[Kahuna]

Over the years I have had a few installs of fuel flow transducers on injected motors. Having followed the instructions, I have not had good results. My main problem has been that putting the fuel flow transducer after the electric pump gives bad data with the electric pump on. Flow increases +2gph with the pump on. No change in any other engine parameter except fuel pressure of course.

This has annoyed me to no end and has even caused me to not run my pump on occasion, when I should be, to keep my FF readings accurate to maintain good time to dry numbers on the monitor. Thousands of RV hours with wrong fuel remaining calculations. Frustrating. But since I have tank indicators, terrible as they are, well I just fudged and guessed on gallons remaining off the FF calc.

Folks for years have been complaining about the same issue. It was always a head scratcher for me. Why does the number go up? The little turbine wheel is not going any faster. The engine is not getting any richer. Im not burning any more fuel. There is no place for the fuel to go after the sensor except to the engine in my installs. SO why on earth does it go up when the pressure increases and no more fuel is going by? I dunno.

After a recent upgrade to 97gallons in the wings, I had a need to really have accurate and reliable FF numbers and hence have had a good reason and need to fix the problem. I can no longer tolerate inaccurate time to dry numbers.

I moved the FF transducer off the fixed mount on the firewall into the plenum area before the spider. This cured the problem. May create others later.

End of report.

 

[Lars]

Interesting. Just as I get ready to drive to the hangar, where I planned to install the thing on the firewall, plumbed between mechanical pump and servo.

My red cube came with the sensor package I got with my GRT display. One tidbit I discovered in reading through the supplied documentation, something not mentioned in EI's own literature, was a paragraph about making a simple accumulator to damp pulsing from the electric pump, in order to prevent the electric pump inaccuracy. Having not flown, let alone tried the accumulator mod, I have no idea if it would work. Just sharing what I read. I think I have the text of the paragraph saved somewhere in case anyone's interested.

I can say that when I was flow testing the EFI fuel pump that I run on my hot-rodded old truck (same pump I believe that AFP supplies), the pulsing flow was apparent on the output. That must somehow excite the guts of the red cube.

 

[petercavallo]

Don from Airflow told me best place on FI engines is between the Fuel controller and the Spider Period, no where else (mount it in the line, cover with Fire sleeve)
Peter

 

[Dan Langhout]

Accumulator

I for one would love to see this!!!! Could you post it or should I PM you?

. . . . . . My red cube came with the sensor package I got with my GRT display. One tidbit I discovered in reading through the supplied documentation, something not mentioned in EI's own literature, was a paragraph about making a simple accumulator to damp pulsing from the electric pump, in order to prevent the electric pump inaccuracy. Having not flown, let alone tried the accumulator mod, I have no idea if it would work. Just sharing what I read. I think I have the text of the paragraph saved somewhere in case anyone's interested. . . . . . . . .

 

[Jamie]

I for one would love to see this!!!! Could you post it or should I PM you?

Ditto that. Sounds interesting.

There are several reasons to not install the FF transducer up front. It's an aluminum box with fuel in it, it's supposed to be mounted horizontally (which can be difficult to accomplish), it's supposed to be as vibration-free as possible, etc.

If you could get accurate FF indications (with boost pump on) without putting the sensor FWF, that would be the ideal situation.

 

[rocketbob]

Sorry for the blurry image. The fitting is a custom fitting that was welded up. Works perfectly. The transducer is in a nice cool place, with no mickey mouse fittings or firesleeve wrap required.

 

[Capflyer]

I had the same issue on my last RV with the transducer mounted in the cabin after the fuel pump. On my -4 the FT-60 red cube transducer is mounted on the lower firewall on a shelf made of a piece of angle between the engine driven fuel pump and carb. For fuel injected it would be between the engine driven fuel pump and spider valve. Mine works perfectly with no change when the electric fuel pump is turned on or off and pretty darn accurate.

 

[elippse]

With an injected engine, the fuel-flow will increase 1% with a 2% increase of pressure TO THE INJECTORS. Does the increased pressure with the pump on translate to higher injector pressure?. If not, then it all must be with the pulsating flow.

