak torque without any loss of low end torque relative to the OEM Duntov cam.
Duke
12-06-2002, 08:11 PM
#23
Burning Brakes
Re: How accurate have you found DD2000 to be? (Chris A)
I agree that DD2000 is most accurate when entering the advertised duration for a cam. BTW, I did re-write this from an earlier post after confirming some things in DD2000. Here's what I found out:
The Crane cam that I was looking at spec'd at 240 @.050, and 306 seat-seat or advertised duration using Crane's cam cards. The lift .558/.558. LSA=110 and intake centerline 110.
DD2000 makes the following adjustments to the @.050 duration figures when it converts from @.050 to a seat-seat duration. Which it must do before it runs the simulation according to the Help file. You don't see these figures anywhere on the DD2000 screen, but I tested this by looking for the same power curves using various seat-seat and @.050 figures. DD2000 adds 34 degrees duration to the @.050 duration to arrive at the seat-seat duration for Roller cams, adds 40 degrees to Solid cam types, and 56 degrees to hydraulic cam types. These conversion values are not very accurate for the cams that I looked at on Crane's website, and it probably causes the program to grossly underestimate the cams overlap when entering @.050 durations.
For example, when I entered Cranes 119651 HR cam with 240 @ .050 duration figures into DD2000 using a cam type of "Roller" the program calculated an advertised duration of 274 and overlap of 25 degrees, and so the cam "looked" like it had a fairly quick ramp rate and low overlap. DD2000 calculated 425 lb/ft of TQ at 2000 RPM, 549 HP at 6000 RPM. In reality this cams advertised duration was 306 which put the program off by some 32 degrees in it's internal calculations and modeling.
When I entered into DD2000 the cams true advertised duration of 306 from Cranes cam card the program was able to see that the cam actually had a lot more overlap (90 degrees) and it more accurately calculated 274 lb/ft of TQ at 2000 RPM, and 542 HP at 6500 RPM. Entering the cams specs using advertised duration (seat-seat) keeps DD2000 from having to make the @.050 to advertised duration conversion and thus reducing a major inaccuracy of the program.
To sum this up: When entering @.050 cam specs into DD2000 your relying on the programs limited ability to calculate your cams actual seat-seat specs which it must have in order to model your engine correctly. The only time the DD2000 power curve might be close with a @.050 duration entry would be if you were modeling a cam with exactly the same @.050 to advertised ratio as the program uses. However there are other problems to consider here as well. If you had to choose a hydraulic cam type in DD2000 in order to get the program to correctly calculate the advertised duration for a roller cam, you'd likely end up with a power band that did not reflect the faster ramp rate of the roller cam once the lift was above .050".
mark
[Modified by 81vette, 7:14 PM 12/6/2002]
[Modified by 81vette, 1:06 PM 12/7/2002]
[Modified by 81vette, 5:38 PM 12/7/2002]
The Crane cam that I was looking at spec'd at 240 @.050, and 306 seat-seat or advertised duration using Crane's cam cards. The lift .558/.558. LSA=110 and intake centerline 110.
DD2000 makes the following adjustments to the @.050 duration figures when it converts from @.050 to a seat-seat duration. Which it must do before it runs the simulation according to the Help file. You don't see these figures anywhere on the DD2000 screen, but I tested this by looking for the same power curves using various seat-seat and @.050 figures. DD2000 adds 34 degrees duration to the @.050 duration to arrive at the seat-seat duration for Roller cams, adds 40 degrees to Solid cam types, and 56 degrees to hydraulic cam types. These conversion values are not very accurate for the cams that I looked at on Crane's website, and it probably causes the program to grossly underestimate the cams overlap when entering @.050 durations.
For example, when I entered Cranes 119651 HR cam with 240 @ .050 duration figures into DD2000 using a cam type of "Roller" the program calculated an advertised duration of 274 and overlap of 25 degrees, and so the cam "looked" like it had a fairly quick ramp rate and low overlap. DD2000 calculated 425 lb/ft of TQ at 2000 RPM, 549 HP at 6000 RPM. In reality this cams advertised duration was 306 which put the program off by some 32 degrees in it's internal calculations and modeling.
When I entered into DD2000 the cams true advertised duration of 306 from Cranes cam card the program was able to see that the cam actually had a lot more overlap (90 degrees) and it more accurately calculated 274 lb/ft of TQ at 2000 RPM, and 542 HP at 6500 RPM. Entering the cams specs using advertised duration (seat-seat) keeps DD2000 from having to make the @.050 to advertised duration conversion and thus reducing a major inaccuracy of the program.
