Fuel flow test fail

[rodsmith]

Did a fuel flow test yesterday. Knowledgeable engine people tell me my modified parallel valve IO-540 should make 290-310 HP so I did my flow calculations based on 300 HP. Using the 1.25 factor for pumped systems I required a flow rate of 34gal/hr. Next decision was what climb angle to use. Others have talked about a max climb angle of 17-18 deg. I added the required 5 deg to that and set the test angle at 22 deg. In my case this required setting the tailwheel 34" lower than the main gear. To perform the test, I disconnected the fuel line at the inlet to the engine driven pump and drained fuel at that point. 3 gallons of fuel above unusable were placed in each tank. With the plane at that attitude the pump inlet was only 4 1/2" below the fuel level in the tank. Tested the left tank first and only got 12gal/hr!. Tested from both and only increased to 13gal/hr. Decided to redo the test at 18 deg, raising the tail to 19" below the mains. This only improved things slightly. 14 gal/hr left and right, 15 gal/hr both, not even half the required rate. I suspected that the Andair fuel pump might be part of the problem. I didn't wan't to tie up my 3 friends who helped position the plane any longer, a couple had traveled quite a ways to help, so we towed my plane back to my hangar. Last night I decided to test some more at the 3 point position of 13 deg. Only got 18 gal/hr each tank 19 both. Decided to test the Andair fuel pump, 46gal/hr with the pump running. Fabbed up a 3/8" line to bypass the fuel pump. That got 24 gal/h, a big improvement but still way below the 34gal/hr required. I'm really perplexed as to what the problem is. My system uses all 3/8" lines, from the Andair fuel selector it goes through the Andair gascolater, then the electric fuel pump. At the firewall is a straight stainless bulkhead fitting. From there a 24" 3/8th firesleeved teflon hose to the pump inlet. Prior to the flow test I had put 7 gal of fuel in each main tank and flushed through a drain just upstream of the electric pump. Pulled the gascolater filter and there were just a few very fine particles in it. I became concerned that my somewhat convoluted vent system might be the problem. I tested with the fuel caps off and there was no difference at all, so the vents are good. I've emailed Andair about the pump, seems like there is too much loss through it. Would love to hear the experience of others, especially someone that's using an Andair pump.

DSCF2222.jpg

[500AGL]

Rod,

I would highly suggest calling Bob on the phone to discuss this. He would be a knowledgeable source for fuel requirements, fuel flow, and perhaps different pump arrangements.

Peter

[Mark Goldberg]

On all my planes with 3/8" lines and Bob's standard fuel system I would get 23-24 GPH from individual tanks and 28-30 GPH using a BOTH setting. Mark

[rodsmith]

What climb angle were you testing at?

[Mark Goldberg]

I don't remember Rod. But obviously something is going on with your system because of the big difference that me and others have seen.

[Nev]

Rod,

My notes show a fuel flow of 34 gallons per hour from individual tanks, electric pump OFF, in the 3 point attitude. This was measured with less than 5 gallons in each tank. With the fuel selector in BOTH the fuel flow was 40 gallons per hour.

For the purposes of testing we disconnected the fuel line at the firewall pass-thru on the engine side, attached a separate hose, and drained it into a calibrated bucket.
I have an EFII type electric pump with bypass, (but no filter in that position).

Hope this helps.

[jaredyates]

Is it typical to test by flowing through the pump, or at the pump inlet? I did the latter, not positive that it's correct. Is there any chance of contamination in the lines? Insects, swarf, etc? This is why we do this test.

[rodsmith]

I tested at the inlet to the engine driven pump, flowing through the electric pump. I suppose there could be something in the lines that wasn't flushed with gravity flow.

[jaredyates]

I am not familiar with the pump that you are using, but it would be totally conceivable that the pump is the restriction. Please let us know if you hear back from the manufacturer about their expectations for whether the fuel should be able to gravity flow through the pump at full rate.

[Bcone1381]

Hey Rod, your getting real close!!