 

[JonJay]

I put mine there

Sorry for the blurry image. The fitting is a custom fitting that was welded up. Works perfectly. The transducer is in a nice cool place, with no mickey mouse fittings or firesleeve wrap required.

[/QUote
mine too works well at this location

 

[Brantel]

I know we are talking injected engines here but the same issue happens on carb'd engines...if...you put it between the electric pump and the mechanical one.

I have found that if you put it after the mechanical one, you can get very accurate readings from it boost on or off.

 

[AltonD]

mine too works well at this location

Same here. Mine is between the spider and the fuel servo. Dead on. I just filled up with 21.5 gallons.

 

[aerhed]

There are several reasons to not install the FF transducer up front. It's an aluminum box with fuel in it,

And it nicely compliments the aluminum spider full of fuel next to it.

 

[Brantel]

After a recent upgrade to 97gallons in the wings,

Good grief Mike!!! How big is your bladder?? Oh wait, I don't wan't to know that...

 

[BuckWynd]

Flow increases +2gph with the pump on... This has annoyed me to no end and has even caused me to not run my pump on occasion, when I should be, to keep my FF readings accurate to maintain good time to dry numbers on the monitor.

To play devil's advocate, I have to ask whether this is really a big enough issue to worry about (or to move an already-installed FF transducer aft of the firewall). After all, on most fights the electric pump is normally turned 'On' just before takeoff and turned off a minute or two later (at a safe altitude)... Then it's turned on again when approaching the destination, and turned off when clearing the runway.

Total time 'On' is about five minutes out of the entire flight. The longer the flight, the smaller the percentage of the flight the Fuel Flow will be reading 2 GPH high. So not only is it a fairly small error, it's an error in the conservative direction, i.e., the engine monitor will say you've burned a bit more fuel than you really have -- which is better than the other way around, IMHO.

5 minutes at 12 GPH is 1.0 gallons
5 minutes at 10 GPH is 0.83 gallons
5 minutes at 8 GPH is 0.67 gallons

...so if your FF transducer read high for 5 minutes, you'd have a little extra fuel in your tanks when you land. And I'm definitely OK with that. Most times, it looks like it would be less than 0.5 gallon difference.

It's like padding your checking account by putting $100 in your account and zeroing the balance in your ledger.

So is the ACTUAL fuel flow increasing when the transducer is located betwen the electric pump and the engine pump, or is this an anomoly with the transducer? Has anybody tested actual fuel burn?

 

[frazitl]

I installed mine above the engine too,

just like the floscan install on my previous Mooney TLS/Bravo (Lyc 540 turbo). It worked well as does my fuel measurement on the 7A. I believe the servo dampens most of the fluid flow pulsations, and though JPI recommends installing between the mech. fuel pump and servo, I like the cooler location (during flight at least) above the engine. I know that the straight tubing requirement after the flowmeter called out by Floscan is violated with my installation, but I'm convinced that the straight flow requirement upstream is much more important. The "U" tube seen between the floscan and the spider below is HEAVY wall SS tubing and the fittings are all steel. Not so much for fire protection, as to prevent cracking due to vibration...

I know that the Floscan is supposed to be installed with the outlet above the inlet. My install in fact does that, but by a somewhat lesser amount that a traditional "wires up" installation. I believe the goal is to force any air or vaporized fuel to pass "THROUGH" the Floscan and not get trapped in the turbine meter thus impacting the measurement. It appears that my less than ideal install does that, as the numbers are always spot on!

Just how I did it, YMMV>

 

[Kahuna]

To play devil's advocate, I have to ask whether this is really a big enough issue to worry about.

Its a huge issue.
I can spend 60 gallons and not go above 1k'agl.
We practice a lot and its all down low. A weekend of flying puts these numbers in the weeds. Or, a flight high also can cause lots of pump time.
It should be right.
The data should be reliable.
The installation instructions should not recommend a place thats puts the data at risk.
Most pilots may not care about being 5 gallons off.
I do.
We fly light on fuel for best performance.
Low fuel is not easy to determine through the caps, particularly on extended 60 gallon main tanks.

