What's new
Van's Air Force

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

sniffle valve required?

William, I have an elbow attached to my intake manifold port, which is then connected to a short piece of fuel line plumbed to a Lycoming LW-75444 sniffle valve which lets the fuel vent out at the aft end of the lower cowl. I just have a drip pan I put under the cowl after the flight.


How is the line attached to the engine at the aft end?
 
Disagreeing with DanH is sure way to getting in trouble but I must disagree as well. The combustible mixture to be worried about is the residual fuel laying in your Sump. I’ve heard reputable stories of catastrophic events like blowing sumps apart while starting an engine with ponded fuel present. I had my own experience before installing a sniffle valve where the fuel/fumes ignited and caused a very loud boom/explosion and upon inspection I found the pressure from the explosion had pushed 2 of the orange silicone o rings off of their respective intake tubes and their shredded remains were laying in the bottom of the cowl. I think it’s foolish not to have some sort of intake drain on horizontal induction engines. I don’t think the factory would be using them if they didn’t think they were necessary. Just because you haven’t had an explosive event it doesn’t mean you can’t have one. The source of the ponded fuel is another discussion but I’ll just mention some injection systems leak/bypass fuel long after their shut down.
Duff

I think Dan already answered your question, but in a root cause kind of way. If he opens the purge valve there is no pressure at the injectors, and no way to push fuel into the intake tubes and sump. Result: no pool. However, if one pushes mixture and leaves on the fuel pump by accident, even Dan will have trouble with fuel pooling.

I still don't understand what Vic is getting at, the physical sequence of events that will lead to some issue. Surely, Dan has had a hundred shutdowns on that engine when the photo was taken.

I am surprised it is so clean though. Not even any oil goo.
 
Last edited:
Dan,
My thoughts exactly on the ability of a wasted spark being able to exert enough force on a rising piston with both valves open on the opposing cylinder. Jimmy?s thought was that maybe it would be enough of a pop to slow down the momentum of a slowly spinning crank during start with higher compression pistons and a light weight prop that doesn?t have enough momentum to overcome a little extra resistance. Maybe that?s not possible, but I didn?t have any more kickback problems after the change.

Earlier Superior composite intake manifolds did have a problem with collected fuel causing some sort of problem (kickback or hydraulic lock like you mentioned) and fracturing that composite manifold.
 
Jimmy?s thought was that maybe it would be enough of a pop to slow down the momentum of a slowly spinning crank during start with higher compression pistons and a light weight prop that doesn?t have enough momentum to overcome a little extra resistance.

True, there is a margin case. Pressure rise in the waste spark cylinder would not be significant, but it would not be zero either....and opposing cylinder pressures would be additive. Assume ignition in both the compressed cylinder and the waste spark cylinder, at (for illustration) 5 degrees BTC. Your system could conceivably have enough energy (starter force and prop inertia) to overcome the usual pressure rise in the compressed cylinder alone, but add some small additional pressure from the waste spark cylinder, and the total may be enough to stop progress past TDC.

True or not, the root cause is still ignition too early. You mentioned using an EI. What was the ignition timing at cranking speeds?
 
Vertical induction slop drain - RV6A

I pro-sealed a little hat fitting onto the bottom of the FAB, with about a 1/4" hole to the inside of the filter, and direct the goo as shown. I get between zero and 3 or 4 drops after a flight. It usually is a blue tinted oil. I exclusively use the purge valve on shutdown. I believe I get more when I've flown around locally with low MAP - probably sucking some oil in through the valve guides. Did this with the old cylinders as well as with the second set of new cylinders.

r-vDpi3T3JqpZou018qBMl98B5VFV7GUac2XRbIe4SgFctjiAiDK0LEMlisDnxh1HwuvhEi_OrACZe65aOJxD9i3S_aGRwe-gZBpJ78k2iUdpRMSQjFHG4fqMDoDmOajBM8fFBaUIpHHTwLy65oVpvquovAVQlo3kouPpPPoPYa4vPKVOD7qaF0NI1FLIMKh9QL7bijzA04HJR7h9sqn6KBIuTPZsNpGvZ8nsBF3tQcNkOMkuUOFlO-8jZLzymI-nwmNH0bLOAfh3KI0uA76TXpcrpYz3224X8eYSuGn9BfeYeWsCVWmxD3Qq6eIHNJIBKd9RDbX9HLvQYEweIc7NMSqmXhLv8dF1uMWLeaPPSTaNVUOBg074GGMFiOqvGI_jDZX9vjl00Hezu3jRQ7vQ6nBbf0Titbkxn5FepUsofiS4qE8uszy99xzIojylu6sUBIaHoAJFd0obCZNffD94WDw68Hkv-3ZjnxeNpIt_GkEn_8hI76VFYS_MmnmwM21fdGBIt5zKlxKsqk4Bny9vJa-192QLw5d149dAxt7rcWhT4U77uwJFg3N1LVUqONT9_FjysFT8UtdZwVmrvbK7zynLMAigKQp6gjnQYI5zYvdPkCGTtL531ieq4GktHp5mt0LjdM7L4Lt1Q68gneevKc8x13hmA8kV0wqjr8nF2ZgSP4uWRJzFXgEUOgrprUL7K7FDo607anvufs_pxDEhjfgSm-sFndBf63MsMrqIuJE0Ylf=w480-no
 
I've ignored all recent "pooled fuel can cause an exploding sump" anecdotes, as I wanted to do some reading first. After all, challenging the collective wisdom of the airport front porch is a task not to be taken lightly.

