Fuel System Questions

[whee]

I read through Z's fuel system design in the kit build section but it didn't really answer the questions I had and since I'm doing things a bit different than him I thought I'd start my own thread.

My engine has Continental continuous flow fuel injection. I'm installing a duplex fuel valve (SPRL) and a boost pump (EFII).

I'm trying to decide where to mount the fuel pump. I want to keep it away from heat so the engine compartment isn't a great choice. I was thinking under the floor just forward of the fuel valve but that will place it right above the tunnel. Does the tunnel get hot? I would think not but don't know for sure.

For the return lines: I'd like to avoid welding in another fitting just to keep frank penetrations to a minimum. Would returning the fuel to the upper port used for the fuel sight gauge cause major error in reading the fuel level? I think it would cause some error but it wouldn't be significant.

[tailwind]

FWIW l feed my transfer pumps into the fuel sight Gage's and during the time that the fuel pump is on the fuel gage is not accurate.

[Blackrock]

The tunnel does not get hot on mine, but I have 25 degree downturns off the end of the pipes. There isn't a lot of clearance above the tunnel. Maybe next to the tunnel would give you more room. I barely had enough clearance for a single-stack Andair valve and fittings. A double stack (duplex) is quite a bit taller so finding a location for that may be a challenge as well.

I connected the aux tanks to the mains by t-ing into the lower sight gauge port. There is a big change in fuel level while the pump is turned on. I suspect you'll see a significant error connecting to the top, too. You have probably thought about T'ing into the tank front pick-up port?

[whee]

Thanks guys! Good to know that feeding the return into the site gauge port will cause a large error. I thought about t-ing into one of the pick-up ports but I want to eliminate all chance of recirculating warm fuel and air. I think in that long of a run that the fuel would cool and bubbles would dissipate but I would feel better if the fuel went back into the tank.

Thanks for the pictures BR. They are helpful. I know the duplex valve will be difficult to fit in there but I'll find a way.

[Bdflies]

Thinking about my response, it might not prove helpful, but might be 'food for thought'. I built a Glasair TD with a Continental IO520-D. That system was very simple, with an un-divided (1) wing tank. I routed the return line back to the flange that mounted the level gauge stand pipe. The return flow, directed onto the float, resulted in inaccurate readings. The system was dirt simple, in that the valve was on, or off. But the gauge lacked accuracy. This convinced me that there was significant return flow. My understanding is that Bonanza's just route the return to one tank (left, I think.). You take off, on that tank, switch to the other, at 1/4, then go back to the original. When you get used to it, it's not a big deal. Personally, if I were designing a system, I'd mimick the one Cessna uses, on the 185. The L/R valve feeds a small header tank, which feeds the electric pump. The return line goes to the header, which allows warm fuel to mix with cool fuel, from the wings. A vent line, from the header, is plumbed to the tank vents, to allow any vapors to escape the header, insuring no vapor lock. Cessna adds the inconvenience of the header, but eliminates the trouble of switching the return flow.
As stated at the beginning; I doubt if I helped you come up with a definite solution. But, hopefully, I've illustrated some options.

Bill

[Zzz]

Was just talking to a friend about this exact thing tonight in my garage. Funny. I also wondered about whether the tunnel gets hot. Another option is the void between the boot cowl and the fuselage tubes. Also, the cabin side of the firewall higher up.

It's a puzzle for sure and I continue to think about it. Where's your gascolator going? Robert Paisley told me to definitely not put it on the firewall as I'd originally planned, just like Blackrock's installation.

[Mark Goldberg]

The normal/best place for the gascolator is just behind the fuel selector valve. That is the low point of the fuel system. Mark

[whee]

Bill, Sharing ideas and experience is what these forums are all about. Thanks. My original plan was to feed the return only to the right tank but I decided a duplex valve would be a better way to do it. In reading about the Continental fuel injection I learned that as much as 50% of the fuel could be returning to through the return line. That's a lot of fuel and I think it would be better to put it back in the same tank it came from.

I looked up the dimensions of the duplex valve and the boost pump again and the valve should fit to the left of the flap handle and aft of the tunnel. The boost pump should fit above the tunnel just forward of the valve. The valve will arrive next week so well know for sure then. I haven't ordered the pump yet and probably won't till I start final assembly.

The pump requires an inline fuel filter so I don't plan to use a gascolator at all. I think creating a low spot in the fuel line between the valve and the pump then placing a T and a drain valve in the low spot will be sufficient to collect and drain any water.

The side of the boot cowl or the inside of the firewall are possible locations for the pump but I'd like to keep it hidden if possible. Since the pump is only 1.9" thick I think it will fit above the tunnel.

