Does Acetone Addition to Fuel Lower Consumption?

[djmeier1]

Nick's finding of a strong correlation of efficiency with coolent temperature should be expanded upon a bit in discussion. One wonders, for instance, what that optimal temperature is and how precicely and reliably that temperature is maintained by the thermstatic controls.

 

[Ernie Rogers]

They're nice girls

Those BYU cheerleaders are proper girls.

I always wondered how they got so many fans to crowd into the Cougar stadium at each home game. I guess seeing would have answered my question.

Now, about coolant temperature and engine efficiency-- I don't know, I think there have been scads of discussions about how TDIs get more efficient as they heat up good. This is consistent with theory. A major energy loss path is heat transfer to the cylinder walls. The futuristic engine concepts often include "adiabatic engines" that don't have any coolant at all.

Ernie Rogers

i really want to reply with some sexual innuendo, but I'll keep it to myself.

 

[nicklockard]

I sent the data file (Excel) to Ernie so he could take a look. I also PM'd GoFaster and TdiMeister (engine guys) because I need to be able to answer the critical question: could pinched fuel lines (but still flowing sufficiently to maintain 60 mph) increase fuel consumption by 6%?

If we can answer this question to our satisfaction, then we can learn a lot from the data. (I know this is torturing some of you and I'm sorry for that! I just want to give this information and data as fair a treatment as I know how, and that takes time and knowledge. )

Cat out of the bag soon then.

At any rate, this is interesting. I hope some of you are interested in doing the road tests; I'm eager to learn more.



Nick

 

[jackbombay]

Who is going to get a higher temp thermostat?

 

[Ernie Rogers]

I didn't get it, try again?

Hi, Nick,

I didn't find your email, want to try again?

Ernie / arcologic@aol.com

I sent the data file (Excel) to Ernie so he could take a look. I also PM'd GoFaster and TdiMeister (engine guys) because I need to be able to answer the critical question: could pinched fuel lines (but still flowing sufficiently to maintain 60 mph) increase fuel consumption by 6%?

If we can answer this question to our satisfaction, then we can learn a lot from the data. (I know this is torturing some of you and I'm sorry for that! I just want to give this information and data as fair a treatment as I know how, and that takes time and knowledge. )

Cat out of the bag soon then.

At any rate, this is interesting. I hope some of you are interested in doing the road tests; I'm eager to learn more.



Nick

 

[Wonko the Sane]

Well, I want follow up on my previous post by reporting results from my second tank with acetone.

First tank: Apparent improvement of 5 mpg for the first 5 or 6 gallons. Then no change for the remainder of the tank.

Second tank: Used 3.5 oz acetone at fill up. Saw about 3 mpg improvement at the beginning of the week. Then back to normal.

Overall, acetone has not impressed me much. It has possibly saved me one gallon over two tankfuls. Maybe if we collected a bunch of data from many drivers and did a factorial design we could find the variables (if any) that cause an acetone effect.

I'm going to use Diesel Kleen for my next two fill ups for comparison.

I plan to take a couple of road trips this summer. If I try a tank of "acetoned" fuel on a 500 mile day, I'll report again.

I'll look forward to reading other members posted experiments/results.

Wonko

 

[russell67]

I have to thought about pinched fuel lines. Just putting a valve in that at freeway speeds i could close it off until i am running real lean. Problem i see with this is the O2 sensor would try to adjust for this and possible try to push more fuel thru.

 

[JettaJake]

huh? say what?

 

[TDIJetta99]

Funny, I don't remember seeing any O2 sensors when I took my exhaust system off ..

 

[nicklockard]

I've been reading up on efficiency factors but I am still struggling to find appropriate normalizing factors. I tried an arbitrary operator summing all the recorded temperatures in Kelvin to various degrees...but it doesn't make sense If anyone wants to take a crack at it I will PM the raw data file to you (text or Excel.)

