This has got to be snake oil 1

[GregLocock]

Researcher claims 5%-20% (!!!!) improvement in efficiency by using an electric field to reduce the fuel viscosity and hence droplet size in a diesel engine.

http://pubs.acs.org/cgi-bin/sample.cgi/enfuem/asap/html/ef8004898.html

http://pubs.acs.org/cgi-bin/sample.cgi/enfuem/asap/html/ef8004898.html

This makes very little sense to me, I wasn't aware that the burn time of the droplets was a significant cause of inefficiency.

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[SomptingGuy]

An alternative to higher injection pressures perhaps?

- Steve

 

[Lou Scannon]

A couple of ways to think about efficiency in a recip engine are degree of constant volume combustion and effective expansion ratio , which are really very similar concepts. So alterations in combustion that increase these measures of combustion performance will tend to increase thermal efficiency. But this will come with a pricetag, which is increased NOx. So the question should be, what is the potential efficiency increase, with NOx held constant (i.e. by adjusting other parameters such as injection timing or compression ratio, to offset NOx increase resulting from faster burn)?

 

[carnage1]

I thought cheap whiskey was what you were supposed to use to reduce diesels viscosity

Luck is a difficult thing to verify and therefore should be tested often. - Me

 

[evelrod]

A couple things pop out...

Up to date, electrostatic atomization technology has not been employed on any fuel system, not to say any vehicles.

However, the main function of our device is to reduce the viscosity of the fuel as it passes the electric field.

I'm no diesel expert, but what would happen if you simply altered the fuel viscosity?

Rod

 

[ivymike]

I'm not sure you could do that chemically without changing other properties that you want to maintain.

If this really does allow much better atomization of fuel at a lower working pressure, then it could certainly improve pump life and/or reduce input torque requirements. Fuel economy and low emissions seem to be pretty closely tied to atomization (via ever higher pressures, typically).

 

[Bribyk]

I don't know how they can claim their dyno is accurate to 0.02 hp when the noise in their steady-state measurement is roughly +/- .01 hp. I believe the minimum significant change would be .021hp (3-stdev)with an RMS noise level of .007hp and they only measured .075hp difference. If you wanted to detect a significant change in fuel consumption you'd need to test at a higher fuel consumption rate or have extremely accurate sensors. There's no discussion of an engine test standards and even if they were within, say, ISO 3046-1, I'd expect you could easily have the parameters within tolerance (such as oil temp) and achieve a .075 hp difference.

Is this even relevant on a ~10-15 year-old diesel engine anyway?

 

[JSteve2]

I have to say, I looked at it expecting complete snake oil. However, there may be something to this. Better atomization is good, and this is an alternative to higher injection pressures. I'm not sure he's using a system I can relate to - he lists the diesel fuel pressure at 13.79 bar which is nowhere similar to injection pressures I'm familiar with. I would expect the modest viscosity decrease (9% I thought I read?) to be more effective at lower pressures. At higher pressures, that actual viscosity change may not be realized as the shear is higher in a high injection pressure engine and diesel is not a Newtonian fluid (nothing is) at extreme shear levels.

In the end, I'd have to see the energy inputs and I'd have to see this on a high pressure application before I would say it's a realizable success in an application. I'm very dubious of the size of the claimed efficiency improvements, either the 12-18% or the 20.4%.

 

[JSteve2]

Evelrod,
Most viscosifier adjustments (like chemical additives) cause incompatibilities, cause emissions impact, cause lubricity problems, or are just too expensive.

Heating the fuel can require signficant energy input, I'm guessing there is a reason this is not done beyond the extent of heating above "cold." There is a diminishing return with higher temperatures, with viscosity gains decreasing and heat losses to the environment increasing.

 

[ivymike]

Heat is cheap and easy to come by. Hot fuel tends to be hard on injectors and pumps.

 

[jmw]

Is there not a power factor correction that manufacturers apply when line testing diesel engines during manufacture that requires they measure the density and viscosity at the fuel temperature? (J something or other); which implies an effect of the fuel kinematic viscosity on power output.

Large diesel engines (Residual fuelled) have an inline viscometer which is used to control the fuel heater so that the viscosity is optimised.

If the fuel viscosity is too low then a fine non-dispersive mist is formed which does not fully mix with the air flow; the fuel burns incompletely and in the region of the injectors.
If the viscosity is too high the large drops formed project too far across the burning zone, again do not mix well with the air flow and incomplete combustion results, and in the wrong place.

In either case optimising the viscosity makes the engine more fuel efficient and less polluting.

Theoretically, the same situation could apply to smaller diesels burning commercial diesel or turbines burning aviation fuel.

BUT:
How variable is fuel quality?
Commercial diesel of a the approved grade for the engine shouldn't exhibit that much viscosity variation nor should it vary much over the normal range of fuel temperatures with the engine operating normally.

Hence I would have expected that the engine design would be that at normal working conditions of temperature etc. the viscosity ought to be pretty much optimum or is this wrong? How significantly does diesel viscosity vary according to source, garde and fuel temperature?
At what point does viscosity control become desirable? I'd certainly like to know.

Am I right in thinking that simply reducing the viscosity isn't automatically a good idea?



JMW

http://www.ViscoAnalyser.com

 

[JWaterstreet]

I think you guys may be missing something very simple in your opportunity to critique the paper. Perhaps the old Mercedes engine (& the newer Iveco engine) they tested had suboptimal injection/atomization to begin with, and the simple reduction of viscosity of the fuel (which could have be simply done by ordinary resistance heaters) allowed better atomization, hence the improvements they observed.

