Fuel boiling at hot idle

I have 60 hours on my RV-8, Mattituck IO-360 (vertical induction) with Silver Hawk injection and two P-mags. After a few teething pains (to be expected), the thing runs great. The only outstanding issue is that when taxiing with hot engine, it starts to miss at low RPM settings after 3-4 minutes. This only occurs with a hot engine, on the ground, at low RPM, with ambient temperatures > 60F or so. I'm pretty sure this is due to fuel boiling somewhere in the lines, but I'm not sure what to do about it.

Here is some relevant information:

• Engine only misses when hot, on the ground, low RPM, outside air temperature > 60F or so.
• Increasing the RPM to 1100+ makes the problem go away.
• Temperature under the cowl, at the top near the oil filler door runs about 50-60F higher than outside air temperature. So it's 110-120F in there when this is happening, although it might be cooler above the cylinders on the forward side of the baffling.
• Turning on the electric boost pump has no effect, so the problem is not occuring upstream of the mechanical pump.
• Adjusting the mixture (manually and/or idle mixture screw) has no effect.
• Every fuel line firewall forward is firesleeved except the lines to the injector nozzles.
• No gascolator installed.
• Hot starts are never a problem -- only idling when hot.

I've talked to some experts and read just about every post I can find (on multiple forums) regarding this issue, but I still have some questions.

1) If this only happens on the ground with very little air moving over the engine, why would installing a blast tube to the mechanical pump make any difference? There wouldn't be much air miving through it...

2) Is there an easy way to tell where the fuel is boiling? Between the mechanical pump and servo, servo and divider, or divider and injectors?

3) At what temperature should 100LL be boiling? My under-cowl temperature probe is moveable, so I can put it just about anywhere to take measurements.

4) Has anyone tried putting firesleeve on the injector distribution lines? Do they even make it that small?

5) Should I just resign myself to the fact that a tightly-cowled engine is going to do this at idle sometimes? I never experienced this when operating similar engines in much hotter climates, but they weren't nearly as tightly-cowled as the RV-8.

6) Does anyone else experience this problem?

Any and all comments/ideas welcome.

Thanks.

Interesting

I have an RV 7a with a Mattituck and an AFP system. I also run mogas exclusively and I don't experience this problem.

The only difference between my system and yours is that you run a mechanical pump (I don't have one), so while there may not be vapour locking going on at the inlet to the mechanical pump...It may be the engine block is transferring enough heat thru the pump to the fuel so that the fuel is warmer than it is in my system when it hits the flow divider.

Almost certainly the fuel is boiling in the SS feed lines to the injectors as there is a lot of surface area and very little volume...This is also the area of lowest pressure in a system that is at idle.

I would go get some small bore (1.8th or 3/16th) rubber hose and slit it down one side and slip over the lines..You can add a few wraps of fine safety wire (like the 020 wire) to keep it in place.

My bet is this will help a lot.

Frank

Geoff,

Here are my thoughts in no particular order.

The boiling temperature of 100LL is not fixed in your system, because it's not all at the same pressure. Areas of low pressure will have a lower boiling temp.

If you have any peaks in your tubing--a high spot followed by a low spot-- you invite boiling at that peak. It's a locallized low pressure area both because of it being higher and because of the flow characteristics. This is particularly true on the low pressure side of the system, but is also possible when the other parts are running at low pressure. i.e. the small tubing lines from the spider to the nozzles while at idle. These small lines are also more sensitive to little bits of vaporization, since the gas (vaporized fuel, not gasoline) bubble can easily be big enough to fill the whole diameter of the tube instead of just a small portion of it.

Frank's comments about the relationship between surface area, cross sectional area, and flow rate at this location make a lot of sense. Basically, you've got a little bit of fuel spending a lot of time (low flow rate), in contact with a lot of hot tubing surface area.

Any chance you could move your rich mixture stop a bit richer to prevent this? It might increase the fuel flow enough to prevent it.

Have you ruled out an ignition problem?

Cheers,
Guy

Geoff,

Here are my thoughts:

1. I second the thought that the boiling is probably occurring downstream of the fuel servo, causing "burps" of fuel to go in the cylinders in this low pressure area. I don't know the Silver Hawk system, but check with the manufacturer to see if you can put smaller restrictors in all the injectors. (I did this with my AFP system, and it greatly improved hot idle.) These smaller injectors will result in an increased pressure downstream of the servo. However, the idle mixture needed to be adjusted after this. Also, you will need to verify that the system will still deliver full throttle, full rich fuel flows without the restrictors limiting it.

2. I would caution against putting any insulation on the small injector lines. It could change the vibrational characteristics such that a line goes into resonance. It goes without saying how problematic that would be.

