Selecting the right mechanical fuel pump

[travis]

Trying to find a mechanical fuel pump for a few different projects I’ve got going on. I found a formula that seems to be well accepted for determining fuel pump requirements...((GPH x 6.2 pounds per gallon)/0.5)) equals HP that a fuel pump will support, with 0.5 being the BSFC and 0.5 being considered as “average”.

So...using this formula, a stock replacement 25 gph pump should support about 300hp. Seems reasonable for at least short bursts. A 40gph performance stock replacement should support nearly 500hp...seems unreasonable to me but 400hp should be safe, correct? And then your performance mechanical rated at 110gph should theoretically support well over 1000hp...but nobody runs that kind of power with such a small mechanical pump...

So...assuming the lines and fittings and what not are sized appropriately, how do you select a pump based on free flow ratings (what does that even mean?)? Is this formula just flat incorrect, or does everyone just way over build their fuel systems?

[treyrags]

You need pressure vs flow data for mechanical pumps, not the "free flow" GPH. You need to know what it will flow in the 4-7 psi range. Holley used to have pressure vs flow graphs for their mechanical pumps in their catalog. Many times they are less than half of their open flow rating. The manufacturer of whatever pump you are considering should be able to provide this information.

[F-BIRD'88]

The "free flow" typical advertized flow rating of a fuel pump does not tell you anything about how much fuel that pump will actually move to the carb, into the carb's needle seat(s).
So , what really gets into the carbs fuel bowl.
You have to flow test it. Best done on the car through the actual fuel lines, pick up, fittings and fuel filter(s) and needle seat(s), and any fuel pressure regulator,, at the working rpm (mechanical) or working supply voltage (electric).

A test orifice of equal AREA can be substituted for the combined AREA of the needle seats. EG: 2 x.110" holley needle seats equals a .156" test orifice.

Not that hard to test a electric fuel pump... A little more involved to test the actual working flow volume rate (GPH_ and pressure (PSI) of a mechanical fuel pump flowing thru the actual fuel system. Requires a mechanical fuel pump test jig machine to work the fuel pump arm at the correct speed (engine crank shaft rpm /2) . You could rig up such a fuel pump test jig with a engine block+ camshaft and a electric motor to spin the camshaft in the block at the correct rpm.
That is if you really want to know how much Horse power a mechanical fuel pump can support.

[F-BIRD'88]

Another practical method to test a mechanical fuel pump is to test it on the actual engine in question. But substitute a separate, temporary (electric) fuel pump system to the engines carb, to actually run the engine-car for during the test.
A quick test of say 15 seconds test flow burst at required max engine rpm, thru a suitable test flow orifice.
into a container and a little math to convert to GPH will tell you the tale.
You'd need a ball valve to open and close the flow from the fuel pump into a container ( EG: 15 seconds). I'd do 3 short burst flow tests and average the amount.

You could use gravity to feed the carb for during the flow test.
Since there is little engine throttle load on the engine it won't need much flow for during the test.

you'd want to measure flow and pressure.

EG: If it can actually move 1 quart of fuel from the car's tank, thru the cars fuel system and pump, thru a suitable test orifice, into the test container in a 15 second test,, that is 60 GPH. Observing a in-line fuel pressure gauge during the 15second test will tell you the matching flow pressure.

[Mattax]

Yes, your numbers are about right.
Like the other guys posted, work off of pump graph done around 4-5 psi.
The output flow is dependent on rpm. As Treyrags wrote, at least for a while, Holley had graphs of their mechanical pump outputs in the catalog from something like 1000 to 8000 rpm.
On a mechanical pump, the internal pressure is limited by the spring.
This is the other specification to consider. A pump that cuts off at 7.5 psi (or higher) has the potential to force open the carb's needle and seat.

If you're doing circle track stuff, or for any other reason really need to know what a mechanical pump does, send it to Ryan Brown for testing.
Its certainly preferable to keep the bowls filled going down the track. The fuel level does effect the carb circuitry.

