What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Fuel Boost Pump Lesson Learned

Mandy

I'm New Here
Bottom Line Up Front: I've been flying with a one-way check valve incorrectly installed upstream of the electric fuel pump.

I purchased an RV-4 about 7 months ago. I'm a military pilot by background, so I had very little (~10 hours) experience in small, piston aircraft. I'm the third owner of the aircraft. I had the aircraft inspected by an A&P IA who is familiar with RVs and owned one for several years. When I purchased it, the previous owner always remarked that he hadn't quite figured out how to start it correctly. He would switch on the electric boost pump for ~5 seconds, it would whir away, and then it took about 5 seconds of cranking with the starter for the motor to fire. Since he had only owned it for about 6 months and flown it for about 15 hours, I just chalked it up to technique.

A couple of months after I had purchased it, I was still experiencing difficulty starting. It would always fire, but it would necessitate mixture at full rich and several seconds of cranking the engine. Upon closer inspection, I wasn't getting any fuel pressure turning on the boost pump when the engine was off. However, when I got the engine online, turning the fuel boost pump on would raise the PSI (normally ~21 - 22 psi) by about 4-5 psi. Several possibilities sprang to mind: 1) I had a leaky fuel sump, so the boost pump was just sucking air. 2) The boost pump was functioning properly, but maybe the fuel pressure was sampled after the engine pump and didn't include the boost pump. 3) The boost pump needed to be primed by the engine pump each start and then it would run.

Whatever the case, I figured the boost pump was definitely working with the engine running (bump in PSI) and I could stumble along using the engine pump to start until condition inspection time.

This past weekend, I was changing the oil and decided to see if I could troubleshoot the problem. I made a diagram of the fuel system and tried to piece how it worked together. My next door hangar neighbor is AX-O, so he came over and watched the fuel system while I ran the boost pump. Sounded like it was working, but still wasn't getting any pressure. I told him I was going to take off the fuel outlet hose from the boost pump to see if anything was getting through and then maybe clean (what I thought) was an inline fuel filter. He immediately corrected me and identified it as a one-way check valve. Upon closer inspection, with a mirror and flashlight, we saw the arrow on the valve was pointing the opposite direction from desired flow. We took off the check valve, installed it correctly, and now everything works perfectly.

There were a lot of lessons learned, but the biggest thing I took away was ownership of maintenance. The plane had been inspected by 2 A&Ps and a couple builders, but the problem still hadn't been identified. I seriously doubt the boost pump would have maintained fuel pressure in the event of an engine driven pump failure. It was my mistake to accept an off-nominal condition just because the airplane had been safely flying for 20 years.

I hope this helps someone out in the future who is also experiencing weird boost pump symptoms.

Thanks for the help AX-O!

Below is a link to the fuel diagram with the check valve installed backwards:

https://drive.google.com/file/d/1f_c-6XgFb6JNbccXTveeEvYHhSD0Q0YJ/view?usp=sharing
 
Fuel Check Valve installed incorrectly.

Excellent Troubleshooting & Easy Solution.
Good work guys.
 
Good job and that’s why I have my condition inspections done by different mechanics each year. A little more expensive but a different set of eyes might catch what the previous one missed.
 
I don?t even know where to start. I asked Mandy to post this as it has numerous lessons learned that should be shared. I also have some personal heaviness with this.

Mandy has mentioned to me several times that the plane gives him some trouble starting and described the symptoms he posted. Originally, I thought that there is a learning curve as he is not used to GA (bad assumption). However, he persisted with the same description (good on him). I did not think much of it because the airplane was flying fine, fuel pressure was shown when the engine war running and it was in line with expectations for a fuel-injected system (bad assumption). The aircraft has been flying for 20 plus years and it has passed numerous inspections by professional mechanics (bad assumption). It passed pre buy inspection (bad assumption).

