Thoughts about vapor-lock & mechanical fuel pumps

[ntsqd]

Boiling fuel is vapor lock. Can achieve boiling fuel at sea level if the temperature is high enough, and it doesn't need to be a record setting heat wave to do it. What I recall from looking into that 10-ish years ago or so was that the light ends of fuel will start to boil in the mid 90's at sea level pressure. As the atmospheric pressure drops with an increase in altitude the boiling point temperature goes even lower.

Mechanical pumps put the whole supply line under a slight vacuum. That lowers the boiling point temperature. Going up in altitude lowers the ambient pressure. I suspect that mechanical pumps create a fixed vacuum relative to ambient pressure. Which means that at higher altitude the pressure on the fuel in the supply line is even lower. I'm working backwards from vapor-lock occurs easier at altitude than at sea level to deduce how a mechanical pump behaves.

That could be wrong, but it fits what I know about when and where vapor lock occurs.

Pumps at the tank or near it are great because they put the supply line between it and the engine under pressure, thereby raising the boiling point temperature. Electric pumps aren't immune to vapor-lock. With an inline pump the section of the supply plumbing between it and the tank is under a vacuum too. This is one of the better reasons for an in-tank fuel pump.

 

[73azbronco]

mmmm, but fuel out of tank is not hot so, lower pressure might not effect it. I think vapor lock occurs either inside the carb or engine pump to carb.

 

[Jetfixr320]

I had a issue several years ago when I first got my Fairlane on the road. After driving and the engine warming up the car was stalling at lights, hard to start after stopping for a few minutes etc. I ended up curing it with a phenolic carb spacer and DEI heat sleeve material around the fuel lines in the engine bay.
I had read or was told the newer fuels with alcohol in them would boil at a lower temperature? This was around 2005 time frame?

And then, when I got my Falcon on the road around 2009 or 10, same issue. Fixed it the same way. No issues since with either car.

 

[steelworker77]

Boiling fuel is vapor lock. Can achieve boiling fuel at sea level if the temperature is high enough, and it doesn't need to be a record setting heat wave to do it. What I recall from looking into that 10-ish years ago or so was that the light ends of fuel will start to boil in the mid 90's at sea level pressure. As the atmospheric pressure drops with an increase in altitude the boiling point temperature goes even lower.

Mechanical pumps put the whole supply line under a slight vacuum. That lowers the boiling point temperature. Going up in altitude lowers the ambient pressure. I suspect that mechanical pumps create a fixed vacuum relative to ambient pressure. Which means that at higher altitude the pressure on the fuel in the supply line is even lower. I'm working backwards from vapor-lock occurs easier at altitude than at sea level to deduce how a mechanical pump behaves.

That could be wrong, but it fits what I know about when and where vapor lock occurs.

Pumps at the tank or near it are great because they put the supply line between it and the engine under pressure, thereby raising the boiling point temperature. Electric pumps aren't immune to vapor-lock. With an inline pump the section of the supply plumbing between it and the tank is under a vacuum too. This is one of the better reasons for an in-tank fuel pump.

I always have some sort issue with this as soon as it gets in the low 90's and I am at sea level basically. On a nice edelbrock mechanical pump, bypass regulator, nylon braided PTFE fuel lines that are insulated near the engine. Lines run on top of frame away from exhaust, holley heat shield under the carb, and vented wheel wells. Not really seeing anything else I could do besides go to an in tank electric pump. Just undecided to stay carbureted or go the pro flo 4 route.

Never really an issue until we hit the 90* mark and you can usually watch the afr gauge when you come to a stop at idle and watch it climb. As a daily driver it can get annoying at times.

 

[Broncobowsher]

The vacuum in the fuel line from the tank is very minor. Inches of water (as compared to inches of mercury that we measure engine vacuum in)
The post pump return line 3-port filter works because of flow. With a steady stream of fuel going back to the tank, you don't have (nearly) stagnent fuel in the supply line. The fuel is still receiving heat, but it is spread across a larger volume. This deals with vapor lock on the supply side, the pump will always get cool fuel.

