The Free Fuel Trial - Low-tech Filtering Methods

A pair of jeans with the legs tied off in a knot (separately) makes a good 10 micron filter. Just hang the jeans inside a bin and fill the legs with oil!

A J-Cloth in a simple frame (see below) also makes a very effective prefilter, but for smaller amounts at a time.

Unbelievably, the simplest and most effective cold filtering method for refining we've found so far consists of a sheet of newspaper! (Broadsheets are best because of the large surface area - at last we've found a second, non-hygeinic, use for the Daily Telegraph). We made up a simple wooden clamp with wing nuts to hold the edges of the newspaper page together (see the picture), then hung the filter inside a plastic bin and carefully filled it with oil. It didn't tear or split and the oil filtered slowly through by capillary action. The paper sheet held 3 litres of oil at a time and filters it down to 2 microns, which is very suitable for use in a diesel engine. Old, clogged, used filter sheets just get used to light the fire, or composted. A J-cloth in the same frame gives a quicker coarse pre-filter.

The newspaper filter	The newspaper filter in action
A J-cloth in the same clamp	A J-cloth filter at work

Pictures above courtesy of Neil Park, inventor of the device. Neil has contributed a follow-up, based on his experiments to date:

"I thought we might be interested in this photo, which shows the thick gunky stuff that the filtration process takes out. This is the stuff that complicates the whole process, as it will block up the filter and slow the filtration to a crawl. In the oil I have been working with about 10% of the volume will end up in the gunk category. When you receive the oil and start filtering you think it is speeding through, but as you get down the drum it gets thicker and slower as more of the heavier gunk gets into the filter with each top up. The fact that it is heavier is an advantage though, and suggests that for larger quantities you should start with settlement, followed with coarse filtration (in my case J cloth), followed by fine filtration with newspaper to finish. Most of the gunk in the photo was scraped off the inside of the J cloth as it went through and is a testament to the strength of the J cloth that it allows this kind of treatment.

You asked about filtration rates at one point, and you can see from the above why it is difficult to give a straight answer to that question. If the oil is settled and then taken off the top, ie fairly clear oil, then it will go through the J cloth filter as fast as you can pour it. Obviously the thicker it is going in the slower it becomes. The J cloth filter will take out the vast majority of gunk, so after it has been through that it will go through the newspaper filter at about 3 litres a day. I have one running just now that was the one in the previous photo sent to you, and it has put through about 20 litres and still going on the same sheet of newspaper.

Wet or Dirty Oil causes big problems. Here's Neil's description:

"Filtering was going well with the oil from my first supplier, so I decided to expand and try a second source of oil. This has shown how good the quality of the oil from the first source is. I was estimating that the proportion of gunk in the first source was somewhere around the 10% mark, whereas the percentage of gunk in the second source is in the region of 50%! What a difference. At this level of fats it is semi solid and is very difficult to pour. After a chilly May night recently it would not come out the drum at all!

Having said that, it is still filterable although it does take a lot more time and attention to keep it moving through the system. The J cloth filter is remarkably good at taking out the heavier fats from the mixture. However, when I came to filter it through the newspaper filter I found out the second major problem with this source of oil, it is contaminated with water. Water is the thing that weakens the structure of the newspaper, and this will not last for long if there is water in the oil. As the oil is quite dark, and any contaminating water is similarly dark, it is very difficult to see water that is there as a separate layer (water is heavier than oil, so should form as a lower layer). The first newspaper filter to fail on me was expected, since I noticed the lower layer going into it when I was filling it up, only realising what it was when it was already in the filter. After about ten minutes the filter broke and the contents streamed out into the holding container, which is an advantage of this design of filter in that it does fail safe!

I then collected up more of the J filtered oil in a container, to allow it to settle out so that I could take the oil from the top, keeping well clear of any water that would have settled to the bottom of the container. Unfortunately this failed as well! Although the oil that went into the newspaper filter did not have a layer of water with it, close inspection of the oil showed a dark particulate held in suspension. After the second newspaper filter failed, this time in one sudden splurge which does produce some splashing to the surroundings, I realised that this fine "particulate" is in fact small droplets of water held in suspension. The fact that these droplets are as stable as they appear to be, makes me wonder if there is also some soap or detergent in the mixture that is helping to stabilise the droplets.

Prior to this new batch of oil I have only had one newspaper filter fail on me, and that was when I put my thumb through it when transferring it from one container to another which is not really a failure, except perhaps on my part! So the newspaper filter has proved to be a reliable and high quality filter on dry oil. However it is just extra work that goes nowhere if water contamination is present.

There are various possible ways of removing the water, but that will require further testing to see what works and what does not. What does stand out quite clearly from this is that the quality of used vegetable oil is very variable, and the ease or difficulty of cleaning it up depends to a large degree on the quality of the oil to start with. As with all recycling and reuse projects, the cleaner it is to start with, the easier it is to recover to a usable product. The more contaminants you have to start with the more complex the process needs to be to remove them. The best solution of course would be to prevent the contamination in the first place, so perhaps I should have a word with the source of the oil, as it may take only minor changes to their routine to prevent all water and detergent contamination."