 

[Mel]

Just as a data point, my transducer has been "ahead" of all pumps (right after the selector valve) for almost 18 years and my totalizer is within 1/2 gallon at every fill up.
47 gallons total including ER tanks.

 

[Kahuna]

Just as a data point, my transducer has been "ahead" of all pumps (right after the selector valve) for almost 18 years and my totalizer is within 1/2 gallon at every fill up.
47 gallons total including ER tanks.

Mel,
This thread was for injected motors with high pressure pumps.
I assume you do not have a high pressure pumps before your selector valve.

 

[Mel]

I appologise!

You are correct. I do have a carburetor.

 

[BuckWynd]

Its a huge issue.
I can spend 60 gallons and not go above 1k'agl.
We practice a lot and its all down low. A weekend of flying puts these numbers in the weeds. Or, a flight high also can cause lots of pump time.
It should be right.
The data should be reliable.
The installation instructions should not recommend a place thats puts the data at risk.
Most pilots may not care about being 5 gallons off.
I do.
We fly light on fuel for best performance.
Low fuel is not easy to determine through the caps, particularly on extended 60 gallon main tanks.

Mike,

I see that it's a huge issue for you. The type of flying you do (formation, low altitude, light fuel load), the fuel tank configuration you have, and your personal limits all add up to good reasons to need perfectly accurate data.

I was just trying to ask if anyone out there has ever verified that an extra 2 GPM of fuel is actually flowing through the transducer with the pump on, or whether there's a sensing anomoly when the unit is mounted just downstream of the electric pump. My transducer is mounted in accordance with the manufacturer's instructions, but I'd like to figure out all the options and issues. I'm not trying to criticize anyone's installation, just trying to get smart.

Being 5 gallons off after 4-5 hours of flying is definitely not acceptable, in my opinion. But being 0.5 gallons off in the same timeframe might be OK if the alternative is to re-plumb my airplane...

 

[Kahuna]

I was just trying to ask if anyone out there has ever verified that an extra 2 GPM of fuel is actually flowing through the transducer with the pump on, or whether there's a sensing anomoly when the unit is mounted just downstream of the electric pump.

No it is not flowing the 2gph. If it were there would be no problem. Time to dry would be accurate.
Also, if it were accurate, other engine parameters would also be affected by the huge increase in fuel flow.
Thats the crux of the problem. Its wrong. The device is adversly affected by the pump being on.

 

[flion]

So the transducer mounting kit for the -10 would be in the worst possible place then? Any feedback from -10 owners with this setup before I get to this point?

 

[rleffler]

So the transducer mounting kit for the -10 would be in the worst possible place then? Any feedback from -10 owners with this setup before I get to this point?

I'm just getting ready to install mine. I believe that between the spider and the fuel servo is where most of the EIS vendors are recommending as oppose to the location tha Van's specifies. If I recall correcly Rob Hickman had some photos of his RV-10 on his support site.

 

[Kahuna]

I'm just getting ready to install mine. I believe that between the spider and the fuel servo is where most of the EIS vendors are recommending as oppose to the location tha Van's specifies. If I recall correcly Rob Hickman had some photos of his RV-10 on his support site.

Well right in their(AFS) current engine monitor install guide.
"AFS has seen good results with the following mounting:
1. The transducer in a stationary location in line between the
electric boost pump and the engine driven pump."

 

[Danny King]

Where to mount the cube

Just as a data point, my transducer has been "ahead" of all pumps (right after the selector valve) for almost 18 years and my totalizer is within 1/2 gallon at every fill up.
47 gallons total including ER tanks.

Mel,
I'm fuel injected with the I0360A1B6 and I also mounted my transducer after the fuel valve but before the boost pump, and the Doll's EI fuel flow is within a half a gallon at each fill up. I used an Airborne electric boost pump that is a continuous rotary vane electric pump, not a diaphragm type of pump. Maybe that is why there is no surging. My fuel flow will jump when the boost pump is turned on or off, but the flow returns to normal in a few seconds. Like you, I have many years of flying success with this set up.

