BIODIESEL & SVO DISCUSSION FORUMS





Sponsors    Biodiesel and SVO Forums Home    Forums  Hop To Forum Categories  Biodiesel  Hop To Forums  Biodiesel Quality    Falling Ball Viscometer
Page 1 2 3 

Moderators: Shaun, The Trouts
Go
New
Find
Notify
Tools
Reply
  
Falling Ball Viscometer
 Login/Join
 
Member
posted
Over the last few years we at the infopop forum have tackled quality issues very successfully. Jan Warnqvists brilliant 3/27 test, later developed into the calibrated 30/270 test improved conversion rates enormously among homebrewers.
The University of Idahos Wash Test along with Soap titration has helped us assess and improve the contaminate levels in our our fuel.
The Carbide Manometer, my own contribution, has helped us reduce water content in our fuel.

I believe that Viscosity is an issue that is becoming more important. Common rail injection systems are now the norm in diesel engines. This system uses pressures of around 30,000psi, some ten times the pressure of injection pumps a few years ago. Pump failure is now a relatively common, and very expensive, problem with new diesel engines. Using a fuel of significantly higher viscosity than normal must, by definition, place a greater strain on these pumps.


The falling ball viscometer was the original viscometer that George Stokes in 1851 used to test and prove his viscosity formula. It is beautifully simple and can be constructed from commonly available materials. It is essentially a transparent tube filled with a liquid. A ball is dropped into the top of the tube and the observer times how long it takes for the ball to fall 1 metre.

The time, along with the density of the liquid, is entered into Stokes formula and the result is the viscosity , calculated in Centistokes.

Below I have shown a sketch of the apparatus.



Tomorrow I will post a video on Utube about making and using the falling ball viscometer and will post a link here to it.

www.imakebiodiesel.webs.com

This message has been edited. Last edited by: imakebiodiesel,
 
Location: Lismore Ireland | Registered: November 25, 2007Reply With QuoteReport This Post
Member
posted Hide Post
Very nice, you've done it again, kudos !!

This will be my third favorite piece of DIY test equipment right up there with the carbide manometer and the 30/270 dish-soap bottle-cap / syringe assembly.

Thanks.



 
Location: coldest N.America | Registered: May 03, 2005Reply With QuoteReport This Post
Member
posted Hide Post
Thanks John,
The secret to this apparatus was finding a suitable ball. Metal ball bearings are uniformly round but are so heavy they fall far too fast. After much searching I found that 6 mm air soft pellets are perfect. They are uniformly round and consistent in weight and are just a little heavier than water. The ones I settled on were 6mm diameter and 0.12 gms. These are very common and can be bought almost anywhere in the world. They are also very cheap. By working out the volume of the ball and multiplying it by the weight I calculated the density to be 1061 kg/cubic metre. This means we know the value of every term in Stokes formula except the density of the liquid and the time it takes the ball to fall 1 metre.


P1 = density of the ball = 1061
P2 - density of the liquid
D = diameter of the ball in metres = .006
t = time for ball to fall 1 metre
H = height of fall = 1


so with the known amounts put into the formula it reads ...

0.545 x (1061-P2) x 0.036 x t

to find out P2, the density of the liquid we weigh one litre in grams.
To find out t we drop the ball into the tube and time its fall.

I would really appreciate it if some one would check over my figures and make sure Im not making any mistakes. I got my Stokes formula from the University of Middlesex library and it seems to result in the decimal point being one place to the right.

Viscosity is very temperature dependent so I have carried out my tests at 20C.

These are some of the results I have recorded so far..
water, P2 = 1000, t = 11.8, viscosity = 14.12 centistokes
Kerosene, P2 = 775, t = 4.6, viscosity = 25.81
Synthetic kerosene, P2 = 788, t = 4.85, viscosity = 25.98
Biodiesel, P2 = 890, t = 9.17, viscosity = 30.76

I will be buying some petrodiesel (uugh!) and testing various blends.

Video coming up soon.
 
Location: Lismore Ireland | Registered: November 25, 2007Reply With QuoteReport This Post
Member
posted Hide Post
Well done mate,a very useful and valid test.(and yet again made with everyday materials)
I currently run on B80 for one reason,because I worry about the viscosity of the biodiesel on the high pressure injection system that's now in most modern diesels.

But with this test I should be able to measure and compare the difference between dinodiesel and bio and take out the 'guess' work when blending and reduce the the risk of injector failure.

Especially useful as we start to come into winter when our bio thickens with the lower temps.

Thank you...
 
Location: County down, Northern Ireland | Registered: August 10, 2008Reply With QuoteReport This Post
Member
posted Hide Post
Is the change in viscosity with temperature the same for all fuels? Otherwise wouldn't we have to compare the fuels at the temperature entering the IP?



