What you need to know about Brake Specific Fuel Consumption

[Ecky]

Originally Posted by drmiller100

...Lets say we take a multi port fuel injection engine.

And we take the throttle body and wire it wide open.

Then we tell the computer to read a fly by wire throttle plate. And we shut off the injectors 98 percent of the time, but every once in a while we give a specific cylinder the "right" amount of fuel to run at WOT.

But most of the time the cylinders are not getting any fuel.

Then we determine "once in a while" to be decided by the throttle pedal - the more we push on the throttle, the more often a cylinder gets a shot of gasoline.

No pumping losses. Full efficiency of the cylinder. The engine can run at very efficient loads, and we can use the air pumping through the engine to be a pseudo carnot cycle.

Sounds essentially like a diesel with a spark plug. I'd be interested in hearing some educated opinions as to why we don't do this. Emissions? I find it interesting that this system could potentially do-away with an O2 sensor.

Honda (and probably others) has tackled this with variable valve timing. Their newest Civic engines don't have a throttle plate, but instead leave the intake valves open into the compression stroke and close when the right amount of air is left in the cylinder to combust @ their ideal AFR, which solves to problems at once - no pumping losses, and vastly improved emissions.

I know there are also engines (Honda V6's?) that shut off 3 cylinders and leave both intake and exhaust valves closed when driving under low load conditions. This additionally reduces the energy lost pushing air through those cylinders.

[serialk11r]

The firing the injectors only part of the time idea is interesting, and the pumping loss you get from any compression/expansion mismatch would likely be smaller than the loss from restricting the flow at the throttle.

But I'm guessing engine smoothness would take a big hit...

[drmiller100]

Originally Posted by Olympiadis

I have been doing that for a very very long time, but it is irrelevant to this discussion of efficiency.

Please define "doing this". If you mean thinking about it all, then I'd probably concede.

If you mean trying stuff, then I doubt your comment.



A good tune can absolutely increase gas mileage measurably. And a "good" tune is relatively easy.

A good tune is NOT just leaning out the mixture by "adding oxidizer". If you desire added "oxidizer", increase throttle opening of lean out the mixture.

A lean mixture only burns faster then an efficient mixture right on the ragged edge of detonation due to the free radicals. It is pretty hard to count on running this unless you take into effect the humidity, subtle variations in coolant temp in the head, history of load (affecting head temperature), oil temperature, and piston crown temperature. All this is pretty hard to measure, and no one to my knowledge goes to this extreme.


As a favor, could you summarize your responses - I'm willing to continue discussions, but your posts could be shortened and still contain the same information.

[drmiller100]

Originally Posted by Ecky

Sounds essentially like a diesel with a spark plug. I'd be interested in hearing some educated opinions as to why we don't do this. Emissions? I find it interesting that this system could potentially do-away with an O2 sensor.

Honda (and probably others) has tackled this with variable valve timing. Their newest Civic engines don't have a throttle plate, but instead leave the intake valves open into the compression stroke and close when the right amount of air is left in the cylinder to combust @ their ideal AFR, which solves to problems at once - no pumping losses, and vastly improved emissions.

I know there are also engines (Honda V6's?) that shut off 3 cylinders and leave both intake and exhaust valves closed when driving under low load conditions. This additionally reduces the energy lost pushing air through those cylinders.

I was not aware of Honda doing this. Thank you.

At the moment this is a thought experiment, but I am planning to try this on one of my wounded old honda civics. I was going to run it with an O2 sensor, and leave it closed loop.

But, now that I think about it, I cannot run an O2 sensor. That is probably one problem. Another problem is probably keeping the CAT lit.

[drmiller100]

Originally Posted by Olympiadis

If you really have this belief, then I ask you to please set up an experiment with a small sacrificial engine. An oxidizer is easily purchased in the form of O2 or N2O. Put a wideband O2 sensor and a K-type thermocouple in the exhaust. Make sure your AFR starts out at the chemically perfect 14.6:1 ratio, make your observations as complete as possible, then add your oxidizer. Note everything that happens from rise in RPM, to the visual condition of the combustion chamber and spark plug upon tear-down.
If you have a way to measure a fixed partial-load power output, then all the better.

I can only share results that I have collected and highly encourage you to collect your own. There's no substitute for learning by doing. Yes you can absorb spoon-fed information, but it's not the same impact, and is often done too selectively.

My small, sacrificial engines are 2 stroke snowmobile engines making 200 horsepower per liter off squeeze, and 250 on the squeeze, honda car engines (stock) and chevy small blocks.

What do you use?

[Olympiadis]

Originally Posted by drmiller100

My small, sacrificial engines are 2 stroke snowmobile engines making 200 horsepower per liter off squeeze, and 250 on the squeeze, honda car engines (stock) and chevy small blocks.

What do you use?

My preference is the Chevy small-block SBC1, because that's what I have the most of, - most parts and tools for. There's about ten complete ones in my shop right now, and lots of spare components. Why would you ask that?

Of course I work with other configurations, but there are so many out there that to be as competent with each is unrealistic, and given the special tools involved for everything from engine and cylinder head work to ECM tuning, it is not financially reasonable for an individual.

If your question was to determine range of experience with engine tuning, then I first started tuning work with electronic engine management in 1994 using the GM 165 ECM and Accel DFI. I am familiar enough with a few other stand-alone systems, but my preference is with the older factory GM ECMs (OBD1) because of the level of control I can achieve with them.

