83 and 87 Subarus, fuel contamination

2008 Mar 31

I have experienced two fuel contamination incidents with older Subarus, 1983 and 1987. Does anyone have information on whom to contact for further technical guidance, or has anyone had a similar experience?

At about age 20, both cars exhibited progressively worsening fuel starvation bucking and kicking symptoms after being run on the road for awhile, and particularly when driving uphill for the first time in a trip. Two notes:

1. They always started readily, and ran well when standing still, or driving slowly on flat terrain and very smooth roads.

2. Once being driven on the road for awhile, after the bucking and kicking started, I pulled over and stopped. The characteristic fuel starvation bucking and kicking soon stopped, and the engine ran smoothly. The bucking and kicking resumed after the car began moving again – varying with terrain, speed and hill grad.

3. The symptoms grew gradually worse over time, as if the cause was accumulating or compounding.

A couple of years ago, the 83 went to Shop #1 for this. The shop had previously solved this problem with a carburetor replacement and other unknown work; although I asked fairly pointed questions, I could never get an explanation for the cause of the fuel starvation. More recently, after I paid them $800, the shop gave up, disposed of the car for me, and gave me credit against future work. The shop claimed they had replaced the fuel filter (how many fuel filters on an 83 GL 4WD wagon?).

A few months ago, the 87 DL 4WD wagon had the same symptoms (which also were solved previously by Shop #1 with no clear explanation). I took it to Shop #2, a Subaru independent repair specialist. They replaced the fuel pump and fuel filter, and it worked fine for 3 months. Then the fuel starvation symptoms recurred, growing gradually worse. By telephone, my mechanic rejected water in the fuel but did not rule out fuel contamination. He suggested the possibility of chemical interaction of different sources of fuels over time. I purchased a plastic fuel siphon, opened the tank top and siphoned gasoline off the tank bottom into a 2-quart clear glass jar, and let that sit for a week.

There were only a few drops of water in the bottom of the glass jar. To my surprise, the bottom was covered with a thin layer of off-white flakes, apparently a precipitate of some kind. I took the sample and car to Shop #2. They found the fuel filter totally blocked with tiny white flakes, which they described as looking like paint particles from washing out a brush with acrylic house paint. They were stumped on the source and too busy to clean the fuel system. Back to Shop #1, which subbed out steam cleaning the tank to a radiator shop, which said the gunk was almost 1/8 inch thick on the bottom of the tank, tenacious and hard to remove. They described the white particles as similar in appearance to “plaster dust” and theorized someone had vandalized my fuel. Unfortunately they disposed of all materials so nothing was left for analysis.


1. How likely is it that I experienced vandalism to both fuel tanks years apart? Note: the 87 was being kept in a locked garage when it first starting exhibiting fuel starvation.

2. Could the precipitate be caused by chemical interactions among different fuel brands and changing fuel chemical compositions (a lot has changed in fuels over 20 years)?

3. Were the fuel tanks coated with something white internally at the factory, and could that coating have flaked away over time? What was the steel treated with inside the fuel tank, if anything?

Thanks for your thoughts.


in answer to your questions, i would say no vandalism, fuel blend problems probably not, and the tanks are not coated. my guess is that the white stuff you see is just corrosion forming on the inside of the tank. the only way to get rid of that is pull the tank and clean it out or replace it.

as to your fuel filter question, which filter did they replace? carbed subarus from 1981 and up use 2 of them but subaru does not consider the filter under the hood an actual fuel filter. it’s referred to as a vapor separator. still a filter to me. the main filter is underneath the car but hesitation and surge problems can be caused by the vapor separator.

matter of fact, the vap. seps. often come from the factory connected backwards and this affects fuel flow.
ther are some other things that could cause these problems but i’ll give you a chance to digest this first. :slight_smile:

(just curious, top of the carbs blackened badly inside the air cleaner housing?

Ask Subaru of America what coating, if any, was used in gas tanks in the mid 1980s. Examine another mid 1980s Subaru tank. A bore scope could be used.

A: dont care B: dont care at the factory 1983 1987? come on.

I was advised by the shop that although most Subarus of that vintage had both a particulate filter and a “water” filter (probably the vapor separator you describe), the 87 was unique in several ways – including that it had only one fuel filter, which was white plastic. Following cleaning of the fuel tank, the car runs fine now. I have not inspected inside the air cleaner housing but will take a look and get back – what are we suspecting here? I forgot to say that the carburetor pre-heat hose was perforated and loose at the catalytic converter, so I replaced it and made sure the clamps were tight – this helped warm up the carb sooner in cold weather, but had nothing to do with the fuel starvation.

Thanks for the suggestion. I’ll research to find a technical support contact at Subaru of America.

the vapor separator was not in place to weed water out of the gas; it was one of several things done to fight vapor lock problems. the early/mid 80s subarus were very prone to this and the vehicles in hot clmates really suffered badly from vapor lock.

the vapor sep. will flow fuel faster one way than they will the other and this is why i mention they’re installed backwards. look at the arrow on the sep. and note which way gas flows. subaru says it’s not a problem; i disagree becase straightening them out has solved a number of poor running problems.

the reason for looking inside the air cleaner is to determine if the carb is excessively sooted up. this problem was caused by defective ASV valves (air suction valve). it’s an emission device used on many subarus of this era and is actually part of a recall since these things were very problematic. exhaust gas is funneled into the air cleaner housing and this plays havoc with the carburetors. hope that helps