My internal combustion engine is more efficent than 30%

Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
Sign up Log in
78 replies · 19K views
smokingwheels said:
eg why is my peak idle (no load) rpm achieved when I fire 30 - 40 degrees BTDC?

Because that releases the most energy the most efficiently for those conditions. It also proves that having the spark occur well before TDC is not evil but necessary. You need to accept that fact and then learn why, so you can move on.
 
Physics news on Phys.org
smokingwheels said:
Does anyone know of a place on the net that would list all variables I need to measure and test for the improvements I have made to my engine, oh and free?

Easy.

Under steady state conditions, measure the power your engine is producing (or measure the speed and torque). Under those same conditions, at the same time, measure how much fuel is being consumed by the engine. Repeat this until your between-measurements variation is less than, say, 2% of your measured figures.

Then post both those figures, along with the fuel type. If you want to make comparisons, run the same engine under the same conditions with your modifications made.

I'll tell you your efficiency, and I won't even charge you for it. Deal?
 
smokingwheels said:
Does anyone know of a place on the net that would list all variables I need to measure and test for the improvements I have made to my engine, oh and free?

Buy, rent, or borrow one of these: http://www.taylordyno.com/catalog/engine-dyno" . (An engine repair facility in your area should have one. Whether they'd let you use it is a different matter.)

Hook it up to your engine.

Hook up a fuel metering device.

Measure the volume of fuel consumed for a given period of time, at a given power output.

Power x Time = Energy

Energy/Fuel Amount = Efficiency.

Energy/ Energy in that fuel = % Efficiency.

Repeat a few times for each load level and at different load levels (output power).
 
Last edited by a moderator:
jambaugh said:
Buy, rent, or borrow one of these: http://www.taylordyno.com/catalog/engine-dyno" . (An engine repair facility in your area should have one. Whether they'd let you use it is a different matter.)

Hook it up to your engine.

Hook up a fuel metering device.

Measure the volume of fuel consumed for a given period of time, at a given power output.

Power x Time = Energy

Energy/Fuel Amount = Efficiency.

Energy/ Energy in that fuel = % Efficiency.

Repeat a few times for each load level and at different load levels (output power).
Thanks for the tips

I don't repeat a test until I have made an improvement.
When this condition is false I will repeat a test.

Buy, rent, or borrow and steal...
I won't be able to afford to fuel to run my engine for 4 months at the moment so I will have to work on something else until then.

I would need a machine that measures the power of the torque pulse from every cylinder instead of a dyno and would need similar type of measurements to see what the drive train is doing.

Thanks Physics Forum peoples for all your help, I am working on my ignition controller now for a while.

In electronics if you know or measure 2 variables you can then calculate the the 3rd there is a law that covers that.

Why in an engine any different?
 
Last edited by a moderator:
smokingwheels said:
In electronics if you know or measure 2 variables you can then calculate the the 3rd there is a law that covers that.

Why in an engine any different?

It's not, it's a maths thing. For a problem to be solvable you need as many equations as you have unknowns.

So if you have an equation with three variables. If you know two, leaving one unknown you can solve it. If you have four variables and know two, you have two unknown and can't solve it until you have a second equation with one of the unknowns in.

In te case below of.
Energy / Fuel amount = Efficiency.
Power * Time = Energy

We have an equation that looks like this:

Power * Time / Fuel Amount = Efficiency.

If we measure power on a dyno and we measure fuel amount and we measure time. We know three of the four variables, so can solve the equation.

I would need a machine that measures the power of the torque pulse from every cylinder instead of a dyno and would need similar type of measurements to see what the drive train is doing.

An engine dyno measures torque and rpm and calculates power. As:
Power = Torque * Angular Velocity.


Whilst you are not able to conduct more practical test, here is some food for thought.

Comparing things like power output or fuel consumption are poor ways to compare different engines in terms of efficiency. You want to use things called 'non dimensional' comparisons. This is where outputs and variables are converted to a standard forn that can be directly compared.

Two to read up on are.
Mean Effective Pressure - http://en.wikipedia.org/wiki/Mean_effective_pressure
Brake Specific Fuel Consumption - http://en.wikipedia.org/wiki/Brake_specific_fuel_consumption

These two variables can be used to directly compare engines. They use easy to measure variables.
 
Last edited:
smokingwheels said:
I would need a machine that measures the power of the torque pulse from every cylinder instead of a dyno and would need similar type of measurements to see what the drive train is doing.
No, you need a device to measure the power output of your engine under load. Note that dynamometers can be used for either engine output measurements to calculate fuel to work conversion efficiency, or power train output to calculate transmission efficiency. Both are the same problem... what is the output power vs input power?