Stand alone systems give a great range of calibration control and some very nice extra features, all within the range of the user interface limits.

Some of the tuning suites like EFI-live and HP-tuners give a remarkable amount of access to most of the tuning calibrations involved with OBD2 systems, but not all, and no access to the code or algorithms themselves.

With the older GM OBD1 ECMs I have complete access to the running code.
This is an advantage to me because I learned machine language programming for the 6502 processor when I was in school, and the knowledge translates directly over to the M68HC11E processor family. This facilitates my ability to both read and write in the language of the code that runs in the ECM. Being able to write code, or "patch" as some call it gives far more power and control over tuning ability. Being able to read it allows you to see what is really happening behind the scenes with electronic engine management, and gives much insight as to what the automotive engineers, particularly the programmers were trying to do, and what they were physically able to do. That is why my preference in each case to engines and ECMs.

Does this answer your question fully?

[Olympiadis]

I'm willing to extend an olive branch in the form of shared information if you (whoever) is willing to admit that you really don't have a clue when it comes to engine tuning, and in particular as it relates to electronic engine management as is used in fuel injected gasoline vehicles.

As dangerous as it may be, I'm a believer in shared information and individual experimentation. I think that if some of you had a look at the code that runs in the engine management system that there would be a great many revelations, which should move forward every discussion of engine efficiency, to include the use of BSFC maps as this thread is supposed to be about.

(I'm referring to sharing a file such as those I use as a reference, - a copy of the binary code that runs in the ECM converted to hexadecimal as it was used during programming.)

It will also become clear that there is a big difference between knowing people in high places and actually knowing what is going on in terms of engine function.

If you are comfortable working in hexadecimal, or programming in machine language then the file will be easy to read. For others you can use the commented text provided, and if you wish to verify function you can learn machine language and trace it out. For further verification of function you could set up a test bench with an ECM running the same code and perform functionality tests (several tuners use this method), or you could use a working model in a running vehicle, using datalogging software to verify functional changes.


I am a human though and have my pride, so the next move needs to be the retraction of some of the statements and claims made thus far, and an honest willingness to learn. Short of that, there's no reason for me to post more in this thread.

[drmiller100]

Originally Posted by Olympiadis

then I first started tuning work with electronic engine management in 1994 using the GM 165 ECM and Accel DFI.
?

that is a good place to start. I invite you to look into Megasquirt.

Megasquirt has a lot of control available, including boost, multiple maps, full support of different sensors, etc.


This is turning into one of those "who has the biggest dick" contests which oldmech has been asking ever so nicely to stay out of.


when all done, if you have a solution using HHO which gives better mileage, and it is reproducable, feel free to post your own thread.

Meanwhile, the rest of us are chasing our own dreams.

[Ken Fry]

Originally Posted by drmiller100

So to the other dreamers out there, step up. TRY THINGS. You learn so much more by TRYING

Geeez, here I am again, going off topic. I agree with everything you say, except perhaps for your take on my support of academia. In most respects, I think academia sucks. Completely. Its function (for the most part) is to stamp out droids to fill slots in corporate America, where independent thought is (duplicitously) encouraged by word and strongly discouraged by deed. It is somewhat coincidental that in later life, I've run into one helpful prof after another, and many do all sorts of things (building not one but several aircraft, for example) in addition to publishing reams of academic stuff.

Core values, innovation, "Our most important asset is our people" (of whom we just laid off 10,000 to drive up our stock price)... yada yada -- all that corp speak makes me gag. But again, I've met some guys from Japan who treat hands-on engine work as art. I went to a race prep school at Honda, where a guy showed me how to hone a cylinder the "right" way... by hand, with emory paper. He was making love to the cylinder: slow, smooth, in-out, in-out with carefully matched twist... No surprise that while the American manufactures were complaining about how hard it was to meet EPA standards, Honda was exceeding them and making it look easy.

Then there is the maturity thing. I am a kid at heart, and every bit as susceptible to "immature" acts as you, I'd think. Given that I am still this way at age 61, I doubt that I'll ever grow up in any meaningful sense.

Actually, amazingly, I don't think about sex all the time, but I like this true story: I was working for a huge tire company, and talked with a seasoned industrial engineer about a newby going to interview the plant's best tire builder (fastest production rate, lowest scrap and repair rate). The newby asked the builder how he was able to produce good tires at such a high rate. He said he thought about sex all the time. The newby sought clarification: "Surely you must think about the techniques you use on the job a lot." "No. I think about sex: all. the. time."

The newby wanted stuff like "I repositioned the hot knife holder, and I cut from left to right instead of right to left." But no. Got the truth instead. Where does that fit on the form?

I wish this forum had an "off-topic" facility, like CR4 does. You can vote yourself off topic, and your post is minimized, so can be easily skipped over. It permits jokes, side discussion that are prompted by a thread but which are not really needing their own thread, etc. etc. You have hit on a lot in your short post, and it is valuable stuff about which I might otherwise say more... but I've already posted a ton that has little to do with the original thread topic.

Thanks.

[drmiller100]

i've been lazy on my trike lately. need to get going.

want to do a challenge Mech and Ken? each of us have a car running and photoed in public by June?????? pure prototype, not finished, but Having someone piss me off pushes me to get things done.