Common sense (I wish I had an authoritative document to refer to) tells me that the electric fuel pump should be on for the test but it seems (with a 52% confidence rate) like you had it off to me.

[rodsmith]

My understanding is that you need to test for gravity flow to the engine driven pump inlet. The electric pump should just be backup in case of an engine driven pump failure.

[rodsmith]

Nev, that fuel flow rate is great, did you test it at a higher angle? Tells me that I have a major issue somewhere.

[Nev]

Rod,

My main focus was checking that the fuel lines were clear, and that my particular fuel system set-up hadn't introduced any unintended consequences. We did test level flight attitude, and a nose low attitude, all with similar results.

I had cleared all fuel lines with an air compressor first. I did this after first disconnecting fuel pumps, transducers (don't blow air through the transducer ) etc.

Also, I don't have a fuel filter in line with the pump, so I would expect my resulting flow numbers to be slightly higher.

One suggestion might be to disconnect the single fuel line just before the electric fuel pump, and test the flow rate at that point as a start, in the 3-point attitude, initially on one tank. This will show if you're in the ball-park, and whether some downstream devices might be slowing the flow rate.

We also discovered that on my aircraft, when adding a small quantity of fuel to an otherwise empty tank, the fuel line developed an air lock and wouldn't flow at all. Cycling the fuel selector to any other position was always sufficient to clear the air lock.

A few more thoughts
Because our Bearhawk's are not "first of type" as such, I was largely checking that mine conformed to the fleet standard. At the time I was of the opinion that the 3-point attitude was very close to the stalling attitude (and it is for a normal stall). However during flight testing I performed a series of power on stalls and logged the pitch attitude at up to a maximum of 28°. To get the nose that high requires a power setting of over 20" MAP at minimum speed ( and an accompanyingly higher fuel flow rate). It is not a maneuver I would normally be intentionally performing, (and certainly not close to the ground).
However, actual flight testing with the boost pump off, reassuringly shows no fuel flow related issues at that attitude, even when down to only reserve fuel in one tank, so long as the ball is centered at all times.

[marcusofcotton]

Hey Rod, your getting real close!!

Common sense (I wish I had an authoritative document to refer to) tells me that the electric fuel pump should be on for the test but it seems (with a 52% confidence rate) like you had it off to me.

Regulatory aspect reference: FAR/AMT (2007) 23.955 Fuel flow. If it is a pump system, 125% system flow rate required for each main and each emergency pump when the pump is operating as it would at takeoff.

Also see Tony Bingelis Firewall Forward pages 171-177 for more discussion.

That's not to say it wouldn't be desirable to have plenty of flow without pumps running...

Nice to hear your getting close Rod!

[Bcone1381]

Rod;

I wrote this before I saw Mark's post.....

The FAA fuel flow test has an objective to confirm that an aircraft has adequate fuel flow to the engine under normal flight attitudes and assumes all systems function normally (FAR 23.959). Your fuel system has a main and aux pump, and you dont rely on gravity if the backup pump fails. So I call this a pump system just like an RV.

You get 24 G/hr with no pump which is similar to what Mark and others with 3/8" lines get similar results. I like that. You are getting 46g/hr with the Andair pump on And I like that. When the Andair pump is off it produces drag and reduces flow. That seems about right to me. 3/8 " lines are very common for 300hp engines but not without a fuel pumps. Lancair legacy has 3/8" lines with 350 hp. The only small change I saw researching it is I would do the test with the fuel exit from the line going to your fuel servo, second best place would be the engine driven fuel pump exit.

I pasted the text below from the FAR's and AC. The FAR's apply to certificated aircraft..
_______________
CFR 23.955
§ 23.955 Fuel flow.

(a) General...

(b)Gravity systems. The fuel flow rate for gravity systems (main and reserve supply) must be 150 percent of the takeoff fuel consumption of the engine.

(c) Pump systems. The fuel flow rate for each pump system (main and re- serve supply) for each reciprocating engine must be 125 percent of the fuel flow required by the engine at the maximum takeoff power approved under this part.