This weekend he had the cowl off and was changing the oil. He asked me to swing by and check the FWF for anything that may not look correct. I started looking at his FWF install and notice it was very complex with fuel lines (when compared to mine). I started trying to figure out what was going on and I was not able to. I was tied up with my own project and may have not given the check the attention required. He mentioned he was going to remove the fuel pump. I asked him not to do that as the chance of breaking something or not finding a part may make things worse. I decided to look the pump up next morning. I was able to find info but the pump looked to be from the 90s and no longer produced. He also mentioned he was going to work on the FWF fuel drain.

After arrival on the second day (before I started work on my project) I stopped by his hangar. Looked at the fuel lines again and was able to get a better understanding of fuel flow. The plane was plumed to completely by pass the mechanical and electric fuel pump if failure to one or the other occurred. When Mandy mentioned that he was going to change the filter and pointed at a very short aluminum cylindrical part, things clicked in my head. I realized there was a check valve. After explaining the concept he told me that he made a fuel system diagram. Before we dove into that, I told him I thought the check valve may be backwards. I looked for an arrow on the part and did not find one. I gave Mandy a mirror and asked him to find an arrow (it was hiding behind other lines, engine mount tubes, and was tiny). In the mean time I second guessed myself and went on the internet to find if the arrow on a check valve indicated the direction of flow, It does. He came back with the direction of the arrow and the diagram. Sure enough, the check valve was backwards.

He swapped the check valve direction and we did a quick check with just the electric fuel pump to on. He immediately got fuel pressure and 2 seconds later had fuel coming out of the fuel servo (flooded the motor). We could hear the fuel going into the cylinders. Success!!!!!!
The system was basically plumed to have a loop around the electrical fuel pump in error. If the mechanical fuel pump failed, fuel would have not made it to the motor by using the electric fuel pump.

We also took the chance and changed the fuel drain valve. In the process I also removed the bowl and found debris as well as a disintegrated rubber gasket.

In the process of trouble shooting, I made some bad assumptions:
-Aircraft was flying. so aircraft and maint history was not questioned.
-I made assumptions on Mandy?s system knowledge even though he is a smart dude.

Learned some lessons:
-I was not focused on helping him as I was tied up with my own projects.
-Time away from a problem is good. Especially with rested eyes.
-Bouncing ideas between peers is a good thing as it may stimulate different thoughts or troubleshooting techniques.
-Post your findings as they may help the next person, Mandy came to VAF researching ideas.

This post is not about Mandy?s or my capabilities. It is also not about pre-buy inspections or conditional inspections. It is about making bad assumptions and learning from those. Those other subjects are for a different discussion.
 
So I'm curious...that design is clearly different than what virtually everyone else is doing, so does it actually solve a problem (rather than creating new ones)?

Assume the check valve is correctly installed. If the engine-driven pump fails, would the electric pump solve the problem, depending on the failure mode?

And how was the electric pump increasing pressure when used while the engine was running? Back pressure from the electric pump to the engine-driven pump?

Enquiring minds want to know :)
 
I think you may want to consider re-plumbing the two pumps in series. With the check valve in backwards, and the electric pump off, there was a huge leak flowing backwards thru the electric pump, back to the input to the mechanical pump. Turning on the electric pump simply stopped that leak path, resulting in higher net pressure. I'm surprised you got as much pressure as you did, electric pump off. The current set up (valve the right way) seems to have the same potential fault. Most mechanical pumps have a "check valve" (some times just a flap) to keep fuel from going backwards. If the mechanical pump's check valve fails, running the electric pump may see the fuel flowing backwards thru the mechanical pump, to the electric input, around in a circle, not going to the engine. If you don't want to re-plumb, at least install a second check valve in the mechanical pump circuit, so it would take a double failure to short-circuit the gas flow from the electric pump.
Edit: see post above, too.
 
Last edited:
This is not my plane. If mandy wants to change things around we can certainly do that.

This may be a result of the plane being built a while back and technology changing. Or it may be that a certified aircraft was built in the same manner and it was copied. not sure.

