Water In the Gas


Trying to solve the hesitation problem in his '73 Super Beetle, Dave picked Rob's brain on this (and many other!) subjects. What follows is a "conversation" between Dave and Rob regarding the subject of water in the gas.

Dave wrote - The problem of water in gasoline seems to be fairly common in the U.S. The auto places sell a lot of Heet in the winter (Heet is a water-getter consisting primarily of methanol). I've had a problem with gas line freeze up in virtually every car I've owned. Of course I've always looked for the cheaper gas, which may be my problem. I told my son that from now on he's to buy the good brand name stuff, 92 octane.

Rob responded - This is interesting. Water and gasoline are immiscible (won't mix) except in minute quantities. Methanol will mix with gasoline, but not much by itself -- it normally requires a mixing agent (benzene I think) to enable a reasonable quantity to be added to gasoline. If there was free water in the fuel, methanol would help soak it up, as it mixes very readily with water. (The Indy cars use straight methanol for fuel, and all fire equipment at their tracks use straight water to dilute any methanol fire below the 'proof' range for ignition.)

I can't see how any fuel company can mix much water with the gasoline though -- sounds a bit suspect to me. And it would have to be fully mixed with the fuel, if not, it would sink to the bottom of the tank and you'd have straight water going to the engine -- so it simply would not start or run in this situation.

Methanol has an octane rating of 133, so adding say a litre of this to 41 liters for a tank full will raise the 87 octane to about 88.2. As this is a R+M/2 figure, the equivalent RON figure would be about 90... getting very close to the 91 RON recommended by VW. So the methanol would help on two fronts.

The third effect of methanol is that it has a VERY high latent heat of vapourisation, (3.8 times that of gasoline) meaning the fuel-air mix (using gas/meth mix) will be a bit cooler as it enters the engine, resulting in fractionally lower combustion temperatures, which also helps overcome the lower octane fuel problems. It's faintly possible the icing problems might occur under the carburetor though, if the heat riser is not working fully.

The down side is that methanol contains 49.9% oxygen by weight, so your fuel/mix (with 1 liter per tank full) moves about 1% to the lean side, which makes it more important than ever to try that 130 jet (VWs don't like lean mixtures remember). Running on straight methanol requires jets 300% larger than gasoline jets.

A trick used in the aviation industry to reduce water contamination in fuels (rusty tanks usually results from water condensing in the tank overnight) is to fill up at the end of the day, so the tank has less air space overnight when the problem occurs. Of course aircraft usually go through a lot of fuel with each trip, unlike a car which only needs filling each week or so, so it's not likely to help much I suppose.

But the only sure way to avoid condensation is the aircraft way fill it up to the brim every night so there is no airspace above the fuel, so no water vapour to condense overnight. Parking indoors helps too. I guess it would help if you only allowed the car to half empty before filling it up, and fill it on the way home than in the morning, so there is the least head space in the tank most nights. It wouldn't be a problem if the temps stayed below freezing all the time that air would be dry anyway it's the constant warming/cooling in the roughly 30 50f range and wet days that's the real culprit.

If the fuels in your area contain any ethanol they will absorb water vapour and may worsen rust problems too. What you really need for the VW is fuel which is high in aromatics, as these have higher natural octane numbers, and don't need so much/any alcohol or MTBE (or TAME or ETBE) to enhance the octane number. We still have leaded fuels available (no MTBE or alcohol), so it's not an issue for us yet.

Dave wrote - I agree that it's very unlikely that gasoline companies are intentionally mixing water into their gas. I can't imagine how Amoco could stay in business if they were actually doing something like that and it became widely known. But there is an undercurrent amongst the prolatariat that the cheaper brands have water in them I've heard this from various and sundry people any number of times.

Regarding "Heet" -- The label says is "Caution: Contains methanol. Don't drink it or you'll go blind!" It doesn't list any other ingredients. The bottom line is -- it worked. And it's cheap. So if we have this problem again a little Heet in the tank will be my first course of action.

Rob responded - One possible cause of water in the gas is simply the climate. When you have wet conditions you do get overnight condensation in the fuel tank -- water droplets in the fuel. There are a lot of posts about rusty fuel tanks, and this is usually the problem (rust on the inside, not the outside).

Dave wrote - My son says that with the Heet in the tank the car is running considerably better. Our climate here is quite dry, but with all the irrigation the relative humidity has increased over the years. We do get a considerable amount of dew some mornings. Could be a little Heet in the tank occasionally would be beneficial.

My son made an interesting discovery tonight. As he was on his way to the gas station (brand-name) with a near-empty tank the car died at every corner; once the tank was full with the good stuff the car ran fine.

The apparent lesson is that the cheap stuff is just that -- and it probably contains water. The more expensive brand-name gas is higher quality and the Bug seems to like it a lot more. Our van and Nissan truck handle the cheap stuff just fine, but the VW seems to be more sensitive to the quality of the gas.

A response from the RAMVA Newsgroup - When you run the tank down near empty you are also allowing any sediments and general crap to enter your fuel lines. Almost any gas tank in any car has depression areas in the bottom of the gas tank. The crap and water that has condensed is supposed to settle into these low areas and not enter the system.

I recommend that you buy some Dry Gas or STP type stuff and add it to your next tank of gas. This will help remove the water and some of the crap that might be fouling your fuel lines, carb, etc.

As for my experience with gas and octane levels, my '74 Super Beetle always has the most amount of pep when I use Sunoco's 94 octane gasoline. It's expensive, so I usually buy Mobil or Hess gas. I try to avoid those minute mart type gas stations. I filled up at one and the car ran pretty poorly. I bought a can of octane boost and some dry gas and it ran better after about another 20 miles. I can only guess that the gas was of poor quality. After that tank of gas, I cleaned the varnishy looking carb and filled up with Sunoco 94. It ran much better.

Two possible causes of the problem come to mind:

  • The gas is poor quality and possibly has water in it, which would be especially noticeable when the tank is near empty. Have a bit of a hard time accepting this one our other cars run just fine on this gas, and I wouldn't think a petrol company could stay in business very long if their gas was watered down. Is the VW carburetor particularly sensitive?
  • More likely (at least in my mind) is your explanation that there is crap and possible condensation in the tank that gets into the fuel lines when the tank is near empty.

Dave responded - Our problem has been the worst when the tank is nearly emply. It sounds to me like a yank-a-tank project is in my future.

I'm becoming more and more convinced that our hesitation problem is primarily caused by water in the gas, probably from condensation inside the tank. (We later learned that this was only contributory to the main problem, which was leakage of air into the intake manifold.)

So I guess the bottom line is buy only good quality, high octane gas, keep the gas tank full, and had a bottle of Heet now and then.

Rob responded - The water problem seems to rear it's head in your winter. I doubt there are anything but trace amounts in the fuel itself water and HCs are not normally miscible. Even alcohol (ethanol or methanol) can be added in only small quantities unless a mixing agent like benzene is added (I bet HEET has a mixing agent in it too).

It's my guess that condensation in the tank would occur most in similar conditions to carburetor icing high humidity/wet conditions with temps just above freezing. The car cools down over night and the moisture in the air space in the tank condenses on the cold insides of the tank. At the same time, the cooling air in the tank would contract and suck more in from outside worsening the situation.

Many US VW owners talk about rust in their tanks this could only come from moisture in the tank. I've never had this problem in our drier climate, and I have never needed a fuel filter in my fuel lines. The earlier fuel pump design had a small filter inbuilt, and this never had any contaminants in it, so I had no real incentive to install an extra filter.

* * * * *




Design by Erin