(1) This flow rate is required for each main pump and each emergency pump, and must be available when the pump is operating as it would during takeoff;
(3) The fuel pressure, with main and emergency pumps operating simultaneously, must not exceed the fuel inlet pressure limits of the engine unless it can be shown that no adverse effect occurs.

§ 23.959 Unusable fuel supply.

(a) The unusable fuel supply for each tank must be established as not less than that quantity at which the first evidence of malfunctioning occurs under the most adverse fuel feed condition occurring under each intended operation and flight maneuver involving that tank. Fuel system component failures need not be considered.



From AC90-89a Amateur Built and Ultralight Flight Testing Handbook
Ch. 11 1. f. Fuel Flow and Unusable Check.
f. Changing Fuel Flow or Pressure: If the aircraft’s fuel flow rate is less than planned, there is a volume or pressure problem. An increase in the fuel flow volume may necessitate installation of larger fuel line fittings on the fuel tanks, fuel selector, and carburetor in addition to larger internal diameter fuel lines. To increase fuel pressure, install an electrically driven or engine driven mechanical fuel pump prior to the first flight.

[rodsmith]

For a fuel injected high wing aircraft, it seems like there should be two flow tests performed. One to see if there is sufficient gravity flow to the engine driven pump. Then to see if the running electric pump provides sufficient flow to the fuel servo.

[BradW1062]

I noticed that the AC90-89 is now on revision C, as of 14Feb2023.

And, this FAR 23.955, along with everything in the whole section from 23.1 to 23.1589 and the appendices, was deleted from the law in August 2017. I guess that's when they were removing the "prescribed" design criteria in favor of "performance" criteria - or whatever it was they were intending to do.

Now, the replacement, FAR 23.2430(a)(3) just says: "Provide the fuel necessary to ensure each powerplant and auxiliary power unit functions properly in all likely operating conditions".

HA! That seems like a tall order to me - I mean, how can the necessary fuel flow ensure that the magneto functions properly, or how can the necessary fuel flow ensure that the carb-heat functions properly, or how can the necessary fuel flow ensure the piston rings function properly - all of those things are also necessary for the powerplant to function properly, and fuel flow has nothing to do with any of them...well, maybe it could be argued that fuel's cooling effect has something to do with the rings, but still.

And I swear I read somewhere in the FARs where it said that the fuel system shall provide 100% of the requirements of the powerplant in any approved operation or maneuver. But now I can't find it. 100% is way different than 150%, or even 125% so it caught my attention at the time. Anyhow, just where do we draw the line? Since these are not certified aircraft by definition, then probably AC90-89C (2/14/2023) is going to be our only guideline of choice.

[rodsmith]

Still waiting to hear from Andair, probably going to have to make a call to England. Looking at my pump specs, it only flows 16gal/hr at 30psi. They say it is adequate for both O-360 and O-540 but seems like inadequate flow for a high HP IO-540. Also when off there is a .5psi pressure drop at 30gal/hr. That explains the significant increase in flow when I bypassed the pump. They make a 35gal/hr pump with 1/2" fittings, but it has the same pressure drop when off. Thinking I may replace it with the Airflow Performance boost pump. That is 32 gal/hr at 30 psi and about 1/4 of the pressure drop when off. I can see one area where I can clean up the plumbing. If I turn the gascolater 90 deg I can save a loop of tubing and eliminate a low spot upstream of the gascolater. For some reason I didn't think of that initially.

DSCF2224.jpg

[alaskabearhawk]

I installed the EFII pump. I really like the clean installation and the straight through design when the pump isn't on. https://www.flyefii.com/products/boost-pump/

[rodsmith]

Thanks Paul. Looks good, not an option I was aware of, will be talking to them.

[triumphantduke]

May I ask what the function is of that vent screen?

[jim.mclaughlin924]

On my patrol with a Titan IO360, the boost pump is not used for takeoff. I have a check valve bypass around the pump and got way more than flow than required in all attitudes with the pump off. Almost double that with the fuel pump on. I have a 10 micron filter on each tank and 40 micron filter on on the inside of the firewall going to the engine. The check valve and final filter came from earls. .5 psi with gasoline is 1.6 feet of head, a lot to lose in a gravity system, probably half of the total available head.

