4 gal E85, 5.6 gal E10, 2.5 gal E10

I fully appreciate there are materials tests that long term use requires a different testing protocol.

My current interest is to find an E ratio that does not throw a P0171 code. Right now, I have a full tank of minimum E74 based upon my region and winter. Now that I'm at the maximum local ethanol ratio, I'm finding 4 of 5 trips throws a P0171 code. But I didn't see the first one until the temperature reached +50F.

Actually you provided two references and only one was a PDF:

• Paul Bedard's "Study: New E15 gas can ruin auto engines" - built around an American Petroleum Institute study, his lay interpretation left a lot of technical details out. Worse, his article about E15 does not answer my questions about E85 in a Prius. In fact, I started with E30 and E35, much greater ratios that appeared to work just fine in the 2010 and 2003 Prius.
• "Final Report CRC Project: CM-136-09-1B Intermediate-Level Ethanol Blends Engine Durability Study" - the table of contents only goes to E20 and I'm already in the E85 range. Once again, not very helpful for my current study.

Still, you might want to read this Bob Glicksman study:
http://www.liquidsunenergy.com/learning/ppt/ice.pdf

So far, everything in the "Ethanol and Internal Combusion Engines" by Bob Glicksman has been confirmed with my current studies. Feel free to read Bob's study. Meanwhile, my ethanol experiments continue.

Bob Wilson

 

Hi Mike,

High power settings in the 2003 are supposed to go into an open-loop, enriched mixture. However, my hill climb test data shows there remains some closed loop operation. Still, at high power, the engine is supposed to use an enriched mixture and the oxygenated ethanol will make it somewhat lean . . . yet alcohol has a cooler combustion temperature.

In theory, a hot, lean exhaust gas might lead to more exhaust valve erosion . . . literally eating the metal seat and exhaust valve. It remains on area I'm still thinking about how to get useful metrics. Regardless, this is something that will take more than the current test is likely to reveal.

In contrast, the 2010 uses cooled, exhaust gas, recirculation to reduce exhaust gas temperatures. I have not modeled it but I suspect this will mitigate the higher oxygen content of E85 better than trying to use fuel-enrichment of the 1.5L 2003 Prius. However, this is a hypothesis and not a formal model nor hard test results.


Bob Wilson

 

No problem, I read it tonight:

• GOOD NEWS
  • "2004 Scion xA, 1.5L I4, Tier 2 Bin 9, Direct Acting, Mechanical" pp. 23 - this is the same engine block as our 2003 NHW11 Prius and similar to the NHW20 1.5L engine. It has no relationship to the 1.5L 'Prius c' engine.
• BAD NEWS
  • "The detailed results are masked in order to ensure OEM confidentiality, including their technical data. The provided vehicle list is not in order within the table below to maintain this confidentiality." pp. 37 - preserved exactly the formatting in the original PDF. Did you read this? What this means is nothing in the report can be traced exactly to our 2003 Prius, 1.5L engine with the possible exception of the common test configuration description provided in the earlier pages.
  • "The exhaust back pressure level can be considered equal to the vehicle as the muffler is not contributing substantially to the overall exhaust back pressure level." pp. 32 - yet we know from the New Car Features of the 2010 Prius, the 1.8L Prius, that the muffler has a spring operated, back-pressure relief valve. This report has removed itself from the performance of the 2010 Prius.
  • "Figure D.2.1: Durability Cycle" pp. 26 - the problem is their test cycle limits rpm to 3,500 rpm which does not follow the operating line of the 1.5L Prius engine in either the NHW11 (maximum 4,500 rpm) nor the NHW20 (maximum 5,200 rpm.) So their "Simulated 1-2-3 WOT accels" fails to follow the operating line of our Prius engine. Although not mentioned, the NHW11/NHW20 Prius use an Atkinson cycle engine with an extended intake valve period. This is a critical omission in their test suite.
  • "The additive treat rate was specified at a level three times higher than legally required to for deposit control to reduce the possibility of deposits affecting the test results and to compensate for the dilutive effect of the ethanol." pp. 20 - HUH? Here is a documented change in the test fuel mix without an independent analysis of the effects of these additives. A+ for honesty, F for failure to do the obvious, testing the additive effects!

