Fuel usage

[Jens]

I haven't spilled any fuel lately nor has anything broken on Thumper for a while so I am at loose ends as to what to do next. I started to think about fuel consumption and costs of fuel ( I paid 10cents/litre for veg oil this summer) and how this compared to buying my electricity.

I have no real accurate way of measuring fuel consumption but I figured I would try an approximation - the height change of the fuel in the tank over a one hour period.

I am running a 20/2 at a constant load which at the moment seems to be about 4500W.

This afternoon I started Thumper, and ran her for 2 hours at the fixed load. Fuel consumption in the first hour was approximately 426 ml and a whopping 4047 ml for hour number two ! Remember, there was no load change !

There is no obvious fuel leak and there doesn't appear to be any transfer of fuel from the diesel tank to the veg oil tank.

The most obvious reason for these screwy readings is that I made an error measuring things.  For now I am discounting these readings as being too far out of whack.

I wonder if I could get some thoughts about the kind of fuel consumption I should see - remember, 20/2, running at a constant load and at approx  850 rpm. The electrical power produced by the attached generator was approximately 4500 watts.
Anand rates fuel flow at 268gms/kw/hr or 197 gms/bhp/hr for diesel at full load and standard reference conditions whatever that may be but I am having a hard time converting that to a number in real life.

Jens

[dubbleUJay]

Jens, this is how I measure my consumption at the moment as in the picture.
A clear 32mm ID PVC pipe, I then took a measuring cup of 25ml and filled it to the top, marking every level as it filled.
The markings are unclear in the picture, but I have about a liter worth of fuel in it. About just a little more than a hours worth of run on my CS single.
Its a bit of a "schlep" to take the reading every few minutes or so, but it works for me as I don't have to run it for long times. Hopefully we can automate it in the future!

Wilhelm

[Ronmar]

Not sure exactly what to expect for fuel use for your setup, but an accurate fule measurement is just a piece of clear fuel line away.  I put a T into my line before the filter for the injector return line. For my fuel comsumption tests, I added another T at this same point connected to a piece of clear line running up to a container turned upside down with the bottom cut out as a funnel.  I put a mark on this line and add a little fuel at the funnel  to get the fuel above the mark, then shutoff the fuel from the tank.  The fuel level in the clear line then drops as the engine consumes this fuel and as it reaches the mark, I start the clock and add a measured ammount of fuel to the funnel.  When the fuel again reaches the mark, i stop the clock and do the fuel use calcs based ont the now known load, quantity of fuel and time...  This didn't really cost me anything as it was put together with scrap.  It is still hooked up, I can post a pic if you want to see it.

[dubbleUJay]

Thanks Ronmar, it doesn't show in the picture, but I have a 3-way valve that I change over when I start to measure.

Jens, I'll have to look again to make sure, but IIRC approximately 800ml/hour with 2.5kW, but don't quote me on that please!  

Modified:
Maybe I'll do another test run 2day, its been a while since we had any power failures and I've been busy with all the other stuff on the forum
I'll let you know later on.
Wilhelm

[dubbleUJay]

OK Jens, lets see if these figure would help you:

I ran my Lister CS 6/1 at 650rpm connected to a 3kW 220Vac 50Hz Gen-head.
I used WVO as that is what was in the metered fuel supply at the time.
I've only got a 1800W space-heater to connect to it as a sort of reliable load. This one has 4x elements which can be switched on individually, i.e. 600W increments at a time. (Needless to say, its not in the house anymore!) 
I didn't use the last element, so 1.8kW & not 2.4kW were used.
BTW, I let the engine get up to temperature first before I did the test.

It used 900ml of WVO in one hour with a constant 1800W load.
So if my calculations are correct:
900/1800 x 900 = 450ml of WVO were used to produce 1kW constantly in an hour or am I not thinking straight!

With this in mind and I directly convert it to your engine capacity without the rpm difference:
450ml/kW x 4.5kW x 2cylinders = 4050ml 

You used 4047ml in your second hour and your running faster than me!

BUT, there's probably a mistake somewhere that I'm not seeing! 

[dubbleUJay]

Thanks for posting the usage figures. I will do a couple more runs to average my figures and see where that ends up.

Jens

Glad I could help out Jens, it was about time I did some testing myself, but I've been so busy with all the electronics or automatic monitoring stuff that I just don't get around to it.
On the other hand, once I have a data recorder in place, it will all be automatic and easy to get to all the figures I hope.
I've got problems with my gen-head as well, I get quite a voltage drop if I exceed about 2kW, that's why I kept it at 1.8kW for now.
All things in the pipeline, but I'll get to them eventually!
I want to do test runs with WVO, WMO & diesel to compare the lot and hopefully the data recorder will work.

Your figures seems to be more or less constant now, so it looks like you're on the right track.

dubbleUJay

[dubbleUJay]

Jens, just to put you into the picture over here in ZA:
1 ZAR = 0.134438 USD

Electricity cost from utility per kW = R0.70 ($0.09) (Your Hydro-power $0.10)
I need 450ml of fuel to generate 1kW:
Diesel = R6.65/liter ($0.89/liter) that's R2.99 ($0.40) per kW (Your Diesel $0.45)
WVO  = R1.50/liter ($0.20/liter) that's R0.68 ($0.09) per kW (Your cleaned Oil $0.09)

Interesting how our costs are very close to each other and we're not even in the same country!
That's enough maths for one day I think! 

We've had a 130% increase in electricity cost the past 2 years and there another one coming soon, that we expect to be about 40%.
I'm not calculating the cleaning of the WVO, but I should at least break even when the next price hike comes. Now if I can get all my ducks in a row for WMO, that would be something! 

Have a great day guys!
dubbleUJay

[mobile_bob]

we gotta get you fellars thinking in terms of fuel weight and not volume
volume is fine for general or relative measurement but you will soon find that
it become problematic trying to compare to various fuels or other folks findings.

measuring by weight accounts for temperature differences and btu content for the most part
and gets things to a more "apples to apples" comparison.

it really takes very little to work things out in weight, a simple scale such as a postal or food scale
graduated in grams is all that is needed, and are readily available for very low cost.

fwiw, doing it by volume is not bad to get a rough idea, but as soon as you start working on efficeincy
it doesn't take long before you will become frustrated by shifting numbers as a result in shifting temperatures
or fuels.

its very hard to determine how something you change effects consumption when you got more than one variable
to contend with in my opinion, and the fuel by volume variable can be maddening at times.

having said all that, i am very happy to see others working on fuel consumption testing of the listeroid twins, i haven't seen
much done in that regard, at least that i recall.  it would be really nice to get some really good testing done that was
done to a standard that could be widely accepted.

good luck

bob g

[BruceM]

It's awfully easy to adjust the flow calculations with real time temperature data; that's what micro-controllers are good at; tedious repetitive calculation. 

All you need is the weight and volume measurements for the fuel's range of temperatures that you will operate in. That you an easily use the postal scale for.   The relationship is likely a straight line.  And a compiler with 32 bit floating point math is a wonderful thing.

[dubbleUJay]

BobG, you are correct, weight is the way to go, but the way that Bruce explains doing it with a micro is essentially the same.
If I had a way of weighing fuel usage automatically to interface to a micro, I would do it that way, but it seems that getting the volume used and calculating the weight with a "look-up table" with a micro is going to be easier. I say "easier", but we're even battling to get that right!
The table shouldn't be to difficult to setup, one would get volume & weight of a specific fuel at a set altitude at say 10degC & then again at say 30degC, with the weight as a constant. Then connecting the 2 reading on a graph should give everything in between.
It's a lot of work now, but will save tremendous time when its done IMHO.