In electronics if you know or measure 2 variables you can then calculate the the 3rd there is a law that covers that.

Why in an engine any different?
An engine is a converter of one form of energy to another. What you are doing is comparable to designing a better solar panel or electrical generator, or electric motor. You measure the input, measure the output and get a ratio efficiency.

You want to calculate efficiency but the comparable calculation for an electronic device is quite involved typically invoking some substantial physics and typically only an idealized calculation, not reflecting real imperfect materials. Similarly with the engine.

The ultimate proof is in the pudding, what power is delivered to the crankshaft under load as compared to what rate fuel is being consumed. You cannot circumvent the need for this empirical test to determine actual efficiency. Only such an empirical test can support a claim of improved efficiency. Only such will tell if your modified engine will get you from Atlanta to New York using less fuel.
 
jambaugh said:
No, you need a device to measure the power output of your engine under load. Note that dynamometers can be used for either engine output measurements to calculate fuel to work conversion efficiency, or power train output to calculate transmission efficiency. Both are the same problem... what is the output power vs input power?An engine is a converter of one form of energy to another. What you are doing is comparable to designing a better solar panel or electrical generator, or electric motor. You measure the input, measure the output and get a ratio efficiency.

You want to calculate efficiency but the comparable calculation for an electronic device is quite involved typically invoking some substantial physics and typically only an idealized calculation, not reflecting real imperfect materials. Similarly with the engine.

The ultimate proof is in the pudding, what power is delivered to the crankshaft under load as compared to what rate fuel is being consumed. You cannot circumvent the need for this empirical test to determine actual efficiency. Only such an empirical test can support a claim of improved efficiency. Only such will tell if your modified engine will get you from Atlanta to New York using less fuel.

"The ultimate proof is in the pudding" ok done a rough estimate if I could drive fairly constantly without too many hills at 60km/h I could do it Atlanta to New York 1418 km if I had a 20 L jerry can on board. It begs the question I will use overdrive on my next test to see if it is any better and will recalibrate one of my load sensors as well but that will be at the end of july now before I can retest any of my theory's.
Oh I can also push my engine to hard knocking without piston damage but the crank takes a beating though.

Unless some one sponsors me I will never get my engine on a dyno.
 
Last edited:
smokingwheels said:
"The ultimate proof is in the pudding" ok done a rough estimate if I could drive fairly constantly without too many hills at 60km/h I could do it Atlanta to New York 1418 km if I had a 5 L jerry can on board. It begs the question I will use overdrive on my next test to see if it is any better and will recalibrate one of my load sensors as well but that will be at the end of july now before I can retest any of my theory's.
Oh I can also push my engine to hard knocking without piston damage but the crank takes a beating though.

Unless some one sponsors me I will never get my engine on a dyno.


I really don't want to seem like we're just bashing you.

A 'rough estimate' is not proof or evidence in any way, shape or form. By continuing to test in the way you have been doing, you are just wasting time and money. By not doing all these useless tests that tell you nothing concrete what so ever, you could have saved the money to do a proper dyno test.


Let's look at the bottom line here.
How much have you spent?
What data have you gathered?

By data, I do not mean speculation or estimates. What concrete data do you have?
 
Last edited:
smokingwheels said:
Oh I can also push my engine to hard knocking without piston damage but the crank takes a beating though.

This is not a smart thing to do; forcing your engine into detonation is a sure way to deplete what little funds you have. It also doesn't improve the efficiency, so there is no reason to continue doing that.

Stop it!
 
ok Change of subject Gensets and there fuel consumption.

Rough figures

2.5 kva 5.5 HP 163 cc engine at 3600 rpm with 2kw load 15 L will last 6.5 Hours

So that is 38.46 cc/min with approx 2.4kw output (80% generator efficiency??).

The turbocalc program calculates fuel for 2.4kw at 13.26 cc/min.
Thus 30% efficient so times 3 to get 39 cc/min of fuel for 2.4 kw output out of engine.
So what I am trying to say, I think this program is correct.

Anyway a normal 4 cylinder engine uses 5 hp of fuel 20cc/min just to get to 750 rpm.
To output 5hp@750rpm would it need another 20cc/min = 40cc/min.
or is is 3 times 20 cc?

My engine uses 12cc/min for 750 rpm.
to output 5hp@750rpm it would need another 12cc/min =24 cc/min
or is it 3 times 12cc?

