Sds efi

[Johnp]

Hello,

I am finally back to working on my Beahawk kit after a several year hiatus. One of the things that has changed since I last worked on it is my decision to install the SDS EFII system. In particular the fuel injection system has some characteristics that pose challenges to the fuel system Bob designed. I suspect that there are others that are interested in the SDS EFI system so I thought I would add this for discussion. Ross from SDS also just said that he has joined the site so that he can watch for questions. He post regularly on the RV forum. Between designing and manufacturing his products and posting on the RV forum I don't know when he finds time to sleep.

The first is that the SDS system is not a returnless system. The return flows may be close to a gallon a minute. While returns can be relativity easily added to the main tanks, feeding the high pressure, high flow, electric pumps is another issue. The system Bob designed was not intended to flow this much fuel. The risk of getting air in the system is also another real concern that the system would likely never experience gravity feeding a carburetored engine.

After thinking about this for some time I decided that a small header tank would likely be the best option for making a reliable and safe fuel system. The other major change from Bob's design that I believe is a good idea is adding a vent between the main tanks that the header tank is also connected to. I discussed adding the SDS EFI system with Svyolo and it appears that we have come to very similar conclusions about what is needed to safely add the SDS system to the Bearhawk.

My system is also a little different from a typical return type system in that I plan to have an independent return valve. This requires more attention to fuel management if the fuel is not set to return to the header tank. However, it does allow me to use the pumps to transfer fuel on the ground or at cruise. It also could cause an unwanted fuel pressure spike of up to 110 PSI, according to Ross at SDS, if the valve position is changed while the pumps are operating. To alleviate this spike I plan to install a check valve with a crack pressure of approximately 60PSI. The crack pressure is adjustable via shims. In the unlikely event that the highly reliable Borla regulator were to fail and cause a drop in fuel pressure it might be possible to close the return valve and utilize the check vale to increase fuel pressure until a landing can be made. A more typical system might be one where the fuel returns only to the header tank and a return or duplex fuel valve, and the associated return lines to the main tanks, would not be required.

I asked Ross at SDS if he thought a 1 gallon header tank would be adequate. His prompt reply was that he though it would be good. A 260HP IO-540 should be using close to two quarts a minute at takeoff power. This means that the fuel in a one gallon header tank would be replaced in less than a minute at take off power and every thee to four minutes at cruise. Since the SDS system does not return any fuel from the injectors and only from the small manifold that is located away from the hot cylinders I do not believe that fuel heating will be a problem. Additionally, the main fuel tanks and fuel lines only have to provide the amount of fuel that the engine is burning, just as Bob designed them to do.

I just obtained a set of fuel pumps and filters from SDS. The machining of the SDS manufactured parts is excellent. I am now working to figure out where I wan to install the pumps, the main filters and the header tank as well as my Dynon pitch servo. At the moment I think I can get them all under the left seat. This protects then in case of a gear failure and also makes the main filter easy to access for removal and inspection at every condition inspection. One thing I will have to consider is that I suspect Bob designed the seat rails to deform and reduce the G's experienced during a very hard landing. Having items under the seat would obviously interfere with this design.

Attached is a preliminary diagram of my proposed system.


Disclaimer.: This is all theory and I have not done any testing of the ideas presented here. It is up to you to design, test, and decided what is safe for your installation. The discussion also applies ONLY to the SDS EFI system as other systems that I am aware of will likely add much more heat to the fuel before it is returned due to their injector and fuel manifold locations.

V/R,

John Poulter IMG_20200513_110630833.jpg

[zkelley2]

Ross makes great stuff. I really like the CPI that I put on my Bearhawk. He's also the best customer service I've ever dealt with in the Aviation industry.

[svyolo]

In the end, I ended up copying, as best I can, Cessna's fuel system logic used with injected engines.

I "borrowed" Cessna/Conti fuel system logic, but am using SDS. If I was using EFII, Rotax, or ULpower, in a BH, I would probably redo their fuel rails with the same logic that Conti/SDS use.

[rv6ejguy]

Hi guys. Good to see this thread. I'm happy to be a member here as of today and to answer any questions on SDS.

[ThreeD]

Great to have you here Ross!!


Dan
Kamloops

[svyolo]

Hi Ross. Welcome

[Bcone1381]

Hi guys. Good to see this thread. I'm happy to be a member here as of today and to answer any questions on SDS.

Ross,

Would you reach out to Bob Barrows and partnership with him to design a bullet proof fuel system that supports your product? Your product has special needs. A high wing with gravity feed to a carb is very different from the needs that your system requires.

Brooks

[svyolo]

Ross;
There have been a few pretty lively discussions here about different ways to run a fuel system on EFI (and this also probably applies to Continental mechanical FI). That was really my only hesitation initially. Myself and JohnP who started this thread have come up with similar systems, independent of each other. A small header tank under one of the front seats.

The BH tanks have a front and rear feed from each wing tank. Each pair meet down low in a T, and feed into the fuel valve, usually just in front of the two from seats, which is near the leading edge of the wing, fore and aft. Usually the valve is a 4 way, with a "Both" setting, which seems to be the preferred way to run most of the time.

Bendix/Lycoming style run returnless. Most BH's running IO Lyc's just take fuel from the fuel valve, no header tank. Seems to work.

One builder (flying now) has a Conti, and his engine uses a fuel return similar to EFI. Most BH's use 3/8" fuel lines, he calculated he needed 1/2" lines to run with no header tank, and returning fuel to the mains.