First I want to thank you for identification of this report. Sad to say, the report by its own words shows it has no relationship to my current experiments. The one possible link would be our 2003 Prius BUT they deliberately obscured their findings so the data from the nearest engine, the Scion xA 1.5L engine could not be found.

You are reminded that I posted the URL for another ethanol study. Please, feel free to add your comments to this thread citing specifics in Bob Glicksman's study.

Bob Wilson

 

Just to test the fuel pump. It has the greatest amount of contact with the ethanol.

I have a converter but not yet installed. Looking at the kit and installation instructions, I'm thinking this only makes sense for those who really want to run on ethanol. Even before the pending MPG vs mph measurements, the preliminary numbers I'm seeing show ethanol is priced too high to be an economical alternative to E10. One might as well burn premium for an equivalent cost per mile.

A couple of weeks ago, I was paying $2.95/gal for E85 while E10 was $3.15. Now E85 is $3.35/gal while E10 is $3.55/gal. The ethanol retail price is apparently 'locked' relative to gasoline and not relative to the energy or MPG equivalent nor the cost of production. In effect, the distributors have priced ethanol to support gasoline sales.

From a practical standpoint, the converter doesn't make a whole lot of sense at these ethanol prices and I don't see it as technically that interesting. However, it can answer the question of whether MAF or injector timing is used or most effective to calculate fuel burn.

My current question remains to find out how high we can go with the stock car. I understand there are reports of E50 and I have both cars at the maximum local ratio, at least E74 up to E85. So far, I've confirmed:

• cold-start problem - after a cold-soak to just below freezing.
• P0171, harmless - it sets the "Check Engine" light which inhibits seeing other codes

My thinking is find what ethanol level works with the stock, Prius. This becomes the maximum level for those who like me are curious. Then use this ratio for my standard mph vs MPG test. Only then does it make sense to install the converter and repeat the mph vs MPG test on that ratio.

I have used two methods to measure MPG: (1) MAF and the stometric ratio, and (2) rpm with injector timing. I need to quantify both before installing the converter. Then I can quantify the converter effects and go to the maximum ethanol ratio without codes.

Bob Wilson

 

Already done that experiment using a 1 gal container of E85 that was unused for four months. It worked just fine, no problems. But the 12V battery discharge is more likely to cause a Prius problem after a month of nonuse.

Bob Wilson

 

One of the common threads is the engine should be tuned for the fuel and ethanol has some unique characteristics including high octane. Our 13-to-1 expansion ratio Atkinson engine is pretty close to the type of 'high compression' engine needed to exploit ethanol which is one area my current experiments are investigating. Sure, the engine could be better tuned for ethanol but I'm currently interested in how to operate a stock Prius on a high ethanol ratio.

For my current tests, I not worried about the long-term, materials issues of E85. With the exception of some plastics, I am willing to take a short term, 2-3 month, risk. If the plastics are at risk, they are already 'toast' and the failure should happen any day now . . . to my cars.

As for using E85 in the future, it looks like it would need to be ~75% of the price of straight gas to make economic sense. So $4/gal straight gas would need to have $3/gal E85 to make economic sense. Yet even $3/gal is at least $1/gal higher than the reported, $2/gal. manufacturing cost. Running E85 is a luxury I can't afford after my testing is done.

The thing this paper provided was assurance that burning ethanol is not like trying to burn 'diesel' . . . nearly instant Prius death. The P0171 error is just the engine ECU seeing the long term trim has gone beyond the pre-defined limit. I'm not seeing them under high power operation so I'm going to take a different approach.

The Prius takes a 'snapshot' when a code is thrown with a collection of data points. I'll start collecting them before clearing the codes. This is safe because it only throws one P0171 per drive. In a week or so I'll have enough points to see what is going on.

Bob Wilson