Pooled fuel in the intake plenum does not cause sump detonations...at least not fuel pooled in the quantity imagined by most.

As noted way back in post #11, fuel in liquid phase has no practical explosion risk. The fuel of interest is that which is in vapor phase, above the liquid, and combined with air in any proportion within the combustible range.

Take a look at anyone's 100LL MSDS. It includes fire risk, expressed at Upper Flammability Limit (UFL) and Lower Flammability (LFL), said limits also describing explosion risk. The UFL for avgas in air is about 7.6% by volume. The LFL is about 1.3% by volume. Mixtures with fuel vapor percentages richer or leaner than the limits do not support combustion.

Ok, so what does it mean in terms of liquid fuel quantity? At standard pressure and temperature (call it 15 psi and 15C), the liquid-to-vapor ratio for avgas ballparks about 1 to 145...one unit of liquid, fully evaporated, becomes 145 units of vapor. That number rises with temperature; it would be about 1 to 180 at a temperature of 200F.

Assume a horizontal Lyc intake plenum is 10" x 10" x 4", a volume of 400 cubic inches. The UFL is 7.6%, so 30.4 cubic inches of vapor (400 x 0.076) is the maximum; more is too rich. 30.4/145 = 0.2 cubic inches of liquid fuel, assuming full evaporation. That's 0.11 fluid ounces, or 3.2mL...about 1/10th of a shot glass, for a sump at 60F. At 200F it's even less, about 2.6ml.

Now for the interesting part. One reason gasoline has been a practical motor fuel is because it is relatively safe to keep in tanks, large and small. At typical temperatures, its physical properties, notably vapor pressure, result in mixtures above the UFL in the headspace of a tank. It's why gasoline tanks rarely explode, regardless of what Hollywood portrays every time the bad guy drives off a cliff. The fuel/air mix above liquid fuel is just too rich.

Here's a figure from a university paper which nicely illustrates reality. The added red graph lines are mine.

Vapor%20Concentration.JPG


The researcher carefully measured vapor percentage in standard 5 gallon rectangular plastic fuel containers, charting liquid quantity and temperature. The risky fuel level is not much more than a thin film, 15mL (half a shot glass) spread across the entire bottom of the jug, and then only at lower temperatures. 30mL pushed the mixture above the UFL at all temps above freezing. The 500 and 1000ml lines? They tell you anything more than a pint of liquid fuel in a 5 gallon space makes that space practically inert. There is not enough oxygen in the liquid or vapor to support fire or explosion.

So, want to worry about blowing your sump to bits? The risk is found when cranking at low temperatures, with very, very little liquid fuel on the plenum floor.
 
Last edited:
Dan,
My start timing, running 2 Pmags was set at TDC, but with the newest software version, which I had, the start timing was automatically delay to 4* after TDC. If it really did that, I couldn?t see any way you could get a kickback. We were fishing for an answer and came up with a possible situation caused by fuel pooling due to improper sniffle valve placement.
I had also considered the possibility of the Skytec starter spinning the engine faster because of the higher voltage earthex battery. An impulse mag kicks out of the delayed start mode to the normal run mode at about 400 RPM, but the Pmag makes this transition (for some reason) at a lower RPM - I think it was around 200-250 RPM. Not sure about that number, but it was something like that. I consulted with Hartzell, and was assured that their starter couldn?t turn the engine at more than about 180 RPM, even at a much higher than normal voltage. When the kickbacks happened, it sure seemed like the start timing must be off, but I couldn?t come up with a reason for it. If it was an internal timing snafu in the Pmag, it cured itself.
 
BIG SNIP

So, want to worry about blowing your sump to bits? The risk is found when cranking at low temperatures, with very, very little liquid fuel on the plenum floor.

Interesting stuff, Dan. There is a time component not depicted or discussed - the time it takes for fuel to evaporate. Concentrations above the upper limit likely had to pass through the combustible region, if only during a short time.
 
Interesting stuff, Dan. There is a time component not depicted or discussed - the time it takes for fuel to evaporate. Concentrations above the upper limit likely had to pass through the combustible region, if only during a short time.

True, and stratification too, assuming static conditions. Some of the reported experiments identify stratification in taller volumes.

