What's new
Van's Air Force

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

Fuel System Design Review

R

rickmellor

Well Known Member
Hey guys ... I've finally got the wings attached to the fuselage so it's time to start routing all the fuel lines. Before I commit to my planned system I figured I'd post it up here to see if anyone notices something blatantly wrong. I believe the configuration is a pretty standard Vans/Airflow Performance fuel injected setup with just a couple minor tweaks. The details are:

RV-7A, Superior IO-360
Andair duplex fuel selector
Van's standard FI kit (AFP parts)
Dynon fuel flow meter (floscan)
Dynon fuel pressure sender
AFP FI engine components

I'm thinking it'll be good divert from Van's plan for a tee fitting for the pressure relief valve returning to the fuel supply line, and from AFP's callout for the purge line to return to the tank vent line (or supply line) and just run them both all the way back to the currently selected tank. This should give me the ability to completely circulate fuel through the system with no risk of returning vapors to the supply lines.

What do you guys think?

-Rick

FuelSystemR1.png
 
Imho, put the flow meter in the metered fuel line. That is, inline in the hose between your fuel servo and the flow divider (or purge valve in your case).

There are numerous advantages to having it there...not the least of which is that you will be less prone to vapor lock. The fewer things between the fuel supply and the engine driven fuel pump, the less opportunity for pressure drop on the suction side of the engine driven fuel pump.

Just nit picking the diagram, but the fuel pressure sensor is not inline with the flow, it's kind of an "offshoot." Again, just being nit pickety. Great diagram!
 
Good points!

Dan, thanks for the feedback! Did you end up moving your floscan up to engine compartment? I was planning to copy your installation location near the fuel pumps, but I was going to see if I could find a way to avoid any blisters on the cover.

With the floscan up front, I suppose I'd want to mount the floscan on the firewall and run hoses to/from it as appropriate. It could be a bit of a pain getting the 6" straight hose length prior to entering the meter. I'll think on it a bit more ... just seems a bit more complicated.

Btw, were you at Hayward Exec the weekend before last? There was a 7 blasting out of there that looked just like yours.

-Rick
 
Fuel flow to tank...

If the flow meter is located as the diagram shows, don't you need a "negative" flow meter to subtract the fuel quantity that is returned to the tanks?

I presume the flow meter is also a fuel quantity gauge...

gil in Tucson
 
az_gila said:
If the flow meter is located as the diagram shows, don't you need a "negative" flow meter to subtract the fuel quantity that is returned to the tanks?
Click to expand...
No. The purge valve is the only mechanism returning fuel, and it's only operating for 30 sec at a time. At 3.8 gph flow, you're only talking about 0.03 gallons flowing during the purge. Even if you purge 10 times between top-offs, you're still only talking about 0.3 gallons. And on average, realistically, how many purges are you really likely to do between top-offs?

Imho it's SO NOT WORTH BOTHERING to calculate the returned fuel in with your "fuel consumed" totalizer! Considerable complexity for negligible benefit.

Rick,

I didn't move my fuel flow transducer, because it's "working fine" for me where I have it. I have the K factor dialed in and it's accurate to well within half a gallon on 30+ gal top-offs. Good enough for me.

As far as vapor lock potential is concerned, it's something I'm aware of given where it's installed, but I haven't had any issues. I'm not trying to endorse the way I did it...just saying I'm comfortable with it there...knowing full well that it would be BETTER off installed in the metered fuel line. My next airplane will of course have fuel injection as well, and that airplane will definitely have the fuel flow transducer installed inline with the metered fuel hose.

I'm basically saying "do as I say, not as I do." :rolleyes: I would have done it differently with the knowledge that I have today (I didn't know diddly when I built my plane).
 
I'm wondering how folks install the FF transducer in the metered line. The mounting instructions from FloScan are sparse but the one thing they DO mention is mounting it with the wires pointing up. The reason is to allow any air in the system to escape the transducer. If the transducer gets air in it, the readings are affected dramatically because the FF transducer is an optical device.