I would like to say that you mostly do RV's, but that would be massively understating the facts.. You have supplied EFI to just about everything imaginable. As a matter of fact I found my "header" tank by reading some of your other web sources. In cars they call them surge tanks. header.jpg

[rv6ejguy]

We've supplied systems for a number of high and shoulder wing aircraft types and some of these have used a small header tank mounted lower, gravity fed by the wing tanks with the pumps below the header. Seems to work just fine. 3/8 lines feeding the header will flow more than enough via gravity, even for a 540.

[rv6ejguy]

Ross,

Would you reach out to Bob Barrows and partnership with him to design a bullet proof fuel system that supports your product? Your product has special needs. A high wing with gravity feed to a carb is very different from the needs that your system requires.

Brooks

I think with a couple BH guys going the header tank direction here, we can wait for the results but I see no issues going this way as it's been done before on other similar airframes. I used to have a header tank in my RV and it worked fine with a Facet pump feeding it from the low mounted wing tanks. This did have the return line from the regulator feeding back through the header and then on to the selected wing tank again.

Once proven here, I'd say just copy the design.

[Archer39J]

Main idea with header or without is to keep the stock fuel system supplying only what the engine burns, no different than carb or mechanical FI.

I'm going the headerless route, don't see the need for heat with good routing, vapor lock isn't an issue if the inlet is at a positive pressure head. Air is a concern so keeping those runs short is the goal, but with BOTH that risk seems minimal. Thinking I might make use of those extra ports and put the regulator on the pump outlet and return right to the inlet manifold. Has anyone else run this kind of setup or something similar? That pump seems almost made for it

[svyolo]

I spent quite a bit of time in front of my computer the last couple of months. By accident. I did look into the fuel systems layout for some of the other EFI engines. Specifically Rotax, ULPower, Lycoming, and Viking.

I thought the ULPower was probably the best thought out engine I have seen. They look expensive, but you get a lot inside the box. EFI, ignition, integrated alternator(like a motorcycle), starter, exhaust. On a Lyc that is all "extra". Their fuel system was very simple and well laid out. The whole engine was. If I was looking for an engine in that horsepower range, I would have a serious look at UL. Their rated power they advertise is at a very high rpm. Drop 10% off rated for 2700-2800 rpm.

The Rotax, was, well, Rotax. Why use 4 parts when you can use 24. But I bet it works well. The one thing I really did like about the Rotax was it comes standard with an electrical system to run the engine electronics, ignition, pumps. I think having that come from the factory reduces risk for a lot of installations. The thing that made little sense to me was they had 2 fuel injectors per cylinder for redundancy. Massed produced electric fuel injectors are probably some of the most reliable things every made. Maybe they wanted bragging rights for "most redundant". Their 100 hp EFI engine is only 3k more than their carburated version and burns a bunch less gas. I bet it is a very good engine.
.
It does look like almost every EFI installation in high wing LSA type aircraft is using a header tank. Some bigger than others. Lots of different locations. Some on the firewall, some behind the front seats (no back seats). Zenith has tall one on the pillar behind the door on the left side.

I think I prefer the fuel feed system on SDS better than any of the others for aircraft style engines. Less heating of the return fuel, and less mechanical stress on the fuel connections.

But I am sure all the others work as well.

[Bcone1381]

The Bold below is "copy/past" some above posts.

....lively discussions here about different ways to run a fuel system on EFI

Usually the valve is a 4 way, with a "Both" setting, which seems to be the preferred way to run most of the time.
(meaning a lack of standardization is resulting in one persons preference in absence of an industry best practice.)

Seems to work.....One builder (flying now) has

Seems to work just fine.

we can wait for the results

Once proven here, I'd say just copy the design.

I think the Group owes it to ourselves to have a fuel supply design that is bullet proof, a design we can call "Industry Best Practice." One we can count on.

Lets not settle for anything less.

[svyolo]

Cessna 185 fuel system is industry best practice for high wing/Continental/EFI fuel system logic. IO Lyc is much simpler as it is a non-returning system. The BH fuel system, stock, has a great track record for non-returning fuel system, which is Lyc/Bendix FI/carb. If you have to return fuel from your FI system, the BH system needs to be modified
.
Whee used 1/2" fuel lines, which seems to be working. If I remember right he has a disclaimer that he can skid the airplane and cause the engine to vapor lock. The Conti FI system clears the air/vapor quickly.

Another other option is to mirror Cessna's fuel systems for IO470-550's. Return fuel to a header tank. I did my best to follow Cessna's logic.

No one, in all of human history, has built more high wing airplanes than Cessna. They built them with Cont (returning) fuel systems, Lyc (non returning), and carb (non-returning).

I am going with Cessna. If someone can come up with a better "best practice" for high wing fuel systems, I am all ears. I am trying to make the SDS fuel system work happily with the Cessna logic. SDS is similar to Continental. SDS returns the same volume of fuel continuously, Conti volumetric fuel flow does change with engine rpm.

[svyolo]

I believe EFI can also be done return-less. Very easily. But SDS's fuel system the return fuel is very easily returned to a header tank. I already have it installed, so that is what I am going to run.
If Bendix can be run return-less, I cannot for the life of me figure out why EFI cannot be run return-less, using the exact same part numbers. This is EXPERIMENTAL, right?

I will test the return-less on the ground, and take videos and post them. For my own plane, I will return fuel to a header tank.