In a moment of complete stupidity, I started distilling the methanol from the fuel without removing the glycerol. The distillation was allowed to happen for a few hours before my error was noticed. I'm assuming that I must have reversed the reaction to some degree by removing the methanol with the glycerine still in the tank, but I'm not sure because I havent been able to do the 27/3 test yet due to the fuel being too hot to pump out the glycerine. Does anyone have any advice as to how I can save this batch? Could it be as simple as draining the glycerol and then adding 1 gram of KOH for every liter of oil? If so, how much methanol should be used?

Thanks for reading this. Hopefully this post about my stupid mistake will prevent someone else from accidently doing it too.

Can't answer your questions sorry, but can you please keep us posted on whether you discover a reverse reaction? You'll fix it - it's all just time and attention.

There is an entire thread dedicated to this subject.

http://biodiesel.infopop.cc/eve/forums/a/tpc/f/44410893...581097391#8581097391

They are trying to make it work. You did it by accident. I don't know much about it, but there may be something useful in there for you.

-Jim

Hi Folks,

I work with D'son Biod.

We had recovered about 3 gallons of methanol from the combined biodiesel/glycerol blend before we realized that the gly had not been drained. Normally we only get 1.75 gallons of methanol from a 54 gallon batch of fuel (we get roughly 2-3 gals from the glycerol). When we had passed 2 gallons (and the boiler temp was still in the 170's F) I got concerned and did some backtracking, wherein we discovered the mistake.

I ran a 27/3 test on the fuel in question after we drained the glycerol. It passed. However being skeptical I did a reprocess test in a jar, with 1 gram NaOH and 50 ml methanol per liter. A small amount of glycerol dropped out... a thin line at the bottom of the jar.

I should have quit here. There was a haze also near the bottom of the test jar, little whispy white stuff. I mixed up the whole reactor's worth of fuel with the equivalent of 50 ml methanol and 1 gram NaOH per liter of fuel. After mixing things up in a repeat reaction, I noticed that the wispy stuff in my test jar was actually soap clumps... a partial gelling, indicating that the reaction was almost full conversion.

Well, I got some soapy biodiesel out of the reactor. Lots of soap scum on the top of the settling tank, and some defninite snot visible in the tubing as we pumped. I pulled a sample in a beaker and D'son said she saw gelling in the bottom today.

The situation was further complicated by a washing error... we wash with a 1 GPM head in order to give our condensor cooling water somewhere to go. We were busy with students in the shop today during the wash, so allowed the fuel/water level in the wash tank to rise above the 1GPM "mist" nozzle. The high pressure of the mist coupled with the very soapy fuel resulted in a full tank emulsion...! Rats

I now have about 20 gallons of BD sitting on top of 100 gallons of white, soapy water. Hoping this will settle back out in the next few days when the weather warms, otherwise its back to the drawing board.

The moral of this story?

We now have a checklist on our reactor for each batch, so that multiple operators can run a batch through various stages without question of what has or has not been done

• If we do a reprocess test, we'll make it large enough to get an accurate sample, and watch for any wispyness above the glycerol dropout
  
• We'll probably just run the incompletely converted fuel rather than goof around with reprocessing in the future.

Well, that's the latest on batch # SU-07-17

Thanks for your interest,

Farmer