Fuel test results

Attached is a pdf of the spreadsheet containing the test results of  
the fuel testing Zan and I performed from 2/8-2/12/07. The results  
indicate that when a significant amount of water is within the source  
oil, no single factor can lead to a good reaction.

Here is what we found:

1. Our KOH on-site is good. We successfully reacted fuel with it and  
passed the 3/27 test well.
2. The reasons for the last three problem fuel batches have been found:
	a) mayonnaise was caused by deficit of lye in a batch; incomplete  
lye = incomplete reaction
	b) incomplete reactions caused not enough methanol AND high amounts  
of water in oil
3. We have confirmed the correct proportions to get good results on  
the table top. This includes 7.78 as our base amount for lye; 22%  
methanol per volume of oil
4. Heat of reaction is not as critical as water content of oil,  
although its still a factor.
5. Ambient heat during the settling period after reaction is not  
critical. In fact, it made no difference.
6. Water content in oil has been our CHIEF culprit since the weather  
cooled; significantly less amounts of water have settled to bottom of  
our oil totes. We MUST get this water out of the oil before reacting.
7. There are two types of precipitates possible from a 3/27 test:  
globules of unreacted oil and water-soluble impurities that appear  
like tiny grey fuzzy material. Presence of these is not a sign of a  
failed 3/27 test, as the fuzzy material dissolves in water and can be  
removed by washing.

Our methodology:

We started with pure, unused soybean oil and first experimented with  
varying amounts of lye, methanol and changing heat of reaction,  
ambient settling temperature, etc. We wanted to see what factors  
independently affected the quality of reaction. Additional methanol  
did not improve any reactions.

After determining that only less lye made a significant difference in  
reaction quality, we moved to good quality WVO with the same  
proportions we used for good results with unused oil.

We also tested the lye from our site to determine efficacy, and we  
also tested phenol red versus phenolthalene to compare these two  
indicators for quality of titration. Phenol red is a poor indicator  
and resulted in much lower titration amounts, which led to poorly  
reacted fuel.

We then tested extremely poor quality WVO which contained a  
significant quantity of water. No matter what variables we used, we  
were unable to get a good reaction.

Further tests could be conducted to try a combination of factors for  
better reacting poor, water-laden WVO, but it is advised to remove  
the water from the oil before reaction.

_______________________________________________
mixmasters mailing list
[email protected]
http://lists.gobiodiesel.org/mailman/listinfo/mixmasters