This paper was written by me and originally published in VINTAGE FORD Volume 28, Number 5, Page 33. The fuel pump in the application discussed here finally failed during a 1400 mile trip down Baja California and back. The bypass feature described worked as designed so that the trip could be continued without replacement of the pump (but with the need for care regarding maintaining the fuel level at a half tank or more).



Fuel starvation is a continuing problem, particularly in 1925 and earlier Model T Fords. Many people have expressed ideas for solutions, but none seem to have addressed the primary causes of the problems. I would like to set forth some observations and thoughts on the basis for fuel starvation, outline what I feel is a good solution, and then apply a fuel pump as a refinement to the solution.


We were driving a recently purchased, restored 1924 Model T in the October 1992 Baja 500 Model T Ford Tour. All 21 cars started out with full gas tanks, and all sailed up the first good-sized hill in Tijuana; all except ours, which coughed to a stop about 2/3 the way up. After a tow to the top of the hill, we were on our way again and 30 miles later the same thing happened while driving down a flat highway. The car has a glass sediment bulb, and both times we observed a quantity of air in the glass bulb. Loosening the bulb allowed the air to escape, and we were on our way again.

This was our first real Model T Ford tour, and we were still pretty green about the cars. At the end of that day's drive in Ensenada, we talked to a number of the other Model T fans and found that many of the cars were equipped with 5/16 and even 3/8 inch fuel lines and that a number also had fuel pumps. People referred to our problem as "vapor lock," but it was clear that air, as well as fuel vapor, was involved.

Our immediate solution was to obtain a length of 5/16 inch fuel hose and some fittings. We replaced the 1/4 inch fuel line with the hose tied along the top of the right hand frame member. The rest of the trip, which included crossing a mountain range, was made without further fuel starvation problems. Air still continued to collect in the sediment bulb, but it did not stop the car. We did not have to drain out the air again on the trip.


This experience set me to thinking about Model T fuel lines and the possible cause of the air that we were seeing in the system. I found literature to be lacking any clear answers.

The only definitive Ford material on fuel lines is given in three figures in the April 1922 Ford Service Bulletin. One of the figures is for open cars and the other two are similar figures for closed cars. No explanation or discussion of the figures is given. Murray Fahnestock repeated the figures in his writings, but he also did not elaborate on them.

Let's put the Ford fuel line picture on a common stylized basis as an aid for the rest of this discussion. In Figure 1 (and stylized Figure 2) the 1/4 inch diameter fuel line bends downward from the sediment bulb, and passes inboard of the exhaust pipe. It then is routed forward, crossing below the exhaust pipe over toward the frame and is secured to the inboard side of the frame at the pillow block for the hand brake system.

The three service bulletin figures differ a bit at this point. The open car figure shows the fuel line being routed below the transmission support arm, while the two closed car figures show it routed above the support arm as the line goes onto the carburetor.

This gravity feed fuel system provides only about a foot of fuel head pressure when the tank is full. This is about a third of a pound pressure. As the tank nears empty, the fuel head is reduced to only a few inches.

Prior Tinkerin Tips tend to recommend a straight, direct fuel line routing as shown in Figure 3. Note that the quick fix we used in Ensenada was essentially the fix shown with the exception that we used 5/16 inch hose rather than tubing.

If the line is routed on top of the right hand frame member, it tends to be shielded from the exhaust pipe heat. The Ford Service Bulletin routing below the exhaust pipe provides a similar, but unstated, benefit in that it is cooler below the pipe than above.


We have purchased two Model T Fords, both in a restored state, and found both to have a rather casual fuel line routing. The line was routed, generally above the exhaust pipe, with a number of high and low spots. This accentuates the exhaust pipe heating effect noted above and provides an ample heat source for fuel vapor lock. This is similar to that shown in Figure 4.

The up and down pitching of the car and the inertial surging of the fuel in the fuel line causes the carburetor inlet valve to allow air to pass back into the fuel line. This air works its way back up the fuel line, collecting in the high spots of the line and, ultimately, in the fuel sediment bulb below the fuel tank.

Surface tension around these air bubbles causes added restriction to fuel flow. The fuel head pressure is already very low, and these added losses make fuel starvation even more probable.

The Ford Service Bulletin fuel line routing has the line at a relatively low position where it is clamped to the frame. This tends to limit the ingress air to the section of fuel line nearest the carburetor. Air in this section can easily bleed back into the carburetor when fuel flow resumes. This appears to be the other unstated reason for the Ford routing.

The Ford sediment bulb has a cross section as shown in Figure 5.

Note that the inlet to the sediment bulb extends down below the level of the fuel line outlet. The air working back up the fuel line from the carburetor collects in the top of the sediment bulb. When fuel is actually flowing around the air bubbles in the fuel line, this bubble in the sediment bulb is partially drawn back into the fuel line so that the fuel level in the bulb rises high enough to flow out the outlet. If the bubble size becomes too large, the fuel level in the bulb drops, thus stopping fuel flow. A sufficiently large amount of air will lower the fuel level until the air can escape up the bulb inlet tube into the fuel tank. This is particularly true when the engine is stopped and the full air bubble returns to the top of the sediment bulb.

The air bubble is fairly permanent and one cannot get rid of it by draining of the sediment bulb. As the bulb refills after draining, the air escapes up through the inlet tube until the fuel level rises to the bottom of the inlet. The remainder of the air becomes a trapped bubble. If one tries to purge the air from the system by disconnecting the fuel line at the carburetor, the low fuel head pressure does not provide an adequate flow to entrain all of the air bubbles and carry them out of the line. The fuel simply works its way around the bubbles and dribbles out of the open pipe.