To sum this up: When entering @.050 cam specs into DD2000 your relying on the programs limited ability to calculate your cams actual seat-seat specs which it must have in order to model your engine correctly. The only time the DD2000 power curve might be close with a @.050 duration entry would be if you were modeling a cam with exactly the same @.050 to advertised ratio as the program uses. However there are other problems to consider here as well. If you had to choose a hydraulic cam type in DD2000 in order to get the program to correctly calculate the advertised duration for a roller cam, you'd likely end up with a power band that did not reflect the faster ramp rate of the roller cam once the lift was above .050".
mark
[Modified by 81vette, 7:14 PM 12/6/2002]
[Modified by 81vette, 1:06 PM 12/7/2002]
[Modified by 81vette, 5:38 PM 12/7/2002]
12-07-2002, 02:12 PM
#24
Burning Brakes
Re: How accurate have you found DD2000 to be? (gkull)
gkull,
I like the first cam best. It still produces great TQ down low, and a lot HP up top. Do you know what the actual advertised specs are for these cams. After what I wrote from above I'm reluctant to go with the @.050 specs. If you don't have the advertised figures maybe you could design these same cams with the same power curves but by setting DD2000 to seat-seat instead of @.050.
mark
I like the first cam best. It still produces great TQ down low, and a lot HP up top. Do you know what the actual advertised specs are for these cams. After what I wrote from above I'm reluctant to go with the @.050 specs. If you don't have the advertised figures maybe you could design these same cams with the same power curves but by setting DD2000 to seat-seat instead of @.050.
mark
12-07-2002, 07:50 PM
#25
Team Owner
Re: How accurate have you found DD2000 to be? (81vette)
Monty and i designed our own cam profiles. I have never ran a dial indicator to find out the Seat to seat on mine. I just ask Crane for very steep ramps. they had like 4 levels and mine is less than all out race. I also ask for the cam to have the longest max open period. I think that's why it runs so well without massive duration
12-12-2002, 12:05 AM
#26
Melting Slicks
Member Since: Feb 1999
Posts: 3,078
Likes: 0
Received 0 Likes
on
0 Posts
Re: How accurate have you found DD2000 to be? (gkull)
Nothing so specific for me as some of you guys, but on one of my engines, I did seem to get pretty agreeable results with what I would have expected. What makes my situation complicated is my high altitude.
I built a pretty moderate 355 Chevy small block, with 10:1 compression, valve job, 221 degree cam at 0.050", headers and full exhaust, and DD2000 estimated roughly 350 hp and 400+ lb-ft.
Skeptical, I took it out and g-teched a couple of mile-high quarter mile runs. I got 14.9 seconds at 101 mph, and repeated that a few times. Then I took the g-tech's horsepower estimate, added one third for the altitude effect, and took into account a 20% driveline loss, and arrived at almost exactly what DD2000 predicted I'd make at the flywheel.
The Dragstrip program that Mr. Gasket sells to go with their DD2000 seems a LOT more optimistic. Taking into account I was running on narrow SR-rated street tires, I can understand the et being "slow" for that trap speed. But no way was that a low-13 performer!
Even worse, my 68 Firebird, which IS a 13 second performer was predicted to run high 11's. :bs
And my 427 Corvette is supposed to be running mid-10's ...on 3.36 gears.
:rolleyes: Yeah right!
I built a pretty moderate 355 Chevy small block, with 10:1 compression, valve job, 221 degree cam at 0.050", headers and full exhaust, and DD2000 estimated roughly 350 hp and 400+ lb-ft.
Skeptical, I took it out and g-teched a couple of mile-high quarter mile runs. I got 14.9 seconds at 101 mph, and repeated that a few times. Then I took the g-tech's horsepower estimate, added one third for the altitude effect, and took into account a 20% driveline loss, and arrived at almost exactly what DD2000 predicted I'd make at the flywheel.
The Dragstrip program that Mr. Gasket sells to go with their DD2000 seems a LOT more optimistic. Taking into account I was running on narrow SR-rated street tires, I can understand the et being "slow" for that trap speed. But no way was that a low-13 performer!
Even worse, my 68 Firebird, which IS a 13 second performer was predicted to run high 11's. :bs
And my 427 Corvette is supposed to be running mid-10's ...on 3.36 gears.
:rolleyes: Yeah right!
12-12-2002, 08:11 AM
#27
Drifting
Re: How accurate have you found DD2000 to be? (MoMo)
The Dragstrip program that Mr. Gasket sells to go with their DD2000 seems a LOT more optimistic. Taking into account I was running on narrow SR-rated street tires, I can understand the et being "slow" for that trap speed. But no way was that a low-13 performer!
Even worse, my 68 Firebird, which IS a 13 second performer was predicted to run high 11's. :bs
And my 427 Corvette is supposed to be running mid-10's ...on 3.36 gears.
:rolleyes: Yeah right!
Even worse, my 68 Firebird, which IS a 13 second performer was predicted to run high 11's. :bs
And my 427 Corvette is supposed to be running mid-10's ...on 3.36 gears.
:rolleyes: Yeah right!
-Greg
Found it again at http://stangstable.com/downloads.htm , called Dragstrip Plus. File is race406s.zip
[Modified by GregP, 8:14 AM 12/12/2002]
