Biodiesel NOx Emissions

A week ago I started burning Biodiesel in my 99.5 Golf. I have spent the week comparing my emissions running both Biodiesel and Petrol Diesel, and found some interesting results. I've posted the charts below
Biodiesel has the reputation of creating increased NOx emissions, which it certainly does, but not by much. It is a little high at a 60 mph and 75 mph cruise..but is significantly lower at WOT.
1300+ppm with diesel,
Under 1000 ppm with Biodiesel.
I also believe that Biodiesel is lower under any type of acceleration, but I can't superimpose my graphs from the scanner and gas bench. I have been trying to display a chart of engine load and gas numbers at the same time…but it hasn't worked out yet….they have different display rates.
I haven't tried any modifications on my car in order to address the high NOx but I will in another week or so. The tests shown here without EGR connected(there are enough variables without having to worry about EGR percentages too.) Also, please note the differences in O2 levels. The numbers on Petrol diesel look normal to me for a Diesel, but the numbers on Biodiesel are high. I thought that my O2 sensor had failed but it calibrated OK. I tried another sensor, and then another gas bench. They are correct as far as I can tell. The only reason I can give for the high O2 is that maybe there is a significant amount of Methanol left over from the procedure, making it an oxygenated fuel.
Both HC and CO numbers are very low as is expected with a Diesel. They are always at or near 0 once operating temp is reached. I don't have a measurement of soot levels(opacity), but there is quite a bit less soot with the Biodiesel. I have 250 miles on 1 gas bench filter, and still going.
I have additional graphs on WOT runs if anyone wants to see them.


This one is the Biodiesel run...


This one is the Petrol Diesel run...



[ May 17, 2001: Message edited by: Joe Rappa ]

Cool! I wish I had your equipment.

I've been reading everything I can find online regarding biodiesel for the past year. Your results don't surprise me. The studies I've seen have shown everything from a 16% increase in NOx to a slight decrease.

Biodiesel's oxygen content is probably to blame for the 02 readings. If I recall correctly, biodiesel is almost 10% oxygen. I've heard that switching to straight biodiesel has trigered check engine lights on some cars because of this.

[ May 17, 2001: Message edited by: Dante Driver ]

Joe,
Great to see that you are doing this. I am not anywhere near as knowledgable on this as you but just as interested in reducing our emissions. I wish we lived closer together so we could get together and run some tests on our car which has some modifications. I am considering some larger injuectors to increase milage but am concerned about the possibility of added emissions. What do you think about this? What is WOT? Just where do you live in NY? I might consider running your way for some tests if you are up for it.

Hey Joe,
Where are you getting your fuel?

Is oxygen (O2) considered a pollutant? I thought oxygen is the part of air we need in order to live. If so, does reporting O2 levels indirectly measure some other operating parameter?

Hi CarsWanted,
O2 is not a pollutant..but in gas analysis we use it to determine combustion efficiency, usually in gas engines. 5 Gas benches are not usually used for Diesel emissions because they get really dirty, really fast. Maintenence is a pain. Diesel emissions are traditionally measured with a particle tester.
O2 levels on a diesel are usually higher than gassers because they don't have a throttle plate and take in lots of air all the time. I can sample all 5 gasses with my bench, so I report them...I think the high O2 is a good thing with Biodiesel.

RC,
I live near Syracuse, NY(central NY). I'd love to test another TDI. Especially another Biodiesel car and a chipped one too. We should be able to get together some time. I am probably coming down to D.C. to visit my sister this summer. How far are you from there?
I make my own fuel.
WOT is Wide Open Throttle.
I think anything that increases mileage will reduce emissions...but I've been wrong before.

Dante Driver,
I've never seen any numbers as far as oxygen content on Biodiesel...but I'd believe 10% I suppose. 16% emissisions increase is just about what I averaged in all my cruise readings (I averaged 18.5%). One cool morning, it was 38 degrees F, I had the lowest NOx numbers I ever had at 60 mph, and I was running Biodiesel. Here's graph of it.
Joe Rappa



[ May 18, 2001: Message edited by: Joe Rappa ]

Joe,
When are you planning to visit your sister? We live two and a half hours east of DC near Ocean City, Maryland. I`d consider meeting you somewhere or be glad to have you as our guest here. My tuning box is adjustable also so we could fiddle with that to see some things.

Originally, I wanted to say 11%, but I went to www.google.com and searched "biodiesel" and "oxygen" and got the 10% number.

