I thought it might be of interest to expand the discussions of air flow for the PSD started here https://www.ford-trucks.com/forums/607368-disconnecting-the-ebp-sensor-isnt-a-good-idea-3.html by giving an overview of a model that I'm working on for estimating it. I'm also installing a gauge for measuring air flow so I can compare measurements with predictions from the model.

When the engine is operated at light loads the boost gauge reads 0 psi, but there's still a "gravity boost" of about 14 psi MAP forcing air into the cylinders during their intake strokes. It's gravity that attracts the air to the earth's surface with enough force to produce normal atmospheric pressure. Under these conditions the flow is given by cfm=(0.1285)*rpm*VE, where VE is the volumetric efficiency, which for now I estimate as 0.75 for lack of better info. The VE actually varies with rpm slightly, and for a PSD peaks in the vicinity of 2K rpm. Using this equation gives 50 cfm at idle, and 300 cfm at max rpm.

At higher loads the turbo starts producing boost which adds to that already produced by gravity. The air pressure first increases at the turbo outlet, and this pushes more air toward the inlet manifold so that MAP increases above 14 psi, and this forces a higher flow rate into the cylinders during their intake strokes. The problem is to determine how much higher the flow rate is as a function of increasing boost.

Boost provides an additional pressure difference (delta P) from the turbo outlet to the inlet to the cylinders. If the air flow through this path is Laminar, the additional flow rate (cfm) is directly proportional to the additional driving pressure (delta P). In this case, a boost of 14 psi would increase the MAP by a X2, and double the flow rate into the cylinders.

As boost increases to higher levels the air flow becomes more turbulent, and a disproportionate increase in (delta P) is required to increase flow rate. For pure turbulent flow a MAP increase of a X4 is required to double the flow rate into the cylinders.

The boost (B) increases the MAP by a factor of [(B+14)/14], and this increases the flow rate (cfm) by a factor of [(B+14)/14]^(1/n), where the value of n depends on the amount of turbulence in the flow, ...n=1 (Laminar), n=1.5 (Intermediate), n=2 (Turbulent).

The Table below gives some tabulated values of [(B+14)/14]^(1/n), and doesn't include 2nd order effects like the pressure drops across the IC and air filter, which are included in my detailed model. It also gives some representative values of rpm along with the range of possible cfm flows depending on the degree of turbulence. When I start towing again in 7 weeks or so I plan to collect data to compare with these predictions.

B (psi)---n=1-----n=1.5-----n=2--------rpm-------cfm(n=1)---cfm(n=1.5)---cfm(n=2)

05-------1.36----1.23------1.17-------2,000------262---------237-----------226

10-------1.71----1.43------1.31-------2,200------363---------303-----------278

15-------2.07----1.62------1.44-------2,500------499---------390-----------347

20-------2.43----1.81------1.56-------2,700------632---------471-----------406

25-------2.79----1.98------1.67-------3,000------807---------572-----------483

In addition to just understanding better how stuff works, the reason I'm doing all this is to try and determine how hard to drive the puny little turbo on my early 99 for obtaining optimum overall performance when both air flow rate and air flow temperature are taken into account. As the turbo is driven harder to increase boost and therefore air flow, the air is also heated which reduces its density. If the turbo is overdriven, the additional small gains in flow that are limited by turbulence are not enough to overcome the increases in temperature, and the MAF into the engine actually decreases. I've preciously posted my temperature model for the turbo and intercooler, and various SOP measurements of HP while towing in various temp conditions.

Turbo & Intercooler Analysis!
https://www.ford-trucks.com/forums/604191-turbo-and-intercooler-analysis.html

 

What kind of guage are you going to install to measure air flow?

 

I've thought about two ways to measure cfm flow rates. Measuring the flow velocity with a small anemometer and then calculating cfm using the known cross-sectional area. Unfortunately, the flow velocity isn't uniform across the cross section, and the anemometer might get ingested unless it's placed upstream of the air filter.

Therefore, I'm figuring ways to use the second method, by measuring the pressure drop across a restriction to the flow. I'm for sure going to install a 30" vacuum gauge in the place where the restriction gauge normally goes on my AIS. I've been told (but I'm still searching the web for it) that there is published data for my AIS that gives cfm flow vs restriction for a clean filter. A number I saw quoted in a thread was 16" at something like 600 cfm, but I can't find the thread.

I'm also considering measuring the pressure across the IC, as I've seen published data from Banks of cfm vs delta P for their power pack units. Even though I have a stock IC, I think I might be able to get some reasonable cfm estimates that way.

Independent of absolute cfm accuracy, either of these methods will allow real-time measurements of relative cfm so that I can tell how this decreases at higher boost due to turbulence, but at this point it's all still a work in progress which I mess with when I've nothing better to do.

 

Ok, I was thinking you might try measuring Delta P across a device with a known performance curve, but after seeing your very liberal use of duct tape in other threads, I figured you would just tape a rotating vane anemomitor in the intake pipe.
Just kiddin' with ya man.
Seriously though, I think measuring static pressure drop with a differential pressure manometer may work. However, to get an accurate CFM may prove to be difficult using the air filter. A clean filter will start to get dity enough to throw off your measurments pretty fast. I have seen pressure drop, venturi type balancing valves for use in labratory exhaust and other critical environments, but I think they might not be very practical because it will add restriction to the engine and not sure of price, but they look expensive! Let us know what you figure out. Keep up the good work.
Nick