What's new
Van's Air Force

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

Fuel flow fluctuation when hot

MacCool

Well Known Member
edit: Sorry...the title is misleading. I meant to type that the gauge-displayed Fuel Pressure fluctuates when the engine is hot....

Background: RV-9A with IO-320 D1A. Fuel flow read from a Red Cube (new) driving an EMS part of an Advanced Flight 5400 EFIS. Red Cube mounted on firewall next to a Weldon 8163A boost pump. Engine driven pump is a Tempest, and was replaced about 70 hours ago.

Issue: When the engine is hot, I often get low fuel pressure when running on the engine-driven fuel pump while taxiing or on approach with manifold pressure at about 15 inches. No stuttering or anything from the engine, just low fuel pressure at the very bottom of the green arc according to the EFIS fuel pressure gauge. Doesn't do this during flight, nor when the engine is not heat-soaked. Works fine in cold weather and first flight of the day.

Trying to decide if this is a problem. The engine performs well otherwise. When I first picked up the plane, the engine-driven fuel pump went out...noted when the engine kept dying while taxiing. Replaced the original Lycoming engine-driven pump with a Tempest at that time. That was about 70 tach hours ago.

Any thoughts...?
 

Attachments

  • fuel pump.png
    fuel pump.png
    3.2 MB · Views: 396
Last edited:
Does the low fuel condition happen with the electric fuel pump on? Have you considered the proximity of the red cube to the exhaust? I would start with shielding the cube and fuel line in that area and see if anything changes. Running at idle lowers the fuel demand and it could be boiling the fuel in the line. Not sure what indications that would give you but it could be showing less liquid flow through the cube if vapors are also going through it.
 
Does the low fuel condition happen with the electric fuel pump on? Have you considered the proximity of the red cube to the exhaust? I would start with shielding the cube and fuel line in that area and see if anything changes. Running at idle lowers the fuel demand and it could be boiling the fuel in the line. Not sure what indications that would give you but it could be showing less liquid flow through the cube if vapors are also going through it.

Sorry, I titled this thread wrong and can't change it. I meant to say that fuel pressure fluctuates when hot.

You raise some important points though...the fuel pressure fluctuation does seem to be prominent only when the engine is hot and I suspect that that plays a role...might be that some insulation would help. I've noticed this on hot days (this is Minnesota, so we don't see much oppressive heat) and see it mainly on taxiing after a hot start, or on a real hot-day 5-mile approach when I pull back the manifold pressure to 15 inches to start slowing down the airplane. Although the EFIS fuel pressure gauge fluctuates and dips into the red, I've never seen the engine to stumble at all. Likewise taxiing...low pressure when hot and taxiing, but the engine never quits, or even acts like it wants to. This is unlike when the engine-driven fuel pump did die a couple of years ago...the engine wouldn't stay running while taxiing.

Pressure does come back into the green on hot-taxiing if I hit the boost pump. On take off, I generally flick off the boost pump at 1000 ft AGL and the fuel pressure remains normal.
 
Problem? Maybe . . .

Mac, Although it could be a marginal mechanical pump, IMO it is just heating of the fuel as you mention, some vaporization causing the lower pressure on the business side of the pump.

I noticed some actual FF and pressure wandering at cruise and installed 5 thermocouples from the fuel entry into the cabin to pump inlet. It increased much more than I had expected. Sorry don't remember the numbers, but your exposed pump FWF will certainly add to the temperature rise.

Is it a "problem" ?? . . . . if the standard boost pump procedures don't completely address it, yes. Otherwise just follow the procedure. Insulation of the pump head, cube, and tubing will certainly help lower the threshold for the pressure drop occurrence. You can get sheets of insulation at a Speed shop and make covers with RTV seams pretty easily. If you wanted to experiment, or more margin.

A source for firesleeve. Lower prices, several steps in the order process and and slower to fulfill but good prices and products
 
Last edited:
Thanks guys. I am inclined to suspect that fuel is boiling, but I've yet to see any actual functional problem with fuel flow or pressure (other than hard starts when hot). I'm just a little gun-shy since I did have the engine-driven fuel pump go TU on my first flight while ferrying it here from St. Louis. I am going to figure out an insulation scheme and give that a try.
 
Last edited:
Two ways to do it, stand off heat shield on the exhaust and insulation wrap on the pump and lines. I do both whenever I have exhaust near fuel lines. You can get the stand off shields from ACS and one brand is anti splat. Theirs works well.
 
Mac,

I suggest you have some basic issues that need to be fixed before you go chasing possible fuel line heat causing your pressure drop.