I've also seen different NOx numbers for different kinds of biodiesel. Unfortunately the NOx numbers for biodiesel from recycled vegetable oil--which is what I have--tend to be high.

Thanks for posting this info!

RC,
We haven't set a date yet, but it will probably be mid-summer. I've also been trying to talk my wife into a vacation in south Jersey...maybe Wildwood NJ.
I also sometomes do training classes down that way. If I head south at all this year I'll be in touch. Having a tuning box and a gas bench to play with sounds like fun to me.
Joe Rappa

Joe,
At this point June weekends are filled but July is wide open. My job alows me a good deal of flexiblity so a weekday is not out of the question either.

Wildwood is a few hour drive and a ferry boat ride north of here. Ever consider the Delaware shore? It`s much more layed back. We are likely to have another GTG in Berlin this fall, maybe you can attend.

I`d love for you to show me what it is you are doing and to begin to figure out more about biodiesel and emissions with these TDIs. Looking forward to getting together with you.

[ May 19, 2001: Message edited by: RC ]

The results are in!
I ran my TDI with the timing retarded 2 degrees in order to lower NOx with the Biodiesel. These results are from the same tank of fuel, and still no EGR in order to minimize fluctuation in test results. I don't notice any difference in power once the engine is warm. It starts and runs the same.
Here are 2 charts, one cold and the other 60 mph hot cruise, same as before.

Here is the 60 mph cruise, with Biodiesel and 2 degrees less pump timing.



Notice that the NOx emissions are lower now with Biodiesel than with Petrol Diesel. I found this a little surprising, so I did the same run 4 times. They were all basically the same. Peak NOx emissions are also lower now as can be seen by the peak at the left side of the chart. That is a full throttle run in third gear. Notice the 2 peaks…more on that in a minute.


I did a single cold run. I don't usually do them because it fills my sampling hose with water too quickly. The results are interesting.



Note the HC, CO and CO2 numbers. There are two reasons the HC and CO are higher than when it is hot. One is that when the engine is cold, there is incomplete combustion. The other is that the converter hadn't warmed up yet. Both probably played a role. Once again, notice the two peaks on the first NOx increase. Those humps appear to follow boost. When I accelerate full throttle my boost peaks around 18 psi, drops to 10 and then climbs to around 14psi where it stays. On occasion my injector #3 fails (the needle lift sensor). When it does, my boost will drop off to about 4 psi and climbs to about 10 psi where it stays. When this happens those peaks become more defined (that's how I first noticed them) and NOx emissions drop off as well(so does the power). I can't say conclusively that boost and NOx are related because the ECU might also be cutting fuel delivery during this failure, but it is something to look into.

The bottom line is; If you run Biodiesel in a TDI and want NOx emissions to be the same or lower than when running Petro Diesel, retard your pump timing a little.

Joe Rappa

[ June 01, 2001: Message edited by: Joe Rappa ]

Hey Joe,
You and I need to get together on this soon.

Hi RC,
There is a GTG scheduled in NYC area the weekend of June 23,34. I'll be down that way to see my father graduate from high school(it's part of that WWII vet program that allows vets that went overseas instead of finishing high school graduate.) He's 78 yrs old and happy as an 18 yr old about it.
In an earlier post you mentioned that June was a busy time for you, and that you would consider coming up here...whatever works is fine with me.
We're still planning a trip to DC, but haven't picked a date yet.
Joe Rappa

Joe,
June is too busy for us but July is wide open for now. We might go to Vermont in the beginning of August but there will be no time to stop by and see you. Looks like your trip to DC might work out best for me at this time, we`d be happy to have you as our guests if you have an extra day or two. This is a fine place to visit summer and fall. We are also looking to have another GTG at the end of September. I`m sure we can work something out, I`ll give you a call soon.

Hey Joe

This is some tidy work. It'd take 'da gubmint' a year to do what you've shown us here. How have you come upon ownership or access to such equipment? Come to think of it - what equipment is it, exactly?

John

Joe,

You mentioned that before retarding the timing 2 degrees you found NOx at WOT to be less with biodiesel than petro-diesel and more at part throttle. There is a pretty simple explanation to this that has its roots in the cetane of biodiesel and the resulting ignition delay period, pre-mixed combustion fraction and rate of heat release.

Due to Biodiesel's high cetane, the ignition delay is shorter than petro-diesel (neglecting differences in atomization quality). By extension, the pre-mixed combustion regime is lessened, and more of the fuel is being burned with a diffusion flame. If this were the end of the story, you should also experience more smoke, because soot generation is most prevalent in diffusion flames... but you don't. I believe the oxygen content of biodiesel has a lot to do with this.