For us, the classic dynamic condition is probably the flooded start. Mixture rich, WOT, and run the boost pump much too long, so a lot of fuel runs down the pipes into the sump. Now throttle to cracked, mixture to ICO, start cranking. It doesn't light because the air/fuel arriving at the cylinders is above the UFL...the little bit of air entering the plenum via the cracked throttle plate isn't enough to offset the ongoing evaporation. So the pilot correctly thinks "Darn, flooded it", opens the throttle to WOT, and cranks again. The greater quantity of fresh air flowing through the plenum headspace dilutes the vapor-rich contents, and the delivered mixture progressively gets leaner with every revolution. Eventually one or more cylinders takes delivery of fuel vapor/air within the flammable range. At that point pretty much the entire contents of the plenum headspace and intake pipes has been leaned to somewhere between the UFL and LFL.
 
Sump (explosion)

Originally Posted by DanH View Post

So, want to worry about blowing your sump to bits? The risk is found when cranking at low temperatures, with very, very little liquid fuel on the plenum floor.

Always enjoy reading DanH?s posts as he always does his homework ( his math calcs remind me of some long study sessions in college Algebra and Calculus).
So back to the purpose of the sniffle valve. Dan makes the interesting point that they prevent hydraulic lock and not necessarily to prevent explosions in the Sump. I?m trying to come up with the scenario that caused my Sump (explosion). I?ve had backfires that have been severe enough to split the welds on a stainless exhaust but this incident was different. For a backfire to blow the O rings off the tubes are the same forces at play but just timed differently with the intake valves open ?
 
I?m trying to come up with the scenario that caused my Sump (explosion).

Understandably. Lycomings do pop in the sump. Right now I don't know precisely why.

So anyway, tell us more about the setup....mags or waste spark EI, OAT and engine temperature, priming, cranking speed, etc. Obviously the throttle was closed.

Blowing out the intake tube o-rings is pretty impressive. Each is trapped between two steel rings on the tube, and when the tube is then shoved up into the plenum spigot, there is very little room for escape. Not something I've heard before, but that means little. Is it a common occurrence?

Intake%20Tube%20O-Ring%20Socket.jpg


Intake%20Tube%20O-Ring.jpg
 
My Horizontal Induction Set Up

I run a 10:1 compression IO-360 parallel valve with a magnesium Sump from an IO-360-C1C. I machined the front pad to match an AFP FM 200 servo and it is equipped with a purge valve. The intake tubes were modified from the original C1C configuration to both fit a parallel valve engine and fit within the confines of an RV-4 cowl. I’m equipped with one Bendix mag (non-impulse) and a light speed electronic ignition with crank fired mini sensors. I run a fixed pitch wood propeller and a 4 into 1 exhaust. The sky-tec inline starter has always turned the engine over fairly briskly.
The circumstances of the explosion were a hot start after fueling with what I consider a normal start procedure with mixture full lean and throttle forward. I believe I did prime the engine for a few seconds with the boost pump and both levers full forward. Note: I rarely shut the engine down with the purge valve as the mixture control usually provides a positive shut off. I did not use the the purge valve to circulate cool fuel for this start as I prefer not to have the inaccuracies it causes in the fuel flow totalizer.
I hope I’ve remembered all the perimeters for this years old event. This was a one time event and after the installation of a purge valve was never repeated. This is the only explosive Sump occurrence that I’ve witnessed first hand the rest were repeated to me from “ The Front Porch “.
Revised: After thinking about it my start Procedure then was prime than mixture lever full forward and throttle lever at idle. Sorry it’s been a few years.
 
Last edited:
The intake tubes were modified from the original C1C configuration to both fit a parallel valve engine and fit within the confines of an RV-4 cowl.

The outboard o-ring flange was removed so the tube could enter the sump a bit further?

Revised: After thinking about it my start Procedure then was prime than mixture lever full forward and throttle lever at idle. Sorry it?s been a few years.

I hear 'ya brother.

Perhaps prime followed by mixture at ICO and throttle full forward,which would be a standard flooded or salvage start? A hot start at mixture full forward and throttle closed would be unlikely...way too rich.
 
Dan,
My start timing, running 2 Pmags was set at TDC, but with the newest software version, which I had, the start timing was automatically delay to 4* after TDC. If it really did that, I couldn?t see any way you could get a kickback.

I don't either....if it really does that.

BTW, from your post a few months ago:

After I installed two Pmags on my previous (RV8) airplane, we had two kick backs with a lightweight Whirlwind prop. The second one broke a few teeth off my starter ring gear. I reset the timing to an extra 2* after TDC. It started fine and I never had another kickback event.

http://www.vansairforce.com/community/showpost.php?p=1382197&postcount=8
 
O-ring Flanges

DanH
?The outboard o-ring flange was removed so the tube could enter the sump a bit further? ?
Negative. The inboard ends the tubes are as manufactured. We went so far as to fabricate some aluminum half circles to hold the intake tube ends perfectly in the center of the Sump sockets as we mocked up and tacked the tubes together. My intake tube inboard ends look exactly like the picture you posted.
Years ago my hot start procedure wasn?t optimal and may be related to the Sump explosion. The only difference between cold start and hot start procedure then was the length of the prime and yes would sometimes require some long cranking.....like 2 or 3 10 second rounds.
 
Back
Top