Looking at the photo, I note that you have the fuel line going into the (new) red cube, then into your electric boost pump, then I assume to your mechanical fuel pump. I also note a bunch of 90 degree fittings.

I refer you to the EMS install instructions on placement of the red cube (as in between the engine driven fuel pump and the engine and not mounted on the engine) and fuel line considerations on the input and output of the cube.

Carl
 
Gonna add a little to the comments. As mentioned, add radian shields where appropriate. Closer to the source would probably be better here; accessibility, no hinderance to local airflow, etc. Insulating the lines is fine. Insulating the pump is not. Even with fuel flow, pressure relief recirc, good mechanical efficiency, etc. it generates heat as a byproduct. Insulating it will affect its life.

Gonna open myself up here but I suspect the upper line is too short. I've never found a spec but it would be very reasonable to assume minimum length should be no shorter than 1/2 max unsupported length. Short and straight is the worse combo. I'd personally ditch the 90 and steam loop it or get a flex line from TSflightlines.

Keep us informed.
 
I run basically the same Weldon pump but mounted in the cabin due to heat concerns, even with a purge valve I didn't feel comfortable with the pump FWF.
 
I run basically the same Weldon pump but mounted in the cabin due to heat concerns, even with a purge valve I didn't feel comfortable with the pump FWF.

I replaced that one just last year. The first one started leaking a little after 10 years. It does get hot, so I limit the amount of time I run it...pretty much only in the pattern.

I did just order a heat shield for the exhaust...need to get some hose clamps. Anybody know the diameter of a Vetterman exhaust used on an IO-320? Save me a trip to the hangar...
 
Red cube

Per the instructions that came the red cubes (I have 2 one for the return to tank) “the wires exiting the cube body must point up !” Yours is 90* off going directly through the firewall and the little counting wheel is laying flat instead of being vertical.
I’am suspecting what’s happening is with the boost pump off the cube is in a vacuum situation as the mechanical pump is drawing fuel through the pump, when everything including the fuel gets hot forward of the firewall some fuel vapors form and hang out in the top of the wheel chamber of the cube and fail to make the counter wheel spin consistently and the incorrect orientation compounds the problem. When the boost pump is turned on the red cube’s situation changes to being a pressurized condition and fuel flow can exert enough pressure to make the counter wheel spin appropriately.
 
I apologize for mislabeling the title of this thread. It says feel flow, but the actual variations and fluctuations are in fuel pressure and I was unable to go back and change the thread title. The red cube itself appears to be working just fine. Over the last several weeks, my fuel computer showing indicated “gallons used” matches the pump gallons within 1/10th of a gallon using the default K factor of 680. The previous red cube likewise worked great in that location until it went TU recently after 10 years and I replaced it. Displayed fuel pressure however does demonstrate fluctuations when the engine is hot. I do think it’s heat related, and I am going to put a heat shield on that exhaust as well as insulate the fuel lines. I really appreciate everybody’s help and interest and I apologize for screwing up the title.
 
Last edited:
Where in your system is the pressure sending unit ? The red cube only reads flow it has absolutely nothing to do with pressure. Relationship of pressure sender position in the fuel lines configuration could have the same issue with being in a vacuum/suction versus pressurized condition.
 
Pressure senders and oil temp sit on the firewall toward the top.

There’s a service bulletin from Advanced Flight on those oil and fuel pressure sensors. I have the new Kavlico sensors but haven’t installed them while I wait to get this particular problem sorted out.
 

Attachments

  • 2BEBD2A9-D9BF-46A9-8286-B5FE9873A387.jpeg
    2BEBD2A9-D9BF-46A9-8286-B5FE9873A387.jpeg
    82.9 KB · Views: 153
Last edited:
Mac,

I suggest you have some basic issues that need to be fixed before you go chasing possible fuel line heat causing your pressure drop.

Looking at the photo, I note that you have the fuel line going into the (new) red cube, then into your electric boost pump, then I assume to your mechanical fuel pump. I also note a bunch of 90 degree fittings.

I refer you to the EMS install instructions on placement of the red cube (as in between the engine driven fuel pump and the engine and not mounted on the engine) and fuel line considerations on the input and output of the cube.

Carl
Thanks Carl. I’ll have to look at it again (I didn’t build it) but my recollection is that fuel goes from the
engine-driven fuel pump —> boost pump —> red cube —> engine. My understanding is that the firewall-forward was done by Don Swords at Don’s Dream Machines.
 