I'll post Ernie's comments and add my own:

Hi, Nick,

There's a lot of numbers in that sheet. I studied the graphs. Here is what I saw--

On the control run, I marked the point on the fuel consumption curve where the fuel temperature hit 60 C-- that appears to be the end of the test. I went across to the acetone test and marked the fuel consumption there at the 60 C point. The two fuel consumption numbers are very close, slightly lower for the acetone run. My guess is 4.4 for the acetone fuel and 4.45 for the control. The difference is less than the scatter.

Air temperature for the two points differs slightly, maybe a degree higher for the acetone run. This may surprise you, but fuel consumption is very sensitive to intake air temperature, higher temperature giving better fuel economy. I can model this effect with my engine simulation program, but I don't see much value here since you are in the noise band already.

A person would be tempted to say the acetone fuel economy is better, but both fuel temperature and air temperature are higher where the fuel consumption becomes noticeably better. That's above the 60 C point, where no data are available for the control run.

In my view, the only good data point in the control run is the 60 C point, at shutdown. That's where the coolant temperature is under control, telling me that the engine metal is just starting to stabilize its temperature. You can't use data from an engine that's still warming up. Anyway, all of the data are just too close to the startup to be very reliable.

How about the effect of fuel temperature? That's a critical parameter. We know the engine should run much better with warm fuel (in most cases). But, I wouldn't be able to calculate the effect as I might with the air temperature. I only model energy, not combustion effects.

There you are-- IF you want to draw a conclusion from the data, the conclusion seems to be no detectable benefit from the acetone. But, really, better data are needed.

I hope this helps. Ask questions if you want.

Ernie Rogers

I agree, but was hopeful that normalizing factors could account for different temperatures. So far I've been frustrated. I'm really sorry guys for letting you down. I will get some road runs with Brian and see if we can flesh this out.

Also, here is the plot of raw data, for what it's worth:

[/IMG]

Fuel consumption is plotted in red and scaled on the right side Y-axis in red letters as L/100 Km. Coolant temperature is in pink. The primary trends visible to me are the strong correlation of fuel economy to coolant temperature evident on the control run. In fact the slopes are nearly symetrical to one another. Looking over the acetone run, you can see that although the coolant temperature has stabilized within a band, fuel consumption continues to drop as fuel and air temperature continue to rise.

In reading over engine efficiency factors in a diesel constant pressure heat addition cycle, we see that compression and expansion ratio's are most important. What I need is to see these factors related to temperatures. I think if I understand the mathematical importances of temperatures, I could take another stab at normalizing the data. The units don't really matter and can be entirely arbitrary, since we are only concerned with relative comparisons. If anyone feels up to smacking me upside the head so I can see the relationships PLEASE feel free

TdiMiester hasn't answered PM's yet. I think he is very busy at school. Brian and I will gather some road data to add to this and see what we see. Thank you, everyone, for your patience.

Nick

 

[nicklockard]

Nick's finding of a strong correlation of efficiency with coolent temperature should be expanded upon a bit in discussion. One wonders, for instance, what that optimal temperature is and how precicely and reliably that temperature is maintained by the thermstatic controls.

I'm also waiting for you to comment on temperatures now that you can see the plot.

 

[PDJetta]

Funny, I don't remember seeing any O2 sensors when I took my exhaust system off ..

The O2 sensor is there on the PDs. Its pre-catalytic converter and its sole purpose is to provide feedback to the ECU for NOx control, to optimize the EGR valve and throttle plate cycling to lower NOx. It does not have any affect on fueling of so called "mixture control".

--Nate

 

[Dangerous_Dan]

Soooo close...

Nick, et. al:

Nice work! It is a shame that you had to shut down the control run before you could get data with fuel temperature for straight diesel ( D ) in the same range as the diesel + accetone ( Da ) data.

It looks to me that there is only one point where you have fuel consumption at the same water, fuel, and air temperature, and while at that point it looks like fuel consumption for Da is a little lower than for D, to me there is just not enough data.

But the thing that does jump off the page at me is the fuel consumption vs temperature. It looks like your coolant temperature is maintained at 100C+- 5C. Assuming the coolant temperature data is from VAGcom and not from an external sensor, have you taken temperature readings while driving?