I can't imagine how one could mount the floscan transducer in the metered line and have the wires pointing up. I would love to see some pictures.
 
I did.

Mounted between the rear baffle and the distributor. I have only test ran the engine, but it is working. This is where JPI recommended and an AP. Lots of cooling air in the plenum too.
 
Fuel system schematic comment

Since you are running dedicated return lines to each tank and the relief valve from the boost pump is plumbed to these returns through the duplex selector valve, make sure the returns in the tank are BELOW THE LEVEL OF FUEL in the tanks when that tank is selected. The reason: the relief valve in the boost pump has a #76 bleed hole in it to dump the fuel pressure when the pump is shut off. If the return line is exposed to air when the boost pump is not running there is a possibility of sucking air into the feed line. This is not a good thing.

That?s why we keep it simple and just return the relief valve flow to the inlet side of the pump. No way to get air in the system as long as there is fuel in the feed line.

Don
 
Thanks for this important info

Thanks Don for this important piece of information!
Too bad that wasn't documented anywhere...

in this case, we will have to replumb our setup. grrml :(
pumpassembly.jpg

as it feeds the pressure relief valve directly into the return line as in Rick's setup... And the return lines are put in high in the tanks, as probably with most return line setups...

Also, we plan on installing a check valve in the return line where it penetrates back through the firewall. That way there is no chance for any reverse flow from the return line.

regards, bernie

P.S. Any other comments/constructive criticism regarding our setup? Anyone?

The mount (with the intended position for the return check valve (not yet installed):
P1010660.JPG
 
Last edited:
Fuel System Comment

One more thing I just thought of is if you have the selector in the off position and turn the boost pump on you will put 125 PSI on the fuel system, another not good situation. This is because your relief valve is dead headed by the selector valve. I assume that in the off position the return ports are also shut off.

Its hard to anticipate everything you guys can do here. Our manual is pretty complete with approved methods on how to install our system that we know works, but that doesn't mean that that's the only way to do it. It's best to contact us for customer support before straying from the manual or getting "expert opinions". This can save you a lot of extra work and $.


Don
 
hi don,

appreciate your input!
this was in no way meant to be a rant...

another interesting point about the return line blockage / over pressure situation.
the fuel selector is a duplex valve, so any time the return line is blocked, there is no supply either...

of course it is not intended, but how do you expect the pump to behave if it is running and the fuel selector is switched off? i imagine this being a pretty unhealthy situation? the problem lying more on the suction side than on the overpressure side (as there is sufficient tubing for the fluid/pressure to spread).

regards, bernie
www.flyvans.com
 
I love this forum!

Don, thank you very much for the feedback. I'm thinking that in this case I can have my cake and eat it too (when will I learn?! :rolleyes: )...

If I revert to the original plans for having the relief valve return to the inlet then the air bleed-back issue and the over-pressure situations will go away (yes, the Andair will block both ports in the OFF position). It looks like I can then keep my purge line running back to the tanks via the returns and still get the complete fuel system purge capability that I want.

For the returns going high into the tank ... I was planning to put them low in the tank to get a smoother diffusion of fuel when it leaves the return tube. I don't know if it happens but I've seen comments that if the return fuel spills into a partially filled tank from the top that there can be foaming. Again, not likely that I'll have this problem with the AFP setup just purging back to the tanks but it's still something to consider.

-Rick
 
I have never done this but common sense says with the duplex valve set up with the valve in the off position on initial start up of the boost pump the fuel pressure would go up to 125 PSI (the internal relief on the pump) this might pop the breaker as the pump will pull in excess of 10 amps. But until the breaker shuts the pump off the pump will make pressure until the fuel is expelled from the suction side of the pump (it can pull about 1.5 PSI negative head), then the pump will run dry and burn itself up as fuel is what cools the motor. So the amount of time it runs depends on how long it takes to get rid of the fuel on the suction side of the pump. With the throttle at idle the flow rate is less than 2 GPH, so it could take a minute or so. With the pump relief returned to the inlet side of the pump there is more fuel in the loop, so running with the selector valve off won?t cause excess pressure and there is more time the pump can be run before the fuel runs out. If there was no fuel used then the pump could run quite a long time just re-circulating the fuel back to the inlet of the pump. Eventually the pump loop will heat up to boil the fuel on the inlet. We have done this and it takes 30 minutes of running or more before the pump gets too hot. In actual use there is always some new fuel being passed through the pump as the engine is running. Most situations don?t make you idle the engine for an hour, so the pump loop heating is not an issue.