[marcusofcotton]

The pump requires an inline fuel filter so I don't plan to use a gascolator at all. I think creating a low spot in the fuel line between the valve and the pump then placing a T and a drain valve in the low spot will be sufficient to collect and drain any water.

I'd put a gascolator ahead of the inline filter. Too much moisture can flow right past that T at flow rates and turbulence in the flow, a gascolator would catch more of that. More water capacity as well.

[Battson]

Strongly agree - both a filter and gascolator is a good idea. Fuel injection needs fuel pressure.

You would not believe how much junk the gascolator can hold while your filter stays clean and free-flowing.

The "last chance" filter in the fuel servo is small in size and extremely fine gauze mesh, and can be blocked very easily.

The filter is larger, but when it starts to block up, you can see the fuel pressure drop almost immediately.

[Blackrock]

Interesting discussions on gascolator locations. The firewall location was selected because to put it under the floorboards meant it would protrude below the fabric. I worried, that if I ever dropped it on its belly, that the gascolator or attached fuel lines would rupture. So, I put it on the firewall to eliminate that possibility. I'm not convinced it needs to be at the low point, but agree, that is preferable from purely a gasclotaor standpoint. But there are always compromises depending on what your priorities are. Mine functions well, and traps water at that location; The most water I've ever recovered from it was about 1/4 a teaspoon. I did put in a low-point drain aft of the fuel selector valve and have recovered about the same amount of water at that location, too. All in all very minimal, but I live in a dry climate. A lot of the fuel I use is pump gas that comes out of a 100-gallon transfer tank (filtered at the pump), and sometimes from fuel bags or Jerry Cans which have a higher chance of contamination.

[JimParker256]

One caveat, if you're using an in-line fuel filter, is to make sure it has a bypass capability if it gets clogged with debris. I attended a seminar at OSH by a guy who designed fuel systems for a living (worked on several current-generation fighters and bombers). He was adamant that the gascolator SHOULD be installed at the lowest point in the system, and that any in-line fuel filters MUST have bypass capability. He said the problem with just using a p-trap type location for a drain instead of a gascolator is that the fuel flowing through it will tend to sweep any debris in the fuel along with the fuel, and not allow it to settle in the p-trap (low point). Draining fuel there may or may not drain water from the system. The gascolator design ensures that the higher-velocity (relatively) fuel flow is greatly reduced in the body of the gascolator, and therefore sediment and water will tend to settle to the bottom, where they can be drained more efficiently. If you don't use a gascolator ahead of an in-line fuel filter, you're almost guaranteed (according to him) to have sediment flow past your low point drain, and be trapped in the fuel filter. Thus his insistence on using only bypass-able filters. Furthermore, he strongly recommended using transparent-body filters, and installing them in a place you can readily see them, so you could visually inspect them before each flight. He then went on to say it would be a "real good idea" to use multiple filters, one downstream from the other, so that any debris that bypasses the first filter would be trapped by the second before it clogged the carburetor jets or injectors. It all seemed very logical, and made a lot of sense to me.

The fuel system Bob Barrows designed looks EXACTLY like the illustration he used in the seminar for a best practice, simple and elegant design. Gave me even MORE respect for Bob's engineering and design expertise! I'm using the Bearhawk gascolator, and it looks like it fits exactly between the aluminum skin and the floorboards. Gee - almost like it was designed to do that! ;-)

[Zzz]

For closed circuit fuel injection systems with return lines, the pressure and velocity of the fuel are enough that water doesn't have much of a chance to drop out of the fuel at the gascolator, or so I have read. As a low point trap they might work well but during actual engine operation I wouldn't bet on it.

[whee]

In really enjoying this discussion. Thanks everybody. I hope it keeps going and more share what they think.

My original plan was to use a gascolator and not a fuel filter because in airplanes in not a fan of inline filters. In my reading and discussions with some of the experimental fuel injection suppliers I discovered the same as what Z said above: at the velocities my fuel system will be experiencing the gascolator will not slow the fuel enough for the water or debris to settle out. EFII specifically says a gascolator does not remove enough of the sediment so a inline filer must be used or fuel pump damage could occur.

I'm trying to design the simplest, most reliable fuel system that will perform the job required. I think Bob designed a great fuel system for carbureted engines and I'd like to only make the changes required for it to work with my Continental FI and EFII boost pump.

Thinking about the system logically this is what my brain tells me:

The velocities of the fuel are high enough that water and most debris will not settle out in a gascolator. I'm not even sure that water will settle out in a gascolator with a carbureted system because it takes time for that to happen.

A gascolator is placed at the low point of a system to catch water while the plane is sitting on the ramp. A low spot in the fuel line with a drain should be able to accomplish the same thing.