Glass sediment bulbs generally have the same features as the Ford sediment bulb except that the outlet is located in the top of the unit. The straining screen is horizontal across the unit below the outlet and surrounding the inlet tube, which extends well down into the glass bowl area. These units have the virtue that the glass can be loosened with the inlet valve open; thereby permitting most of the entrapped sir to be bled out of the bowl.

After reviewing the available information and arriving at the air bubble conclusion, I re-plumbed our 1925 touring in a somewhat more radical manner as shown in Figure 6.

The fuel line, which is 5/16 inch tubing, is routed directly from the sediment bulb (now a glass unit because the original Ford bulb had a persistent leak) to the outboard side of the right frame member. The line bends downward and forward along the outside of the frame member and is clamped at the forward end of the hand brake pillow block. At this point the line bends down a bit and crosses over to the inboard side of the frame. Two minimum radius bends then carry the line up to the carburetor inlet. This provides about a five inch drop in the fuel line immediately upstream from the carburetor; air admitted to the line by the carburetor inlet needle valve is confined to the local area and readily bleeds back into the carburetor. This characteristic is similar to that of the Ford Service Bulletin routing. On the other hand, the routing shown in Figures 3 and 4 will cause most of the air in the system to accumulate in the fuel sediment bulb.

The fuel line being outboard of the frame member is fully shielded from the exhaust pipe heat in all areas except where the line crosses at right angle as it leaves the sediment bulb and for the radiant heat from the manifold to the carburetor area. This routing coupled with the use of a glass sediment bulb permits most of the entrapped air to be bled out of the line by the previously mentioned loosening of the glass on the sediment bulb.

The elimination of the trapped air in the fuel line, coupled with the use of a larger diameter tubing, provides a much improved fuel delivery to the carburetor and minimizes fuel starvation due to flow resistance problems. It does not correct the other fuel delivery and starvation problem the stoppage of fuel flow due to very low tank levels and/or climbing of hills.


I elected to correct this low fuel head problem by adding an electrical fuel pump to the modified fuel line system. The pump selected is a 6 volt, inline unit that is rated to deliver 4 to 6 psi. This is too high a pressure for a Model T carburetor to reliably hold off, so a regulating means is required. The method normally used with electrical pumps is to provide a return line from the carburetor inlet point, and to control the flow by a throttling orifice or a pressure regulating valve.

It was my desire to have a gravity flow fuel system in which the fuel pump use would be optional for times of need. The fuel pump selected does not have a free forward flow, so other means had to be developed to permit the gravity fuel feed with the pump turned off.

The arrangement used is shown in Figure 7. Initially the pressure was controlled only with the use of a bypass line connected closely around the fuel pump, and no return line was used. The actual pressure was set through the use of an orifice in one of the bypass line fittings.

This orifice was formed by standing the fitting on end and melting a 3/16 inch thick solder plug into it. The plug was then drilled out to the desired size and the edges of the resulting orifice were slightly beveled. With the pump running and without the return line, the following results were obtained:

The 0.125 inch orifice was found to be very satisfactory. The system could be totally purged of air by connecting a temporary plastic tube from the carburetor inlet back to the fuel tank filling port and running the fuel pump.

The 1/4 inch return line was added to assure that the system would be self purging of entrapped air.

The glass sediment bulb was mounted to the side of a 1/8 inch tee fitted to the bottom of the fuel tank, and the return line extends through the bottom of the tee with a re-entrant fitting and standpipe extending to within an inch of the inside top of the tank. The standpipe inside is about 9.5 inches of 3/16 inch copper soldered into the fitting on the bottom of the tee. This standpipe assures that fuel does not flow back through the return line when the fuel pump is turned off. All fuel delivered to the carburetor flows through the glass sediment bulb.

The combined flow resistance of this standpipe and a 0.104 inch orifice in the pump bypass line provides a 3 foot fuel head at the carburetor inlet when the pump is turned on. With the pump off the fuel flow is adequate for all normal, non-steep-hill driving.

The actual fuel pump assembly used is shown in Figure 8. It can be seen that the bypass line is very closely connected around the fuel pump. These bends, like those at the carburetor and where the line first comes over the right frame rail, are minimum radius bends. It is vital that all such bends be made using the proper tube bending equipment to avoid the creation of added flow resistance by collapsed or crimped tubing. All tubing fittings used are compression fittings.

This type of fuel line and fuel pump installation may not be attractive to the real antique car purist, but it sure gets one up those hills, even when low on fuel.

The appearance under the hood is pretty well stock, other than for the 5/16 inch fuel line. All of the plumbing and pump system is under the car along the outside of the right frame rail. One has to slide under the car to see it, but it is also very accessible for inspection and work in that location.


This type of orificed pressure regulation causes the fuel pump to run continuously. This cuts into the effective pump life, and this pump failed after about five years service. Note that the car was able to continue on a long tour because the bypass line allowed fuel to flow around the inoperative pump and on to the carburetor. It is recommended that one use a pump followed by a pressure regulator so that the pump only runs when needed. I have installed such a system with a check valved bypass line so that fuel flow can continue when the pump is shut off or inoperative. The use of the pump permits one to freely use all of the fuel in the tank, rather than have to back up a hill when the tank is only half full.