3. I don't know the typical idle pressure with the Silver Hawk system, but if it similar to the AFP system, the boiling temp of 100LL at that pressure (around 1 to 2 psi? - can't remember) is only around 145F. I have a chart of vapor pressure vs temp for 100 LL somewhere, I'll look for it.

Good luck - anywhere (except for the little injector lines) you can insulate things will help.

You were wanting to know what the boiling temperature of 100LL is. Why not put some 100ll in a pot and place it on the stove with a thermometer and watch till it boils and note the temperature. You had best make sure your wife is not around she might object to you using her pots and stove.

Um, gas vapor + air = boom

I surely would not boil fuel inside. It should be a handbook thing anyway. Rick 90432

Norman CYYJ said:

You were wanting to know what the boiling temperature of 100LL is. Why not put some 100ll in a pot and place it on the stove with a thermometer and watch till it boils and note the temperature. You had best make sure your wife is not around she might object to you using her pots and stove.

Click to expand...

do what

Why would a partial vapor lock go away by increasing rpm unless the mechanical fuel pump pressure is low. You said the electric pump ON does not change the situation so it is not vapor lock due to low fuel pressure.

It may not be a fuel problem at all.

David-aviator said:

Why would a partial vapor lock go away by increasing rpm unless the mechanical fuel pump pressure is low. You said the electric pump ON does not change the situation so it is not vapor lock due to low fuel pressure.

It may not be a fuel problem at all.

Click to expand...

David, while the fuel pressure leading up to the fuel servo is around 20 to 25 psi, the regulated pressure to the injectors at idle is probably in the neighborhood of 1 psi. That's all a fuel servo is, a precision pressure regulator.

What Geoff is probably experiencing is quite common on FI RV's. The fuel in the small lines above the engine boils, momentarily burping out some fuel. It is not vapor lock, but an inconsistent fuel flow. The engine gets too much fuel, then too little, and so forth.

The problem goes away when the rpm is increased simply because the pressure increases accordingly. Additionally, the increased fuel flow reduces the residence time of the fuel in the "hot" area, reducing its temperature.

Geoff,

I have no solutions, but just know you are not alone. I have the same set up as you, IO-360, Silverhawk FI, and I've been flying my plane for over 4yrs. Every summer I get the exact same symptoms you describe, exactly. I don't like it, but I've learned to live with it. Just too tightly cowled and no way for heat to escape while on the ground.

AlexPeterson said:

1. I second the thought that the boiling is probably occurring downstream of the fuel servo, causing "burps" of fuel to go in the cylinders in this low pressure area. I don't know the Silver Hawk system, but check with the manufacturer to see if you can put smaller restrictors in all the injectors. (I did this with my AFP system, and it greatly improved hot idle.) These smaller injectors will result in an increased pressure downstream of the servo. However, the idle mixture needed to be adjusted after this. Also, you will need to verify that the system will still deliver full throttle, full rich fuel flows without the restrictors limiting it.

Click to expand...

Silver Hawk is the same thing as Precision (Bendix RSA) injection with the word "experimental" stamped on the front of it... I thought about smaller restrictors, but Mattituck advised against it unless I could be sure that the smaller ones would still deliver enough fuel at full power. I've read about smaller restrictors helping the hot idle problem, but I'm hesitant to do that right off the bat because my injectors are currently very well balanced (< 0.2 gph) for LOP operations and I don't want to mess that up unnecessarily.

2. I would caution against putting any insulation on the small injector lines. It could change the vibrational characteristics such that a line goes into resonance. It goes without saying how problematic that would be.

Click to expand...

Understood. There is even an AD on the books about proper mid-span clamping of these lines to prevent vibrational cracking. On the other hand, Mattituck suggested that I try firesleeve here.

3. I don't know the typical idle pressure with the Silver Hawk system, but if it similar to the AFP system, the boiling temp of 100LL at that pressure (around 1 to 2 psi? - can't remember) is only around 145F. I have a chart of vapor pressure vs temp for 100 LL somewhere, I'll look for it.

Click to expand...

According to the manual, "...metered fuel pressure at the nozzle is negligible at the low fuel flows required at and just above idle." Also, the installation guide for the servo says this about fuel temp: "120F recommended max at inlet to servo to minimize vapor formation downstream of the servo....It is assumed that temperatures may exceed the above listed temperatures under some conditions, but these temperatures may result in degraded performance at low engine speeds."

120F is about the temperature under my cowl during taxi in, but I find it hard to believe that fuel at ambient temperature (60F cooler) could heat up that much so quickly. Then again, maybe it could... Maybe I should clamp my under-cowl temperature probe on one of the pushrod tubes to see how hot it's actually getting above the cylinders.