Don
 
That Picture Is Scary

That photo is enough to cause me to stop thinking of ever putting on FI.
 
Does AFP FI really need a return?

Don,

It was my understanding that your system did NOT need a return like other systems. If this is not true, could you please explain?

Thank you
 
Why a duplex valve??

az_gila said:
If the flow meter is located as the diagram shows, don't you need a "negative" flow meter to subtract the fuel quantity that is returned to the tanks?

I presume the flow meter is also a fuel quantity gauge...
Click to expand...
dan said:
No. The purge valve is the only mechanism returning fuel, and it's only operating for 30 sec at a time. At 3.8 gph flow, you're only talking about 0.03 gallons flowing during the purge. Even if you purge 10 times between top-offs, you're still only talking about 0.3 gallons. And on average, realistically, how many purges are you really likely to do between top-offs?

Imho it's SO NOT WORTH BOTHERING to calculate the returned fuel in with your "fuel consumed" totalizer! Considerable complexity for negligible benefit.

....
Click to expand...

OK Dan ... that makes sense... but if we are only talking about 0.03 gallons a purge, why bother with the duplex valve?

It's a lot more cost and a lot of extra plumbing/lines.

Always start on the left tank when the tanks are full, and just return the purge line to the left tank.

Any other time, no big deal that the purge return goes to the left tank.

This alone would save a few $100 and reduce the plumbing needed.... Less complex is usually better.... :)

gil in Tucson
 
Last edited:
az_gila said:
OK Dan ... that makes sense... but if we are only talking about 0.03 gallons a purge, why bother with the duplex valve?

It's a lot more cost and a lot of extra plumbing/lines.

Always start on the left tank when the tanks are full, and just return the purge line to the left tank.

Any other time, no big deal that the purge return goes to the left tank.

This alone would save a few $100 and reduce the plumbing needed.... Less complex is usually better.... :)

gil in Tucson
Click to expand...
As one interested in the fuel injection systems I am very interested to hear responses to this question. As dbuds2 said all of that complicated plumbing sounds a little much for my taste. If I can simplify the FI installation I am very interested. I will anxiously look to read any responses to Gil's question on this.
 
Hi,

our thinking really goes this way:

avgas is gonna go sooner than later...
even if the return line might only be used for purging at this point in time (current plans are for a tmx-iof-360), constant return systems will be required to run automotive gasoline. and the various fadec's and related systems coming to market all have their individual requirements as well.
how diesel/jet-a systems will look like is not yet quite clear, but chances are that they will need some form of return line as well. in any case, expect to see changes firewall forward within the next 10 years.
also, avgas is way more expensive over here in europe, so motivation to move to alternative fuels is even bigger when the time comes...

with our setup, we're pretty much ready for whatever may come down the road. at least no hassle in removing the wings, installing a return line etc...
it's a bit of extra cost (fuel selector $$$, very little added weight), but peanuts compared to the overall tab :-( the AFP FI setup we need anyway for the fuel injected engine.)

regards, bernie
www.flyvans.com
 
Bernie,
Can you give more details on what all of that weighs? Also you posted a picture of the fuselage but I am not quite clear where your fuel systems will be mounted. Can you provide some more details on where exactly all of that will be mounted in the fuselage? Is the Andair fuel selector valve going to be in the same location as the standard Van's selector valve location? I would appreciate any additional information you can provide as I am looking to do the FI and am concerned about some of the same issues with running various fuel types you mentioned above.