The next place it can be an issue is where the fuel pressure drops again, the fuel bowl. This falls into a couple of potential issues. The fuel line from the pump can take in heat, but not boil since the pressure is higher. Once through the needle and seat, pressure drops, maybe below the boiling point. Mix in heat transfer from a hot intake manifold soaking up into the carb.

Even seasonal fuel can be an issue. Spring is often the prime vapor lock season. Winter blends with a high RVP boil lower (so they start/run good when cold) in a hot season are not good. Fill up at a station at the end of winter blend season and let it sit a bit (hobby cars do this a lot) then drive on a very warm spring day. The fix is really simple. Drive more and use fresh gas that is in season.

 

[ntsqd]

mmmm, but fuel out of tank is not hot so, lower pressure might not effect it. I think vapor lock occurs either inside the carb or engine pump to carb.

Every liquid has a unique to it boiling point curve, graph it and one axis is temperature and the other axis is pressure on the liquid. Gasoline is a mixture of a bunch of different liquids, so it has more what I'd call a boiling point band where the lightest components boil at a lower temperature at any pressure curve and the heavier components boil at a higher temperature at any pressure curve. The mixture ratio of these different components is a large part of the difference between "winter fuel" and "summer fuel".
Lower pressure will affect the fuel in tank as the fuel will be at or near whatever the ambient temperature is. If there is a return system it will soon be hotter than ambient, but the usually higher pressure in the supply line more than offsets this. The supply line between the engine mounted pump and the carb is as Broncobowsher points out below, at a higher pressure and not likely to boil.
When there is a return system, and it has been running for a while the fuel in the tank is hotter than ambient temperature. Carefully (!) put your hand on a fuel tank in such a situation if you don't believe me. That has caused me to start thinking of putting a power steering sized cooler in the return lines of my future project vehicles.

The vacuum in the fuel line from the tank is very minor. Inches of water (as compared to inches of mercury that we measure engine vacuum in)
The post pump return line 3-port filter works because of flow. With a steady stream of fuel going back to the tank, you don't have (nearly) stagnent fuel in the supply line. The fuel is still receiving heat, but it is spread across a larger volume. This deals with vapor lock on the supply side, the pump will always get cool fuel.

The next place it can be an issue is where the fuel pressure drops again, the fuel bowl. This falls into a couple of potential issues. The fuel line from the pump can take in heat, but not boil since the pressure is higher. Once through the needle and seat, pressure drops, maybe below the boiling point. Mix in heat transfer from a hot intake manifold soaking up into the carb.

Even seasonal fuel can be an issue. Spring is often the prime vapor lock season. Winter blends with a high RVP boil lower (so they start/run good when cold) in a hot season are not good. Fill up at a station at the end of winter blend season and let it sit a bit (hobby cars do this a lot) then drive on a very warm spring day. The fix is really simple. Drive more and use fresh gas that is in season.

It is not much of a vacuum, but when the temperature and barometric pressure are already close to the boiling point it won't take much. When I did that research I found that it was possible to have the lighter ends boiling off at sea level at an fairly common (for this Coastal Desert) ambient temperature.

I've experienced fuel boil in the carb bowl too. I wasn't ignoring that or implying that it isn't a problem. I was setting it aside as a different issue.

For that problem the DZ302's heat baffle plate under the carb worked far better, particularly with the AFB design's float bowl placement, than just the thermal spacers available at the time (the phenolics were not). The first one that I built was small, very much like the DZ302 part. The second one was much larger, it was the length and width of intake manifold, less the area around the distributor. I made both from 1/8" thick aluminum, and used a dremel to cut out holes where bits of the carb extended below it (usually linkage and Holley accel pumps). Works exactly the same as the exhaust heat-shields recently discussed in the perforated inner fenders thread.

 

[jamesroney]

OK, now you are just making stuff up...

I wrote a really long and detailed engineering assessment of your vapor lock nonsense in the other thread. Then I realized that it might be a hijack, so I deleted it. I wish I had archived it so I could post it here.

PLEASE take my feedback as sincere, constructive, and light-hearted criticism. I think you can handle it, and I trust that it is the kind of feedback you were seeking.