I have a problem!
Im giving up an going to collect data from other peoples cars I have a draft form at http://smokingwheels.dyndns.org"

I was thinking of betting 1 dollar if you can beat me on my test track in similar EFI car...
Gives me something to do instead of playing with figures.

...have att form...What am I missing from the drivers point of view?
 

Attachments

Last edited by a moderator:
You've been told this already, but the amount of fuel used at idle is just not a way to compare what the fuel efficiency is like on load. To infer the fuel consumption on load from a fuel consumption at idle is irrelevant, meaningless, negligent, incorrect, wrong, and completely pointless.
 
smokingwheels said:
...2.5 kva 5.5 HP 163 cc engine at 3600 rpm with 2kw load 15 L will last 6.5 Hours

So that is 38.46 cc/min with approx 2.4kw output (80% generator efficiency??).
The official conversion used in BSFC (brake specific fuel consumption) maps is

100% efficiency ≡ 83 grams gasoline per kW-hr.

See http://en.wikipedia.org/wiki/Brake_specific_fuel_consumption

"Certification gasoline = 18,640 BTU/lb (0.01204 kW·h/g)" = 83 grams per kW-hr

So a 100% efficient 2.4 kW engine should use 3.32 grams per minute, or about 4.5 cc per minute. This implies that your engine is ≈12% efficient.

The BSFC map in the above pdf shows a 40% efficient engine, with a minimum fuel consumption of 206 grams per kW-hr.

See table in above pdf for table of engine efficiencies. In particular note the Toyota Prius engine at 37% efficiency.

Bob S
 
smokingwheels said:
"The ultimate proof is in the pudding" ok done a rough estimate if I could drive fairly constantly without too many hills at 60km/h I could do it Atlanta to New York 1418 km if I had a 20 L jerry can on board. It begs the question I will use overdrive on my next test to see if it is any better and will recalibrate one of my load sensors as well but that will be at the end of july now before I can retest any of my theory's.
Oh I can also push my engine to hard knocking without piston damage but the crank takes a beating though.

Unless some one sponsors me I will never get my engine on a dyno.

The drive test may indicate a qualitative improvement (or disimprovement) but you really need some numbers. So if you can't afford a dyno, make one. Probably the cheapest thing to do is fix a setup so your engine is vigorously stirring a big drum of water sufficiently to generate an appreciable load.

To calibrate it,
* measure the volume of the water in cubic meters,
* measure the initial temperature in deg. C,
* run the apparatus for a minute or so but measure the time to the second (100 seconds is a nice round number),
* measure the temperature afterwards and determine the change in temperature.

1 Calorie = 1000 calories = energy needed to raise the temp of 1m^3 of water 1 deg C.
= 4.184 kiloJoules = 4.184 kilowatt seconds.

This will allow you to calculate power output at a given engine RPM with your setup.
Follow this procedure for each run and carry out several runs at different RPM's.

Once you've calibrated the load you can then run the engine at constant load calculated for that RPM. Run it for 30min or so and measure the fuel consumed per calculated Joule of energy output and thus calculate explicit efficiency.

This setup should be quite accurate but the precision depends on your measurements. Try to find the calibration time which gives you a good temperature change but be sure your initial temperature is the ambient temperature, and it will help if you be sure the can is covered as much as possible to prevent evaporation loss. Insulation around the water drum would also help. Also the stirring apparatus should agitate the water but you want to make it turbulent, something like an egg beater or butter churn. But be sure no water sloshes out. (I recall once seeing a fellow tuning up the the outboard on his boat by running it in an drum of water in his back yard.)

Basically this is reproducing Joule's experiment measuring the http://en.wikipedia.org/wiki/Mechanical_equivalent_of_heat"
 
Last edited by a moderator:
Our dyno uses twin eddy current retarders.

https://www.youtube.com/watch?v=5VkDLfmGH7Y

That's only a 1300hp pull but it was enough to go from 0 to 252.9mph in a mile from a standing start.

We're finishing a 2500hp engine for some world record runs.

Engine management systems of today's EFI cars have the ability to display a relatively accurate mpg...unless the manufacturer purposely skews the data to show more favorable mileage.

They use injector pulse width, rpm, grams per second (from mass airflow sensor), gearing, downstream monitoring of exhaust gases, and the fuel trim correction values.

Miles per gallon increases require optimizations in many areas but the hardest area to optimize is the driver.

I've seen guys get 18mpg in a prius because the driver drove in such a way that he defeated any of the benefits of his car's design.
 
Last edited:
Would exhaust manifold temperature tell how efficient an engine was?