Thanks,
 
steve,

sorry, we did not weigh the components...
our pump package looks a lot more complex than it actually is.
we riveted two pieces of angle to the floor longerons. on these pieces are 4 nutplates each. now the fuel pump assembly plate is laid onto these angles and attached to the angles using screws.
we also took great care to attach the fuel filter adel clamps using an-3 nutplates and run the various fuel lines for easy serviceability later on.
compared to the standard vans/afp mount, we added some aluminum tubing for the return (really lightweight), the dual selector is probably twice as heavy as the standard andair selector, which is again heavier than the default van's el cheapo part. but even that isn't that bad, as a large part of the body is made from composites/plastic.
The piano hinge by the way is intended for the mounting of a cover.
And this is a picture from an earlier stage of construction:
P1010044.JPG

but it shows the approximate location nicely. The package is even a bit closer to the spar in the final location.

the selector sits low on the floor, about level with the tank supply lines, which is another feature positively affecting the head pressure to the fuel pump/probability for vapour lock.
since we plan to build a center console between the flap channel, the point where van's default fuel selector is located and the panel, we will integrate the fuel selector knob with the extension, all from andair. www.andair.co.uk

regards, bernie
 
Last edited:
flyvans.com said:
Hi,

our thinking really goes this way:

avgas is gonna go sooner than later...
even if the return line might only be used for purging at this point in time (current plans are for a tmx-iof-360), constant return systems will be required to run automotive gasoline. SNIP[/url]
Click to expand...

There is an ongoing confusion between what the Airflow Performance purge valve line's and what classical return lines' in some FI systems functions are. Their functions are unrelated.

This thread started with a question about a method to plumb an AFP purge valve. All that tubing and a duplex valve are not needed to get the benefit of a purge valve. My advice is to plumb it in exactly as AFP has outlined in their manual for RV's. As has been pointed out in earlier posts in this thread, hazards to creative fuel system design have been identified.

Please, this is not a flame, but fuel system in-flight design testing is not something to take lightly.

Regarding where to put the flow sensor, I have had no issues with mine between the AFP fuel servo and the flow divider. It is mounted on the firewall, horizontally with the wires pointing up.
 
AlexPeterson said:
There is an ongoing confusion between what the Airflow Performance purge valve line's and what classical return lines' in some FI systems functions are. Their functions are unrelated.
Click to expand...

Unrelated functions but compatible requirements. The purge line must NOT be returned to the fuel inlet. AFP recommends either returning it to the tank or dumping it into the fuel tank vent line. The purge valve allows you to remove hot fuel and vapors from the fuel system so a full return of some sort is required.

AFP's default configuration essentially gives you a duplex selector by requiring you to select a specific tank (left) while you're purging so that the purge output can discharge to the opposite tank. With a duplex selector you don't have to worry about that. Fewer checklist items are good.

For me the deal maker with the duplex selector is that I already have the Andair. I was originally planning an ECI fuel system, which uses a return, but have since settled on the AFP setup. Since I've already spent the money why not take advantage? I understand that there are simpler/cheaper ways to get the same result but this one works for me. Also, a nice bonus that the Andair + fuel tank return gives you over the vent line returns is that you don't risk dumping gas (small amount as it may be) out on the ground.

As for the pressure relief line running to the Andair ... Don pointed out that this is not such a good idea. You won't run the pump with the selector set to off in any case other than a mistake, but why risk the torture on a critical component? Much better to stick with the plans on that bit.
 
I've modified my original diagram to show the flow sensor between the fuel servo and the purge valve. I've also corrected the relief valve output to return to the pump inlet. I'm posting the updated diagram for posterity.

FuelSystemR2.png


AlexPeterson said:
Regarding where to put the flow sensor, I have had no issues with mine between the AFP fuel servo and the flow divider. It is mounted on the firewall, horizontally with the wires pointing up.
Click to expand...

Alex or others who have mounted the flow sensor in front of the firewall, would you guys mind posting a picture of your setup? I could use some ideas to get the wheels turning.