First and foremost, "boiling fuel" is NOT vapor lock. Vapor lock is vapor lock. Boiling fuel is fuel that is at a temperature higher than it's boiling temperature. If you transport liquid in a tube, and it exceeds the boiling point, it just boils. If the tube is sufficiently long, and the temperature gets lower, it just condenses back into fuel. Mass is conserved. So you can certainly have a situation where the fuel turns into "steam" and back, and nothing happens. This situation almost never occurs. But it can.

Vapor lock occurs when the "steam" is allowed to accumulate in a location where it prevents pumping or movement. So yes, if you boil the gasoline, and you allow the vapor to accumulate in the mechanical diaphragm, it simply will not pump. Diaphragms are designed to transport incompressible liquids. If you surround one with boiling vapor...it stops pumping.

This can also happen in a fuel transport pipe, tube, or vessel. If you run a steel tube full of gasoline near a concentrated heat source...like an exhaust pipe you can boil the fuel in a particular location. IF that happens, and you are relying on vacuum to pull the fuel...you can theoretically create a situation where the volume of vaporized fuel exceeds the displacement capacity of the diaphragm. But I don't believe it has ever happened in real life. Because if you model the fuel transport dynamics in absolute pressure, the atmospheric pressure of 14.7 psi in the tank is always "pushing" the fuel. But none of that matters.

BECAUSE...the fuel in the fuel bowl after the needle and seat is sitting at atmospheric pressure. Sure, you can pump at 5 psi, or 50, or 500. Once you regulate it down to 3 psi at the seat, and then 0 psig in the bowl...if it is hot, it will boil. The VAST majority of "vapor lock" diagnosis is due to fuel boiling in the heated aluminum reservoir that you call a carburetor. The carburetor relies on intake air to provide cooling, and the coolant and exhaust heat riser provides heat. Once the intake air is stopped, the carb reaches 250 degrees, and the fuel in the bowl will boil. When the fuel boils, it either evaporates, or condenses back in the intake plenum. At that point, the engine will not restart...because it is flooded, and simultaneously starved. In some cases, the carburetor is so hot that any new fuel introduced past the needle valve is immediately boiled, and vented. We also call this condition "vapor lock" but it really isn't. It's really just a boiling carburetor and it often resolves itself by simply waiting until the engine is cool.

So back to the actual vapor lock part of vapor lock. This is the condition where the fuel on the vacuum side of the diaphragm is vaporized, and the pump simply doesn't pump. This condition is often remedied by allowing the liquid fuel in the tank to recirculate. Insulation, heat shields, and re-routing fuel lines helps. And yes, pumps at the tank prevent low pressure cavities at the fuel pump which prevent fuel starvation. 100% agree on that.

Now I forgot what we are solving?

 

[ntsqd]

I can say without exception that every vapor-lock problem that I've ever experienced was the inability of the fuel pump to deliver liquid fuel to the carb. Call that whatever you want, I call it "vapor-lock."

I have also experienced fuel boiling out of the bowl(s), but not while driving, only when heat-soaked from a hot shut-down.
Once the pump can't pump a liquid, then the bowls go dry. How would you know that the bowls boiling dry was the first part of the whole problem? The engine quit because it has no fuel, but did it quit because the bowls boiled off or because the pump couldn't re-fill them. When cranking said stalled engine or forcing the electric pump to run results in no fuel delivered, liquid or liquid immediately boiled to vapor, nothing, then I'm pretty comfortable with saying vapor-lock in the fuel supply system was the cause. It was the cause the times that we were bored while waiting for things to cool down and spent part of that time diagnosing the condition.

If you look at it from an absolute pressure perspective then the tank is under ~14.7 psi pressure (assuming sea level-ish), but so is the destination (unless a blow-thru carb'd blown engine). I'd expect that to cancel the tank 'pushing' on the fuel since it is effectively pushing back with the same pressure.
I'm sure that there's a vaporization aspect to that 4-6psi fuel being dropped to ambient pressure at the needle and seat. Most of the time it likely condenses back to liquid nearly immediately, but when hot enough....... I'm equally sure that considering that is getting really deep into the weeds.

We weren't solving anything. I had that thought and almost posted it in a thread, but then decided it really didn't belong there. Hence, this thread.

 

[jamesroney]

I can say without exception that every vapor-lock problem that I've ever experienced was the inability of the fuel pump to deliver liquid fuel to the carb. Call that whatever you want, I call it "vapor-lock."