 

[jmw]

By the way, I have to confess my ignorance and caution. I was previously aware of electro-viscous fluids e.g. as used for shock absorbers, but I thought they were fluids filled with ferrous particles so a bit of checking was in order.

From this US Patent (

http://www.freepatentsonline.com/EP0387857.html):

An electroviscous fluid comprising an electrically insulating liquid and solid electrolyte particles dispersed therein.
And we learn bout the primary electro viscous nature of some salt solutions and, apparently, some observed effect in polystyrene. But a significant effect in fuel oils or even diesel? That takes some getting used to.

There are some viscometers that depend on magnets in some way; some use strong magnetic fields to self centre the rotating element and thus also do away with bearings (and bearing drag) e.g. the Levimetrix (

http://www.levimetrix.com/).

More relevant is the Stabinger SVM 3000 (

http://www.anton-paar.com/ap/apinternet/html/default/cxsn-5qsl4r.en.0.jsp

and select brochure for screen).
This is approved to ASTM D7042 and is of equivalent precision to the capillary viscometer ASTM D445. It also makes use of magentic levitation of the rotating element. The primary application for this viscometer is fuels, and lubricants. So how can you employ magnets without affecting the performance of the viscometer?

Yet here we have it:

"....the new physics principle that proper application of electrorheology can reduce the viscosity of petroleum fuels

The article is in the references:

http://pubs.acs.org/cgi-bin/sample.cgi/enfuem/2006/20/i05/pdf/ef060072x.pdf?sessid=4888

and the authors are: R. Tao and X. Xu, and R. Tao is one of the authors of this report so it is a natural progression from discovering the principal to looking at applications.

OK they talk about pulsed electromagnetic fields and our viscometers presumably rely on stable magnetic fields.... but I'll need to think about it some more.




JMW

http://www.ViscoAnalyser.com

 

[dicer]

Think valence energy levels, then it isn't so funny. Gota get the old head outa the box on occasion.

 

[jmw]

Referring back to JWaterstreet's post, it may well be that there is some improvement in fuel efficiency by reducing the viscosity but the questions are:
[ul][li].... reduce it to what?[/li]
[li]How stable is viscosity from one batch of fuel to another, [/li]
[li]how well does the existing injector design atomise the fuel i.e. at normal fuel temperature (how much does that vary when the engine is hot?)[/li]
[li]what is a significant variation in fuel viscosity?[/li]
[li]how do you know by how much the magnet has reduced the viscosity? Too low a viscosity can be as bad as too high a viscosity, at least, it is for large diesels burning HFO, is it also true of a rail or truck engine, a car engine?[/li]
[li]If there is an optimum viscosity and if the variation in viscosity for whatever reason can vary significantly then surely some form of feedback control of fuel viscosity (e.g. using a viscometer to control an inline heater) would be better than simply sticking a fixed size/performance magnet in place.[/li][/ul]


JMW

http://www.ViscoAnalyser.com

 

[btrueblood]

This quote from the paper bugs me...

"The typical laboratory test result of the Mercedes-Benz with a dynamometer is shown in Figure 7, At a fixed fuel consumption rate close to 500 g/h, the dynamometer measured the engine output. When the device was turned off, the average power output was 0.3677 hp. It increased to 0.4428 hp after the device was turned on. This indicates that the power output was improved by about 20.4% at the same fuel consumption rate. In other words, if the engine on the road is under the same condition as our laboratory test with the dynamometer, the fuel mileage will be increased by 20.4%. The laboratory test was repeated for 3 h and had an error within 5%.
"

The engine they are testing has a rated bhp of several hundred horsepower (hp), and probably uses at least 25-50 hp in normal driving at low speeds...so why test it on a dyno at ~1/50th of its expected minimum operating output power? (I have trouble even imagining the engine maintaining a steady rpm with that low of an output power). Irregardless of the signal-to-noise problem with a dyno operating near its zero, pointed out by others, how can they make any meaningful statements about fuel mileage at any reasonable speed/operating power level based on tests at 0.4 hp output?

The stated mileage numbers are dubious for all the usual reasons (small sample size, uncontrolled drive cycle, etc. etc.).

 

[Bribyk]

Not to mention their generic uncertainy statements and use of significant digits (.3677 hp +/- 5% !?)... According to the dyno, fuel mileage will still be zero, cause the car's not moving, but it looks like if you turn one headlight on you'll save 20% in fuel while you're at the stoplights.

 

[jmw]

Hang on a sec guys...
500g/h and only 0.4hp?
That is gallons per hour is it or is it grams per hour?

Road tests: well now, I have found my fuel consumption varies according to the amount of attention I pay to the on-board computer.

If this were a car to which the device were attached without the user's knowledge and was switched on on alternate days or there was some way to conduct a double blind test, we might be getting somewhere. As it is One has to wonder if we are actually measuring device effect or operator effect (Deren Brown is pretty good reading on ouija boards and how they work..something the same could be working here. Then again, it could all be pucker gen................couldn't it?


JMW

http://www.ViscoAnalyser.com

 

[btrueblood]

Pretty sure they meant grams/hour...

 

[Lou Scannon]

No one has responded to my post above. We all know there are plenty of adjustments to be made to a modern emissions controlled engine that will increase thermal efficiency, but all of these I know about increase engine-out emissions of regulated pollutants (especially NOx) and most of these increase stress on the engine also.
So, how is this method different, assuming it works in the first place?