Water is drained from the system during preflight so little is any water will be in the fuel during flight. If there is water in the fuel I don't think the gascolator with catch it and the engine will past the fuel without a problem. My cars do not have water separators and they have never stalled because of water contamination.

Fuel flow and pressure will be monitored via a digital engine monitor. Paying attention to trends should clue me in if my filter is getting plugged.

No my limited experience (400hrs and 10yrs of airplane ownership) this is what I know:

I've drained water out of the fuel system only a handful of time during preflight and never during post flight.

The cumulative amount of debris the gascolator caught over ten years would not have plugged a single fuel filter.

I never had a safety of flight issue related to fuel contamination.

This experience was with an airplane stored outside and fed a steady diet of mogas from a gas can.

If my line of thought is wrong I don't have a problem installing a gascolator upstream of the inline fuel filter but I'm not sure that I'd gain any real benefit from it. Also, I think the hump in the belly for the gascolator is unattractive and it sucks to crawl under the plane during preflight to drain it.

[Battson]

I'm trying to decide where to mount the fuel pump. I want to keep it away from heat so the engine compartment isn't a great choice. I was thinking under the floor just forward of the fuel valve but that will place it right above the tunnel. Does the tunnel get hot? I would think not but don't know for sure.

http://i202.photobucket.com/albums/a...a7bf0d5.jpgFor the return lines: I'd like to avoid welding in another fitting just to keep frank penetrations to a minimum. Would returning the fuel to the upper port used for the fuel sight gauge cause major error in reading the fuel level? I think it would cause some error but it wouldn't be significant.

I put the EFII fuel pump exactly where you describe, as did Fellman I believe, there are no problems with doing that. This picture will show you how. The tunnel doesn't get hot enough to cause any issues.



Sorry about the poor quality, I hope you can still see the detail.




I would look at using the fuel vent / aux tank port to route the return line.
You can also see from the picture, how to mount the gascolator at the lowest point (without getting far enough aft to get into the belly fabric).

As with the fuel pump, the simple hardware store Al angle section plus adel clamps is a strong and light way of permenantly mounting the part, without additional welding.

I subsequently installed the fuel transducer in the long straight run behind the selector valve. I don't seem to have a photo of the complete set-up for some reason.

[mswain]

What fuel pump did you use? Also, what is the cylinder mounted next to fuel pump? Last, what fuel filter is that inline?

[Battson]

EFII, that cylinder is part of the pump and so is the filter.

[whee]

Great pictures Battson! Thanks.

I was hoping not talk about the fuel transducer because I'm not sure my idea is going to work. Because of the continuous flow system GRT says I will need two transducers; one in the supply and one in the return and their system will calculate the actual fuel used. My plan is to install the transducer just before the fuel distributor which will place it after the return line so all the fuel it is measuring will be fuel used. The tech I talked to at GRT said this won't work. He couldn't tell me why it wouldn't and he didn't know of anyone that had tried it. Because it not working won't be a safety of flight issue I think I'll give it a try.

[Battson]

Just for the sake of clarify, I will restate a few things which are already very obvious to everyone here, in order to build to a logical conclusion:

The fuel system isn't designed to operate with anything more than the very occasional droplet of water in it, during flight. The only *safe* option is to remove the water pre-flight.

Gascolators are good traps for catching heavy particles at any time, and they have a coarse screen inside them to prevent big stuff getting past and blocking downstream filters. But water is not *that* much heavier than fuel, can't be screened out, and the gascolator is really only designed to catch droplets during flight, rather than anything more serious.

So, the gascolator could be installed anywhere in the system at achieve that end. But it is installed at the lowest point to catch water before flight begins, which naturally drains downhill.

That water can come from three main sources:

1. Due to condensation overnight while the plane is parked. That water sinks, flows downhill, and pools at the lowest point - either the tank drain or the gascolator. It can be drained before flight.
2. Leaks in the tank cap or elsewhere in the system, during rain while parked outside. Again - drained as above.
3. Water entering the system during re-fuelling. That needs to be given time to settle, and drained from the tanks (and gascolator if there's a lot) before flight.

If you have no gascolator, or its not at the lowest point - then you have no way of draining any water which gets past the fuel tank's low point. That water then has no choice but to be injected into your engine.

So my point is, there are basically two different functional requirements of that item within the fuel system:

1. In-flight water trap + strainer function
2. Lowest point water drain pre-flight function

In conclusion - even if the current thinking is you don't need to fulfil the first requirement (which is a big call), you still need to install something to meet the second requirement anyway, at least a drain valve - and it might as well be a gascolator. You get the added benefits of the strainer, bigger water trap, and no loss of fuel pressure even if a huge volume of nastiness ends up in your system