 
Location: coldest N.America | Registered: May 03, 2005Reply With QuoteReport This Post
Member
posted Hide Post
This is the kind of thing we can investigate using the viscometer.
My guess is that the heavier fuels( diesel, biodiesel) increase in viscosity as they get colder more than light fuels. Certainly as they approach their gel point the viscosity increases dramatically.
The tubing and fittings are good to about 80C so we could test hot fuels.


I redid my calculation for water because it was the first one that I did and I only took one reading and the temp wasnt at 20C. This result is better at 11.47@20C.

I figured out where I went wrong with that pesky decimal point. The diameter of the ball is 0.006 metres. 0.006 squared is not 0.036 as I stated but 0.000036. So If I recalculate the equation for water the result is 0.01147. However this result is in stokes, multiply by 100 to get centistokes and the answer is 1.1 which is almost exactly what you would expect.

This message has been edited. Last edited by: imakebiodiesel,
 
Location: Lismore Ireland | Registered: November 25, 2007Reply With QuoteReport This Post



Member
posted Hide Post
Some more interesting results.

I bought some LSD (low sulphur diesel) today ( boy that hurt, 1.59 euro a litre) and tested both Lsd and B50. The results are as follows...

LSD P2 = 832, t = 6.14, viscosity = 2.76 centistokes @20C

B50 P2 = 857, t = 6.96, viscosity = 2.78 centistokes @20C

Note that the viscosity of B50 is much closer to LSD than it is to B100 (3.76 centistokes @20C). This may explain why B50 runs so well in these new high pressure diesel engines.

I will try to get the video posted very soon as I am keen for others to set up viscometers so we can compare and check results.
 
Location: Lismore Ireland | Registered: November 25, 2007Reply With QuoteReport This Post
member
posted Hide Post
Pair that up with a couple Light Beam Sensing Switches triggering the timer and you could get some really accurate time readings. Maybe even accurate enough to use a metal ball.
 
Location: Southern WI, USA | Registered: May 18, 2006Reply With QuoteReport This Post
Member
posted Hide Post
Thats exactly how the viscometers used in industry work. The only thing I dont like about them is the price. This one from Cole Parmer http://www.coleparmer.com/buy/...ball-viscometer.html costs 5498 euro. Yearly calibration is extra of course.
Maybe some bright sparks here on the forum could improvise a light sensing system.
 
Location: Lismore Ireland | Registered: November 25, 2007Reply With QuoteReport This Post
Member
posted Hide Post
hmm, your data doesn't seem to match what i've seen in various literature sources. water should be closer to 1.0 than 1.1, so you have some error.

http://lib3.dss.go.th/fulltext...,no12,p1511-1513.pdf

this article shows b100 >7cSt (20*C), b50 >5cSt (20*C).

i think it's pretty well established that b100 for most feedstocks is >4cSt at 40*C, while petroleum diesel is closer to 2 cSt (40*C). also, b50 should lie approximately between b0 and b100. so if b0 is 2 cSt and b100 is 4 cSt, then b50 will be close to 3 cSt.

if you really wanted to be accurate, you could purchase some viscosity standards, and calibrate your system against those. they are kind of expensive, but much more accurate.
 
Location: PEI, Canada | Registered: September 27, 2012Reply With QuoteReport This Post
Member
posted Hide Post
this may be useful





Imagekinematic_viscocity.gif (47 Kb, 15 downloads)
 
Location: coldest N.America | Registered: May 03, 2005Reply With QuoteReport This Post
Member
posted Hide Post
Great suggestion for using a plastic 'airsoft' ball.

I use these convenient 300ml salad dressing bottles for my fuel mix samples.
[BD soaks the labels off]



Using a sample of diesel fuel as a standard and measuring the drop time I can easily make a mix of jetB [JP4] or kerosene+petrol with biodiesel or vegetable oil that has the same relative viscosity as the diesel fuel I'm using. With the diesel sample and the fuel mix sample side-by-side, I drop a ball into each at the same time and see if they hit the bottom at about the same time.

I'm not as concerned with the viscosity in centitokes as much as I am in making a fuel mix with about the same viscosity as diesel fuel. This is a great way to check it.

Thanks for the useful suggestion.



 
Location: coldest N.America | Registered: May 03, 2005Reply With QuoteReport This Post



Member
posted Hide Post
Your article shows the viscosity of No2 diesel at 4 centistokes @20C while Johns chart says 3.5. Engineering toolbox site gives a range from 2 - 6. A quick survey of online sources shows a range from 2 to 8 and these tests are all carried out with certified calibrated viscometers. Assuming that all these viscometers were working properly we would have to accept that there is a wide variance in diesel fuel samples. Even the diesel I bought yesterday had a SG of 832 when the European DIN standard is 845. The variation in results for biodiesel is even wider due to the range of feedstocks available.

Improvised equipment can never be totally accurate and at present to be within 10% is acceptable to me. The real value of this type equipment is to allow the home brewer to make valid comparisons between his fuel and blends of diesel/ biodiesel. I was suprised at the results I got with the B50 yesterday and will want to repeat the tests.