 

[rleffler]

Well right in their(AFS) current engine monitor install guide.
"AFS has seen good results with the following mounting:
1. The transducer in a stationary location in line between the
electric boost pump and the engine driven pump."

In the 4500 manual, AFS has expanded the locations to:

AFS has seen good results with the following mounting:​

• 1. The transducer in a stationary location in-line between the electric boost pump and the engine driven pump.​
  2. The transducer in a stationary location in-line between the fuel injection servo and the distribution block.​
  3. The transducer in a stationary location in-line between the Engine driven pump and the Carburetor.

Here's a picture of Rob's installation from this thread on his support site:

 

[elippse]

I have to agree with Buck's analysis. I think the case he laid out makes the most sense!

 

[Kahuna]

Mel,
I'm fuel injected with the I0360A1B6 and I also mounted my transducer after the fuel valve but before the boost pump, and the Doll's EI fuel flow is within a half a gallon at each fill up. I used an Airborne electric boost pump that is a continuous rotary vane electric pump, not a diaphragm type of pump. Maybe that is why there is no surging. My fuel flow will jump when the boost pump is turned on or off, but the flow returns to normal in a few seconds. Like you, I have many years of flying success with this set up.

Danny I think you and Mel have it right. I miss read Mels comment and took his word "ahead" to mean after. You used the word "before". Now I get what you both meant.

This makes the most sense of a location. Keeps that expensive device outta the firewall and in a location thats accurate. I like it. I shoulda moved it there. May still do it yet.

It sure makes me nervous to see those things sitting out there in high vibration areas waiting for a problem to happen. Just makes me nervous.

 

[AlexPeterson]

I mounted mine on the firewall (forward side), and it is between the servo and the flow divider. This requires one extra flex hose as compared to no fuel flow transducer. Accuracy is amazing. Pump on/off, doesn't matter. Usually within a couple tenths of a gallon per fill up.

 

[Brantel]

Keep in mind the makers of these things recommend you push fuel thru them not suck it thru.


I know several people do it that way and they have a million hrs to prove it but there is some reason why the manufacturers don't like it...

 

[Auburntsts]

OK, not arguing the point, but how does it know whether the fuel is pushed or pulled through it? I'm not an engineer so really don't have a clue.

 

[Bubblehead]

I have almost 200 hours on a Floscan unit mounted on the firewall between the boost pump and engine driven pump on my IO-360. When I turn the boost pump on the fuel flow goes up initially but within a few seconds it is back down to the original flow rate or maybe .1 gpm higher. It has been accurate within .3 gallons every time I've filled up. Admittedly most of my flying is of 1 to 4 hour duration so any inaccuracy caused during boost pump operation is masked by the small duration of pump operation compared to the flight time.

I use the standard Airflow Performance boost pump and filter and RSA-5 system.

The Floscan is level with wires up and the inlet is pretty close to a 90 degree fitting but it sure has been accurate.

I am finishing up some maintenance on my -8 and will do a test when I test fly to see how accurate it is for a flight where the boost pump is on a significant amount of time.

 

[Brantel]

The transducer has no idea.

Some things that come to mind as to why you would want to avoid this situation are:

The orifice size thru these things is pretty small and creates a pressure drop and lowers the vapor point of the fuel which can lead to vapor lock.

There are a few places on these devices where a seal could fail. If you are sucking fuel thru them, it would be like trying to suck fuel thru a straw with a hole in the side of it open to the atmosphere. If the leak were bad enough, you might have a fuel starvation issue.

Thats a couple....

OK, not arguing the point, but how does it know whether the fuel is pushed or pulled through it? I'm not an engineer so really don't have a clue.

 

[hydroguy2]

I have a "Red cube" transducer installed between the AFP servo and the spider. After I read this thread earlier I figured I should test my FF.

Boost on or Boost off= no change in FF indication. works for me.

 

[frazitl]

The difference between

sucking and blowing is important, especially when the fluid is near its vapor pressure.

In any fluid system, good design takes careful analysis to ensure that local cavitation does not occur. Normally this is in the pump itself, but any source of pressure drop (valve, flow meter, elbow, etc.) is MUCH less of a problem downstream of the pump than upstream. Cavitation in pumps causes erosion of impellers and the pump body itself. Cavitation in a flowmeter (especially turbine type) causes inaccurate readings (often too high readings), and over time, damage to the flow meter itself.

