[culick]

"g-force" is somewhat of a misnomer. first of all "g" is not short for gravity or gravitational, it is used to designate a constant, i.e. the acceleration of gravity at the earth's surface (1g =~ 9.8 m/s^2). acceleration is not force, force is mass times acceleration.

if one says g-force = amount of force it takes to accelerate a mass, of value, m, 1g then that would be correct, g-force = mg.

If you say your car "pushes" 1g down the track, you mean that it accelerates at approx. 9.8 meters per second per second (or 22 mph per second), not that the lateral force (from your drive train) is 22 mph/s.

it's bonehead freshman physics dude.

[Brian Baker]

My head hurts hurts. I need an aspirin aspirin.

[Tom Vaught]

Without getting too picky on terms I believe most
people, who read these posts, will get the basics well enough to understand how to fix the fuel
system on their vehicles.

There are very few Scientists on this board, therefore you must take the thought they are trying to put forth over pure engineering terms.

I know a lot of engineering types who could not
set up a NOS system if their life depended on it
and I work with them every day.

Just my opinion! Tom V.

[Engo]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by culick:
"g-force" is somewhat of a misnomer. first of all "g" is not short for gravity or gravitational, it is used to designate a constant, i.e. the acceleration of gravity at the earth's surface (1g =~ 9.8 m/s^2). acceleration is not force, force is mass times acceleration.

if one says g-force = amount of force it takes to accelerate a mass, of value, m, 1g then that would be correct, g-force = mg.

If you say your car "pushes" 1g down the track, you mean that it accelerates at approx. 9.8 meters per second per second (or 22 mph per second), not that the lateral force (from your drive train) is 22 mph/s.

it's bonehead freshman physics dude.<HR></BLOCKQUOTE>

culick. besides the physics thermology, would you agree with Tom McQueens description and conclusions?

Street ´69 Firebird. 3800 lbs with driver. 455 with D-ports. 275x50 drag radials. Stock suspension. [email protected] (motor only)[email protected] (+150hpNOS)

[culick]

ah come on Tom V, you getting soft on us?

I think Tom McQueen is going in the right direction but I think he confuses pressure and force (pressure is force over area) and grossly overestimates the amount of fuel in the line. Here's my stab at the prob.

the volume inside the fuel line can be calculated as a cylinder w/ the formula:

v = pi*length*inner radius^2, where

pi = 3.1416
length = 12ft = 144"
radius = inner diameter / 2 = 0.25"

so,
v = 3.1416 * 144" * 0.25"^2 = 28 cubic inches

there are 231 cubic inches in one gallon so there are:

28 cubic inches * (1 gallon / 231 cubic inches) = 0.12 gallons of fuel in the line, which weighs,

0.12 gallons * 6 pounds / gallon = 0.72 pounds of fuel

now, according to Newton's first law when the car begins to accelerate at 1g all the fuel wants to stay put and needs its weight in force to get it moving along w/ the car at 1g (as Tom M. stated) but the fuel is a fluid and not a rigid body which means that the force that needs to be exerted to get the fuel moving along with the car is spread out over some area, which is the cross sectional area of the fuel line at the pump (it is in a plane orthogonal to the direction of motion). Force applied over area is pressure (psi) so the pressure that the fuel exerts on the pump is:

pressure = 0.72 pounds / (pi * 0.25"^2) = 3.7 pounds per square inch

this means that 3.7 psi needs to be added to the 6.5 psi that you want the pump to maintain at the regulator, thus you need to test the pump flow at 10.2 psi. 300 lb/hr of fuel is 50 gph, so if the pump is 77 gph at 6.5 psi it may well be below 50 gph at 10.2 psi. Also, with slicks you are probably accelerating over 1g (for a short amount of time), at 2g's you would need to test pump flow at 14 psi.

Also the walls of the fuel line apply a "drag" force that would help to get the fuel moving so the pressure figure may be overestimated somewhat.

[Dave A]

Interesting discussion thus far. (I will comment briefly that 'g-force' is acceptable slang around here; its meaning is never misinterpreted).

The short version of the rambling below is I think the system is complex enough to warrant testing on the actual vehicle, perhaps with the WB O2 sensor, to figure it out (or trial and error), but you may be able to improve the bench test somewhat. I am not an expert, so please take my comments below for what they're worth.