Last edited:
The way mine is built, the boost pump is inside, from the feed tanks to the electric fuel pump, mechanical fuel pump, throttle, red cube, divinding block, cylinders. Never had any issues with that set up.
 
Last edited:
Do you know for sure the fuel pressure indications are accurate?
What is the min pressure required by the engine (from the book)?
You could always change the limits displayed on the EFIS.
I tend to agree with the others, fuel pump FWF and lost of 90 deg fittings could be your problem.
Can you fabricate a cooling shroud fed with cool air from the baffles to cover the pump and try to reduce temps?
 
Right now, the only actual problem I’m having with fuel pressure is a wonky fuel pressure gauge reading when the engine is hot, as well as some mildly hard starting. I see those fluctuations primarily on taxiing back after landing, although the engine runs fine. It’s not an issue on climb out after I turn off the boost pump…pressure readings solely on the engine driven pump at that point are rock steady and in the green.

I think that the fluctuating readings and hard starting when hot are likely attributable to fuel vapor in the lines and from what I’ve read here, I’m hopeful that some exhaust shielding and fuel line insulation will help with that. That’s where I think I’ll start. The engine has run reliably for 500 tach hours with the current fuel configuration, about 250 of those hours in the Atlanta area. I did replace the Weldon boost pump at 450 hours due to a small leak, and the Red Cube at about the same time due to failure.
 
Do you know for sure the fuel pressure indications are accurate?
What is the min pressure required by the engine (from the book)?
You could always change the limits displayed on the EFIS.
I tend to agree with the others, fuel pump FWF and lost of 90 deg fittings could be your problem.
Can you fabricate a cooling shroud fed with cool air from the baffles to cover the pump and try to reduce temps?

It’s a fuel-injected engine so needs higher fuel pressures…Lycoming wants 14 psi minimum, and taxiing while hot drops me down to 8 -12 psi on the gauge. I don’t know if the readings are accurate in an absolute sense, only that they are repeatably wonky/low at 900-1100 rpm when the engine is hot, repeatable rock-steady in the middle of the green when the engine is cold/cool.

Fuel line routing and configuration, and especially boost pump location, might be an issue, but it would be a PITA to re-configure and although I appreciate the input, I’m gonna put those fairly far down the list of things to try.
 
Last edited:
As an AP I would like to chime in here.

First of all I see a HUGE fire risk with the piece of soft aluminium tubing between the pump and the Red Cube. It is simply not good practice and should be taken out before further flight.

Then like others have mentioned.
1) Get rid of those 90 degree fittings. You just have to.
2) Make sure you use proper fuel lines and fire sleeve them.
3) I would install the electric boost pump and the red cube behind the firewall.
4) Install the Red Cube with the leads pointing up.
5) Make sure that there is a restrictor installed between the fitting on the engine pump and the line running to your FP sensor.

Questions:
1) Where is your main fuel filter installed?
2) What is your fuel line like from the red Cube onwards to the injection body and how is it routed?
3) What is your fuel line like and how is it routed from the injection body to the spider on top of the engine?
4) How is your system configured/plumbed?

One good example of the configuration is: Fuel selector - fuel filter - Boost pump - Red cube - engine pump - FI body - spider. there are also others.
 
From the FT-60 (Red Cube) install manual:

Mechanical Interface:
If the aircraft has a fuel pump(s), the flow transducer MUST be installed downstream of the last fuel pump.
Installing the transducer upstream of the fuel pump(s) can cause vapor lock and jumpy/inaccurate readings.
The Red Cube FT-60 has 1/4" NPT ports. Apply thread sealant to fittings, assemble and torque fittings to
8-10 ft. Lbs., DO NOT EXCEED a torque of 12 ft. Lbs. The Red Cube FT-60 should NOT be installed with the wires pointing DOWN (the best situation is with wires pointing UP). Also, the fuel line on the outlet port should not drop down after exiting the transducer. Both of these configurations can trap bubbles in the transducer causing jumpy readings. The inlet port, outlet port and flow direction are marked on the top of the Red Cube FT-60.

Here is one way to meet the rules (and yes I know the steel fitting need to be aluminum):
BD6-AD8-CB-3-F6-A-431-D-A110-EBF840515-A50.jpg
 
Last edited:
Ahead of the firewall; best practice is to use steel fittings only. That should include the fittings on the ends of the hoses also.
 
What happens if you fly inverted? And the RV10 stock had the transducer in the tunnel. Kinda goes against the "guidelines"
 
What happens if you fly inverted? And the RV10 stock had the transducer in the tunnel. Kinda goes against the "guidelines"

Yep - when inverted I suspect having an error in fuel flow reading will be low on your priority list.

