I use them for things like cam selection, etc...they're great for seeing trends in power for things like that. On the Desktop Dyno 2000, you sort of have to ignore the low speed torque numbers when choosing a large duration vs displacement cam. It automatically assumes port velocity is optimal at low rpm's, which inflates the numbers. However for peaks, and power curves, it's surprisingly accurate. Accurate meaning within a good sized ballpark =). You have to, and I mean HAVE to get accurate flow numbers though. The more real the flow numbers you get, the better. Inflated numbers flowed with a huge bore sizes etc won't help your accuracy any.
Just my thoughts...most people either love them or hate them.
The output was probably within 5-10% accurate, I have only chassis dynoed my car but it did make me decide on a smaller cam. I was able to input bigger cams only to see minimal if any difference in HP & torque numbers. A pretty cool tool IMHO.
I use Performance Trends EA. It is a lot of fun to try different combos and see what they do. I have learned a lot about cam selection and different aspects of engine design. From what I have seen it is within 5% of what true dyno numbers show.
Only thing about Perf Trends seems to be that if you run a set of unknown heads, headers etc, you have to totally guess at the values, which makes things frustrating to say the least. What good is guessing? lol. That's the whole point of a 'dyno' simulation =).
lol, every motor I've ever built in RL that I tried on this thing does that =/.
Shows [email protected], and 377lb/[email protected] for the 347 I'm putting together with Pro Comp 210's, 10.8:1 compression, 750 Holley, Single plane intake, 286-296 Solid roller cam, and 1.75-1.875 stepped tube headers.
Desktop dyno shows 505hp and 460lb/ft at the same rpm's, same combo lol.
For comparison, a relatively mild flat tappet 331 with the same heads and 10:1 compression here in town recently dynoed at [email protected] and 398lb/[email protected] I know the builder somewhat and believe the numbers. This was with a 670cfm Holley.
Which program to believe? I'm not sure, I may not be setting up the perf trends one right, as I can never get 'clean' numbers, and I know these motors run lol.
If you get those comments on all your tests you are using too much compression for the octane rating of the fuel, ports and valves that are too large, and too much rpm for the design parameters. You may find that if you drop the size of your ports and valves you will lose the "poor mixture control" and the "likely valve toss". You can increase quench area with smaller valves and you will likely lose the prone to detonation warning with added quench and lower compression.
Side note: You can hear pre-ignition but you don't always hear detonation. You find out about it when it burns the side out of a piston or when spark plugs show melted or burned electrodes. Detonation is most likely in low turbulence, high compression engines where the ports are large enough that fuel falls from suspension in the air column moving into the chamber. You get a lean mixture that burns to fast at RPM ranges from peak torque to peak HP. With the throttle open about half way after making a hard run shut the key off and see if your engine continues to run. If it does then you are detonating.
I have used the performance trends dyno for several years and run the drag simulation the offer also. Real work test on many different cars and always with in a .1 of real track times. The beauty of the program is you can make a change.maybe correct or incorrect and it will point you in a right direction , if it shows a loss in power / torque you dont buy the part.
I used the program on 10.5 outlaw motors, pro mods, super street. ect........and always been with in .1 of real track times.
I will just say that using computer programs to build an engine, or choose components is likely to leave you with a bad taste when it comes to true life performance. They might be accurate within a percentage, but those programs will never predict engine accelleration, load variance, engine recovery, or wave pulsing in intake and exhaust tracts.
Find you a GOOD engine builder with experience, and credits to his/her name. THAT will be the best money spent. Use programs like video games...to kill time!
I dont disagree with you about the curves, but I believe programs are a poor way to design an engine. You need experience building engines to understand the importance of things that I mentioned earlier.
Programs are fun, but I've seen 500Hp combos at the track not running nearly as well as they should. Granted some of it can be chassis, but the cars I am refering to had good chassis and descent 60ft.s. MPH was down, so that is an indicator POWER was down.
Having said that, I did an engine for a guy that wanted to follow his computer aided design, and finally talked him into changing the cam...that combo ran very well. He has talked about changing the cam to what the computer had recommended...but still has not done it. Maybe because he now understands what I was getting at.
EA 3.0 was a huge factor in my engine buildup. If nothing else, it gives you some insight on how various variables affect an engine. It predicted my engine nearly right on. I'm using it to predict gains from improvements i'm in the process of making.
The bad thing about desktop dyno's is that they can be very good if your entering correct data or they can be very bad if your entering incorrect data. IOW, you really need to learn to use it and double check your input. EA 3.0 makes fewer assumptions that desktop dyno...it requires more input. Therefore, a better chance of being accurate ond conversely a higher chance of being way off.
Tracy Blackford: Corona, Ca
'65 FB Mustang 331, 282S cam, ported 351W heads. T5z, 3.50 9" posi.
346 [email protected] on a warm spring day (335 RWHP SAE corr.)
<font size=-1>[ This Message was edited by: blkfrd on 11/5/06 4:20am ]</font>