Thinking out loud, culick's calculations seem reasonable for the pressure required to accelerate fuel contained within the fuel line at a rate of one g.

It seems we are ultimately trying to consider the 'response' of our fuel system...how quickly an it go from delivering idle fuel flow to WOT fuel flow while simulataneously being accelerated with the car? This discussion seems to be focused on the response time; the assumption is made (and probably validly) that steady state fuel delivery is sufficient (and if it was not, it would show up as a pressure drop towards the end of the track).

(Just theorizing here) - the difficulty in diagnosing a fuel system response issue using a vehicle mounted pressure gauge is that the guage itself has some 'mass, damping, and capacitance' and acts essentially as a filter to some extent. In fact, damping may be purposely added to the guage to prevent the needle from bouncing to the point of being unreadable.

Not only do we need to accelerate our slug of fuel in the fuel line at the rate the vehicle is accelerating, we also need to accelerate it from launch condition fuel flow (as low as idle for some cars but very close to WOT for others) to WOT fuel flow. One must also consider of fuel feed into the pump (even for a rear-mounted pump); if too much pressure is required the fuel will simply cavitate no matter what the pump's pressure side is capable of (probably good reason to mount the mount behind the tank?).

The actual acceleration profile of the vehicle is somewhat complex; it can start with a momentary spike as it absorbs the shock of unleashing the drivetrain on it, maintain a high acceleration level for a moment, and then taper off quickly (especially at a shift). The fuel system is subjected to this acceleration curve, and also to the 'flow' acceleration curve as it transitions from pre-launch to WOT fuel flow. Fuel flow must be analyzed, under these varying conditions, both before and after the pump (granted, if flow after the pump is satisfactory all the time, then it necessates flow before the pump being adequate; it's when there is a problem that one must determine if its before or after the pump)

My two cents is that one would need some sort of data acquisition system and electric transducers to measure fuel flow and fuel pressure accurately enough versus time to really distinguish the response time of one fuel system setup versus another - it is possible to make 'worst case' assumptions about the vehicle environment the fuel system operates in for the first 60 or 330 feet or so and test response times under these conditions; but one still needs accuracy in measuring flow and pressure.

When dealing with a fuel flow issue at lauch, on a system that is otherwise adequate to meet the demand of the engine further down the track, it seems reasonable to suspect that we're dealing with volumes of fuel of significantly less than a gallon for most cars. In determining a 'response time' of the fuel system, one would need to measure instantaneous fuel flow with a fairly high degree of accuracy.

That being said, for a bench test, perhaps the following MIGHT work (these are my thoughts, I have not done this before but am just thinking out loud)
-Assume one g constant acceleration by inverting the fuel system 90 degrees (run straight up and straight on the bend on the supply side to the pump)
-at the end of the fuel system, have a large, low restriction 't' and a valve (preferably and electric solenoid valve) to switch between them. Connect your data aquisition unit to the solenoid so that you can record the exact time of solenoid activation and deactivation.
-On one side of the 't', insert a restriction that results in a steady state flow rate that seems reasonable for your pre-launch condition. Run this side of the t into a large resevoir for fuel.
-on the other side of the t, run the line through a restriction that results in a fuel flow rate appropriate for WOT steady state (you'll have to measure it and adust it until it's close)
-have another reservoir at the outlet of the second side of the t

to run the test, perform the following steps:
1) empty the WOT reservoir completely and weigh it on an accurate scale
2) set the solenoid so that fuel is routed through the rep-launch circuit
3) energize the fuel pump and pump fuel through the pre-launch circuit
4) start your data aquisition and energize the solenoid to pump fuel through the WOT circuit
5) after a pre-determined time (say two seconds - enough to reach steady state flow, but not long enough to render the response time insignificant) switch the solenoid back to the pre-launch circuit. Use a stop-watch with fractions of a second if it helps.
6) stop the pump
7) weigh the fuel in the WOT reservoir
7) look at the data recorded and figure out exactly how long flow was routed to the WOT reservoir. Add or subtract the steady state mass flow rate times the difference between the actual time interval and the desired time interval.
8) the corrected mass in the WOT reservoir divided by the time interval is the mass flow rate of the fuel system for the first second(s) of operation under a simulated launch into steady 1-g acceleration with near-instant transition from pre-launch to WOT fuel usage.