Yep - the RV-10 plans do put the transducer in the tunnel, and yep it gives erroneous reading when the boost pump is on, and yep that placement provides an unnecessary fuel pressure drop to the inlet of the mechanical fuel pump. All this is first hand knowledge.

Now look at the RV-14 FWF instructions that put the cube between the fuel pump and the Fuel Manager on the engine. Good placement except the instructions have you hard mount the cube to the engine in a high heat area (goes against the guidelines). Here too I have first hand knowledge of the fuel flow going to ~11GPH when above 7K’, regardless of mixture setting. Moving the cube to a proper location solved the problem. Note - other RV-14 builders have not reported this problem but I have second hand information that the same issue presented itself on an RV-12.

We changed out the red cube, no joy. Engaging the appropriate vendors yielded no technical feedback. We can only postulate that the combination of vibration, heat, atmospheric pressure along with inlet and outlet hose lengths set up some kind of resonance that the little spinning wheel did not like.

We do know that meeting the guidelines eliminated the problem - so that is how I install the flow meter.

Carl
 
Fuel Pressure

I too run a Wheldon pump from within the cab through the Facet pump to the gascolator I have no engine driven pump so there is no vacuum in the line in the engine bay. Has been very reliable and no problems, at take off pressure drops from 6 psi to 5psi which activates Low Pressure warning, you could change you Red line settings in the Efis but manufacturers warn that below 4 Psi these sensors become out of tolerance. I don’t like turns bends and loops in a vacuum line with heat as long as you have booster pump on at take of and landing it is probably ok👍
 
Mac,

I suggest you have some basic issues that need to be fixed before you go chasing possible fuel line heat causing your pressure drop.

Looking at the photo, I note that you have the fuel line going into the (new) red cube, then into your electric boost pump, then I assume to your mechanical fuel pump. I also note a bunch of 90 degree fittings.

I refer you to the EMS install instructions on placement of the red cube (as in between the engine driven fuel pump and the engine and not mounted on the engine) and fuel line considerations on the input and output of the cube.

Carl
Carl, I pulled the cowl a couple of days ago to review the fuel line routing. I found that you are correct...the fuel line goes firewall -> Red Cube -> boost pump -> engine-driven pump -> fuel servo -> engine. Fuel pressure readings come from a T-fitting off the engine-driven pump and routed to the fuel pressure sensor. This was apparently the routing chosen by the guy (Don Swords) who did the firewall-forward work for the original builder 11 years ago. Performance on fuel flow readings isn't affected. I'm not sure if this arrangement causes or contributes to the hot-engine fuel pressure issues I'm seeing (gauge fluctuation and mildly hard hot-weather starting), but I think I'm going to start with insulating the exposed fuel lines and figure out a way to insulate the Red Cube. I appreciate your input.
 
I have stayed--

away from this discussion, because of our transducer mounting kit. BUt in this case, I'll bring up something that may have been overlooked. I know I did, until last year. Vans engineering when designing the RV10 and the RV14, went against the 'recommendations' of EI on the location and plumbing of the FT60.
In the RV10, the outlet of the transducer isnt 2+ inches and the OE Rigid tube makes a 90* vertical bend. Against EI's recommendation.
The RV14 uses a AN823-6D 45* adapter on the inlet port, again going against the recommendations.

So here is a kit manufacturer that has several engineers that decided that placing the transducers in those locations was fine. And---I kinda think that for the most part, they work just fine. Yes, everynow and then you'll get an anomalous unit, but givent he number of these that are flying, I think the crazy reading numbers are relatively small. If they werent, I think Vans would have put out an SB on the subject.

YES---we went against the 'recommendations' too, bu mounting it at the flow divider. (Actually my reason was laziness.) In mounting in in a lower location like near the servo you have to have 2 hoses, 1 long to the flow divider and one short, from the servo. The short one is ALWAYS a PITA to get right---ask Dan Horton, and a few others. So gee, lets eliminate 1 hose, but you still have to connect to somehow to the system. In our case, YES the wires are NOT pointed up, but evereything seems to work just fine. What happens it you have them pointed UP and make a banked turn? UH---the wires are not up. Does the unit quit, or stop flow? Doesnt seem so. Even sustained inverted flight, it doesnt stop the flow.

So 'recommendations' are just that in my opinion for whatever thats worth. But, if you are going to deviate, then its kinda important to think out of the box and look at things from all angles, including inverted.