My recommendation is to put the flow meter downstream of any pumps or metering valves when possible. A straight flow path before the flowmeter is very important for accurate metering. A straight flow path after the flowmeter is also important, but less so than before the meter...

Bottom line, is that if it works for you, great. Just be aware that any failure causing an obstruction on the suction side of a pump is much more of a problem than an failure that causes an obstruction downstream of a pump...

Be safe!

 

[Lars]

I for one would love to see this!!!! Could you post it or should I PM you?

I don't have access to a scanner at the moment, but here is the text of the documentation GRT supplied to me, verbatim, from document "fuel flow EI FT-60.cdr" (note to the pedants- "dampener" is incorrect; it should be "damper", but I'm a bit pedantic myself so it was hard to ignore):

Pulsation Dampener

A pulsation dampener may be required if the fuel flow readings are erratic, or if they read higher when an electric fuel pump is turned on. This is mostly likely when the fuel system is composed of mostly metal fuel lines, and there is no trapped air in the system. A pulsation dampener can be fashioned by installing a tee in the fuel line between the electric fuel pump, and the flow sensor, and connecting a 1 or 2 foot piece of tubing to this tee, oriented vertically up. Cap the end of this tubing to trap air in this line. Most often no provision is required for a pulsation dampener, as the fuel line leading to the fuel pressure sensor traps air and performs this function.

I spent the day installing my red cube on the firewall, plumbed between the mechanical pump and the servo. The advantage to me being that I was able to make that architecture work with the hoses I had on hand. If for some reason I decide I hate the arrangement, I can re-plumb the thing so it's between servo and spider by buying throwing money at the problem... er, buying some hoses.

 

[rv72004]

Mine is on the center floor between selector and firewall. Hardly any difference. I in any case, never fly with below 10 gallons on board. Ran out of fuel once [A collapsed bladder ,not RV] and will never forget it! One day I must try run a tank dry to see what it actually goes to.

 

[Dan Langhout]

Thanks!

Sounds like exactly what you do to solve water hammer problems in residential plumbing systems. Not sure if I'd want to do that on my aircraft fuel system though.

I don't have access to a scanner at the moment, but here is the text of the documentation GRT supplied to me, verbatim, from document "fuel flow EI FT-60.cdr" (note to the pedants- "dampener" is incorrect; it should be "damper", but I'm a bit pedantic myself so it was hard to ignore):



I spent the day installing my red cube on the firewall, plumbed between the mechanical pump and the servo. The advantage to me being that I was able to make that architecture work with the hoses I had on hand. If for some reason I decide I hate the arrangement, I can re-plumb the thing so it's between servo and spider by buying throwing money at the problem... er, buying some hoses.

 

[rleffler]

I don't have access to a scanner at the moment, but here is the text of the documentation GRT supplied to me, verbatim, from document "fuel flow EI FT-60.cdr" (note to the pedants- "dampener" is incorrect; it should be "damper", but I'm a bit pedantic myself so it was hard to ignore):

I spent the day installing my red cube on the firewall, plumbed between the mechanical pump and the servo. The advantage to me being that I was able to make that architecture work with the hoses I had on hand. If for some reason I decide I hate the arrangement, I can re-plumb the thing so it's between servo and spider by buying throwing money at the problem... er, buying some hoses.

Matt Draille at Matronics sells one. The Pulsation Damper (PD) is a stainless steel sphere with a fitting on one side that connects to a 'T' AN fitting and is placed in-line with the fuel flow either just before or right after the fuel flow transducer. The PD acts like a common 'air shock absorber' isolater in a typical sink facet. The air is compressed inside of the sphere in time with the pulses in the fuel line caused by the boost pump and/or the mechanical fuel pump. This extremely simple device is incredible effective at reducing the fuel flow reading fluctuations common in turbine transducer-based electronic instrument installations. Below is a paraphrased testimonial from a customer on the use of the PD in his 0360 power RV-8 installation.