[Skip Fix]

I was pondering a "fuel reservoir" concept to use EFI with and un baffled stock gas tank.

Coffee can size "tank" inlet in the top, pickup in the bottom(like a dry sump oil system tank). For this it could be mounted in the rear and before the electric pump(non pressurized).

For drag racing one in front of the engine, but then the whole tank would be pressurized,and possible sealing problems(leaks) with gasoline.

[Dave Schiffers]

So could we have an example of an ideal system for a 9.90's - low 10's second street car, for us simplist's that want to set up the fuel system the right way the first time.

Thanks folks

Dave Schiffers

[Tom Vaught]

Quote:

"ah come on Tom V, you getting soft on us?"

You and Dave know way too many fancy words compared to me, a poor country boy who barely
can work a TI-35 calculator.

Would like for you to explain how something goes faster due to "drag" though!

To answer the other gentlemen's question:

I have personal experience with a 10/ high 9 second steel GTO (no glass) that used a Mallory
140 fuel pump, a LARGE Holley Billet regulator
(with the 7/16 internal relief ball), and 1/2"
fuel line from the tank to the regulator and 3/8"
on to the carb bowls. It was worth 3 tenths over his previous system and was not overly expensive
to purchase. It did not use a Fuel pump return
line.

This system should work fine for your application.

Tom V.

[culick]

I simplified the calcs made above into an easy fuel stack pressure formula:

fsp = avd/A

where:

fsp = fuel stack pressure (psi)
v = volume of fuel line (gal.)
d = density of fuel (in lb/gal)
A = cross-sectional area of fuel line (in square inches)
a = max acceleration of the car (in "g"s)

this further simplifies because,

v = lA/231

where:
l = length of fuel line (inches)
a = cross-sectional area of fuel line (in square inches)
231 is the number of cubic inches in 1 gal.

substituting la/231 into the fsp equation gives

fsp = (lA/231)d/A

the "A"s cancel so the final form is:

fsp = ald/231

where
l = length of fuel line (in inches)
d = density of fuel (in pounds/gallon)
a = max forward acceleration of the car (in "g"s)

fsp is in units of psi so it can be added to the pressure desired at the regulator, then look at the pump flow rating at that pressure, e.g.

if
l = 144"
d = 6 lb/gal
a = 1.5 g's

then fsb = 5.6 psi
and we want 6.5 psi at the reg. then the pump needs to push 12 psi.

then find the rating of the pump, the mallory 140 flows ~70 gph at 12 psi

70 gph fuel flow (at 6 lb/gal) is 420 lb/hr and then assuming a BSFC of 0.5 lb/hphr, it should be good for 840 hp.

of course a 840 hp car may be able to accelerate at more than 1.5 g's so the calculations may need to take this into account.


I used "drag" as an analogy, it is actually a fluid shear force, the fuel line is being accelerated at the rate of the car, the fuel is almost at rest, the inner wall of the fuel line pulls the immediately adjacent fuel molecules along with, it those molecules pull on other fuel molecules that are more toward the center of the line but they cannot pull them hard enough to keep up so they lag (shear).

[Engo]

Thanks culick!

Street ´69 Firebird. 3800 lbs with driver. 455 with D-ports. 275x50 drag radials. Stock suspension. [email protected] (motor only)[email protected] (+150hpNOS)

[Engo]

culick. I got into a heated debate with a couple of other guys on another board in this issue. I was advocating exactly what you were saying but they refused to believe me. Two of them were board moderators too.
The argument that they relied on was something like this " The weight of the gasoline (thus the size of the fuel line my comment) is completely irrelevant due to the fact that the fuel system is a closed system. The acceleration affects the fuel system as whole."
How do I convince these guys (I believe they are engineers) that they are wrong?

Street ´69 Firebird. 3800 lbs with driver. 455 with D-ports. 275x50 drag radials. Stock suspension. [email protected] (motor only)[email protected] (+150hpNOS)

[ponjohn]

Is there a way to determine the real "g's" for the first 60 feet based on weight and short time?

