Trust, but verify

[PackardV8]

Over the years, I've found DynoSim and other similar computer engine programs to be pretty close in predicting what will happen; assuming head flow is input correctly.

On a recent simulation of a small V8, DynoSim is saying a switch from a "standard flow dual plane" to a "standard flow single plane" will give a 43-horsepower increase; that's 15%. Would you expect that to actually happen?

Part II.
The standard CFM calculator says this engine needs 585 CFM, but the DynoSim keeps increasing horsepower up to 750 CFM, all else being equal. Since this is a vintage road race build, top end horsepower is not as all-important as broad throttle response. What would you use?

[MadBill]

I've always found that cam specs and port CSA had more effect on powerband than carb size. Assuming the engine is in the 283-302 range and 450 + HP, regardless of the manifold I'd dyno it with a range of sizes up to at least 850 CFM, then track test for response with the highest HP size that didn't kill mid range against say a 650.

The 308" SBF in a Vintage road racer I consult on made 630 HP @ 8,000 and 448 lb-ft. @ 6300 with a 1,000 CFM Braswell and had dynamite response down to 4,000. Results may vary, as not every carb is a Braswell...

[randy331]

On a recent simulation of a small V8, DynoSim is saying a switch from a "standard flow dual plane" to a "standard flow single plane" will give a 43-horsepower increase; that's 15%. Would you expect that to actually happen?

We've seen close to that amount, but I wouldn't bet on anything based on a sim.

The crossover point has always been 4500-4800 ish rpm. Below there dual planes have been better, above there single planes.

Randy

[emsvitil]

Don't forget with a dual plane each cylinder can only pull from 1/2 of the carb.

[Geoff2]

When we get around to racing DynoSims, I'll be sure to get that version!

[mag2555]

If the sim does not show low speed torque rolling off when you switch from a duel to a single plane, and or go in steps up to a 750 cfm Carb then I would question its overall accuracy .