Exhaust valves

[GreenSquirrel]

There is some merit to this although it's effect on mileage has yet to be credibly tested but here it is; exhaust valves create backpressure at low RPM's (when the engine is cycling smaller volumes of exhaust per minute) but open to allow full use of the cavernous exhaust pipe, it allows the engine to get a small bump in hp lower in the rev range, my question is has anybody tried exhaust valves to nurse the engine during acceleration at lower speeds for the purposes of mileage/fuel-conservation? I intend to give this a go on my next vehicle but would like to know what others know/have experienced about this subject.

[aerohead]

Originally Posted by GreenSquirrel

There is some merit to this although it's effect on mileage has yet to be credibly tested but here it is; exhaust valves create backpressure at low RPM's (when the engine is cycling smaller volumes of exhaust per minute) but open to allow full use of the cavernous exhaust pipe, it allows the engine to get a small bump in hp lower in the rev range, my question is has anybody tried exhaust valves to nurse the engine during acceleration at lower speeds for the purposes of mileage/fuel-conservation? I intend to give this a go on my next vehicle but would like to know what others know/have experienced about this subject.

At least one production motorcycle incorporated a 'smart' muffler which had a computer-controlled butterfly valve inside.
The motorcycle engine underwent a full-spectrum dynamometer test to evaluate it's best performance at all rpm's as a function of exhaust back-pressure,then a look-up menu was burned into the ROM chip which allowed the computer to modulate a servo motor linked to the muffler's valve; enabling the engine to always remain in the sweet-spot.
If you've got a loose $100,000 you want to throw at a chassis dyno,you'd be able to do this to your car.

[RustyLugNut]

Originally Posted by aerohead

At least one production motorcycle incorporated a 'smart' muffler which had a computer-controlled butterfly valve inside.
The motorcycle engine underwent a full-spectrum dynamometer test to evaluate it's best performance at all rpm's as a function of exhaust back-pressure,then a look-up menu was burned into the ROM chip which allowed the computer to modulate a servo motor linked to the muffler's valve; enabling the engine to always remain in the sweet-spot.
If you've got a loose $100,000 you want to throw at a chassis dyno,you'd be able to do this to your car.

For roughly 10,000 USD, you can purchase an array of pressure sensing spark plugs and the attendant data logging. A set of well thought out acceleration runs will get you most of what you desire.

If you are doing a single cylinder engine, you can purchase the sensing spark plug for about $1K and cludge your own data logger using slower commonly available electronics and do multiple cycle overlays.

But, that's just what I would do.

[niky]

You could actually do the same thing with a piggyback chip like the UniChip-Q, to which you can add auxiliary maps to precisely control things like extra injectors, nitrous and even a flapper valve.

Just tune it once on the dyno, you're done.

[Daschicken]

Originally Posted by aerohead

At least one production motorcycle incorporated a 'smart' muffler which had a computer-controlled butterfly valve inside.
The motorcycle engine underwent a full-spectrum dynamometer test to evaluate it's best performance at all rpm's as a function of exhaust back-pressure,then a look-up menu was burned into the ROM chip which allowed the computer to modulate a servo motor linked to the muffler's valve; enabling the engine to always remain in the sweet-spot.
If you've got a loose $100,000 you want to throw at a chassis dyno,you'd be able to do this to your car.

My grandpa's 1999 yamaha r1 had an exhaust butterfly valve, but I doubt it was computer controlled as the thing had carbs. Although it did have a MIL system that had codes for the exhaust valve, so it could be.

[aerohead]

Originally Posted by RustyLugNut

For roughly 10,000 USD, you can purchase an array of pressure sensing spark plugs and the attendant data logging. A set of well thought out acceleration runs will get you most of what you desire.

If you are doing a single cylinder engine, you can purchase the sensing spark plug for about $1K and cludge your own data logger using slower commonly available electronics and do multiple cycle overlays.

But, that's just what I would do.

The thing about the dyno though,is that you can get BSFC data as you go,which is of great importance,and a dyno's the only place to scientifically acquire the data directly that I'm aware of.

[RustyLugNut]

Originally Posted by aerohead

The thing about the dyno though,is that you can get BSFC data as you go,which is of great importance,and a dyno's the only place to scientifically acquire the data directly that I'm aware of.

It is used daily in industry. Lay people see it as "unscientific and variable". Granted, the lab dyno does allow control of certain variables, however, in the case of exhaust back pressure versus load and fuel use, all one needs is a set of accurately mapped fuel injectors/pump for fuel flow and an in-cylinder pressure sensor to see power potential as well as an exhaust pressure sensor and position sensor on your exhaust backpressure valve.

A baseline run followed by a set exhaust valve run yields a pair of pressure traces that allow you to determine a gain or loss. No need to calculate BSFC as combustion pressure is all the feedback you need. A pair of runs takes 5 minutes or less. A well thought out set of runs can yield a reasonably extrapolated pressure map with which to program an engine map.

Again, there is no need to carry the burden of a dyno cell instrument set. In-cylinder pressure will tell you if the exhaust back pressure is beneficial or detrimental. I would probably add a stepper actuator to the exhaust back pressure valve to make "fanning" the back pressure during runs possible. This reduces the number of runs during your data gathering.

[gone-ot]

...or, you can buy software programs that simulate just about 100%-functions of an engine in operation...at any load...under any weather conditions.

[RustyLugNut]

Originally Posted by Old Tele man

...or, you can buy software programs that simulate just about 100%-functions of an engine in operation...at any load...under any weather conditions.

None of the software programs found at your corner bookstore will be able to do the job, but there are several commercially available programs that would allow you to do most of your ballpark work. Of course, there is the issue of computational power. Some titles need a Cray class computer to run on, while others can run on lesser mainframes or desktops though with very slow results. And in the end, you must vet your model with actual data.

[Xist]

Aerohead, do you remember approximately what kind of a difference that setup made for that motorcycle?