Fuel Delivery

Facts & Figures for Mechanical Pumps, Filters, & Lines

There are three main components needed to deliver the fuel from the tank to the carb; pick up and fuel line, pump, filter.  Getting facts about the fuel pumps seems to be the biggest difficulty.  The other two items, lines and filters are not quite as bad in this respect.  A system should have a regulator as well.  However, many mechanical pumps have sufficient control built into the diaphram spring.  If this cutoff pressure is below what the carburator's inlet can handle, all is well and the additional regulator can generally be dispensed with.

Pump Performance
Pump Fit Issues on Chrysler LA Engines
Some Notes on Lines and Fuel Pick Up

Pump Ratings & Performance

One solution adopted by many hot rodders and racers is to convert to an electric fuel pump located near the fuel tank.  However, because I am a little paranoid of electric pumps in an collision, and because of the extra load electric pumps place on the electrical system, it has been my preference to stick with the stock arrangement using a  mechanical pump. 

Below is a list of mechanical pumps available for the Chrysler LA 273, 318, 340 and 360 cid engines.  The advertised numbers provide little insight because what is unstated is the test conditions they were made at.  Given no other information, assume the pressure given is the cutoff or maximum pressure the pump will produce. 
* Maximum pressure occurs when the flow out is zero, so it is most useful in deciding if a regulator is needed.  link: How mechanical pumps regulate pressure.
* The flow advertised is ususally the maximum.  Maximum flow only occurs if the pump was emptying into an open bucket,  has no resistance to drawing on its supply.
* Pump performance  is rpm dependent.
  We need is to know whether there is sufficient fuel delivery throughout the rpm range the engine is going to operate with a realistic pressure or resistance in the inlet and outlet fuel lines.   Fortunately, Holley was showing such information, in addition we sent three pumps to Ryan Brown for testing.  Testing flow rates at 4.5 psi on the outlet seems to be reasonable.  If tested with 3 psi on the outlet flow rates should be slightly higher.   If the suction side was restricted, the flow rates will be lower. 

Fuel Pumps for Chrysler LA engines

Brand &

Flow (gph)
Max Flow
@ 4.5psi
Flow @4.5




chrome  12-838
failed* 1/4 npt 2# 4.5 oz
80 gph** n/a 
 80 @2000  50 @ 2000

110 gph  12-360-11
110 @5000 70 @ 7000
1/4 npt 1# 8.2 oz



15 @ 500

Super Street M6902

1/4 npt
Competion M6270
7.5 7.5 80
70 @ 6000 1/4 npt 1# 9 oz

Mopar Perf

is a Carter


9.5 7 90
71 @ 6000
3# 7.5 oz
with Edelb bottom plate

3/8 npt 3# 9   oz

Pressure Limit = Cutoff or Maximum Pressure, zero flow.
Advert. Free Flow is the advertised maximum which is free flow, or in the case of the stock pump simply the shop manual specification which is not maximum because it is only at 500 rpm. (Flow will go up with rpm until the internal design restrictions choke it off.)
*The Holley Chrome pump was used and Ryan Brown could not get it to work well enough to produce consistant numbers.  **80 GPH model listed as closest ref to older Chrome pump.

A graph overlaying our test results and Holley's is below.  How comparable Holley's test procedure and setup is to Ryan Brown's is hard to know.  Clearly they are similar.
Test Rig at Ryan Brown's

Also on the graph is an estimate of the maximum fuel consumption for the power curve of an engine.  This is an estimate for a v-8 that peaks around 325 hp just over 5000 rpm.  
A close approximation of how much fuel is needed at a given rpm is based on these formulas.
 1 lbs/hr of fuel required = horsepower /2
 gal/hr needed = (lbs/hr /6)

Finally, before making a final decision, use a good safety factor because fuel consumption or the resistance in the flow might be higher than estimated resulting in less fuel actually making to the carb bowls than estimated.  This is particularly true during extended high G accelerations such as as drag strip launches.

Max Required is based on horsepower for a 340 bored .060 with Competion Cam's "Magnum 280" cam.
Holley data from Holley Website product info. They also have (had) good pump tech if you can still find it.
Carter and Mopar Performance pump data from testing at Ryan Brown Racing.  New MP P4529638 6 valve pump.  Used Carter M6270 'Competition' pump.
Interesting to note that the Holley 80 peaks before or at 2000 rpm.  (This model was not avail for Chrysler small blocks but included in graph for comparison with Holley's data for the 110 Pump.)