Do you concur with my thought that a blast tube on the mechanical pump won't do a whole lot of anything at idle speeds on the ground?

Thanks for your reply..

Guy Prevost said:

Any chance you could move your rich mixture stop a bit richer to prevent this? It might increase the fuel flow enough to prevent it.

Click to expand...

I presently have *zero* RPM rise at shutdown, so I suppose I could enrichen the idle mixture somewhat. However, the last time I tried it I went as far as 8-10 clicks on the adjustment wheel and still never got much of an RPM rise -- but I did get lots of exhaust popping at idle on approach and some during taxi. I backed it off after that until the popping went away. Even with no RPM rise, the engine accelerates smoothly and quickly from idle, so I have left it alone for the past few months.

Have you ruled out an ignition problem?

Click to expand...

Not scientifically, but I have no reason to suspect the ignitions since the engine runs perfectly in every instance except for hot idle.

I suspect I'll just have to live with it. I don't feel that it's unsafe -- just annoying.

tobinbasford said:

Geoff,

I have no solutions, but just know you are not alone. I have the same set up as you, IO-360, Silverhawk FI, and I've been flying my plane for over 4yrs. Every summer I get the exact same symptoms you describe, exactly. I don't like it, but I've learned to live with it. Just too tightly cowled and no way for heat to escape while on the ground.

Click to expand...

Yeah, everything I've read (Lycoming manuals, Silver Hawk manual, Sacramento Sky Ranch manual, etc.) all say that to alleviate this problem during prolonged taxi in hot weather you should just increase RPM until it goes away -- pump more fuel through the lines faster so it doesn't have a chance to boil.

I'm just a little surprised that "hot" weather in my case is 60F. But since that translates to 120F under the cowl for me, that probably counts as hot.

Thanks...

Norman CYYJ said:

You were wanting to know what the boiling temperature of 100LL is. Why not put some 100ll in a pot and place it on the stove with a thermometer and watch till it boils and note the temperature. You had best make sure your wife is not around she might object to you using her pots and stove.

Click to expand...

She also might object to not having a house anymore after I blow it off the foundation...

MSDS boiling point

AlexPeterson said:

the boiling temp of 100LL at that pressure (around 1 to 2 psi? - can't remember) is only around 145F. I have a chart of vapor pressure vs temp for 100 LL somewhere, I'll look for it.

Click to expand...

I found some MSDS information on the web. They all report an initial boiling point and a final boiling point. I assume these are for standard atmospheric pressure, but that information wasn't readily obvious to me by reading the MSDS. Maybe if I were more familiar with the testing methods... Anyway...

Shell reports an initial boiling point of 25C and a final point of 170C.
BP reports 40C-170C.
Conoco-Phillips reports 32C-170C.

I suppose you might be right about blowing the house of the foundation if you weren't careful. The reason I didn't suggest going outside and using a hot plate to boil the 100LL is that if there is any wind blowing it would cause a low pressure area in the pot causing the 100LL to boil at a lower temperature and you would not get be able to get an accurate reading.

AlexPeterson said:

Geoff,

Here are my thoughts:

2. I would caution against putting any insulation on the small injector lines. It could change the vibrational characteristics such that a line goes into resonance. It goes without saying how problematic that would be.

.

Click to expand...

adding insulation (rubber hose) will only lower the resonant frequency of any vibrations and add a dampening effect...Highly unlikely this will be bad...IMHO.

Frank

And besides

If it works it will be a very easy solution and you could add another clamp to the middle of the span if you were nervous about it.

Frank

P-mags are temp sesitive.

I would really think if your P-mags are getting adequate cooling. They clearly are sensitive to overheating. hence P-mag's concern about the blast tubes. Also, they have started storing the highest temp they experience. You can read this with the Eicad software.

frankh said:

adding insulation (rubber hose) will only lower the resonant frequency of any vibrations and add a dampening effect...Highly unlikely this will be bad...IMHO.

Frank

Click to expand...

Frank, perhaps... But, it cannot be said that lowering the resonant frequency is better, unless you know all the various excitation frequencies, and see that it is moving away from any concern areas. You might be moving it right into a sweet spot. It is not an area that I would recommend fooling around with. If the heat problem was anything more than a ground idle nuisance, it would be worth investigating what affect insulation has on vibration of these rather important lines.

AlexPeterson said:

David, while the fuel pressure leading up to the fuel servo is around 20 to 25 psi, the regulated pressure to the injectors at idle is probably in the neighborhood of 1 psi. That's all a fuel servo is, a precision pressure regulator.

