If anyone is thinking of assembling this apparatus I would suggest using 5mm clear pvc hose rather than the 3mm I have used. The 3mm hose is difficult to fill with coloured water and tends to have air gaps can affect the reading.
The calcium carbide I bought on Ebay works really well. It is in the form of stones about an inch in diameter. I break up a stone with a hammer and crush it to a fine powder, fill the plastic drinks bottle top and use it immediately. If left in contact with air it will react with the ambient moisture and be useless very quickly.
Back to my tests. All these results were recorded at a temperature of 15 degrees C, this is important because if the temp is not consistent the results will be wrong.
I started by testing a sample of the same bio again. 308mm this time, that is a very consistent result. I then tested a sample of dino diesel. The result was 50mm. I checked it by doubling the sample size to 100 gms and this time it was 102mm
If I assume the dinodiesel sample is around 500 parts per million then my bio sample contains 3000 parts of water per million! This sample was taken from a batch which had been dried for about 2 hours so that result was possible. I had another batch which had been dried for over four hours so I tried a sample of that, result 80mm
I took some of that last bio and raised it to 120 degreesC for 10 minutes and let it cool. when tested it read 40mm.
These results suggest that the readings do correlate to water content. They also suggest that I am not drying my biodiesel sufficiently.
What I would like to do next is to do a fresh batch of biodiesel and take test samples after every hour of drying to see just how long it takes to dry bio down to 500 ppm.
Note that both samples of bio were sparkling clear.
I was quite alarmed by the results of my test on my supposedly dry biodiesel so I decided to take it a stage further. I decided to put the bio which measures 310mm back into the drying tank and see how long it would take to get it really dry.
My drying outfit is a steel oildrum with a bubble shaped fountain fitting in it powered by a small central heating pump. A 2000w heat gun blows hot air over the fountain drying the biodiesel.. I have found that the bio becomes sparkling clear in 2-3 hours with this setup.
I had previously taken the samples from the bottom of the tank which gave the reading of 310. this time I circulated the bio for 5 minutes before taking the first sample. I then switched on the hot air and took a sample every hour for 6 hours. I put them into sealed glass jars and allowed them to cool for 24 hours.
This morning I tested each sample. The temperature was 15 degreesC in my workshop and all of the samples looked identically clear.
From these readings I think I can make some assumptions.
First the initial reading of 310 was high because I took it from the bottom of the tank. When the bio was mixed up for five minutes it read at 245mm, better but still pretty bad if the ASTM limit is 50.
Second It took 6 hours of drying plus the two hours the bio had already been subjected to, a total of 8 hours to dry the fuel down to the same reading as ASTM standard dinodiesel. i need to rethink my drying system.
Third the readings do strongly suggest a correlation with water content.
Has anyone assembled the apparatus yet, lets call it a Carbide Manometer, If so I want to hear your results as I think I have gone as far as I can for the moment. We need someone independent to repeat these tests.
Nicely done. I have been following your activities and results and find them quite interesting.
Tim C Cook has done of lot of work relating to drying WVO, primarily, but his techniques apply equally well to biodiesel. Do a "find" for Tim and look at some of his posts. You might get some good information related to your drying issues.
I have worked with Tim on a number of other projects and you are right, he is just the man for this problem. However I should really keep this discussion confined to the carbide manometer rather than discussing drying methods.
The calibration of the manometer is worth looking at. The simplest way to do this is to test a sample of ordinary dino diesel and use that reading as a pass/fail level. But this will not tell you the actual water content of your biodiesel.
Another approach is to take two samples of biodiesel, one washed but not dried, the other thoroughly dried, and have them tested by a lab. Using those two reference samples it is a simple matter to calculate the water content of any sample.
The two reference points can be calculated by testing the two samples using the weigh - heat - weigh method which is described elsewhere. Use a large sample, 1 litre or more, to get really accurate results. This will involve you acquiring a triple beam balance or similar accurate weighing scale capable of measuring to a tenth of a gram.