-Rick
 
A new requirement

flyvans.com said:
......
with our setup, we're pretty much ready for whatever may come down the road. at least no hassle in removing the wings, installing a return line etc...
it's a bit of extra cost (fuel selector $$$, very little added weight), but peanuts compared to the overall tab :-( the AFP FI setup we need anyway for the fuel injected engine.)

regards, bernie
www.flyvans.com
Click to expand...

Ahh... a different design requirement not mentioned before... :)

That makes more sense now....

gil in Tucson
 
@AlexPeterson

we're fully aware of the "creative fuel system inflight testing" risks...
i would just like to remind you that, if you mentally get rid of all the return system parts, you end up exactly as per AFP on the supply side of things! So nothing creative there. The one thing that needs change/was overlooked/not fully taken into account, has been brought up by don from AFP in this thread. We will replumb the pressure relief right back to the pump inlet instead of the return line.
Aren't these forums simply awesome?

regards, bernie
 
Bernie,
Thanks for the posts. Please keep them coming. I would like to ask anyone who has pictures of your installs to please post them if at all possible. Many of us out here are interested in the fuel flow issues associated with FI as we are planning for our own systems. This thread has been very helpful.
 
flyvans.com said:
Hi,
...

avgas is gonna go sooner than later...
even if the return line might only be used for purging at this point in time (current plans are for a tmx-iof-360), constant return systems will be required to run automotive gasoline. and the various fadec's and related systems coming to market all have their individual requirements as well.
how diesel/jet-a systems will look like is not yet quite clear, but chances are that they will need some form of return line as well. in any case, expect to see changes firewall forward within the next 10 years.
also, avgas is way more expensive over here in europe, so motivation to move to alternative fuels is even bigger when the time comes...

...
Click to expand...
I am planning for FI and auto fuel. Why would I need a constant return system? Vapour lock?
 
Question about the AFP pump kit

I bought my pump kit 2nd hand, and the only doc it came with was a drawing by Van's. No AFP installation manual (I'll buy one from AFP soon.) So please forgive the dumb questions.

By staring at the intestine-like snarl of tubing I got, I figured out which is the relief valve and which is the bypass valve. So:

1) Is the purpose of the bypass valve to allow fuel to flow to the mechanical pump when the electric pump is off?

2) Is the purpose of the relief valve to set the fuel pressure to some fairly high value (i.e. a bit more than the FI servo wants?) What is that value? I assume the valve cracks above that pressure, and the excess fuel goes back to the T in the inlet line?

3) What's the flow rate in that relief line when the engine is drawing small amounts of fuel?

Thanks,
Martin
 
rickmellor said:
I've modified my original diagram to show the flow sensor between the fuel servo and the purge valve. I've also corrected the relief valve output to return to the pump inlet. I'm posting the updated diagram for posterity.

FuelSystemR2.png




Alex or others who have mounted the flow sensor in front of the firewall, would you guys mind posting a picture of your setup? I could use some ideas to get the wheels turning.

-Rick
Click to expand...

Rick, how did this design end up working for you? I am trying to design mine right now. Do you have any pictures of the installation?

- Matt
 
Not sure...

idleup said:
Rick, how did this design end up working for you? I am trying to design mine right now. Do you have any pictures of the installation?

- Matt
Click to expand...
Sorry for the late response ... hopefully this is helpful.

It was hard. :) There is very little room in there to run six lines around the Andair. Plenty of guys have done it but I found it to be much harder than I'd expected. I ended up selling the plane (life changing event) shortly after attempting this work and wasn't able to complete it. I think I was on to the proper solution at the time but wasn't able to prove it out.

You might check into what the Eggenfellner folks have done since this setup is similar to what they use. You're sure to find some empirical data there.

-Rick
 
idleup said:
Rick, how did this design end up working for you? I am trying to design mine right now. Do you have any pictures of the installation?

- Matt
Click to expand...
Not sure...
Quote:
Originally Posted by idleup View Post
Rick, how did this design end up working for you? I am trying to design mine right now. Do you have any pictures of the installation?

- Matt
Sorry for the late response ... hopefully this is helpful.