[Mattax]

Not turning anything up in a search G**gle and Duck duck. Maybe you'll have better luck. Here's one saved from back around 2003.

f110GPHchrt.jpg

[travis]

Apparently these mechanical pump flow graphs are some closely guarded secrets...I am finding absolutely nothing. So I guess figuring at 1/2 the free flow info would be fairly safe?

The pressures that these performance pumps are rated at really puts them close to the edge of overpowering the needle/seat on a lot of carbs...especially the edelbrocks I run on a lot of these low budget street deals. Your almost forced to use a regulator of some sort.

[treyrags]

Thanks Mattax. They are actually still in the current Holley pdf catalog. With my limited computer skills I have not figured out how to post them. For example on a 130 GPH pump: free flow is 70 GPH @ 1000 rpm, @4.5 psi its 42 GPH. 130 GPH free flow @ 5000 rpm, @4.5 psi its 82 GPH

[Mattax]

Thanks Mattax. They are actually still in the current Holley pdf catalog. With my limited computer skills I have not figured out how to post them. For example on a 130 GPH pump: free flow is 70 GPH @ 1000 rpm, @4.5 psi its 42 GPH. 130 GPH free flow @ 5000 rpm, @4.5 psi its 82 GPH

Well, Heck, the fact you found a current Holley catalog is pretty impressive web skills in my book. Everytime I bookmark a page on their website, it gets changed sooner than later.

For screenshots I've been using a freeware program called IRFAN VIEW, version 4.38 (works with XP as well Windows98). Its pretty small, no ads.
http://www.irfanview.com/
It's not the easiest to learn what does what, but I can walk you through the basic things its good for. For screen capture, hit 'c' on the keyboard, select 'custom' and start. Then save the file where you want, in whatever file format you want. I find jpg is good for pictures, png is a little better for graphs.

[Mattax]

Apparently these mechanical pump flow graphs are some closely guarded secrets...I am finding absolutely nothing. So I guess figuring at 1/2 the free flow info would be fairly safe?

The pressures that these performance pumps are rated at really puts them close to the edge of overpowering the needle/seat on a lot of carbs...especially the edelbrocks I run on a lot of these low budget street deals. Your almost forced to use a regulator of some sort.

We must have been typing at the same time!
Treyrags gives you the scoop on the Holleys. For Carter, you could try to contact them. Edelbrock and and RobMC, same but somewhat more likely to get a result. I don't know why they make it such a secret, but I obviously don't know much about marketing. I sent some pumps to Ryan Brown Racing, I've posted those results on my website - but they're specific to mopar smallblocks.

My experience with that Holley 110 has been its max pressure is around 6 psi and has never given me trouble (with various Holley 4150s). This is in contrast to one of the Carter strip pumps which occassionally would overfill bowls - eventually I figured it out and would 'just' pull over and crack the site plug out and soak up the extra. Lovely - especially on the way into work... Running along steady at interstate speeds (3000 rpm plus minus) and then letting off the throttle was a situation that would easily cause this.

I personally don't like the regulators between pump and the inlets because the pressure is regulated by restricting flow. I think its worth hunting down what you want. And yes, the OEMs pumps are often adequate. Probably the only data point you'll get for them is a free flow test at idle speed in the service manual - but worth a look.

[cgarb]

That's why I run a throttle bypass with mine, no regulator to restrict flow to the carb, full PSI wide open and 2 to 3lbs at idle.

[treyrags]

If its important to you to have consistent pressure at the bowls at wot a regulator is needed, especially if you want to run lower psi. The internal seat in the regulator is the restriction and they vary a lot between manufacturers. A larger seat will not only allow more flow but will cause less turbulence and aeration the same way a larger needle and seat behaves at a lower pressure. The one I use has a 7/16 seat and many of them have a 7/32, which is more than triple the difference in area.

[77cruiser]

Here are some flow graphs, at the bottom of the page for each pump.

http://www.robbmcperformance.com/products/chevy550.html