I’m at the point where I need to start doing fuel system planning. My firewall is riveted, nosewheel is next.

I heard that AM recommends NOT using a header tank, so I’m wondering what and how to route lines and return lines.

Ideas welcome :-)

Views: 3082

Reply to This

Replies to This Discussion

First, you might want to check to see if what you "heard" is accurate. If Aeromomentum really does recommend against a header tank, I would find out why. I cannot think of an engine related reason not to use a header tank. 

Hence the forum question.

If I use a header, the return line just needs to go to the header right? Seems like it might actually be easier that way. This is a part of the build where I have very little prior experience to lean on.

Speaking generically, fuel injected engines pump a lot of fuel to the pressure regulator. Only a small portion is routed to the engine. The rest is unused and must be returned to the tank(s).

If you have two wing tanks and no header tank, you must return the unused fuel to the tank that it comes out of. Otherwise, you can easily overflow a tank. This means you must have an expensive selector valve that will select left or right tank and at the same time switch the return fuel system to return the fuel to the tank that is selected. Think of it as two separate valves stacked on top of one another. With a system like this, selecting both tanks is not an option. With a header tank, you can just plumb both wing tanks to the header and then use a simple ON/OFF valve. No possibility of ever taking off on an empty tank, or unwittingly running a tank dry. All fuel is always available to the engine.

While the fuel is circulating in the engine compartment, it will pick up heat. Heat can contribute to vapor lock. This is more problematic with auto fuel than 100LL because auto fuel is more subject to vapor lock. You do not want that heat to build up. If you are returning to a small header tank, you could heat up all of the fuel in the header tank. So a header tank must be large enough to dissipate the heat that is collected.

Mark from AeroMomentum told me that he had seen issues with header tanks, were plumbing issues or operator error, like accidentally shut valves to the header tank, caused fuel starvation and therefore suggested that the potential benefits will not outweigh the risks.

Keep in mind that AeroMomentum is not selling header tanks and that they therefore have no stake in the game whether builders install header tanks or not.

We debated back and forth whether to install a header tank or not. The main reasons for us to install one would have been to be able to have a 'both' setting on the fuel selector and the additional fuel capacity.

Eventually, we however decided against it:

  • Did not want to add the additional weight very far aft.
  • Potential risk of the fuel in the header tank (particularly if it is small) getting too warm, as it circulates at a pretty high rate through the engine. 
  • Would have wanted shut off valves for each tank, to avoid unwanted cross-feeding and overflow when parked on a not level surface, but figured that we would now have to check 3 fuel valves instead of 1 before take off, with the risk of forgetting one.
  • The header tank would still somehow have to be vented, either through the wing tanks or an additional drain valve, connected to the top of the tank.
  • If installed low and forward it makes servicing the controls and autopilot servos difficult. Installing it high, automatically means that it will have to be moved even further aft, it would also be more difficult to create a sufficient support structure, particularly if the expectation is that the tank stays in place even in case of a crash.

We installed an Andair Duplex fuel selector, which was around $115 more expensive than their simple on/off selectors. This is certainly less than the cheapest header tank system. Fuel lines are 3/8" steel braided hose from the tanks and 1/4" steel braided hoses for the return. Very clean, supposedly unlimited lifetime (doesn't have to be replace every few years like rubber hoses) and a minimal number of connections inside the fuselage.

Are you aware of Mark Pennsenstadler's blog? It is in any case very worthwhile checking, he also has some great details regarding his fuel system. His system is very similar to ours, he also went without a header tank: http://www.zenithowner.com/marks_log.htm

Just to be clear: I am not opposing header tanks at all and can absolutely understand when others come to a different conclusion.

Yeah, I’ve adapted some of Mark’s ideas into my build. He does such a great job.

I’m kinda leaning towards your solution if I end up with an AM15. It’s a toss up now between this and Viking. They are awfully close to the same engine. I really enjoyed hearing Mark talk about the overhaul plan for these.

Any word on the cowling for a Cruzer yet?

I might be wrong, but believe that the Viking needs a header tank, as the engine does not have a fuel return line. Whatever air enters the fuel line to the engine (e. g. after a tank was ran dry or air was sucked in for some other reason) has only one way to go - through the injectors. The header tank is therefore needed to make sure that absolutely no air makes it to the engine.

I am not 100% sure if this is correct, but would recommend that you talk to Viking in case you decide to go with their engine but don't want to install a header tank.

Our cowling is finally in the works and is supposed to be done in the next few months. Mark has posted some pictures of the first cowling, which is made for a RV-12. The basic design will, after some adaption, also be used for the Zenith:

Additional pictures of the installion:



I just noticed you mentioned autopilot servos. Any chance you have pictures and details of what you did there?

