Tuesday, September 13, 2016

Silica Gel Vent Filters--Dry Diesel and Gasoline Makes for a Happy Engine

rev 9-13-2016

The Problem. Ethanol loves to absorb water from the air. For all the talk and trouble water in the gasoline causes--far worse with e10--to me it's rather conspicuous by its absence that neither owners nor builders ever took a serious look at vent filters. Cars have had sealed tanks fitted with both pressure controls and filters since 1971. I considered this a few times--I've installed very large descant traps on very large chemical tanks--but figured if it was so against the conventional wisdom in the boating community, it couldn't be right. Funny.... That's not like me.

For the past 5 years I've had vent filters in the field on my boat and a number of friends. I monitored the humidity inside and outside and found the relative humidity to be 20-40% lower on the tank side of the filter, sufficiently dry to prevent water absorption. I've backed this up with lab testing and found a lot to like.

The value proposition:
  • Drier gas/fuel.
  • Less evaporation, to the tune of $8-$12/year. The unit should last ~ 10 years with very little service *, so that will nearly pay for it. Reduced engine troubles are the cake.
  • Reduced volatiles loss means better starting, particularly in cold weather.
  • Reduced alcohol loss means better resistance to phase separation/emulsion.
  • Reduced volatiles loss mean less gum formation (better solvency).
  • Reduced volatiles loss is good for the environment. Yes, that counts.
  • Less oxygen (less convection) means less gum formation.
  • Less water mean less risk of diesel bug. Just eliminating free water is not always enough. The little buggers can survive on emulsified and dissolved water. In testing, vent filters actually dried the diesel very slightly!

And the big one:
  • Less water means less corrosion. Even dissolved water is an electrolyte, allowing free movement of ions and accelerated corrosion.

Why am I sure? Because I believe in systematic testing.

Evaporation Testing

One liter bottles with 500 ml e-10, starting levels marked with tape. From left to right:
-  Plain 1/8-inch ID vent.
-  10 ml silica gel descant
-  10 ml activated carbon  adsorbent

The non-filter bottle lost 3 times as much fuel as the carbon and silica gel jars. Lost volitiles means hard starting, gum, and money lost.

Both carbon and silica gel are both adsorbents that pull water and organic vapors from the air. Silica gel (the packs you find in with your new DVD player) has a higher affinity for water, while carbon has a high affinity for organic vapors. However, both adsorb reversibly; that is, if exposed to high temperatures, and either clean air or an excess of something else, they release what they have previously adsorbed. Carbon can be flushed by steam and air, while silica gel can be flushed by alcohol, gasoline and air. Both have the effect of keeping the tank drier and reducing evaporation, adsorbing and desorbing with each day/night breathing cycle.

But silica gel is better from the boaters viewpoint:
  • Better water adsorption.
  • Water is self-scrubbing for years with e-10. the alcohol is enough to push the water back off, refreshing the filter for several years.
  • Carbon gets water saturated from dew in a marine environment. After just a few weeks of testing, my carbon filter was soaked and ineffective. Silica gel does not do this.

Corrosion Testing
Then there is the matter of corrosion. In my experience, the leading cause of carburetor trouble in this age of e-10 is not gum, but is plugging cause by aluminum hydroxide gel resulting from carb bowl corrosion. Dry fuel and less oxygen means less corrosion and better fuel stability.

Corrosion samples. No vent filter on the left, silica gel in the center, carbon on the right. This change took only 3 months.

Bottom Line: A gasoline of Diesel vent filter is a cheap investment in reduced engine trouble. The is simply nothing more aggravating, right down to your bones, than an engine that won't start when you are ready to head out.

Although the test filters photographed above were fabricated from PVC, this is NOT safe practice for permanent installation. PVC is not highly resistant to gasoline vapors and the adhesive is quite vulnerable over time.  While it won't fail in this laboratory setting, based on refinery experience with PVC, the joints will fail if I add heat, vibration, and wait several years.

The Solution. Only one company makes these, but they are rugged and none of our test units have failed. A few weeks ago I opened my carbs for the first time after 5 years, just in the interest of product research, since they are still running fine. As expected, they were shiny as a mirror, not pitted as I have observed in the past, in the same time period, with the same model motors. I used to clean carbs all the time, but not since I installed a vent filter, which is nice.

Installation note. They need to be at a high spot where sea water cannot enter and tank overfills cannot contaminate the gel. I put a Parker Lifeguard between the tank and vent for mine, which has worked very well.

My filter first vent filter was custom fabricated so that I could test many options and take measurements more easily, but it works the same. Note the high location and the fuel/air separator to the left.

 *  Regeneration: Experience shows they do required regeneration every few years, but it's free and quite easy. Simply remove the silica gel fill. put it in a pan, and slowly heat it on the grill until it turns blue again. This takes about 15 minutes, there is virtually no smell, and can be repeated numerous times. Allow to cool before reinstalling the gel. Piece of cake.

1 comment:

  1. Awesome post, Drew. I need to add a silica vent filter. . .