It was hard. There is very little room in there to run six lines around the Andair. Plenty of guys have done it but I found it to be much harder than I'd expected. I ended up selling the plane (life changing event) shortly after attempting this work and wasn't able to complete it. I think I was on to the proper solution at the time but wasn't able to prove it out.

You might check into what the Eggenfellner folks have done since this setup is similar to what they use. You're sure to find some empirical data there.

-Rick

Plumbing the six hole Andair valve is a bear but lots of guys have done it. I had to buy about 10' more fuel line before I got it done.

I am using the Andair with AFP and Lycoming after the Subby experience. Basically, it is the same plan as what is published here except the pressure relief by-pass returns to the selected tank rather than the inlet line and I am using the same 2 pumps as with Subby. They are Airtex E8228's, very similar to the AFP pump, and are wired to use one or the other.

It had not occurred to me until reading this thread that if a pump is turned on with the Andair valve close, pressure relief by-pass fuel has no where to go and the system will be subject to 125 psi quickly. That is a problem but on the up side, it sure will check the integrity of the plumbing in a flash. The solution to the issue is do not turn on the pump with the valve closed. All Subaru installations are plumbed that way.

I've had my engine running 3 times and all is OK except no oil temperature indication which should be resolved as soon as I can get a replacement sensor from GRT. The AFP system and purge valve are on a learning curve with me and Don has been hammering me with his procedure and it is beginning to sink in. The only unresolved issue is fuel pressure. I have 21 psi on the engine pump and 29 with the electric unit. This is OK at high power settings where mixture can be manually adjusted but at idle the 29 psi floods the engine to almost quitting. At 21 psi it is smooth. A possible solution is to rework the pressure relief valve to lessen the spring tension and reduce by-pass pressure, or adjust idle mixture to more lean. There may be a happy mixture medium between 21 and 29 psi. I will be working on that as soon as I have oil temperature squared away.

Back to the gist of this thread, if you are in the building stage by all means go with the Andair and get the return lines installed. It is not that big a deal and the system will be ready for whatever is coming down the pike with regard to fuel and fuel systems.
 
Last edited:
Fuel Filter

You might consider a fuel filter in the line from each tank before any compnents of your input flow and eliminate one single point failure location.

Bob Axsom
 
I agree with Bob. I was planning to make this post but he beat me to it. A couple of thoughts about this:

1. The filter is there to protect the AFP pump mostly, although the rest of the fuel system benefits as well.

2. Certified aircraft have strainers ( coarser than a filter typically) and pumps that are not as finicky as to dirt. These strainers are also typically "drained" at each preflight, if there was a bunch of dirt, it would mostly be expelled during draining. These filters are closed. What they catch stays in there until inspection time. Avgas is generally filtered at many steps in the distribution supply chain, and is normally exceptionally clean. But you could get a bad batch, the bottom of a tank maybe, who knows?

3. The way this system is plumbed, the fuel pump sucks fuel through all the lines and the filter. If the filter begins to get clogged, the pump suction will increase to a point. It may get to the point that enough fuel for full throttle operation is compromised. Before that happens though, on a hot day you may generate fuel vapor. The fuel pump can not pump vapor. This is called vapor lock. If these filters had a bypass valve this would not happen, but then unless you knew the bypass was open you could be damaging the pump. Think Chip Detector lights on helicopter gearboxes. Better to have chips and oil pressure for a short while than to shut it all down with a clogged filter without bypass.

4. If you happen to get a bad load of fuel, fly for a while, and gradually the filter builds up dirt and hence differential pressure, what are you going to do? You will not know anything is wrong until you start to experience power loss. You will be looking for a landing spot.

I am building a Harmon Rocket, and have incorporated 2 of these filters in my system. One on each feed to the selector valve. If I have a dirt problem, at least I can switch tanks and also get a new filter as well.

These filters should be opened and inspected and cleaned if necessary at each annual. I also installed little ball valves to isolate the tank from the filter, otherwise you would have to drain the fuel tank for each inspection. I safety wire these valves open between inspections. They are located beside the front control stick base, under a cover.