Clearly the MP six valve pump would be the choice if lots of flow was needed as low as 2500 rpm.  Also it would better handle a restrictive fuel system and it had a reasonable cutoff pressure that would not require a regulator with most carburators.  However, it has drawbacks in weight and in fit.  Fit issues described below.  For this reason, we decided to to try a Holley 110 pump.  First, its claimed curve (if accurate) is a little better in the mid range than the Carter M6270.  Second, our experience had been that the M6270's cutoff pressure of 7.5 psi was just a bit too high.  Occassionally this resulted in overfilling the bowls and flooding the carb when demand was low (coasting, idling).  Ryan mentioned that the (then) new Holley Kieth Dorton series pumps he has tested were generally pretty good.   Anyone dependent on a mechanical pump for racing should consider sending their pumps for testing.   Only drawback is the delay in packages going across the US/Canadian border. 

Fit Issues with the Mopar Performance 6 valve pump (p4529638)

Interference problems occur using the six valve Mopar Performance fuel pump and water pumps with passenger side inlet.  Note that the inlet sweeps back, but it still clearly possible to fabricate a solution to avoid contact and not restrict flow. The casting nub just barely touches the top of the fuel pump housing and can be ground back to clear.
The other problem is that the inlets and outlets are 180 degrees opposite each other.  When the inlet faces toward the fuel line from the tank, the outlet faces right into the coolant hose.  However, the Edelbrock bottom plate offers a solution using gentle 90degree tube fittings.  In the photograph the alternative inlets and outlets are shown.  The ones not used can be simply plugged.


A tip from the factory. Make sure the fuel filter is arranged so the outlet is higher than the inlet. This helps hot vaporized fuel to burb out.

Many of the a-bodies used 5/16" fuel line.  A good way to minimize the resistance in the fuel lines is to size up.  Getting good 3/8" diameter replacement lines is not quite as simple as we'ld all like.  First, the lines need to be made in such a way that the bends do not pinch or get oval and reduce the cross sectional area.  Second, the clips and fittings need to be changed as well.  Third, and perhaps most important, there is not a good source for 3/8" fuel tank pick up for these cars.  The pickup is probably the most critical place to increase the fuel line diameter.  For the new fuel line, three brands were tried.  Only on the third try was the product satisfactory.  Inline tube provided a line that was specifically for the '67 and had very little loss of diameter in the bends, the others both had noticible distortion in the bends, and one did not fit the '67 frame in the wheelwell. The first was returned so I don't know if it would have fit.

Serious high horsepower or high G cars should also consider going to braided aircraft line.  At least AN-6 will be needed, and many prefer -8.  This gets into a whole new area of hooking up AN hose and fittings to a stock fuel tank, or getting a fuel cell.

To convert the fuel sending unit required a little work.  At least in a '67 A-body Plymouth, the original pick up is a different design than the usual replacement units.
The first attempt is documented below.  However it didn't work out so ended up having to use an aftermarket 3/8 pickup for a later A-body and plug the opening for a return line, 

As best possible, the replacement unit was made more like the '67 when the 3/8 tube was installed.  The sending unit was disassembled to get it as far from the soldering as possible.  The it was reassembled and riveted to a bracket. 


Top. 1967 Pickup and Sending Unit.
Bottom. 1968+ Replacement Unit

In foreground, replacement unit with 3/8 tube and sender relocated onto intermediate bracket.  Trial fit showed that tube did not need to come out as far as shown - an 85 degree angle would have kept it closer to tank and further from axle.

Detail of modified unit.  Sender housing barely fits into tank hole when attached as shown.  It had to be squeezed down a little in a vise, and the back plate of same filed down to match.

Home:'67 Barracuda..one curve to another  
More Tech:
Improving Forward Lighting & Electricals
Ignition Wire Comparison
Handling Theory for late-'60s to mid-'70s Mopars

    If you take advice, or copy something I have done, you are on your own.  I'm just trying to be helpful. You need to make your own decisions, and know about or discover the risks before barging ahead on any endevour.

copyright Mathew Grubel, 2003
third edition. Feb 13, 2017