 

[F1Boss]

Sorry for the blurry image. The fitting is a custom fitting that was welded up. Works perfectly. The transducer is in a nice cool place, with no mickey mouse fittings or firesleeve wrap required.

image snipped

I would suggest that you insulate the sensor - it's gonna heat soak at a fuel stop, and give you hot-start headaches. Who needs that? Wrap the hose ends too, while you're at it; not much you can do about the flow divider.

Caveat: don't expect the fitting between the flow divider and the sensor to be able to support the sensor on your vibrating Lycosaur (especially if it is aluminum) - add a support for the sensor.

A sniffle valve will help the hot-start problem too.

Carry on!
Mark

 

[Don at Airflow]

Why the flow reading changes

So far I have not seen an answer to Mike?s question. ?Why does the flow meter read higher when the boost pump is turned on if the flow meter is installed before the engine driven fuel pump??

Here?s why.

The diaphragm fuel pump has 2 sets of diaphragms in it. The lever that runs on the cam actuates one of the diaphragms. This diaphragm has a spring on the ambient side that pushes the diaphragm down, (the lever actually just compresses the spring) thus pumping the fuel through the outlet check valve in the pump. The spring pressure times the area of the diaphragm is what sets the pump pressure to the fuel injection system (or carburetor). When you turn on the boost pump the fuel pressure from the boost pump overcomes the spring pressure in the engine driven pump and pushes the diaphragm up to it?s full travel (like when the cam pulls the lever on the pump arm collapsing the pump spring). This displacement along with all the other compliance in the hoses on the suction side of the engine pump increases the volume, thus an increase in the reading on the flow meter is seen.

As we have said time and time again, we install flow transducers on fuel-injected engines on the metered side of the system, that is, between the fuel control and the flow divider (or purge valve if you have on in the system). Additionally installing the flow meter on the suction side of the engine pump may cause instability in the reading at cruise as some vapor can be formed across the relatively small orifice in the flow meter transducer. Fuel pulsing on the suction side of the engine pump can also give erratic readings. And the higher you fly or the hotter it gets, the more this problem is prevalent.

Hope this helps clear up the mystery.

If you have questions on install of the transducer on fuel injected engines drop me an email. We have done many and they all work great. Many of the installs do not follow the published methods.

Don

 

[9GT]

Don,,,can you possibly explain why some installations seem to work just fine when located between the electric and engine driven fuel pump and others do not? Not trying to be argumentative, but we located the Flo Scan on the Cozy MKIV with a Superior IO-360 XP between the AFP electric pump and the engine driven mechanical pump and there is no error whether the electric pump is on or off with a BM EFIS One display. I just feel real hesitant locating my Red Cube FT-60 anywhere in the engine compartment of my RV-10. I made the decision (right or wrong) to locate it in the tunnel per-plans but made up a replacement fuel line just in case I need to move it due to a fuel flow error.

 

[rvmills]

So far I have not seen an answer to Mike’s question. “Why does the flow meter read higher when the boost pump is turned on if the flow meter is installed before the engine driven fuel pump”?

Here’s why...
Don

Don,

Thanks much for the explanation...good stuff!

I would suggest that you insulate the sensor - it's gonna heat soak at a fuel stop, and give you hot-start headaches. Who needs that? Wrap the hose ends too, while you're at it; not much you can do about the flow divider.

Caveat: don't expect the fitting between the flow divider and the sensor to be able to support the sensor on your vibrating Lycosaur (especially if it is aluminum) - add a support for the sensor.

A sniffle valve will help the hot-start problem too.

Carry on!
Mark

Mark,

What would you insulate the transducer and hose ends with...just wunderin'...I have the plenum off for glass repairs while I work on fairing improvements, and looked at my transducer after reading this thread. I can insulate while I'm in there, though so far no hot start issues (I do have a sniffle valve too). I also have it supported with a bracket that is adel-clamped to the adjacent push-rod tubes, and it seems to be well supported (backing up your point!)

FWIW as a data point Kahuna, my c1998 IO-540 FF transducer install has worked very well for 800+ hours (last half of the hours are mine), and fuel burn/fuel remaining on the VM-1000 is always very close to fuel pump results (on or within a couple tenths...I watch those like you and many others do!) I think you'll be happy!