[500AGL]

Jim do you have photos or a sketch of how and where you plumbed the pump and bypass in by chance?

[whee]

Bummer deal Rod! When these things happen it is so incredibly frustrating. Just when you think you're in the home stretch you get smacked with something unexpected that may not have a straightforward solution.

I wish we could see your entire fuel system but I imagine it is pretty standard; though seemingly small changes can have a big effect. Minimize the use of 90s, make sweeping bends, and pick components that minimize losses.

The EFII boost pump module is nice and has low suction loss when in bypass but it's expensive and it sounds like you'll still have a problem upstream of the pump. I think the Airflow performance pump is just as good and quite a bit less expensive. I have the EFII boost pump in my plane and it works fine.

Take some pictures of your fuel system so we can see the whole thing and we will provide you with some constructive feedback.

[Battson]

Rod - it seems to me you passed the test.

Testing is with the pump running, you need 125% with the pump running. You got that.

[Dpearson]

rodsmith I see a strong similarity between what you describe (low flow/high pressure drop caused by the aux pump in "bypass" mode) and the final update in the discussion of the 2019 PR-ZJO Engine-out incident in Brazil, which is described in this thread. Check out the final update (post # 174) from MattS , which I have copied here:

So almost 4 years after my engine out and hard landing we have some more information. We overhauled the engine and had some wing repairs to complete. After 12 hours or so of flying after getting it back together, I had a couple times when the engine sputtered. It always came back strong but could tell something wasn't correct. It's a Cont IO360 and therefore has a pretty extensive procedure for measuring and adjusting the fuel pressure. Doing this test, we could not get the correct pressures. Wouldn't quite hit the full PSI requirement and was fluctuating a lot. We started troubleshooting and found out if we removed the electric boost pump from the system, we could get the correct pressures and it was stable.

I fully agree with your comment: "For a fuel injected high wing aircraft, it seems like there should be two flow tests performed. One to see if there is sufficient gravity flow to the engine driven pump. Then to see if the running electric pump provides sufficient flow to the fuel servo."

Based on my reading of 23.955 (thanks Bcone1381 for including that text) I think the test limit of 125% is appropriate for both of the test cases you describe.

I've got a bit of fluidics experience, and I fully agree with your "gravity flow" test approach to ensure sufficient flow in the "aux pump off" case. Ideally per 23.955 you would conduct the "aux pump off, engine-driven pump functioning" test at the output of the engine-driven pump with the engine-driven pump "...operating as it would during takeoff" but of course that's impractical. 125% gravity-feed flow at the inlet of the engine-driven pump, with the aux pump off and bypassing, is good evidence that the engine-driven pump won't create negative pressure at its inlet or in upstream plumbing (which can cause fuel vaporization, especially at altitude) as it is trying to "suck" fuel through the upstream obstruction of the aux pump's bypass valve.

It is true that the Tony Bingelis "Firewall Forward" book (Page 177) mentions only to test with the auxiliary pump on, and makes no mention of how to test the case that covers the vast majority of my flying time (aux pump off, engine-driven pump is the sole means of fuel delivery). Without such a test, an aux pump that has limited (or no!) bypass capability could escape testing and become a safety of flight issue. I think that's a potentially hazardous oversight in what's otherwise a very useful book. I think the "gravity flow" test you're doing closes that gap nicely.

[svyolo]

I agree with this 100%. One of our local tech counselors was an old Boeing fuel system guy. I described my system to him (1/2 gallon header tank) and he described it as "converting a gravity fed fuel system, to a low wing, pumped system." I also believe you need to test two ways, gravity and pumped. The pumped so that the fuel system in operation can supply the needed fuel, and the gravity test (to pump inlet, or in my case, header tank inlet) to make sure the pump isn't scavenging fuel thru the system by pressure. By that I mean gravity is feeding the pump inlet, the pump isn't drawing fuel thru the system.

I will be testing both ways.