My engine which has a cast manifold and I have a normal engine but its got extractors.
Problem..
Can I compare the two?

Test setup:
My engine temperature was about 95 deg C and warmed up and idling, the ambient temperature was 12.9 deg C, wind W25 km/h and wind gusts 43 km/h.
The Second engine temperature was below 88 deg C and still warming up, the ambient temperature was 9.5 deg C, wind W 29 km/h and wind gusts 52 km/h.
I melted 60/40 Solder on the exhaust path about 3-5 inches away from the head with the engines idling after going for a drive to warm them up.

My modified engine took some time to melt the solder the second engine did it really quickly with the flux smoking well.

I was wondering if its because of the extractors being thinner and less surface area would it be hotter?
With more surface area for the cast iron manifold would it run cooler?
Will ambient temperature affect the manifold temperature a lot or not?
The ambient temperature dropped after testing the modified engine.

I have a Video play list if you are interested to see the results from 4 short clips.
http://www.youtube.com/playlist?list=PLA7848D3EF5B27FF3

Solder (60/40) melts at 370 deg F and a soldering iron is about 700 deg F.
My engine with the cast manifold took quite a while to just melt the solder with not much smoke from the flux.

The other normal engine with the extractors did it very quickly and the flux was rapidly turned into smoke in a similar way to a soldering iron from my experience.
I know I need 2 of the same to compare so I need to find a few volunteers in Australia if
possible.
 
So you're using solder as instrumentation while doing tests on two different engines under two different conditions at idle?

Firstly, thermocouples are the way to measure exhaust temperatures on engine test. You can get them for less money than you probably spent on your solder. Melting solder (and trying to draw a conclusion about the way it melted) doesn't tell you anything, except that the exhaust pipe is hot. It doesn't even give a good indication of the temperature of the exhaust gas!

Secondly, (and we've been through this before, but I'll say it again because it seems to have soaked in like water off a duck's back) performance indicators at idle tell you so little about what your engine is doing that when doing engine development work, we hardly ever take data at idle. It just doesn't tell you anything about what the engine will do under operation.

Thirdly, what do you mean by 'extractors'? The only exhaust extractors I've come across are ventilation systems for taking combustion products out of the test cell.

Finally, what are you trying to determine now?
 
Thermocouples Ok but I would have to get a hole drilled and a fitting welded to my exhaust.

An engine under load is what's needed.

"what do you mean by 'extractors'?" They are smooth mandrill bent tubing providing a rounded flow for the exhaust gasses also usually larger than cast iron counter parts and they are freer flowing.

"Finally, what are you trying to determine now?"
To see if there is a relative difference between the two engines I guess I will have to find a similar engine setup to compare to mine.

I don't have any money to play with, I guess that is what my trouble is and why I don't give up.
Another reason I don't give up is my car has done a trip at 80 km/h one night and got a fuel consumption of around 4.1 L/100 km, worst case is as high as 5 L/100 km this is dependent on the slope of the pavement where I filled up, its about a 3L variation on a full tank from tests I've done since, so I am discounting if I got less fuel when I filled up that night. My consumption would of been around 3.3 L/100 km, I have a witness or two but I doubt if they would talk to me again (x girlfriend and her x husband). That 71 US MPG I serious do not think I can reproduce that ever. I have shelved my engine thing for a few months now maybe a bit longer is needed. But I could try and contact the x husband some how.
 
smokingwheels,

If you don't want to build/buy a dynamometer, and you have the engine mounted in a working vehicle, I would suggest you hook up an auxiliary fuel tank (which allows you to measure precise fuel usage), take it on the highway and make a few trips with and without your modifications. To be more scientific you could have someone else do the driving, not letting them know whether you do or do not have the mod installed. Try to drive at constant speed and work up the average mileage with and without the mods. See if there is a significant difference in fuel economy which will indicate whether there is a significant difference in fuel efficiency.

Be sure you use the same type of fuel each time. Look http://www.epa.gov/fueleconomy/data.htm" for information on what variables can affect fuel economy so that you can minimize variations and get good data.

And I'll repeat again what everyone has been telling you:
Performance information at engine idle is useless! You must test the engine under load!
 
Last edited by a moderator:
This one got a LD28 with a turbocharger for a while before going back to gas...and a turbo.

Post is related because it's the same year and model as being discussed.

http://princemakaha.homestead.com/files/IMG00069.jpg
 

Similar threads

Is My Internal Combustion Engine Efficiency Too Low? A Table Analysis

  • MMCS
  • · Replies 4 ·
  • Engineering and Comp Sci Homework Help
Replies
4