Sun Break Processing

Hello,

I've just subscribed to this list. Since our company needs to become a
co-op member for this all to work. I figured that Dave and I should be
looking at this list.

I would like to keep you all updated from now on, so we can keep
everyone informed on how each batch is going. Problems encountered,
and what we are doing about it.

A  5% pre-wash should be possible in our reactor. We recently filled
the reactor to absolute max and think we can get between 10 - 14
gallons of water added safely.

I have a couple of questions about how you were doing your pre-wash.

1) 5% per volume of oil or total working fluid in the reactor?

2) At what point during the reaction did you add the water in the reactor?

3) Did you pre-heat your water and to what temp?

4) How hot was the reactor when you added the pre-wash?

5) Did you pour the water in the reactor or induct through a pump?

We have done pre-wash fuel before and had good success. However we
used to add the water directly in the top of an open reactor. The
current reactor we use is a closed system and once the reaction starts
we don't open it again until the reaction is complete, settled and
ready to drain.

I could induct the water, but worried about the agitation incurred
within the pump.


-------------------------------------------------------------------------------
Processing Update
-------------------------------------------------------------------------------

We have processed 4 batches of fuel and have had very good success.

Batch #1 - Titrated at 2.9ml, Used 6.97kg NaOH
Batch #2 - Titrated at 2.8ml, Used 6.86kg NaOH
Batch #3 - Titrated at 1.8ml, Used 5.78kg NaOH
Batch #4 - Titrated at 2.5ml, Used 6.50kg NaOH
Batch #5 - Titrated at 2.9ml, Used 7.00kg NaOH
Batch #6 - Titrated at 2.7ml, Used 8.12kg NaOH - Girlmark Formula
Batch #7 - Titrated at 2.8ml, Used 8.22kg NaOH - Girlmark Formula

All of the first batches were great success with very clean
separations and good looking product.

However with #5 we ended up with too much oil and the separation was
above our standpipe and we got alot of mixed glycerol in the first
barrel drained. We have set this first barrel aside to inspect
further. The remaining 4 barrels looked fine.

Girlmark and others have suggested that the 3.5g base reaction
quantity is wrong and 5g is closer to the real number. These numbers
have been verified with GC testing.

Dave and I were really cautious about this since we have had bad
experiences when you add too much lye. We derived our own formula last
year from our own experimenting.

Our formula is (((titration * 1.1) + 3.5) * total litres).

You'll notice that we always add 10% to the titration. This minor
adjustment was putting us much closer to the 95-96% conversion ratio.

However the GC tests done by people on the infopop forum suggest that
you can increase the conversion ratio as high as 98-99% with the
following formula.

Infopop GC Formula (((titration ^ 1.1) + 5.0) * total litres)

We decided to do a mini-batch with the increased amount of NaOH and
felt comfortable that we wouldn't make glop.

Soon after we processed the whole batch. I decided that to be cautious
we would be a little lean on NaOH. The formula called for 8.32kg and
we used 8.12kg.

The results of batch #6 were great! The fuel looked great but we
noticed that once again we had slightly overloaded the processor as
our separation was above the standpipe. Although this wasn't a major
problem. So once again I set aside the first barrel for later
inspection and drained the additional four.

So now we feel confident that Girlmark's formula is better and we
proceed to process batch #7. This is were things got ... well
different.

There are several variables which changed with batch #7.

1) There was alot of paraffin wax in the oil, we removed as much as we
could but there was still some in there.

2) The titration initially was high, around 4.9ml. So we re-titrated
another 2 times with completely different samples and consistently
titrated at 2.8ml and decided our first titration was in error.

3) We normally make a weak methoxide with 2kg NaOH in 55 gallons and
induct around 10-12 gallons to leave some room in the barrel for
mixing the remaining 3-6kg. There was some miscommunication and the
entire 8kg was added to the already full 55 gallon barrel leaving it
very very very full. Mixing was difficult, so we inducted some of the
methoxide from the top of the barrel in an attempt to free up some
room in the barrel for mixing. Upon inducting we found that the tank
temperature was too hot and any methoxide we inducted would flash and
stall the pump. You really want both the oil and methoxide to be hot
when mixed. By the time the tank temp dropped the methoxide was not
nearly as hot when inducted. I also noted that during our mixing of
the methoxide it didn't get as hot as normal. Usually you know it is
hot enough when you start smelling the paint on the outside of the
barrel ;)

4) We ran out of NaOH and decided to use some NaOH that had survived
our fire last year. It looked alright, but in retrospect.... perhaps
not a good idea.

5) We heat our oil to 180-190F and then let it sit over night to let
the water settle out. The next day we would drain 4-8 gallons to
remove most of the water. The process may be repeated depending on how
the oil looks. The next day after our initial we forgot the drain off
the water. In an attempt to fix the situation we heated again and
settled for 3 hours and drained 4 gallons off. This was not good
enough.

So during the reaction I didn't see the glycerol precipitate like I
normally do. Although I was seeing plenty of glycerol build up near
the bottom of the tank. I knew something was up ... I decided to leave
the system recirculating for about 8 hours instead of our usual 4-6.

I turned off the recirculation and took a sample off of the top of the
tank. I came back 5 hours later to inspect the sample and found a
considerable amount of soap fall out. Much more than normal. I
proceeded to take another sample from the top of the tank hoping for
better results after several hours of settling.

I came back to find that the second sample also had considerable soap
fall out. However there was something else that I have never seen
before happening. The fuel was turning gelatin. Literally like jello
shots.

We drained the system and found that the glycerol had partially set up
in the bottom of the tank .. Ugh ... we normally don't have this
problem. So we started cleaning the tank out, yuck!

We had a definite scum line and lots and lots of soap.

So ..... we are going to investigate this reaction further and see
what we can do about salvaging. Take in mind, this batch could have
been much much much worse. We have fuel, it just has odd properties.

Any ideas?

Currently we are primarily blaming ...

1) Water content; Explains the soap

2) Paraffin; The only way I can explain the fuel settling up like Jello.

So ... what do you all think?

We are halting production until we can fix this fuel. This is beneficial.

a) The co-op can use this time to catch up on washing

b) We will be attempting to remove the soap through methanol recovery
/ fallout. This is the method described by Graham Laming on the
infopop forum. If this works out we can start recovering our methanol
from both the fuel and the glycerol fall out. Check out Graham's
method here.

http://biodiesel.infopop.cc/eve/forums/a/tpc/f/719605551/m/2811039341

I have no idea how to get the paraffin out other than perhaps a
recirculating flow through a filter as the temp drops. A real pain,
but incremental filtering might be the only way.

--
Jeff Brandt
Sun Break Biofuels, LLC
mailto:jeffb-hD0ojMJ1hjDksvZKaRJZc1aTQe2KTcn/@public.gmane.org
503-317-4564