Imakebiodiesel, to calibrate, how about making some bio with know quantities of water? Start with drying a small measured batch of bio to zero water (heating in a shallow pan?), then add to it a known amount of water and shake em up. Would that work?
Once I get some calcium carbide, I'd play with it.
That is really the way will need to calibrate. The idea of compareing to pump dino wouldnt work because you don tknow the water content of the pump fuel, it could be bone dry or it could be soppin wet. On a weight/weight basis .2 grams (close to .2ml) in 400 grams of bone dry Bio would then give .05% or 50PPM water which is the pass/fail on the ASTM spec.
If you can't dazzel them with brilliance, then baffel them with bullchit.
I agree the idea of comparing to dino diesel is not very good. You have to assume it is ASTM standard and it might well not be.
If you have a weighing scales that is accurate to 1/10 of a gram then either testing two samples by the w-h-w method or making up your own reference samples by adding water to a dry sample will work. Both methods are dependent on the accuracy of the scales.
Bear in mind that the readings of the carbide manometer are temperature dependent. In hot weather the readings will be higher than when it is cold. Im not sure how much of a factor this is but it should be easy to find out and correct for any temperature difference.
The temperature adjustment is dead easy, Charles Law states that volume of a gas is directly proportional to its temperature in Kelvin ( or centigrade ).
So if you calibrate your manometer at a temperature of 15 degrees C and later take a reading when the temperature is 20 degrees, divide your reading by 20 and multiply by 15 to get a true calibrated result.
But, your manometer measures pressure, not temperature. Does Boyle's law have any impact on the calibration?
Imake, as always your methods impress me. When you dry bd for two or three hours do you start out with room temp bd and warm it by just using the heat gun on the spray or do you pre heat the bd?
Boyles Law is the principle by which the manometer works. The greater the volume of gas produced the higher the pressure within the vessel as long as the temperature stays constant.Incidentally Robert Boyle was born and lived here in Lismore in Ireland, perhaps thats what led me in this direction. But Charles law states that if the temperature rises than so does the volume and therefore the pressure. So you must adjust your readings if you take a reading at a higher or lower temp than the temp you calibrated at.
Raften , when I am washing my bio I heat the wash water with 200w mat heater attached to the bottom of the tank. By the time washing is finished the bio is around 25 degrees C. I dry the fuel using the setup described earlier for a couple of hours and the temp of the bio goes up to around 30 degreesC. It goes from cloudy to sparkling clear in that time. I turn off the dryer and the mat heater and put the lid onto the tank. I have always assumed that sparkling clear bio must be dry or nearly dry up until now. I do remember Girl Mark expressing doubts about using this as an indicator but the Mercedes engine seemed to run fine so I didnt worry. However the water levels indicated by my tests would probably cause long term problems so I feel I need to correct my drying methods.
I wonder how many other biodiesel brewers do the same as me. I think a simple quick test like the carbide manometer would help a lot of people make better safer biodiese.
Imake, have you done a vapor test or refrigerator test on the fuel that you previously considered dry. Those are my two tests and always pass with flying colors. I may have to make a device like yours to be sure. I bring my bd to 120f, 49c, then bubble and blow dry for three hours.
I have used the refrigerator test occasionally and it seemed ok,
Your drying procedure obviously works better than mine. Your temperature is nearly double mine and you blow dry for about half the time my system takes. I think I have been a bit too cautious about applying the heat. On my next batch I will try preheating the bio before blow drying.
Going back to the subject of calibration, the DIN standard for diesel in Europe is 500 ppm or mg/kg of water and sediment. This means that in a kilogram of biodiesel (roughly a litre) there must be no more than half a gram of water.
The method for preparing reference samples might be as follows...
1. Bring 2.5 litres of biodiesel up to 120 degrees C and maintain for one hour to dry completely.
2. Weigh out exactly 1 kilogram of dry biodiesel and add exactly half a gram of water, mix throughly and store in a sealed jar.