As I mentioned, retarding the timing slightly IMO is beneficial when running biodiesel because it restores the optimal ignition delay when running high-cetane fuel. For the benefit of other readers, it want to point out that the ignition delay is very closely related to the combustion knock. There's no such thing as an absolute lack of combustion knock, and even trying to mitigate it by minimizing ignition delay as much as possible is not always the desired thing to do.

Here's the mechanism of how fuels effect the combustion process and hence affects noise and emissions, with significant oversimplication in some areas:

Low cetane fuel --> long ignition delay --> large pre-mixed combustion fraction --> higher rate of pressure rise and heat release --> greater combustion noise --> higher NOx emissions.

High cetane fuel --> shorter ignition delay --> smaller pre-mixed combustion fraction --> lower rate of pressure rise and heat release --> reduced combustion noise --> lower NOx emissions.

Note that the cetane fuel does not by itself actually determine how fast combustion occurs and how quickly or completely is comes to completion. That mechanism is far more complex and involves far more than simply the cetane value of the fuel.

Also, if you think about it, if you take the concept to the limit, reducing the rate of heat release, while good for NOx emissions and low combustion noise, will also reduce thermal efficiency. Remember that for a given compression ratio, to maximize efficiency, heat addition should occur instantaneously at constant volume. In the real world, of course, that's impossible, but that's the ideal that is striven for if thermal efficiency is the sole objective.

And to answer a question you asked me, Joe, yes, amoung automotive high-speed engines, the TDI is the most thermally efficient. It has a minimum BSFC of 197 g/kWh, which equates to a brake thermal efficiency if 43.2%, and is significantly better than any other in its class, even among state-of-the-art BMW, MB, Renault engines. As you may well know, medium-speed engines are hard-pressed to achieve mid-40% range, and low-speed marine engines achieve in the mid-to-upper 50%.

There has been a recent mania about advancing the injection timing in the TDI, in some cases outside the range recommended by VW. Up to a point, I agree that there are some improvements in power, fuel consumption and soot emissions to be achieved, but I am equally adamant that past a point there's no benefit and only potentialy more problems and much more NOx. This is not a panacea, people! My car was checked over on the VAG-COM recently and the injection timing was found to be slightly advanced but still well within "the-band." I left it alone. Don't mess with things if you don't know what you're doing.

RI_TDI,
I've got quite a bit of automotive diagnostic equipment. The gas bench I am using on my TDI is an Andros bench identical to the OTC Microbench, except in different packaging with custom software. It is piggybacked to a hand-held PC running a DOS kernal software package. The PC is also a scanner, signal generator, 4 trace recoding lab scope, and engine analyzer.
The timing equipment, injector test bench, and injector cleaning stuff is not my personal stuff. I teach automotive at a state university. The school owns that stuff.

TDIMEISTER,
I'm really getting an education from you. I appreciate it. You last explanation was an eye-opener for me. I thought that raising the cetane would delay ignition. Now it is beginning to make sense.
Could you please define "combustion fraction"? I'm not sure I understand that one.

I am now testing the emissions with the EGR system working. I should have some reportable numbers by the weekend.

Joe Rappa

This is really getting interesting, we may really figure out how to run our TDIs in the most ecologically friendly manner right here on Fred`s! Paying attention VWOA?

Hey guys,
I`m looking for a 11mm pump to help atomize the biodiesel better (Oldman got to the one I was eyeing first ). What is your take on that and the possible implications?

Joe,
We have an OTC 5 gas anaylizer and the operating manual says Not to test diesel emissions. I would like to but I don't want to foul the machine. We sent it to OTC to have it repaired a year ago and it cost $1500 for repairs and upgrades. What are you using for fuel stock?. Are you washing your fuel? I've been reading about making bio-diesel from used cooking oil but so far haven't made any.

Joe, with regards to the difference between "pre-mixed" combustion and "diffusion" flames, think of the difference between a bomb with a mixture of atomized fuel and air, and a flame-thrower.

At the very instant fuel is injected into the cylinder, NO combustion occurs. The fuel rapidly vapourizes under the heat of the surrounding air. Also, under the intense temperature and pressure, thermal cracking and dissassociation of the fuel molecules occur, predominantly in the very-rich mixture region near the center of the spray plume. These causes molecular bonds to be broken, first at the weakest unsaturated bonds, which liberate radicals and small saturated and unsaturated hydrocarbons. The saturated HCs can be in the form of straight-chained or branched parrafins or cycloalkanes. Unsaturated HCs may be aromatics or various alkenes and alkynes. The mechanism of their formation from the long chain fuel molecules is beyond my knowledge at the moment by maybe Skypup has expertise in this area. Aromatics tend to be the nucleus for soot formation.

