Almost finished with my second test batch (one 20 gallons and one 18 gallons) and I have removed the soap both times without water or magnesol.

My latest batch (almost done) was:

1. 18 gallons of wvo heated to 130F
2. titrate at .8
3. added 590 grams of KOH to 4 gallons of methanol
4. reacted for 1 hour
5. moved to settling tank
6. settled overnight
7. transferred back to processor and heated to 180F to remove trace methanol
8. Transfer to wash tank
9. spin BD in spinnerII modal 60 centrifuge at 75psi to remove soap

Soap Tests:
12:15 (start) = 1976 ppm
2:00 = 1520 ppm
4:00 = 1140 ppm
6:15 = 760 ppm
9:30 = 380 ppm
Still running to clean the rest out.

The soap caught in the bowl is a very thick, dark brown syrup of soap. So far, I have about 600ml of this substance I have removed from the BD.

No water or magnesol was added. Just time spent spinning in the filter.

I have done this for the second time. The first time was with a 5% prewash batch. The “shake um up test” passed great, just like when I use magnesol.

I am sure I can cut some of the time out of the process as I currently drain the filter into the same tank. I plan on now setting up a 2 tank system to cycle the “clean” fuel to the empty tank and back and forth to clean. This should greatly reduce the time running the pump.

Just thought I would post the results. Comments?

OK, I'll bite. Where do I get the equipment?

I picked up the filter for removing the magnesol from BD. I got it from Ebay. it is the same oil filter trucks use to filter motor oil. I am driving it with a 3/4hp positive-displacement pump turned WAY down. I bet it is running about 500rpm.

The filter was about $200 and the pump it got for free from some guys at work.

The majority of the syrup I removed was liquid, not solid. The solid soaps stick to the side of the filter bowl. The liquid soaps are staying inside the bowl when it is spinning, allowing the BD to pass by. When I stop the filter, I remove the bowl and open it over a pan to clean it where all of the liquid soaps come out.

When I start the filter, my BD at 1976ppm was clear and "clean" looking. By the time I was done, I removed about 600ml of the liquid. I imagine the amount is a little more as I could not get 100% of the soap in the pan. Imagine trying to move maple syrup from one bowl to another.

Forgot, workshop temp was 60F. The BD was still about 120F when I start though.

http://www.spinnerii.com/index.cfm/lev1/401/pressID/534/startrow/1

website. Wonder if you could use the guts from an old dyson vacuum cleaner instead?

quote:

Originally posted by bayshorecs:

I am sure I can cut some of the time out of the process as I currently drain the filter into the same tank. I plan on now setting up a 2 tank system to cycle the “clean” fuel to the empty tank and back and forth to clean. This should greatly reduce the time running the pump.

Just thought I would post the results. Comments?

excellent results and you are right you may see a big improvement using transfer instead of pump around. Be interesting then to see what the PPM are after each transfer. And how long a transfer takes. Keep up the good work and post your results here.

bayshorecs

I have a couple of questions,

How do you know how many ppm of soap is present?

How often did you have to clear the filter of soap?

How about some pictures.

I'd love to get away from waterwashing.

great idea, btw,

Jason

bayshorecs,

Any guesses on the purity of the soap?

Since I use 10g samples it simplifies to:

ppm soap = mL of 0.01N HCL * 304.4

So:
1976 ppm = 6.5mL of 0.01N HCL
1520 ppm = 5mL of 0.01N HCL
1140 ppm = 3.75mL of 0.01N HCL
760 ppm = 2.5mL of 0.01N HCL
380 ppm = 1.25mL of 0.01N HCL


I cleared the filter about every 2 hours which was more often than needed. I am in process of setting up a two tank system. This should greatly reduce the processing time as I am not poluting the "dirty" fuel with the "clean" fuel. Each time I cleaned, I got about 100mL of the syrup in the pan. It looks like very thick glycerin.

I will take some pics of the final spin tonight along with the jar of syrup. I plan to let it run for another 3 hours and see where I end up. My first batch hit 75ppm before I quit and sent it through the 5 micron and 2 micron filters to storage. I figured there was little point continuing without the 2 tank setup as I would never be able to process 100% of the fuel, it would always be a percentage.

quote:

Originally posted by RickDaTech:
bayshorecs,

Any guesses on the purity of the soap?

I still have the soap in a jar. What tests do you want done?

Working it out in my head:

1ppm=1mg/1L

2000ppm = 2000mg/L = 2g/L * 68L = 126g of soaps.

I have about 500ml of the syrup. Anyone able to work this out further?

And I pulled my BD sample from the exit flow of the filter, not from the top of the tank.

Where can one find a Spinner II filter for sale?

It seems impossible to find a price list anywhere.

Bob

quote:

Originally posted by bayshorecs:
Since I use 10g samples it simplifies to:

ppm soap = mL of 0.01N HCL * 304.4

Forgive my ignorance, but what does this mean? I assume that 0.01n HCL is what you removed from your filter? what about the 304.4? I know I'm probably missing something obvious.

I'm looking forward to the pictures of your setup and results.

Jason

That is the conversion formula for calculating soap count.

Refer to: http://www.biodieselcommunity.org/testingforsoap/

I guess a more accurate way to write this would have been:

x ppm soap = y mL of 0.01N HCL * 304.4

You add y amount of .01N HCL and get x ppm soap.

Just chemistry.

I had to search on ebay for about 1.5 months to get mine. You can buy one new for $600 I got mine for about $200 on ebay

quote:

1ppm=1mg/1L

2000ppm = 2000mg/L = 2g/L * 68L = 126g of soaps.

I have about 500ml of the syrup

I wonder if the surplus liquid is polymerised oil or polymerised biodiesel perhaps?

Any good way to tell?

Unfortunately, I no longer have the glycerin from this batch to do any tests on. The next batch I make, I will try to keep some of everything and include samples during each soap check.

You could probably find the purity by using the soap titration test you referenced. Only you would want to use 0.1N HCL instead of 0.01N HCL and cut the sample back to 1 ml. That way the results would come out in parts per thousand rather than ppm. It should take about 1-2 ml to titrate at these ratios. That would give you the ratio of soap to biodiesel.

Of course you could just add some KOH & water and turn the biodiesel into soap as well. Then you have some fine dish soap.

could some of the surplus be gyderine?

It looks a lot like thick glycerin. It sat in the standoff tank for about 5 hours after the methanol removal. I am sure some of it is though.

Maybe I can remove the glycerin through the centrifuge too. That would make for a lot of bowl cleaning though.

think I will try hooking up gthe cyclone from a dyson vacuum cleaner to a pump just to see what happens. Probably craze and melt without doing anything useful but be interesting to find out. At least its a big 'bowl' so would need little cleaning per batch.