3. Prepare a second sample in the same way but with 2 grams of water.
4. Allow the samples to cool to room temperature before testing.
This provides you with a sample at pass standard of 500 ppm and another at 2000 ppm and it is a simple matter to calibrate the carbide manometer with these two samples.
My aplogies, I made a mistake in the temperature correction post. The temp must be measured in Kelvin, not centigrade. This means that you add 273 to the temp in centigrade to get kelvin. So the temperature correction is very small for a change in temp of 10 or even 20 degrees. In fact its hardly worth bothering with since the test is only accurate to 2% at best using a basic weighing scales.
For those without accurate scales, you can make reference samples by volume. Maybe a medical syringe for the water?
You are absolutely right, volume is the obvious way to go. We all have the equipment.
A litre of dry biodiesel and half a ml of water would be right for 500 ppm while another litre with say 2 ml would give 2000 ppm. Im going to have a go tomorrow. I have also found a test lab at a Technology Institute in Waterford city not far away who will test my bio and give me a water content reading accurate to 20ppm. Once I have my carbide manometer calibrated it woud be interesting to have a sample tested to see just how accurate, or inaccurate, it really is.
Today I calibrated my carbide manometer. I began by bringing 2.5 litres of biodiesel up to 125 degrees C and kept it there for 2 hours. Steam stopped rising of the surface after an hour but I wanted to be sure.
Using a clean dry graduated cylinder I measured out a litre and added 2ml of water and immediately put it into a sealed container.
I then did the same with another litre but this time added half a ml of water.
I used a new clean 5ml syringe, I was very careful and I wore my closeup glasses, but I reckon this method cant do better than 10% accuracy. However I would be happy with that.
All of the following tests were carried out at 16 degrees C (289degrees K)
All samples were 100g to maximise accuracy and the vessel was shaken for 5 minutes or until no movement was seen for a minute
As a control I tested some of the totally dry bio and got a result of 9mm.
This may have been caused by the screwing closed of the cap or it may be a true reading caused by atmospheric moisture in the vessel. I dont know for sure but it is so small a reading it does not really matter.
I tested the sample with 2ml per litre or 200ppm first. The reading was 622ml,
The sample with .5ml per litre or 500ppm gave a reading of 183ml.
Considering that the measurement of the water in the syringes could not be better than 10 percent accurate this result is acceptable.
It also agrees with the readings made while drying my biodiesel in a previous post. Bear in mind that those readings were for 50gm samples and should be doubled to match the readings above.
I now feel that this method really is a fast accurate way to determine water content in biodiesel and can help home brewers make better fuel. However I am only too aware that people who develop ideas often cant see the faults and drawbacks. Has anyone put together a manometer? If you have I would really like to hear from you.
Sorry about the typo in the last post. the line should read...
" I tested the sample with 2 ml per litre or 2000ppm first."
Rickdatech has rightly pointed out that I have not properly described the apparatus as it has evolved over the last 3 pages. I will post a full description and how I use it with some photos soon
To make a carbide manometer you will need a plastic drinks cup with an airtight screw on lid. I bought mine in a local supermarket for under 2 euro. You will also need 3 metres of clear 5mm pvc hose. Bore a suitable hole in the lid and push the hose through an inch or so. Seal with a hot glue gun or epoxy resin. Arrange the hose as shown in the picture using 5mm cable clips. I fixed mine to a white board but you can use any light coloured surface or wall.
Mix a little water with food colouring and fill the U bend using a syringe. Tap the hose to remove any air gaps. Your carbide manometer is ready for use. before you can use it you need to get some Calcium Carbide. I bought 2.75lbs on Ebay for about 15 euro. This is enough for hundreds of tests.
It comes in the form of stones and must be kept stored in its airtight tub.
|Powered by Social Strata||Page 1 2 3 4 5 ... 16|