The point is that the ignition delay period during which all this is happening (which is lengthened with a low cetane fuel) allows more time for this breakdown and mixing to occur before combustion begins in earnest. You can understand that when you give extra time for reactive species to form and mix with oxygen before combustion begins, it's going to do so with a bang. In SI engines, pre-mixed combustion predominates, but the reason it doesn't have the violence of a Diesel engine is that the temperatures at the end of the compression stroke is much lower; and in the former there is a point source of ignition which then propagates through the combustion space whereas in Diesel engines spontaneous combustion occurs in multiple locations.

Therefore it should be easy to understand how ignition delay is directly affects the "pre-mixed combustion fraction" as I called it, that is, the fraction of the combustion process that occurs in pre-mixed form as opposed to diffusion burning. By extension, it is the very rapid pre-mixed combustion is responsible for the high rate of pressure rise and heat release that gives rise to combustion knock.

What I've said above would seem to contradict the observed increased NOx and decreased soot emission when burning biodiesel. With it's high cetane, ignition delay would therefore be very short, thus there would be little opportunity for mixture development before combustion commences. I think the reason we don't see the expected phenomenon with biodiesel is due to the molecular structure of the fuel, and the paper you have seems to agree. It had been suggested that the displacement of sulphur and aromatics in the petroleum-diesel might be the main reason for significant soot reduction in biodiesel-blends and B100 applications, but it has been found that the magnitude of soot reduction is far greater than the amount of suphur and aromatics that were displaced. Remember that biodiesel is typically composed of high molecular weight saturated hydrocarbons with one or more esters in the chain, and can contain as much as 11% oxygen by weight. The oxygen content could also be the big determinant for NOx formation.

The other big factor has nothing directly to do with the fuel itself, but injection timing. As I said before there is an optimal timing for a given fuel. Almost all the experiments I've read that have reported increased NOx have been conducted with fixed injection timing, that is, timing is set for baseline ~45 cetane petro-diesel and unchanged for 50-60 biodiesel blends. A few papers, like the one you have, varied the injection timing and found that the rate of decrease in NOx using biodiesel was far greater than with petro-diesel, but startlingly without the corresponding deterioration in soot emissions (see Fig. 7 and 8).

WOW, great work going on here for sure!

To put the diesel combustion cycle in better perspective, download these two 500KB MPG and watch them operate in 3-D to see what Dave is talking about;

http://pics.tdiclub.com/filemaxx/%2E%2E%2Fmembers%2FSkyPup%2FHigh%20Speed%20Diesel%2EMPG


This one if 5.5MB, if you have high speed access it is excellent too:

http://pics.tdiclub.com/filemaxx/%2E%2E%2Fmembers%2FSkyPup%2FVW%2FTDI%2Eavi


Enjoy, that should help put the Direct Injection Diesel Combustion event in the proper perspective!

Hi Joe,
This is great work!

I have one question though.
The Oxygen percentage in air is 21%.
How does your analyzer show 27 and 28%?
I would be interested to know what you think about that.
Do you have any way you could gauge the combustion efficiency or perhaps peak pressure? There could be an explaination for NOx there.

Did you measure the exhaust gas temperatures?


Tdimiester,
Great work!
I would recommend it to everyone to go through what Tdimiester has written (even though lengthy and boring).


Thanks
diesel_desi

CoolWhiteDude,
I don't recommend you use your gas bench on a Diesel. It can be a real mess. I used to work as a factory rep EDGE, and did a bit of R&D stuff for them. These small benches have a tendency to get water in them, fouling the sensors and changing the calibration. I have cleaned many of them in the field, but you want to be careful. It is usually costly if you goof them up, as you know.
I use used cooking oil from a restaurant near my house. I have used both washed and unwashed fuel. I will probably run only washed fuel now that I have gotten better at it. I recommeng getting a pH meter and a good scale if you start making it.