At the same time I am anxious to make my viscometer as accurate as possible and to that end I am having a sample of my synthetic kerosene tested by an independent certified lab. I will be able to compare their results with mine.
 
Location: Lismore Ireland | Registered: November 25, 2007Reply With QuoteReport This Post
Member
posted Hide Post
Because we experience a wide seasonal temperature range from -30°C to +30°C our diesel fuel viscosity can vary quite a bit between the #2 summer fuel and kerosene winter fuel. So the range of 2-8 seems reasonable.



 
Location: coldest N.America | Registered: May 03, 2005Reply With QuoteReport This Post
Member
posted Hide Post
"Diesel fuel" is quite variable both regionally and seasonally. There is no single viscosity for diesel fuel.



Speak for yourself and refrain from trying to interpret what other members have stated.



 
Location: coldest N.America | Registered: May 03, 2005Reply With QuoteReport This Post
Member
posted Hide Post
I never measured D2, it was Andyman who brought up the density of D2.
In fact we dont have D1 and D2 here, we have LSD, agricultural "green " diesel and gasoil heating grade diesel.
The 2-8 result was a google search" density of diesel". My point is that there are many different diesels with different specs in every country and every season. There is no one correct viscosity figure for diesel. You are right that the engineering tool box figure of 2-6 was at 37C but I presume the range would not be any narrower at 20C.

We are not trying to establish an absolute value for diesel or biodiesel because one does not exist. We are trying to devise a cheap and acceptably accurate way of comparing the viscosity of different diesel/biodiesel blends.

I did measure the SG by weighing one litre in a graduated cylinder and I also checked it with my calibrated hydrometer. Both gave the same result. I have to have a calibrated hydrometer because I am an Authorized Fuel Tax Warehouse and must be able to prove the SG of my fuel.

John, your idea of the 2 jars is fine as a rough guide but taller jars will give you a better result. Bear in mind though that without factoring in the density the result will be distorted.
 
Location: Lismore Ireland | Registered: November 25, 2007Reply With QuoteReport This Post
Member
posted Hide Post
Yes, factoring in density will increase accuracy. For approximation I assume that all my fuel mixes are about the same density as diesel fuel. Calculating density is fairly easy to do since the bottles hold an equal and measurable volume to the ring on the neck, and they're filled to that level for the ball drop test. If I need more accuracy I also have 3L clear plastic jugs that approximately double the height. Using those 'airsoft' plastic balls is a stoke of genius.



 
Location: coldest N.America | Registered: May 03, 2005Reply With QuoteReport This Post
Member
posted Hide Post
I think my success is due more to persistence than genius. Like Edison Who tried 700 wrong ways to make a light bulb before discovering the right way, I just kept trying out different balls until I found the airsoft ball was right.

One thing to note. When the ball falls into the liquid it tends to wobble about a bit before settling into a steady rate of fall. I place the timing start mark about 2 inches below the surface to give the ball time to settle. If you time from the moment the ball hits the surface you will get a wide variation in times.

I hope to get the video up today. My director( my son Ruairi)and I are having "artistic differences". He wants to do the whole thing again because some of the green screen effects did not work perfectly. I want to get the thing up on utube as soon as possible.
 
Location: Lismore Ireland | Registered: November 25, 2007Reply With QuoteReport This Post



Member
posted Hide Post
The video is now available to view at this link http://www.youtube.com/watch?v...gXs&feature=youtu.be .
I will repost the formula so than my subscribers will not have to search back for it.
here is a diagram of the viscometer and Stokes Formula.




The secret to this apparatus was finding a suitable ball. Metal ball bearings are uniformly round but are so heavy they fall far too fast. After much searching I found that 6 mm air soft pellets are perfect. They are uniformly round and consistent in weight and are just a little heavier than water. The ones I settled on were 6mm diameter and 0.12 gms. These are very common and can be bought almost anywhere in the world. They are also very cheap. By working out the volume of the ball and multiplying it by the weight I calculated the density to be 1061 kg/cubic metre. This means we know the value of every term in Stokes formula except the density of the liquid and the time it takes the ball to fall 1 metre.


P1 = density of the ball = 1061
P2 - density of the liquid
D = diameter of the ball in metres = .006
t = time for ball to fall 1 metre
H = height of fall = 1


so with the known amounts put into the formula it reads ...

0.545 x (1061-P2) x 0.000036 x t

This will give the viscosity in Stokes so multiply by 100 to convert to centistokes.
 
Location: Lismore Ireland | Registered: November 25, 2007Reply With QuoteReport This Post
Member
posted Hide Post
The video is good but the music is distracting and makes the voice hard to understand for those not familiar with your accent.



 
Location: coldest N.America | Registered: May 03, 2005Reply With QuoteReport This Post
  Powered by Social Strata Page 1 2 3  
 

Sponsors    Biodiesel and SVO Forums Home    Forums  Hop To Forum Categories  Biodiesel  Hop To Forums  Biodiesel Quality    Falling Ball Viscometer

© Maui Green Energy 2000 - 2014