What Geoff is probably experiencing is quite common on FI RV's. The fuel in the small lines above the engine boils, momentarily burping out some fuel. It is not vapor lock, but an inconsistent fuel flow. The engine gets too much fuel, then too little, and so forth.

The problem goes away when the rpm is increased simply because the pressure increases accordingly. Additionally, the increased fuel flow reduces the residence time of the fuel in the "hot" area, reducing its temperature.

Click to expand...

Thanks. That makes sense.

My experience with Lycoming was carbs and Ellison. Both worked quite well, all things considered.

Anyone ever tried this modification:

http://lycoming.com/support/publications/service-instructions/pdfs/SI1502A.pdf

Or, is it even relevant to this problem?

Heinrich Gerhardt

With the exception of AFP injectors and distribution block, I have the exact same setup as you (Bendix FI, dual P-Mags). I get this rough loping idle on a very hot day and a long taxi. What seems to help most is pulling that mixture knob almost to the kill point as soon as I'm off the runway. I only have this situation when its real hot (above 80F OAT). I've pretty much learned to deal with it and have declared it a non-issue. BTW, Napa sells some black firesleeve that is 3/8" diameter (they don't call it firesleeve but it is glass lined silicone tubing), also Summit Racing sells small diameter firesleeve. I used the NAPA firesleeve on my plane as wiring conduit in some FWF locations(sensor lines coming off of EGT's, ignition leads, etc). I have not used it on the injector lines, but if I was going to sleeve the injector lines with something besides the rubber tubing that Frank suggested, that is what I'd use.

Wow that's hot!

Bob Brown said:

With the exception of AFP injectors and distribution block, I have the exact same setup as you (Bendix FI, dual P-Mags). I get this rough loping idle on a very hot day and a long taxi. What seems to help most is pulling that mixture knob almost to the kill point as soon as I'm off the runway. I only have this situation when its real hot (above 80F OAT). I've pretty much learned to deal with it and have declared it a non-issue. BTW, Napa sells some black firesleeve that is 3/8" diameter (they don't call it firesleeve but it is glass lined silicone tubing), also Summit Racing sells small diameter firesleeve. I used the NAPA firesleeve on my plane as wiring conduit in some FWF locations(sensor lines coming off of EGT's, ignition leads, etc). I have not used it on the injector lines, but if I was going to sleeve the injector lines with something besides the rubber tubing that Frank suggested, that is what I'd use.

Click to expand...

Thanks for the data point.. I lean very aggressively on the ground as well.


Today was the first day over 80F here this year, so I took some new measurements. I moved my under-cowl temperature probe so now it's right next to the mechanical fuel pump. I was quite surprised to see this after only a 4-minute taxi-in:
Before anyone asks, the other engine parameters were all fine. CHTs in the high 200s and oil temp 180F.


So after I shut down I got one of those laser pointer thermometers and started pointing it at things:

This is a good 25-30F hotter than I ever saw when the thermometer was up near the top of the cowling by the oil filler tube. That surprised me, as I figured the heat would rise and it would be hotter near the top.

Anyway, I'm contemplating louvers, but I really think the issue is lack of airflow around the engine on the ground. Also, louvers might help the fat, already-firesleeved fuel line temperatures, but they wouldn't do a thing for the nozzle distribution lines that are inside the plenum.

I just can't get over the fact that when I used to flight instruct in Houston, we flew IO-360s and IO-540s all day with lots of taxi time and I never encountered a single vapor issue. Hmm..

Geoff said:

snip
[/INDENT]

Servo body 150F.​

snip

Click to expand...

The servo body certainly makes for a large fuel heater when the fuel is only at idle flow rates. For a while on my AFP servo, I had a radiative heat shield installed. I made it from .032" 2024. It was more or less a cylinder, with cutouts for the various linkages, etc.. I put little blobs of rtv on the inside of the shield, to "stand it off" from the servo. I safety wired it in place - very thoroughly. Obviously, it is important to have the installation bombproof, such that it cannot foul any linkage. There are hot exhaust pipes staring at the servo from three sides, and I know there is some amount of radiative heat transfer going on. I don't believe I have a picture of this install. I took it off during one annual and never put it on (the smaller restrictors seemed to have fixed the problem enough).

I still get a somewhat crummy sounding idle at AOT's of 80 or so and greater, but I just ignore it and occasionally during taxi give her some throttle to clear it out. It is normal to get a small cough during initial throttle up for take off also, but once the rpm is coming up, it quickly clears.