I ask because my Jetta has always been a 40-45 mpg ( 160,000 miles of fuel logs ) machine and I have attributed it to gravel roads, mountain passes, and my wife's inability to drive 55. But it is rare that my coolant temperature is above 85C when monitored with VAGcom. Even after driving over our little pass ( 300 meter elevation gain ) my coolant temperature stays around 80 to 85C. Your data says that is way too cool!

If I read your numbers correctly I should get about a 15% improvement by going from 80C to 100C! Ouch! Double ouch! Does anyone know what the OEM thermostat on an AHU ( A3/B4) opens at? And how hot could my engine safely run? And what about the numbers for all the ALH ( A4 ) engines?

I know that "winterized" fuel has less energy per volume, but could this also be a part of our lower fuel mileage in winter? It looks like a 10C drop in engine temperature will result in quite a bit higher fuel consumption.

I remember that Ernie R. blocked off some air passages and got a signifigant improvement in mileage. At the time everyone was thinking that it was drag reduction. Maybe not? So would all of the over 50 mpg club folks mind checking their coolant temperature and reporting back?

 

[djmeier1]

acetone

Nick, don't be thinking that you should be "sorry". You have not let anyone down. You should be proud of having made the effort to get us all some scientifically sound data. I would think that no one expected immediate conclusive proof one way or the other.

Good science happens in pains-takingly slow steps, most of which advance knowledge only very little, while raising new lines of inquiry. That is pretty much what has happened in this case. The question really is whether, in light of the fact that you (and Brian) have been contributing your time, you want to continue the process. Either way, i appreciate what you guys have done.

I am willing to help defray part of the expense of additional runs, should that seem to be the thing to do. I am also willing to measure and contribute temperature data, if you will explain how I use my Vagcom to do so. Finally, I'm willing to measure actual fuel usage over duplicate runs as suggested by you guys earlier, if the process (cantainers and connections, etc.) is clearly laid out.

I would also completely understand it should you and Brian determine that you wish to abandone the whole thing. Dave.

 

[CinRG]

I also want to give my thanks for the time and effort put out to do these experiments. It is folks like you all that do such a great job advancing our knowledge of TDIs and diesels in general

I track my mileage continously and drive a regular 63 mile daily commute. I am going to try the acetone and report back the results.

One suggestion on the fact that some have noticed the initial mileage improvement being great and then dropping off after a few days. Why not redose with acetone?

Greg

 

[Ernie Rogers]

One suggestion on the fact that some have noticed the initial mileage improvement being great and then dropping off after a few days. Why not redose with acetone?

Greg

Yikes!

Nick, are you watching? I feel squirmy with that note about the acetone effect wearing off. If it evaporates out of the fuel, hey, no problem. IF it goes someplace else, e.g., into the tank wall, yikes!

Ernie

 

[nicklockard]

Ernie,

As mentioned above by Wonko The Sane and discussed previously, the tank temperature during normal driving conditions is well above the boiling point for 'neat' acetone. It is quite feasable that acetone is in fact evaporating off. It can not hurt the tank walls (High Density Poly Ethylene I'd guess.)

Dangerous Dan, you brought up the point on temperature I was hoping for That question is: "why is Nick's Jetta's coolant temp so high?"

And you're absolutely right on the money. During normal driving, my coolant temperature varies between high seventies and low 90's. But during this test, there were only two radial fans blowing air at the front of the car during the test...and they weren't drawing fresh cool air in from outside the shop (although the front bay doors were wide open, we were in the back of the shop.) So I think they were just blowing the same warm air around. If you notice that the temperature is almost sinusoidal it is because of the coolant fans kicking on to bring temps down. The car's ECU is obviously mapped to not allow coolant temperatures much above 100 Celcius.

Like Ernie, I too blocked the front of my car with the VW approved winter front (it's made of black vinyl and just velcro's or snaps into place (since taken off now that it is warmer here))...I wasn't keeping track of fuel economy at the time, but warmup times were reduced noticeably. I would think this would contribute to a more efficient-running engine and better economy too. That is also something we can all test out with vag-com.

Brian, can you explain to people how to log those various parameters in vag-com so they can help with data gathering?