I need to sit down with the diagram again and do a failure analysis. The part that is weird to me is the "T" after the mechanical pump.

If the mechanical pump diaphragm fails, would an electric pump ON just make the fuel coming though the fuel system shoot out of the mechanical fuel pump "failure port"? and not make it to the fuel servo? Maybe the reason for bypassing the mechanical pump.
 
Gents,

Thanks for all the thoughts. Unfortunately, the original builder is deceased. I'd love to spend a couple hours talking to him. He was an A&P IA, which makes me think this is an existing system design on a certificated aircraft somewhere.

I'll break out the documentation I have on the aircraft - I'm pretty sure I remember seeing a receipt for the purchase of a mechanical fuel pump. I'll look at the manual and see what its failure modes are. Adding a check valve after the mechanical fuel pump sounds like a great idea.

The only possibility I can think for the current routing is if the mechanical pump could "seize" or get clogged. Then it would make sense for the electric pump to be able to completely bypass the mechanical one.

I think Bob is right on with how the electric pump "increased" fuel pressure. It simply stopped the leak path. Some data I omitted in my original post which corroborates this theory: I have higher net pressure now with just the engine pump running than I did when the check valve was upside down. ~22 psi before as opposed to ~27-28 psi after.

Thanks again for all the input!
 
Gents,

Thanks for all the thoughts. Unfortunately, the original builder is deceased. I'd love to spend a couple hours talking to him. He was an A&P IA, which makes me think this is an existing system design on a certificated aircraft somewhere.

I'll break out the documentation I have on the aircraft - I'm pretty sure I remember seeing a receipt for the purchase of a mechanical fuel pump. I'll look at the manual and see what its failure modes are. Adding a check valve after the mechanical fuel pump sounds like a great idea.

The only possibility I can think for the current routing is if the mechanical pump could "seize" or get clogged. Then it would make sense for the electric pump to be able to completely bypass the mechanical one.

I think Bob is right on with how the electric pump "increased" fuel pressure. It simply stopped the leak path. Some data I omitted in my original post which corroborates this theory: I have higher net pressure now with just the engine pump running than I did when the check valve was upside down. ~22 psi before as opposed to ~27-28 psi after.

Thanks again for all the input!

Assuming the mechanical pump is the normal tempest that is normally on these engines, if the pump fails, it will free flow fuel through it. If it seized, you'd have the arm of the pump broken off into the accessory case, and I've never heard of that happening provided the pump is installed correctly.

I had a Comanche with the paralleling mechanical and electric pump and check valves like you describe.
 
Why not just plumb it up the way everybody else does, in series? What problem is this (more complicated) system trying to solve?

I think this is a great example of how adding supposed "safety" features into a design can make it *less* safe. Here, the system now has an additional T fitting and a check valve, and guess what? The additional check valve was itself the cause of a "failure" due to improper installation.

Seems to me the best course of action would be to get rid of the T, and the valve, and put the boost pump in series ahead of the mechanical pump like everybody else.

Adding a check valve after the mechanical fuel pump sounds like a great idea.

No, it doesn't. To me, it sounds like yet *another* complication to the system that can fail or be installed incorrectly, and isn't necessary with the simpler, in-series plumbing.

I seem to recall that a pretty high number of engine failure-related accidents are related to fuel systems...why make it more complicated than it needs to be?
 
Fuel plumbing

WARNING This is a serious building Error!

The problem is that the boost pump should be in series with the engine driven pump. I highly recommend you re-plumb this aircraft correctly. The boost pump will push fuel through the engine driven pump before starting, and will give you a pressure bump when turned on with the engine running.

In cruise flight the engine driven pump will pull fuel through the boost pump. In the event of engine driven pump failure, the boost pump will supply fuel to the engine pushing it through the engine driven pump.

The primary reason EAB aircraft crash early in their testing is fuel related problems like this one. The fact that this aircraft has flown successfully for years could be dumb luck! Please fix this!
 
Last edited:
Back
Top