Side question Mike...I looked at your website to see if the long-range tank install and plumbing is described. Have you posted anything on the site or on VAF on that?...interested in how you did it (my engine is thirsty too!)

Thanks...great thread!

Cheers,
Bob

 

[Bubblehead]

When you turn on the boost pump the fuel pressure from the boost pump overcomes the spring pressure in the engine driven pump and pushes the diaphragm up to it?s full travel (like when the cam pulls the lever on the pump arm collapsing the pump spring).
Don

Ignoring the transducer for a minute, when the boost pump is turned on does fuel flow increase and stay above the previous flowrate or does it return to or close to the original flow rate?

 

[roee]

Steady state error or transient error?

Seems to be some confusion about the observed symptoms. There are two very different possible scenarios, and there seem to be references to both in various posts. Let's clarify.

Scenario 1 is steady state error. Any time the aux fuel pump is on, the fuel flow indication reads higher than actual. It continues to read higher than actual as long as the aux fuel pump remains on.

Scenario 2 is transient error. When the aux fuel pump is turned on, the fuel flow indication reads high for a short time, and then settles back down to actual while the aux pump remains on. When the aux fuel pump is turned off, the fuel flow indication reads low for a short time, and then settles back up to actual as the aux pump remains off.

So what does the difference mean? Well... a lot.

Steady state error (scenario 1) would be more difficult to explain (instrument error associated with sensitivity to pressure or possibly reverse flow), and it would also be much more troubling in its implications for the pilot. From an engine management standpoint, it would mean that the fuel flow indications are always inaccurate any time the aux pump is on. And from a fuel management standpoint, it would result in cumulative error in the fuel remaining calculation (totalizer), making it increasingly inaccurate the longer the aux pump is on. Pretty bad.

Transient error (scenario 2) on the other hand, can be explained by a change in effective volume in the plumbing downstream of the aux pump, as Don at Airflow described. And this scenario is also not so troubling for the pilot. Fairly benign, actually. From an engine management standpoint, it means that the fuel flow indication doesn't accurately reflect the engine's fuel flow only for a short time after turning the aux pump on or off, and it will quickly settle back to an accurate reading while the aux pump remains on (or off). And from a fuel management standpoint, it's actually ideal, because it always accurately reflects the fuel flow leaving the tank, and so the fuel remaining calculation (totalizer) is always accurate.

So, let's clarify -- which scenario are people actually seeing? Those who have experienced fuel flow indication error associated with the aux fuel pump -- which scenario is it, 1 or 2?

 

[Kahuna]

This thread started on senerio #1.

Don. if I read your statement correctly, "This displacement along with all the other compliance in the hoses on the suction side of the engine pump increases the volume, thus an increase in the reading on the flow meter is seen."

If the volume is actually increasing, then your saying that the transducer is reading correctly? That is not the behavior Im seeing.

My findings are that the transducer reads a much higher flow, but other indications are that the flow has not changed at all. EGT readings, cht readings, and totalizer readings/actual fuel used.

Now while what your saying Don may in fact happen, this is your life every day. On the many installs Ive dealt with, and planes flown, I have never seen where the flow is actually increasing on a non-bypass mechanical servo injection. Im sure you have seen thousands more installs than me there is no doubt about that. I only relate my experience.

 

[Don at Airflow]

Flow meter readings

You are correct. The fuel flow to the engine does not change. This is because if the fuel control (Bendix or Airflow)if it is working correctly the fuel flow to the engine is determined by the amount of air the engine is consuming, not the inlet fuel pressure. So when you turn on the boost pump you should not see a change in EGT, CHT or engine performance. So if the flow meter is located in the metered fuel line it should read constant as the fuel flow from the output of the fuel control has damped out all the pump pulsations, compensated for inlet fuel pressure, and the device is on the pressure side of the system so vapor should not be a factor. So in the case where the flow meter reads high for a while then comes back to a lower flow after the boost pump is turned on that is exactly what the flow meter is reading. For those cases where the flow meter reads high, or incorrectly when the boost pump is turned on and does not return to a lower flow, I would suspect that there is some kind of EMI or error in the instrumentation due to current load. I am no authority on electronics so do not assume this statement is correct; It?s only a theory. Bottom line is if there is no change in EGT or engine operation and the fuel flow reading changes, it tells me that the actual fuel flow the engine is using has not changed.