Hello Michael,

We're not quite there yet, I think it will be another few months until we're ready to install the servos.

By the way, what I “heard” I got confirmed by Aeromomentum. They don’t recommend header tank design, but the engine will happily work with one if that’s what you have (obviously). 

This is what Mark from Aeromomentum said to me. There's advantages to both, but overall a header tank definitely adds complexity.

"If you are using a header tank you can run the return back to the header tank or to the main(s).  But keep in mind that the return flow is higher than the engine's consumption so this favors returning it to the header tank.  Having the the fuel return to the main tank(s) helps keep it cool due to the large surface area and volume so this favors returning it to the main tank.  I am not a fan of header tanks due to the added weight, complexity, cost and additional failure modes.  Duplex valves are just $200 and allow fuel return to the specific tank being used."

• Dual external fuel pumps
• External single fuel regulator
• Fuel return commonly back to primary fuel tanks
• 6 port duplex fuel selector valve
• Dual in-tank pumps
• In-tank dual fuel regulators
• Header tank based system
• No fuel valve needed
• Dual external fuel pumps
• External single fuel regulator
• Fuel return commonly back to primary fuel tanks
• 6 port duplex fuel selector valve

• UL and Rotax use external fuel pumps for their injected engines. They are a copy of what was used in cars prior to 1998.
• When 2 pumps are used together, for takeoff and landing, the amperage draw can be 10-14. This is a lot since these engines are not equipped with good alternator systems. Even with a single pump running, the draw is scary high if either the generator or external voltage regulator were to fail. The dual fuel injectors and sparkplugs cannot work without electricity.
• External pumps produce close to 2 x the required fuel pressure for the injectors. The excess pressure is regulated back down, using an engine mounted fuel regulator. Excess fuel is returned to the fuel tank.
• All this additional fuel, when returned, brings heated fuel to the tank. The total fuel circulating between the main fuel tank and the engine, then returned, is about 30-35 GPH.
• Due to the excessive fuel circulating, the amount of fuel gravity fed to the firewall mounted pumps is also 30-35 GPH. Hence the large hoses specified.
• Rotax and UL each use a single fuel regulator at the end of the fuel rail, mounted to the hot, air cooled cylinders. The Rotax brand is likely to be Bosch but the UL is a no-name copy. Each is filled with rubber parts and will fail, sooner or later.
• This is now an antiquated way to feed fuel to a fuel injected engine.

What is the big deal about a fuel regulator? Well, what is the big deal about a wing attachment bolt?
• The external fuel regulator is the single point of failure in a UL or Rotax injection system.
• An external regulator requires external fuel return hoses.
• The regulators used are automotive quality and have a definite life span.
• The engine ECU (control computer) knows nothing about fuel pressure. If the pressure is wrong, the engine will not operate correctly, no matter how many sparkplugs it has.
• A single fuel pressure regulator is not consistent with any other dual capability. The engine cannot run without it.
• ASTM was never informed of this.

• The Viking fuel system does not use a 35 gph gravity drain system, a single fuel regulator or a complex 6 port fuel selector valve.
• The Viking system is based on dual, in- tank pumps, each equipped with its own fuel regulator.
• Each pump draw less than 1.6A and has no external fuel return hoses.
• Only the amount of fuel used by the engine is gravity fed from the main fuel tanks to the header tank
• A fuel level gauge measures the exact fuel level of the header tank, providing accurate fuel gauging.

• A fuel injected engine MUST have a solid fuel pressure at all times, or the engine will stop
• Having reliable gravity feed, all the way from the wing tanks to the firewall mounted pumps, of 30-35 gallons / hour is much harder than the 2-12 GPH that the engine is actually using. When this system was used by cars, the pump was mounted right below the fuel tank.
• There is an illusion that these systems work because fuel usually come out of the pump when turned on. What is not so obvious is how small the margin is for the system to stop working.
• It is likely that even though the pump was never designed to pull fuel on the suction side, that in the airplane application it actually does this to keep up with demand. Since there is suction, any air leak prior to the pump has the potential for pump cavitation. An O-ring leak in a fuel drain, gascolator or selector valve would not be good.
• Since the fuel is at low pressure (suction) as it enters the hot engine compartment, any exposed metal fuel pump body or fuel system component could cause the fuel to boil and cavitate the pump, dropping the pressure from 43.5 psi to less than 10 psi. Once the pressure has been dropped, the pump will not be able to re-gain pressure due to the 43 psi fuel regulator at the other end of the engine fuel rail. (Here is where a small bleed bypass around the pressure regulator would be handy but this is not implemented by either Rotax or UL. The bleed would allow some fuel to flow, re-priming the pump with fresh fuel and again be able to make pressure)
• A larger amount of unusable fuel. Because more fuel is being returned to the tank, than is used by the engine, fuel is on the move in the tank and not always available at the pickup location. To guard against this, more fuel must be kept in the tank, shortening the available range. Also, as the fuel level decrease in the tank, the warm return fuel becomes more and more pronounced and pump cavitation is even more likely.
• Serious concern of un-porting the fuel pickup location. As mentioned above, one second without fuel pressure is one second the engine will not operate. Carbureted engines have a fuel bowl from which the engine can draw fuel. Not so with an injected engine. There has to be fuel available to the pump at all times. With low fuel, and in an extended descent, this is usually not the case. Some installations add more complexity to the system by the installations of door-post mounted “sumps” in order to have some protection from this. However, keep in mind that these fuel pockets only last for a few seconds when the system wants to pump 35 GPH in a circle. The pump will draw the fuel from the pocket in no time flat, bringing back the un-porting issue.