Thank you everyone for the support, patience, and forbearance. I would gladly like to repeat it with an upgraded test apparatus on the dyno. Brian and I are going to try to fashion a semi-permanent set of line extenders that pokes through the firewall into the cabin for doing such physical consumption measurement tests...that way we can test various additives and other things.

It's always exciting to learn new stuff.

 

[Nitrowolf]

I'd like to echo what djmeier1 said... I'll contribue a bit financially, and be willing to log data ... although I'm not sure how useful my data would be given the mods to the car.

 

[Dangerous_Dan]

Not so cool

I did a little checking and...

There are several flavors of thermostats for our ( AHU and ALH ) engines.


1) 71C -> for very hot climates.
2) 80C -> for somewhat hot climates.
3) 87C -> for most climates. It looks like this is supposed to be stock.
4) 92C -> for our friends in the great white north.

The fan switch is supposed to turn on at 102C and back off at 95C, which is exactly what you were seeing! It looks to me like VW is not too worried about a TDI at 100C if they turn the fan on at 102C.


I suspect that my thermostat is sticking open or else it is the 71C part ( since mine is hecho en Mexico ) and I am going to replace it ASAP.

I may have a hard time getting valid data for a while since a set of Nokian NRH's are on the way ( 17lbs for 195/60-14 with silica rubber! ).

 

[nicklockard]

Two possible interesting threads to pursue here:

1. Those who would like to test warmer thermostats or the usage of winter fronts as economy improvers (doesn't matter if you're highly modded for this one, I don't think...) I know Ernie started to do this a while back, but I'd like to learn more and would gladly pitch in

2. Those of you who would replicate acetone runs once Brian and I test out our upgraded volume measuring apparatus and set up. We'll show how to do it with pictures...but first we need to practice all the steps

 

[rotarykid]

Want to talk about hot ??

I got a 97 Passat that had been so hot that it deposited aluminum into the cylinder walls and melted the oil pan baffel . In fact the thing got so hot that the baffel was liquified and sucked into the oil pump .

And I would have never believed it if I had not seen it , but the extreem heat didn't do any damage to the head at all . On the old engines the head would crack if you looked at it wrong . So I don't think heat is the problem that it once was .

Also the thing had a 91 * C thermostat just as my 85 TD Jetta did . On my 91 Non-turbo diesel Jetta a 87 * C thermostat was used . The latter of the 2 were bought new in NC and the Passat TDI was bought new in Colorado . It gets preaty hot in NC in the summer and we all use AC so the heat isn't a real issue if you have a good radiator and the fan is working properally on low & high speed .

 

[TDIJetta99]

I'd be willing to do a few tests with a warmer thermostat as well as making up some sort of apparatus to switch fuels while driving.. I drive a pretty much consistant run to work and back every day so I might be able to get something worth looking at with Vag-com...

Where might I be able to get the higher temp thermostat anyway?

 

[Variant TDI]

I tend to believe that coolant temp has less of an effect than other factors.

Obviously, when the engine wasn't up to operating temp, economy was lower... We all know that. But when the coolant in the Acetone test was going through multiple oscillations, consumption wasn't making similar oscillations. It was still dropping, fairly smoothly, as other factors were still heating up (fuel/air temp)

Is there an oil temp reading on VAG-Com?

 

[ofhs93]

Is there an oil temp reading on VAG-Com?

I believe there is but from what I have read they deleted the oil temp sensor in our North American cars.

 

[Variant TDI]

I think my oil pan (2002) has that sensor. I'll look today when I get home.

I was under a friend's R32, and his pan had the mount, but no sensor.

 

[djmeier1]

evaporation?

Well, I want follow up on my previous post by reporting results from my second tank with acetone.

First tank: Apparent improvement of 5 mpg for the first 5 or 6 gallons. Then no change for the remainder of the tank.

Second tank: Used 3.5 oz acetone at fill up. Saw about 3 mpg improvement at the beginning of the week. Then back to normal.

Overall, acetone has not impressed me much. It has possibly saved me one gallon over two tankfuls. Maybe if we collected a bunch of data from many drivers and did a factorial design we could find the variables (if any) that cause an acetone effect.