I have also experienced fuel boiling out of the bowl(s), but not while driving, only when heat-soaked from a hot shut-down.
Once the pump can't pump a liquid, then the bowls go dry. How would you know that the bowls boiling dry was the first part of the whole problem? The engine quit because it has no fuel, but did it quit because the bowls boiled off or because the pump couldn't re-fill them. When cranking said stalled engine or forcing the electric pump to run results in no fuel delivered, liquid or liquid immediately boiled to vapor, nothing, then I'm pretty comfortable with saying vapor-lock in the fuel supply system was the cause. It was the cause the times that we were bored while waiting for things to cool down and spent part of that time diagnosing the condition.

If you look at it from an absolute pressure perspective then the tank is under ~14.7 psi pressure (assuming sea level-ish), but so is the destination (unless a blow-thru carb'd blown engine). I'd expect that to cancel the tank 'pushing' on the fuel since it is effectively pushing back with the same pressure.
I'm sure that there's a vaporization aspect to that 4-6psi fuel being dropped to ambient pressure at the needle and seat. Most of the time it likely condenses back to liquid nearly immediately, but when hot enough....... I'm equally sure that considering that is getting really deep into the weeds.

We weren't solving anything. I had that thought and almost posted it in a thread, but then decided it really didn't belong there. Hence, this thread.

Sure, but can you say without exception that every fuel pump that wouldn't deliver fuel was a diaphragm unit that was "sucking" on the fuel?

Because I can certainly describe the 1952 Willys M38 with L134 engine that ran the discharge pipe of the fuel pump directly beside the exhaust manifold, resulting in fuel starvation. But the fuel pump certainly delivered liquid fuel. And the fuel was around 3-5 psig. Adding an electric pump at the tank did nothing.

The traditional "fix" is to use the WIX 33040 filter from a 76-86 Jeep CJ. (Which is a clone of the Vapor separator from the 68-71 Mopar 426/440.)

The point being that there is a distinction between fuel vaporization due to reduced atmospheric pressure, and vaporization due to excess over-temperature. Fuel which is "baked" between the pump and carb will boil at 3-5 psi regardless of whether you have an electric fuel pump at the tank or not.

I have absolutely no doubt that "vapor lock" can occur in the diaphragm of a mechanical fuel pump. And it can be mitigated by an electric pump at the tank.
I also have no doubt that boiling fuel in the carburetor feed line can result in the other thing you call "vapor lock." And it cannot be mitigated by an electric pump at the tank. (but it can be mitigated by recirculating fuel in the line near the carb)


I think we do not disagree. It's just the matter of calling two different failure modes the same name...and then applying mitigations that only resolve one mode.

James.

 

[ntsqd]

That I've personally experienced, yes without exception. That I've seen on various forums, no. I clearly recall an LS swapped FJ60 that was having vapor-lock like problems. That turned out to be not classically vapor-lock but became the turning point for me with regard to inline electric fuel pumps (which is a related, but whole separate discussion). They were acceptable to me until then. His was collapsing the suction hose. That's far more than a couple inches of water, but that's also not a diaphragm pump.
My 70-something Cherokee Chief had one of those 33040's on it. I think they're a good step in the right direction, but more can be done.

Since fuel behaves the same as any other liquid with regard to vaporizing then increasing the pressure it is under increases the boiling point temperature, the "Vapor phase transition temperature" to use one chemistry book's phrasing. And lowering the pressure, no matter how far or how little lowers the boiling point by a respective amount. My primary goal on my own projects is to never experience vapor-lock. To do that I want supply fuel under at least 20 psi pressure and I want a returning supply system so that fuel is not stagnating in one hot spot and trying to boil. (I've no intention of building a Smokey Yunick Adiabatic Engine.) This is trickier and more prone to potential problems with a carb than it is with EFI, but I've done it so I know that it can be done.

To me "Vapor-lock" is the inability of the supply system to deliver liquid fuel to the engine. The hows and the whys all fall under that phrase umbrella phrase, and there are many usual suspects. I think that a fair number of the accepted "fixes" are acts of desperation that don't really do much or are band-aids that hide the true problem by seeming to fix it. Most of the time. but not all of the time. If all that people get out of this whole thread is what is really happening in different areas in the fuel supply system and how to combat the various problems in those area then I'll consider this thread will be a success.