Summer Idle blues

Reviving this thread as we get deeper into summer. I have the exact same experience as Alex and have the same AFP hardware. There appears to be no doubt we are boiling the fuel on hot days with hot engines.

If I pull the throttle all the way back, I have to pull the mixture all the way back to keep her from stalling from flooding. I have already leaned the idle to the point of getting a pretty decent stumble when transitioning from idle circuit to the power circuit. When I want more throttle, I have to be sure and move the mixture forward again or she will stall from being too lean. When you are in a stop and go taxi situation, the left hand can get pretty busy. Anyway it seems we have a pretty good idea what the problem is, the question is what are the most effective ways to stop the heat.

Has anyone tried turning their AFP Distributor upside down? If so did it help?

Has anyone covered their injection lines? If so, with what and did it make a significant difference?

I am just guessing here, but it seems the two biggest heat sinks are the mechanical fuel pump and the fuel controller. Shields for the fuel pump are available, but are they effective? I think many of the IO certified aircraft use these, maybe that is why they idle better. Shielding the controller is more problematic due to moving linkage and might not be necessary if the fuel pump is handled effectively. Do Certified aircraft use a special gasket material where the pump meets the block to reduce heat transfer? Frank noted he does not have this problem and he does not use a mechanical fuel pump. Again that makes me think if we effectively isolate the fuel pump from its 175F environment, we might be a long ways toward smoother idle.

I will soon be taking steps to mitigate heat to the fuel system, but want to try the most effective things first. So, if anybody out there has already slayed this dragon, please share your experience.

I would also agree with the others here that this is not a safety issue. It is just a pain on the ground. As soon as you put the system in flight mode, it is smooth as butter. I still prefer Injection over carburetors for a number of reasons. If I can just get rid of this summer loppy idle, it will be perfect!

Geoff,

While flying off my 40 hours in my O-290 powered RV-9 I had thee engine stoppages caused by the same thing, one was even on the ground before I figured it out.

I have a 180 degree loop in my fuel line between the gascolator and the fuel flow meter. Once I insulated it with firesleeve, the problem went away.

I hope that helps.

Would Ceramic Coating the Exhaust do the Trick?

I wonder if ceramic coating on the exhaust would fix the problem?

I have the exact same loop as Bill R. in my 3/8" fuel line from the gascolator to the fuel flow sensor. I would like to avoid the engine stopages Bill experienced and cover the line with fire sleeve per Bill's suggestion and am hoping not to have to re-make the tube. So, I wonder what size (inside diameter) fire sleeve could I slip over the existing 3/8" fittings?

Bill, how did you add the fire sleeve to the loop?

Don,

I'm assuming that if you can run the engine up and take off the problem is quickly remedied by the rapid cooling of the inside of the cowling, and the increased fuel pressure and volume of cool fuel?

Anyone had problems on take off run up?

647jc said:

Bill, how did you add the fire sleeve to the loop?

Click to expand...

Joe,

I took an extra "nut" down to the local speed shop and bought hose just large enough to slip over the nut. I then cut it to the proper length and slid it over the existing nuts and zip tied it in place. Unfortunetly I don't recall the size.

It took me all of 10 minutes to remove the loop, slide the hose on, and re-install the loop.

Exactly

Geoff said:

Thanks for the data point.. I lean very aggressively on the ground as well.


Today was the first day over 80F here this year, so I took some new measurements. I moved my under-cowl temperature probe so now it's right next to the mechanical fuel pump. I was quite surprised to see this after only a 4-minute taxi-in:

OAT 85F.​

Temp under cowl next to mechanical pump 175F!!!​

Before anyone asks, the other engine parameters were all fine. CHTs in the high 200s and oil temp 180F.


So after I shut down I got one of those laser pointer thermometers and started pointing it at things:

Mechanical pump 175F.​

Fuel lines into and out of the mechanical pump 175F.​

Servo body 150F.​

This is a good 25-30F hotter than I ever saw when the thermometer was up near the top of the cowling by the oil filler tube. That surprised me, as I figured the heat would rise and it would be hotter near the top.

Anyway, I'm contemplating louvers, but I really think the issue is lack of airflow around the engine on the ground. Also, louvers might help the fat, already-firesleeved fuel line temperatures, but they wouldn't do a thing for the nozzle distribution lines that are inside the plenum.

I just can't get over the fact that when I used to flight instruct in Houston, we flew IO-360s and IO-540s all day with lots of taxi time and I never encountered a single vapor issue. Hmm..

Click to expand...

Thats why I don't have a mechanical fuel pump and my electric pumps are in the wingroots.

And noght idle at any temperature or altitude on Mogas so far.

Frank

oops

No rough idle at any temp or altiude on mogas.