Don

 

[David-aviator]

This thread started on senerio #1.

Don. if I read your statement correctly, "This displacement along with all the other compliance in the hoses on the suction side of the engine pump increases the volume, thus an increase in the reading on the flow meter is seen."

If the volume is actually increasing, then your saying that the transducer is reading correctly? That is not the behavior Im seeing.

My findings are that the transducer reads a much higher flow, but other indications are that the flow has not changed at all. EGT readings, cht readings, and totalizer readings/actual fuel used.

Now while what your saying Don may in fact happen, this is your life every day. On the many installs Ive dealt with, and planes flown, I have never seen where the flow is actually increasing on a non-bypass mechanical servo injection. Im sure you have seen thousands more installs than me there is no doubt about that. I only relate my experience.

I believe what Don said is correct.

Another way to say it, the flow indication goes up with the electric pump ON because there is a flow error due to the suction anomalies of the mechanical pump if the electric pump of OFF. The steady pressure provided by the electric pump fixes all that. The flow reading with the electric pump ON is accurate, the reading with it OFF is not. The indicated fuel used is an average of both readings.

That's also why the system works better with the flow device down stream of both pumps, preferably installed between the controller and the flow divider. I have mine installed about 8" before the controller and indications are perfect but it would be more perfect if it were installed as Don recommends.

Typically, when the tanks are filled the actual add is within .1 or .2 gallon of the calculated add based on indicated previous fuel used. I use the fuel flow system to monitor fuel burned and consider it much more reliable than the float type gages.

 

[F1Boss]

Red stuff!

Mark,

What would you insulate the transducer and hose ends with...just wunderin'...

Hey Bob:

Standard 'firesleeve' material. The sleeve for larger hoses can be sliced and diced to fit around the transducer, secured with safety wire (.025 will be fine). Fill the exposed ends of the sleeving with red rtv so the final install looks sanitary, and stays that way.

My 550 has a Cirrus factory SS box around the sensor - I think its actual heat-soak resistance is minimal, speaking from experience of trying to get the dang thing started when hot. The upside is that this lil' box will cover any of the smaller FloScan sensors we use on our ships.

Carry on!
Mark

 

[airguy]

I believe what Don said is correct.

Another way to say it, the flow indication goes up with the electric pump ON because there is a flow error due to the suction anomalies of the mechanical pump if the electric pump of OFF. The steady pressure provided by the electric pump fixes all that. The flow reading with the electric pump ON is accurate, the reading with it OFF is not. The indicated fuel used is an average of both readings.

That's also why the system works better with the flow device down stream of both pumps, preferably installed between the controller and the flow divider. I have mine installed about 8" before the controller and indications are perfect but it would be more perfect if it were installed as Don recommends.

Ok, so you're saying that during standard operation (boost pump off, sensor placed between the boost and mechanical pump), that the indicated flow is actually lower than true (due to flow errors induced by the pulsations of the mechanical pump) and when you turn on the boost pump to raise the pressure and eliminate those pulsations, that the now-higher reading is actually the true flow?

Easy to test - run two identical flight legs, one without touching the boost pump, one with the boost pump on full-time. Check ACTUAL fuel burn (gallons pumped back in the tank to full) and see which one is closer.

Typically, when the tanks are filled the actual add is within .1 or .2 gallon of the calculated add based on indicated previous fuel used. I use the fuel flow system to monitor fuel burned and consider it much more reliable than the float type gages.

OK - can you tell us exactly what your operating technique is on these flights that produce this close approximation, startup to shutdown? I'm just trying to pin down the differences between what people are seeing, and figure out what is different and where the error might be. If you get this result with the boost pump off and operating on the engine driven pump, then your first statement that this scenario produces a lower-than-true flow indication cannot be correct.