 

[jamesroney]

To me "Vapor-lock" is the inability of the supply system to deliver liquid fuel to the engine. The hows and the whys all fall under that phrase umbrella phrase, and there are many usual suspects. I think that a fair number of the accepted "fixes" are acts of desperation that don't really do much or are band-aids that hide the true problem by seeming to fix it. Most of the time. but not all of the time. If all that people get out of this whole thread is what is really happening in different areas in the fuel supply system and how to combat the various problems in those area then I'll consider this thread will be a success.

That was the point of my comment. Because when you use the symptom to describe the failure mode...you necessarily conflate causation. Especially when you use words that have meaning in English as well. "vapor" and "lock."

"Vapor lock" does NOT mean inability to deliver liquid fuel to the engine (to me!) Because vapor lock implies a root cause. Lack of fuel is an effect. From the perspective of the carburetor float bowl needing liquid fuel...The C&E matrix can have many causes for fuel starvation that are NOT vapor lock.

Your LS swapped FJ is a good example where fuel was not delivered, and it was not due to vapor lock. My example of boiling fuel in the bowl, or in the steel line is also not vapor lock. True Vapor Lock is the hydraulic condition wherein the liquid medium cannot be transported because the liquid has reached the boiling point, and cannot be pumped. Water in a well over 32 feet deep cannot be lifted by suction, and the pump becomes vapor "locked."

If we can differentiate between vapor on the suction side of a pump, and vapor on the pressure side of a pump, (and give them different names,) I will consider it a success. I use the terms "vapor lock" and "boiling fuel" to make the discrimination. My words are no better than any others. I am just providing a distinction. You could call them "fred" and "bob" As long as we recognize that there are different mitigations for "Fred-failures" than for "Bob-failures."
(and yes, I fully understand that the thermodynamic term of "boiling" describes the partial pressure state change between the liquid and gas phase... and vaporizing is indeed boiling)

Minor changes to pressure will have a profound effect on "Fred-failures." Major changes to pressure will have no effect on "Bob-failures"

Once we have a common understanding of the language, then it really doesn't matter what we call it. But until we do...there will be misunderstanding.

 

[ntsqd]

To me "Vapor Lock" means that the fuel has turned to vapor and is not pumpable. This the commonly used phrase to describe this condition, so I'm using it rather than create a sociology experiment trying to get people to accept a new label. As to where the fuel turned to vapor, that is a function of whatever is wrong with the system and while the diagnosis will need to know where it happened, it doesn't affect the end result. I do not think that before or after the pump distinction is necessary to the labeling of the problem, only to the diagnosing of it and only then because there are potentially different factors that need to be considered. Among the casually informed more terms tend to create more confusion, not less.

Even though that LS swapped FJ didn't actually have vapor-lock, the symptoms presented to the operator were identical to it.

Rather than assigning different labels for different locations I'd suggest that instead a logical methodology for determining where the actual failure is occurring be created or adopted. That could be broken down into pre-pump vapor-lock conditions and symptoms and post-pump vapor-lock conditions and symptoms. Hopefully this would then become a reference for people designing a new system so that the resulting design lacks all of the usual flaws that result in vapor-lock.

 

[jamesroney]

To me "Vapor Lock" means that the fuel has turned to vapor and is not pumpable. This the commonly used phrase to describe this condition, so I'm using it rather than create a sociology experiment trying to get people to accept a new label. As to where the fuel turned to vapor, that is a function of whatever is wrong with the system and while the diagnosis will need to know where it happened, it doesn't affect the end result. I do not think that before or after the pump distinction is necessary to the labeling of the problem, only to the diagnosing of it and only then because there are potentially different factors that need to be considered. Among the casually informed more terms tend to create more confusion, not less.

Even though that LS swapped FJ didn't actually have vapor-lock, the symptoms presented to the operator were identical to it.

Rather than assigning different labels for different locations I'd suggest that instead a logical methodology for determining where the actual failure is occurring be created or adopted. That could be broken down into pre-pump vapor-lock conditions and symptoms and post-pump vapor-lock conditions and symptoms. Hopefully this would then become a reference for people designing a new system so that the resulting design lacks all of the usual flaws that result in vapor-lock.