Flame war starts now---

Tom
 
I have fuel pressure fluctuations on my IO-360 equipped RV-7. New Tempest fuel pump about 40 hours ago and original Weldon electric pump in lower part of engine bay. Replaced the fuel pressure sensor. Skyview reads low and fluctuating fuel pressure at various times except on first part of first flight of the day when temps are cool. On very warm afternoons, engine is rough on taxi with boost pump off. Often shows low fuel pressure in cruise flight with boost pump off. Never runs rough, though. Firesleeve on the fuel lines and recently put reflective wrap on the hose between the pump. Gascolator mounted low on firewall and is cleaned regularly. No blast tubes on Tempest pump, yet. Looks like a real hassle to install one of those cooling shrouds. I installed the Tempest pump trying to solve this problem but no change. I probably replaced a perfectly good pump. Does this sound like a heat problem?
 
Greg,

New problem or always there?
Any indication of an air leak (as in the pump is sucking some air with the fuel)?
Are there too many 90 degree fuel fittings?

Carl
 
In my case (Tempest pump/Weldon boost), I found that insulating the fuel lines, a couple of heat shields on the proximate exhaust, and finally replacing the fuel (and oil) pressure sensors with the Kavlico P255-150G-E4A (150 psi) according to the AFS/Dynon Service Bulletin from back in May 2020 solved my fuel pressure fluctuation issue.
 
My fuel pressure issue has been going on for quite some time (5 yrs +) but due to my location, I don't often see it unless I fly someplace warmer and land for a fuel stop mid-day. Turning on the boost pump solves the issue 100%. I have inspected lines for blue stains but am not sure how to find a suction leak in the fuel lines. I also replaced the sensor per Dynon. I'll look for 90 degree fittings.
The braided steel fuel line between the boost pump and the replaced Tempest fuel pump is recently sleeved with reflective insulation and I cannot tell any difference there. How can I pressurize all the lines and fittings to inspect for a leak without ruining some other components?
 
After a recent flight in cool weather with cowl flaps closed and oil shutter partially closed to increase temperatures inside, I had no fuel pressure fluctuations when turning off the boost pump.

I also watched the pressure indicated on the Skyview after engine shutdown with throttle & mixture pulled out. It only lost a few #s of pressure after 30 minutes of post flight cleanup etc.

I then kept the throttle & mixture out or off, turned back on the boost pump, watched the pressure go back up to normal green range (Fuel injected engine 30 psi, if I recall). I used a strong light and looked all around for fuel leaking, particularly around the engine driven pump fittings. I was thinking that any suction leak there would also leak fuel with pressure supplied by the boost pump. After 5 minutes of running that pump, no leaks, no drips, no blue stains.

I'm still at a loss for the fluctuating fuel pressure on hot days. If it is localized fuel boiling, then my hopes of using non-ethanol MoGas aren't very good.
 
Fuel Pressure.

After a recent flight in cool weather with cowl flaps closed and oil shutter partially closed to increase temperatures inside, I had no fuel pressure fluctuations when turning off the boost pump.

I also watched the pressure indicated on the Skyview after engine shutdown with throttle & mixture pulled out. It only lost a few #s of pressure after 30 minutes of post flight cleanup etc.

I then kept the throttle & mixture out or off, turned back on the boost pump, watched the pressure go back up to normal green range (Fuel injected engine 30 psi, if I recall). I used a strong light and looked all around for fuel leaking, particularly around the engine driven pump fittings. I was thinking that any suction leak there would also leak fuel with pressure supplied by the boost pump. After 5 minutes of running that pump, no leaks, no drips, no blue stains.

I'm still at a loss for the fluctuating fuel pressure on hot days. If it is localized fuel boiling, then my hopes of using non-ethanol MoGas aren't very good.

You can get vapor lock with 100LL on very hot days. Same with autogas. The solution is to do what the auto industry did years ago. Keep the fuel from the tank to the engine under pressure. Just about impossible to get vapor lock when the fuel is "pushed" to the engine. Put in a secondary pump that runs anytime the fuel pressure at the carb/fuel body goes below a set point. Even 5PSI in an injected engine will solve this problem

I run 100% 93 OCT Pump autogas in my RV-7A with zero issues using this technique.
 
Fuel pump shroud

Does anyone make a fuel pump cooling shroud that fits a tempest pump for an o-320 lyc? Replaced the original mechanical pump with the tempest and now the shroud doesn’t fit
 
:pDoes anyone know if someone make a cooling shroud for an 0-320 with a tempest mechanical fuel pump
 
Back
Top