John
69 Judge(Sold)

67 Firebird-street/strip car in progress

[johnta1]

G's from 60' times


johnta1
Pontiac Power RULES !!!
www.wallaceracing.com

[Injunjoe]

That calculator will give you the average G's over the first 60 ft. The only way I know of to get a true G vs. T is with a G Tech Pro, or any other means of measuring acceleration and logging the data. I believe the G-Tech has a data logger.

Joe

[Dave Schiffers]

Tom Vaught

Thanks for the info.

I have surfed the Summit cataloge and cannot seem to find the regulator you described.

Might you possibly have a part # or other specific regulators to use and which to stay away from?

Thanks again for the help.

Dave Schiffers

[Skip Fix]

You mean there's a calculator newer than the TI SR-10??And they used to be $120.00!And they actually did square roots, my SR-11 even had pi on it!

Dating myself here but in freshman chemistry we could NOT use calculators, had to use slide rules due to expense of calculators they thought it gave an advantage to some. You could also tell the engineering students by the size of their slide rules, the big ones clipped to their belts.

[culick]

it is true that the fuel delivery system is closed loop, but that doesn't mean that its response will always be adequate. the most commonly encountered closed loop system is a thermostat. you set the thermostat to a desired temp (set point) and the system compares the actual room temp to the desired set point temp and then turns on the heat or a/c in order to bring the actual room temp to the set point temp. when the room reaches set point the system turns off the heat or a/c. the system tries to maintain set point despite external influences, i.e. outside temp, the system adapts itself, by turning on heat or a/c, to changing external changes in order to maintain a constant desired temp. however if the set point is a comfortable 70degF and the house is engulfed in flames such that the actual room temp is 1000degF do you think that the system will be able to cool it down by running the a/c? no way, the system's response is totally inadequate.

the negative feedback in the fuel flow delivery system is the float needle valve assembly. the set point is determined by the float and is the desired fuel level in the bowls. the fuel system adapts itself to the fuel consumption of the engine by allowing fuel to flow into the bowls when the float drops below its set point, but in order for the system to maintain the fuel level set point the pump must supply fuel at a rate that is greater than the engine is consuming it. if the inertial effect of accelerating the fuel in the fuel line puts extra stress on the pump such that the rate of fuel consumption exceeds rate of supply then the set point will not be maintained.

just because the system is closed loop doesn't mean that it magically immune to the laws of physics. as Dave pointed out it is a fuel response problem.

[Dave A]

Tom V., if it's any comfort, I can use only about 5 percent of the functions on my TI calculator as well....

I'm thinking if one chooses a fuel pump that'll meet the flow/pressure requirements of the engine at WOT plus the 'fuel stack' pressure described by culick that that'll be close enough (without having to worry in detail about the response of the fuel system).

One still needs to make sure that on the feed side of the pump, that the 'fuel stack' pressure required to accelerate fuel in the line, plus the pressure required to maintain WOT fuel flow through the feed does not exceed the vapour pressure of the fuel (bearing in mind that with the pump mounted behind the feed from the tank, the 'fuel stack' pressure will actually help).

Engo, it sounds like the folks on the other boards advocating the notion of the fuel system being a 'closed' system are simply mis-applying the term. A true 'closed system' means there is no exchange of energy or matter between the system and its surroundings. When installed in an accelerating vehicle, the fuel system is subjected to force from the fuel system mounts in the vehicle, which make the fuel system (and the fuel inside of it) accelerate along with the vehicle. It is therefore NOT a closed system in the classical definition of the term, because it is receiving energy from the accelerating vehicle.

[KEN CROCIE]

engo--Please refer to my post on the 1st page.note that this car was tested on the chassis dyno and the track for both fuel systems.note that I did not use any calculations come up with exactly the right size or maximum efficiency as the engineer types would have you do.just the same common sense advice I have been giving my customers and friends for the last 30 yrs.the simplest thing to do is to get more fuel pump than you will ever think you need.then run a return type regulator.this keeps the fuel moving at all times@ (except launch) rated pump pressure, avoiding the tragedy of having to accelerated the from zero velocity @ idle..My recommended system is a carter hp6901 and any mallory return style regulator.