Here is a link to the valves, they are made by Summit racing, and have -6 AN connections. Work great.

http://store.summitracing.com/partd...&N=700+4294733687+4294923409+115&autoview=sku
 
Bob & Pvans,
Do you have a suggestion for what filter to use? Do you have a particular type (brand) you use? Where do you get them? I have used the cheap paper filters with the clear plastic bowl on other fuel line setups but do not particularly like them for the RV.

I am also going with the design listed above for my fuel system. I can attest to the spaghetti mess of tubing that it requires. If the Airflow Performance filter is a concern for clogging, why would adding additional filters upstream decrease this risk? I understand about having one in line from each tank and the idea that switching tanks would also switch filters but are these in addition to the existing filter in front of the fuel pump? I am not that interested in removing the existing supplied filter in front of the fuel pump. I also am a little concerned about adding additional filters in line upstream. This fuel system setup I am working on has already gone way beyond the KISS method of installation. I am not sure I am ready to increase the complexity.
 
Last edited:
Hi Steve,

I used two of the Airflow performance filters, one in each feed line from the tanks to the selector valve. There is no additional filter between the selector valve and the Airflow Performance electric fuel pump assembly.

Here is a link:
http://www.airflowperformance.com/html/cat_17.html


KISS is a great design goal, and I agree this has gotten way out of hand as compared to say, a Cessna 150 ( no pump, simple on-off valve, one little strainer and a carburetor) , but then who wants to have Cessna 150 performance? Part of the price we pay.
 
Bob Axsom said:
You might consider a fuel filter in the line from each tank before any compnents of your input flow and eliminate one single point failure location.

Bob Axsom
Click to expand...

Or maybe you'd be adding 2 failure points.

The problem with filters on the suction side is it is a point of pressure drop and vapor lock even if there is no dirt in them. Granted, if the fuel is so dirty it jams up a filter and causes a major pressure drop the entire operation is in serious trouble.

I have the AFP filter installed as per instructions but I would prefer it down stream on the pressure side out of vapor lock considerations, especially with mogas. There are varying opinions on this subject when considering auto world technology. In that world, filters are on the high pressure side of the pump. But with AFP, Don invented the system so I will go along with his point of view.
 
pvans said:
Hi Steve,

I used two of the Airflow performance filters, one in each feed line from the tanks to the selector valve. There is no additional filter between the selector valve and the Airflow Performance electric fuel pump assembly.

Here is a link:
http://www.airflowperformance.com/html/cat_17.html


KISS is a great design goal, and I agree this has gotten way out of hand as compared to say, a Cessna 150 ( no pump, simple on-off valve, one little strainer and a carburetor) , but then who wants to have Cessna 150 performance? Part of the price we pay.
Click to expand...
OK, I gotcha now on the filter setup. You are just adding one filter in front of the valve selector for each tank instead of one singular filter after the valve and in front of the pump.

In terms of comparisons to a Cessna 150, in my mind I am not necessarily trying to emulate the Cessna, Piper, etc. design as much as I would like to copy the Toyota, Honda, Ford, GM, etc. design of fuel injection systems. I know those systems do not have spaghetti tubing running here and there out of a fuel pump. Why are those designs not viable options for us in the aviation world?

Honestly, I really do find it funny that so many aviation guys are so snobbish when it comes to automotive design. For some reason they seem to think aviation design is far superior to anything the automotive world can dream up. The way I see it though, airplane design (at least in terms of what we are talking about here) has an awful lot to learn from those automotive engineer types.
 
David,
Remember, the whole point of these filters is to protect the AFP pump, which has very small tolerances.

Steve, the automotive designs are good, but they now mostly all have pumps installed inside the fuel tanks where they are always flooded and thus have no suction problems. Tank shape becomes an issue for this. And if the pump fails the engine won't run. Airplanes try to have redundant/failsafe modes so as to preclude this.
 
pvans said:
David,
Remember, the whole point of these filters is to protect the AFP pump, which has very small tolerances.

