That has to be a pretty clunky (and smokey) old engine to be able to benefit to the tune of 18.8% just from better atomisation.

This kind of device sounds like it could be after-market as well as new car. Can we believe the claims across a wide variety of injection engines (gas included)? If even 10% average improvement, it would be huge as the device would seem to be relatively inexpensive to make.

It might not have better performance than a modern high pressure injectors, but will be cheaper and so have a large market opportunity in places like India

Agreed that this smells like snake oil.

The testing methodology just isn't scientific. Road tests? Oh jeez. The only way you could verify increse in mileage is to put the same car throught he same controlled lab tests with both the device and without. Just the fact that the driver doing the road tests knows that this device installed could easily account for the increase in mileage due to subconscious changes in driving patterns.

Now the concept makes sense on old-fashioned diesels, and if it really does increase mileage by even 5%, it could be a cheap and useful retrofit for older trucks. But I wouldn't hold your breath until some valid study is done.

Computer controlled dynamometer tests with the same vehicles before and after the installation might be more convincing. The same set of cars either do or do not perform more efficiently under consistent identical tests.

Fuel atomization sounds like a good idea, but I am not sure that this is achieving the results that they say it does. It seems something like the electrolyze water to H2 and O2 schemes. It makes inefficient vehicles improve a bit, but so would a good tune up. When you tune them up the improvements due to the add on devices go away.

I wonder if you get similar effect if you add detergent to the fuel. Detergent reduce surface tension.
So it might be passable to improve fuel atomization in such way.

Gas mileage is an important issue, one of the ways I have been saving money at the gas pump is with Bishop's Original- it increased my mpg by 15%.

They probably did use an old Mercedes-Benz with poor atomization for testing and newer common rail system would not benefit as much. But I have to wonder how many diesel ships and locomotives would benefit from this. A 12% to 15% improvement in fuel efficiency for a ship is huge and would quickly be profitable to implement.

Uhm, if you heat the fuel or add any of a zillion additives, you can also reduce the viscosity. Hate to spoil the excitement, but viscosity has been played with endlessly in the adaptation of SVO; it has to be sufficiently low for the engine (they typically get to within a factor of 3 of standard diesel) and beyond that if low viscosity made a world of difference on diesel efficiency we would have all heard about it. The authors dismiss additives as being inordinately expensive. You can use shorter hydrocarbon chains and alcohols--which are about the same price as diesel fuel.

Further, if incomplete combustion was leaving 19% economy on the table before this mod, you would be seeing ungodly plumes of smoke coming out of the engine. If combustion was too slow, engineers would have advanced the timing (or you'd have heard about a huge efficiency fall-off with high RPM, and fixes for that). High-pressure common rails would be discussed in the literature for fuel efficiency first, not emissions abatement.

The author dismisses high-pressure as unavailable technology--that is true of gasoline only, where it is hard to see how you can benefit. Normally there is plenty of time for gasoline fuel droplets to completely vaporize, or they would be going back to carburettors.

Also odd (back to diesel) is that the researchers are very impressed by their data which show a big boost in very small droplets, the end of the scale where I don't think there's a problem, but don't even mention that their own chart shows an increase in the larger droplets, which you'd think would be a bigger risk of incomplete combustion and particulate formation.

Small droplets aren't a panacea. They also mean that air drag will stop them before they get very far, so they'll cluster around the injector in a small cloud and produce an oxygen-poor zone. Which is why the high-pressure common rail guys need so much pressure and also have gone to multiple orifice injectors as part of the design.

I'd put my money on charged injectors (which the authors dismiss as impractical) with the voltage adjusted for conditions, maybe with an injection-cycle waveform. And then only for widening the spectrum of fuels that work well while still giving optimum emissions. Except that I don't think the engine makers care. That's a free-world public-interest versus Big Oil issue. which the lawyers in Washington don't understand.

It seems like they should have ran a test with uncharged mesh screens in the fuel line. Couldn't the drag produced by these screens just being in the line have effects on the performance of the engine (although I would have assumed poorer performance)?

Why do these miracle devices always test on old s-boxes?
The average 15 year old Merc could see it's efficiency go up by double digits with a normal service (fluids, filters, air pressure and initial injection timing).
Testing over 6 months just brings in a number of other variables, such as temperature, fuel blends, traffic density, etc.

Given the results they claim, which would save shipping fleets millions upon millions of dollars (and let them retire filthy rich just as soon as the patent application clears), you would think that they would be willing to borrow a modern car and rent a dyno for a few hours...

Don't they teach any scientific methodology at Temple University?

Alas, this idea won't work with a modern gasoline engine. The reason is simple: modern gasoline engines already have fuel injectors that atomize the fuel very finely, so adding this device won't provide any more improvements in fuel atomization.