• The UL and Rotax systems usually use a duplex / 6 port fuel selector valve. These are complex, costly and have 6 fuel line connections right inside the cabin. Another style is from a pickup truck with dual fuel tanks. These are of terrible quality and are electrically operated. Cut one open and inspect the construction if considering using one. Again, there is no backup so be sure it works.
• Viking does not use a fuel selector valve. Fuel simply drains from the two wing tanks into a single header tank. From there, no selection is required. The popular C-150 also has no fuel selector.

• The Viking system consists of a header tank and two fuel pumps. That’s it. Fuel fills the header tank by gravity. If the engine uses 6 GPH, the system only has to flow 3 gallons from each wing tank. (only about 1 qt every 5 minutes)
• The pumps are inside the tank and submerged in fuel, just like every modern fuel injected car. The fuel regulators are right on the pump bodies, pre-filer screens are part of the system, etc.
• A precise fuel sender unit and gauge are available, allowing the main tanks to be used to a lower level with complete confidence. The 30 min VFR daytime reserve is in the header.

• The short answer is – yes you can. Viking has such a system and has been successfully tested on a UL engine. The long answer is much more complex and you need to understand it if you are contemplating such a conversion. Here are the details.
• First, it is important to understand how the original system works and the reason for why things are the way they are.
• Fuel pumps: The fuel pumps are big, heavy and draw a lot of current for a reason. In order to reliably produce 43 psi of pressure, additional capacity is used, and then regulated down. Because the pumps are external to the fuel, they are only cooled by the fuel running through them and by convection. If the pump slowly got hotter and hotter during operation, a vapor situation would surely occur.
• In order to use fuel for pump cooling, excess fuel needs to flow through the pump. Most systems run 3-5 times more fuel through the pump than what is used by the engine. The main reason the fuel is returned back to the fuel tank is to cool it. The fuel also cools the fuel rail and purges air from the rail when first primed for starting.
• If we eliminate the large pumps, we also eliminate some of the heat put into the fuel. Some fuel must still be returned to the header tank since these fuel rails are not designed to purge air without a return at the last injector. A small amount of return also helps cooling the fuel injectors.

• The system operates exactly as if used on a Viking engine, with one exception. There is a small amount of calculated fuel return from the fuel rail back to the header tank. The existing fuel regulator is replaced with a bleed large enough to purge air from the system but small enough for the in-tank pumps / regulators to easily maintain 43 psi of rail pressure.
Hi Jan, Good write-up! However, don’t see this covered. It seems in the case of high wing CH750 fuel is delivered to the header tank via gravity. How would fuel be delivered to the header tank in a low wing CH650?


New from Zenith:

Zenith Planes For Sale 

Classified listing for buying or selling your Zenith building or flying related stuff...

Custom Instrument Panels
for your Zenith

Custom instrument panels are now available directly from Zenith Aircraft Company exclusively for Zenith builders and owners. Pre-cut panel, Dynon and Garmin avionics, and more.

Zenith Homecoming Tee:

Zenair Floats

Flying On Your Own Wings:
A Complete Guide to Understanding Light Airplane Design, by Chris Heintz

Builder & Pilot Supplies:

Aircraft Insurance:


West Coast USA:

Pro Builder Assistance:


Transition training:

Lavion Aero

K&S Aviation Services

Aircraft Spruce & Specialty for all your building and pilot supplies!

How to videos from HomebuiltHELP.com

Developed specifically for Zenith builders (by a builder) these videos on DVD are a great help in building your own kit plane by providing practical hands-on construction information. Visit HomebuiltHelp.com for the latest DVD titles.

© 2024   Created by Zenith.Aero.   Powered by

Badges  |  Report an Issue  |  Terms of Service