Steve, the automotive designs are good, but they now mostly all have pumps installed inside the fuel tanks where they are always flooded and thus have no suction problems. Tank shape becomes an issue for this. And if the pump fails the engine won't run. Airplanes try to have redundant/failsafe modes so as to preclude this.
Click to expand...

You are absolutely correct about protecting the pump. The auto installs have a fine screen in the tank to keep junk out of the pump and also a filter down stream to keep the nozzles clean. Recommended filter change is at 30,000 miles, that may equate to 5 years for us. :) I have always had a gascolator or in line filter and quite frankly have never found much stuff in them during the condition inspection. I pour fuel through a Mr. Funnel, that screen is so fine it will separate water from fuel.

We do have good redundancy with a mechanical pump or with 2 electric pumps. (or with my arrangement - 2 electric pumps and the mechanical pump - Don at AFP says that is over kill. He is right. I have them because they were in and wired to start with.)
 
Last edited:
pvans said:
Steve, the automotive designs are good, but they now mostly all have pumps installed inside the fuel tanks where they are always flooded and thus have no suction problems. Tank shape becomes an issue for this. And if the pump fails the engine won't run. Airplanes try to have redundant/failsafe modes so as to preclude this.
Click to expand...
I would like to understand the ideas behind why we cannot have the same fuel pumps inside the fuel tanks in our aircraft. What precludes us from using this design if it will eliminate the vapor lock issue.
 
Low wing aircraft, uncoordinated banked turns with low fuel level = uncovered pump pickup possibilities. Same thing happens to many EFI cars on the race track with fuel levels below 1/4 tank and high lateral G. One reason why in-tank pumps may not be the best idea with fuel injection.

Header tanks ensure a constant fuel supply and are well proven.

Make sure you test fuel feed at altitude during Phase 1 to ensure it works at high bank angles, low fuel levels and with improper rudder use. You don't want to find out the engine quits under these conditions as you turn final at low altitude. This has hurt more than a few people unfortunately.
 
rv6ejguy said:
Low wing aircraft, uncoordinated banked turns with low fuel level = uncovered pump pickup possibilities. Same thing happens to many EFI cars on the race track with fuel levels below 1/4 tank and high lateral G. One reason why in-tank pumps may not be the best idea with fuel injection.

Header tanks ensure a constant fuel supply and are well proven.
Click to expand...
Would it be reasonable to assume then that having a header tank with an in-tank pump could reduce the risk of an uncovered pump? If so, it seems this would be a simple solution to using a submersed fuel pump and therefore eliminating the "suction" pump issues and all the complexities associated with suction type pumps. Am I still off base with this thought?
 
ECI fuel return?

RVbySDI said:
Would it be reasonable to assume then that having a header tank with an in-tank pump could reduce the risk of an uncovered pump? If so, it seems this would be a simple solution to using a submersed fuel pump and therefore eliminating the "suction" pump issues and all the complexities associated with suction type pumps. Am I still off base with this thought?
Click to expand...

Sounds right to me. My qustion is, what difference does it make where the pump is actually located (when using fuel injection)? Isn't air going to be sucked into the intake tube and make it's way to the pump whether the pump is close by or further down downstream having the same effect?

With the Van's fuel injection setup, how does the pump assembly deal with the air?

The newer ECI fuel injection system has a return to the tank. Does this handle any air bubbles better?

Bevan
 
Low wing wingroot pumps

With my wingroot system I have landed with low fuel and it has never been a problem in uncoordinated flight.

to be honest..in the standard system if you have the wrong tank selected for a cross wind landing ..this would be far worse than having two low wing pumps with both pumps running.

frank
IO360 electric wingroot pumps only, now featuring 10 ETOH
 
Fuel Flow Sensor

Hello,

I have a Dynon Fuel Flow Sensor (Floscan) to install in my new ACS-100 SORA. I am wondering if I can isntall it inside the engine compartment and also if it is necessary to have a by pass.

Thanks,

Zaramella
 
You must log in or register to reply here.
Back
Top