I can live with "pre-pump vapor lock" and "post pump vapor lock" for the purpose of this discussion.

What do you call fuel boiling in the carburetor post needle and seat wherin the heat soak causes the fuel to vaporize resulting in a no-start condition?

 

[ntsqd]

Vapor lock



Post hot shut-down I'd call it heat-soak induced fuel boil-off. BT, DT, got the demerit badge.
I assume that it can happen while running, it certainly seems possible. I know of no instance of that happening with any degree of confidence, but that's one experience out of 8.3 billion or so.

Thinking about your post up above, is it fair to say that all vapor-lock is locally boiling fuel, but not all locally boiling fuel is vapor-lock? Where the boiling occurs is crucial to the definition. Should it occur inside of any plumbing (tubing, hoses, pump, etc.) it is vapor-lock, but where exposed to ambient temperature and pressure (fuel bowl) it is not.

 

[Yeller]

I’ll throw fuel on the fire…. I’ve seen anything that creates a reservoir under the hood create boiling fuel and a no run condition. Items like the under hood fuel sumps that house an electric pump for fuel injection, the used to be popular "fuel accumulator", I've seen C5 corvette filter regulators connected to the fuel rail. they all are a hot weather no go.

 

[jamesroney]

Vapor lock



Post hot shut-down I'd call it heat-soak induced fuel boil-off. I assume that it can happen while running, it certainly seems possible. I know of no instance of that happening with any degree of confidence.

Can't happen when running under normal conditions, since the induction air volume provides convective cooling to the carburetor thru the venturi. Many complaints of Bronco "vapor lock" occur after heat soak.

I’ll throw fuel on the fire…. I’ve seen anything that creates a reservoir under the hood create boiling fuel and a no run condition. Items like the under hood fuel sumps that house an electric pump for fuel injection, the used to be popular "fuel accumulator", I've seen C5 corvette filter regulators connected to the fuel rail. they all are a hot weather no go.

The under-hood reservoir is an aluminum tank mounted to the exhaust heat riser and is called a "carburetor."

 

[Yeller]

Can't happen when running under normal conditions, since the induction air volume provides convective cooling to the carburetor thru the venturi. Many complaints of Bronco "vapor lock" occur after heat soak.


The under-hood reservoir is an aluminum tank mounted to the exhaust heat riser and is called a "carburetor."

what's that? hadn't run one since 1998 when I did my first TBI swap

 

[jamesroney]

The HDPE plastic fuel line on my 1967 Bronco with the 170 6cyl is at no time closer than 2 feet from a heat source hotter than a transfer case.

The HDPE plastic fuel line on my 70 Bronco with the V8 only gets within a foot of the exhaust manifold near the steering column.

No Bronco has ever experienced pre-pump vapor lock, or post-pump vapor lock But perhaps heat soak induced boil off. Unless that Bronco has dual exhaust. If so, remove it.

 

[ntsqd]

There's a healthy bias out there against those plastic fuel lines. At least some Scout II's came similarly fitted and I'd venture to say that the bias against them is even stronger in those circles. As a fuel conductor they'll take less heat than a steel tube, but they're also a poorer thermal conductor, so the heat that they are exposed to isn't going into the fuel nearly as fast as it would if those were steel lines.

 

[acropilot]

I don't have any issues with my bronco but i built a 32 ford 3 window coupe hot rod with a 450hp 383 stroker sbc with a 750 double pumper holley and a carter mechanical pump that i had some vapor lock issues but only running on interstate at speed for several hours on a hot day.I installed a carter electric pump at the tank that runs through the mechanical pump and a toggle switch on the dash that is in the off position pretty much all the time but when running on interstate and i start to feel it surging like it's fixing to shut down i turn on the electric and it's all good.With the hood and hood sides on it's a really tight engine compartment.I have a background in aviation and some of the piston driven GA type aircraft have a mechanical pump on the engine and an electric pump at the tank for back up.Standard practice on take off is to turn the boost pump on and once at altitude turn it off then when entering the landing pattern turn it on again as a back up just in the mechanical fails at the worst possible time.