I'm going to use Diesel Kleen for my next two fill ups for comparison.

I plan to take a couple of road trips this summer. If I try a tank of "acetoned" fuel on a 500 mile day, I'll report again.

I'll look forward to reading other members posted experiments/results.

Wonko

Hey, Wonko. I think I agree that the Acetone might be leaving the fuel over time. And, I suppose that the rate of evaporation accelerates with increases of fuet tank temperature and with agitation. But, would you explain how you are determining the relative rate of fuel consumption in the first portion of your tank full vs. the later portion? Are you simply judging by the fuel guage and odometer? Dave.

 

[TDIJetta99]

I think my oil pan (2002) has that sensor. I'll look today when I get home.

I was under a friend's R32, and his pan had the mount, but no sensor.

Mine definitely has the sensor, but no wiring that I could find anywhere...

 

[brucetmoose]

Another reason why acetone may not have been considered - because of the VOC's

I live and work in the Rochester, NY area - home of Eastman Kodak. Every year Kodak gets HAMMERED by the environmentalist's who pound on them about their Methylene Chloride (dichlormethane) emissions. As documented in this post about acetone and here about MC, they both give off plenty of VOC's (Volatile Organic Compounds) if you leave them out there in the open air. In Kodak's case, almost all of it is leaks from the process they use to manufacture the acetate film base (it dissolves the acetate so it can be put onto giant rollers and the base literally rolls off the metal rollers - they look about 16 feet tall - there's one near the Kodak building where that work is still done on the outside which is used as a sculpture so to speak).

Acetone : http://en.wikipedia.org/wiki/Acetone
dichloroethane: http://en.wikipedia.org/wiki/Methylene_chloride

Interesting comment about acetone directly quoted from the site above:

"Small quantities of acetone added to fuel may improve economy and engine performance, though this is only anecdotal and refuted by most automotive engineers".

I can see where the enviro's would go off the deep end if they found out that significant amounts of acetone were getting into the air via the TDI's. Just look at the warnings on a container of acetone from a paint store, you'd think that the world was about to end if it all vaporized. That's all we need, more baloney from the EPA about how bad diesel is.

What I imagine would be ideal is to have a container in the car that has acetone in it that is gas-proof (meaning it doesn't leak VOC's) and that it's mixed with the fuel right before it gets injected. My company makes such mixers (called inline mixers), not sure they make them that small, but it would properly mix the two together on the way into the fuel system. Such as system would also have to be check-valved so you didn't get fuel back into the acetone, etc. That way, you don't lose the acetone into the air or at least you minimize the loss.

Now the neat chem trick would be to figure out how and why acetone works and then develop a substitute that doesn't vaporize and can be put into the tank or even mixed at the point of fuel manufacture.

 

[nicklockard]

Re: acetone sub-thread...We don't even know if it works at this point *shrugs* But, I want to do another run on the dyno if you guys think it will give more reliable results than the road tests. Again, I'll chip in $40 toward a run.

Re: Coolant temperature sub-thread...

I saw an interesting posting by Drivbiwire where he says you can run a 70/30 blend of G12/G12+ coolant/water ratio and that it will aid in faster warmups. This does make some sense, since coolants generally have lower heat transfer coefficients than water (lower Cv, Cp). I'm told that the factory fill is 50/50. Can anyone confirm?

Since our Tdi's are so efficient and generally run on the colder end, I'm going to try draining out a bit and replacing it with the correct amount of G12+ to bring it up to 65-70% or so. If you know something and want to warn me off feel free


Second idea: I'll try to do some vag-com runs with Brian this weekend using my winter fronts and without the winter fronts to see if there's any fuel economy increase.

 

[JettaJake]

Now the neat chem trick would be to figure out how and why acetone works and then develop a substitute that doesn't vaporize and can be put into the tank or even mixed at the point of fuel manufacture.

now that's a hard nut to crack!

essentially all liquid hydrocarbons are VOCs and the reason acetone works at all (assuming) has to do with its being volatile

VOC = volatile organic compound