Diesel_Desi and TDIMeister,
I was pretty surprised by the high oxygen readings on the Biodiesel. I am pretty sure(but not positive) it is because of the high oxygen content of the fuel, and the fact that there is no throttle plate on a diesel. The really high O2 numbers are from unwashed fuel, which has excess methanol in it, providing even more O2 yet. I have some new graphs with washed fuel, where the methanol is removed and the O2 is significantly lower. You should see them in a day or two. I've been running the new fuel with the EGR functioning.
I don't have EGTs, and I look at CO2 in relation to the other gasses as an indicator of combustion efficiency on a diesel, although it doesn't work as well as it does on a gasser.
I'd be interested to hear any other explanations for the high O2. In my first post I referred to it. Note the difference between the fuels.

Skypup,
Thanks for the video clip link. It kicks butt.

TDIMeister,
Very thorough explanation. I understand now. Thanks for the continued education.

Joe Rappa

PS: glad to see everyone finds this stuff interesting.

[ June 05, 2001: Message edited by: Joe Rappa ]

Hi Skypup,
Where do you guys find this stuff? I am surprised to find the cetane is so high. I don't know why, but I thought it was around 48 or so. I must have read it someplace. Anyway, thanks for the info. It is interesting to know about the carbon chain links, I suppose that is a major reason it behaves so differently than petrol diesel.
Joe Rappa

This is the final installment of my Biodiesel NOx testing on my car. This past week I tested it with 2 degrees ignition retard and with the EGR in place. I also undid my 1 turn of VNT twist. The car produced slighty less NOx than with regular petrol diesel. So, the bottom line here is if you run Biodiesel, and you're concerned about your emissions, you have no need to worry, with minimal tuning and negligible power sacrifice you can have your cake and eat it too.
Here's the latest and final chart. You can compare it to the first chart I posted with the petrol diesel numbers for comparison.

I want to thank everyone again for their help on this thread. This board rules! Where else can you do stuff like this?

Joe Rappa



[ June 07, 2001: Message edited by: Joe Rappa ]

BD-100 has a cetane rating over 55 I believe and that is about the highest cetane value which has an effect on the TDI ignition delay, cetane values above that have little to no increase in combustion.

Also remember too, biodiesel fuels are primarily straight chain alkanes with some unsaturated bonds but not alot. This paraffinic structure is ALOT more homogenous than petrol diesel fuel, meaning that most of the molecules are similar length in similar three dimensional conformations, compared to the 500 or so compounds that comprise middle distillate diesel fuels with an abundance of aromatics and double bonds.

What this leads up to is that Biodiesel also lights off more homogeneously, ie at around the same temp, which is higher than petrol diesel. The combustion of biodiesel is almost 100*F higher than diesel fuel, and without the aromatic content too. This, in combination with the higher cetane level, also contributes to the cleaner burn with less unoxidized molecules left over in the combustion chamber at the end of the downstroke.

<BLOCKQUOTE><font size="1" face="Verdana, Helvetica, sans-serif">quote:</font><HR>Originally posted by Joe Rappa:
Hi Skypup,
I am surprised to find the cetane is so high. I don't know why, but I thought it was around 48 or so. I must have read it someplace.
Joe Rappa<HR></BLOCKQUOTE>

The cetane rating for the B100 I get from World Energy is listed at 47.6. Could the cetane rating be variable in different company`s fuels and from different sources, i.e. virgin soybeans vs. recycled vegi oil? I add a bit of cetane with each fill up to try to get mine up to 50.

Joe Rappa, are your charts on your first and last postings taken under identical operational conditions? The waveshapes look so different; on the first petro chart, nothing even happens for about the first minute of display. I'm just asking for clarification here. It is really exciting that you have this equipment and are posting the results of your experiments for us to see.
[edited to correct the emoticon.]

[ September 04, 2001: Message edited by: cars wanted ]

Regarding cetane levels of various biodiesels, yes, the source of the oil stocks from which the fuel is derived yield widely varying properties such as viscosity, heating value, cetane number, and composition. I seem to remember that rapeseed methyl ester (canola is a close cousin of rapessed) yield high cetane but middle-of-the-pack heating value; soy methyl ester give slightly less cetane and less heating value still. Don't quote me on that. I'll find out that info more conclusively.

Hi Cars Wanted,
All the tests are done on the same stretch of road, at the same speed, but not necessarily the same startup. On some occasions I stop and turn around, and run the stretch over again. You can see the accel spikes on those runs. Sometimes I run a little longer trying to acheive the 60 mph, steady throttle condition.
road testing is not always the most scientific test method, so I tried to keep all things equal when doing it. Thanksfully the temps have been pretty constant in the mornings around here lately, so I haven't had problems with that variable.
So, to make a long story short. The cruise sections are as close to each other as humanly possible under road test conditions.
Joe Rappa