Include

[Wapitibill]

Talk about perfect timing! I recently made up my first batches of biodiesel. A good friend of mine is a petrolium inspector for the state of Wisconsin. Last week I gave him a sample and had him run the standard tests as done on commercial diesel fuel. The lab report came back to me this morning.

The sample I submitted was made from new vegetable oil following the "standard" receipe: 200ml methanol, 3.5gm lye, made in an old blender. I did not wash the sample.

The main test done on fuel is what is called the distillation curve test. A sample of fuel is heated. The temperature at which the fuel first starts to boil is noted. Then, the temperatures at which 10% on through 90% of fuel boils, is noted. The resulting profile determines the characteristics of the fuel. Contamination (such as the inevitible traces of gasoline found in diesel fuel) will cause spikes in the distillation curve. In addition, the flash point, specific gravity, cloud point and other parameters are tested.

The test results of my home-brewed biodiesel were "not good" (in my friend's words.) Without going through the sea of numbers he sent (unless you want me to...) my biodiesel started boiling at 192 deg. F. Diesel fuel (D2) starts boiling at 324 deg. F. (This was probably the residual methanol in the biodiesel.) After that, not much of anything happend until the temp was above 600 degrees, then all hell broke loose. D2 boils at a nice, even rate from 410 through 630 deg. F.

To "cut to the chase," I'll post the lab report verbatim.

"The sample was a solid mass when removed from the lab freezer, which was @18 deg. F.

The solid mass rapidly liquefied when placed in the fume hood. Room temp. was @ 63 deg. F. The sample was completely liquefied in about 30 min.

When completely liquefied, the sample was hazy in appearance. When the container was agitiated, and allowed to rest, a substance which was first thought to be water settled out in droplets.

When again allowed to rest, the liquid showed a phase seperation. When the lower phase material (NOTE: "lower phase" refers to the stuff at the bottom of the bottle) was tested with water detection paste there was no reaction.

The liquid was then carefuly decanted into a large glass cylinder, with all of the lower phase material remaining in the original container. To remove the hazy appearence, the liquid was filtered through ashless filter paper. After filtering, the product was clear and bright, free of any undissolved water and sediment.

The resulting properties of the filtrate was as follows:

API specific gravity: 28.0 (regular D2 sample is 30.2)
Flash Point: 85 deg. F (regular D2 sample is 172)
Sulphur: .0066 wt.% (regular D2 sample is .0208)

The first and second distillation attempt failed. The Initial Boiling Point (IBP) occured rapidly (140 seconds) @ 146 deg. F. After 5% recovery @ 157 deg. F, the distillation rate fell to 0.0 ml/min and the [automated] distillation unit aborted the test. Along with the heavy gravity and low flash point, this indicated that the sample seems to consist of both very low and very high boiling range components, with few midrange components. The 10ml of the recovered distillate from both distillations was collected and examined. After settling, the distillate phase seperated into two components approximately 20% lower phase, 80% upper phase. Both portions of the undistilled sample were retained and conmingled [mixed.]

With the low boiling range components removed, a distillation was performed on the conmingled sample. When heated, some popping and somewhat violent boiling occured, with steam and water droplet formation in the receiver. The IBP of 194 deg. F and the later apperence of water droplets in the bottom of the receiver suggest that the initial drops of distillate consisted of dissolved water in the sample. (NOTE: The methanol I used was 1980 vintage. It may have absorbed some water from the air.)

After the IBP, the distillation temperature remained in a tight range of 622 deg. F to 626 deg. F @ 50% recovered, at which point the distillation was halted. This suggests that some thermal cracking [refining] may have occured, resulting in smaller, lower boiling point molecules forming. The distillation residue in the flask was dark and viscous, and was somewhat difficult to dissolve in tolulene."

(End of lab report quotation.)


I tested some of my homebrew in a 15KVA Kohler genset powered with a Perkins 4-108 engine. It put out full power on the loadbank, and didn't produce any unusual smoke. It seemed to run cooler than when I tested it on D2. The exhaust had the characteristic "french fry" smell. In my opinion the engine actually ran quieter than it did on D2. This is strange because I would think that the "all at once" boiling point characteristic of the biodiesel would cause the engine to run noisier.