Ethanol percentage flex fuel

[wsteele]

For me, without actually testing the ethanol content of the fuel I purchased, I am just happy to see anything under 10% ethanol content since that is what I am purchasing.

It would definitely be more accurate if I bought one of those ethanol test kits and actually checked the fuel but I don't really care that much as long as it continues to calculate at something under 10%.

I will continue to post my calculated ethanol percentage over the next several fill ups so we can see if this was the actual fix.

Having used my test bottle on a number of samples so far, I have come to a few conclusions.

First, the gas stations are pretty reliable as far as not having more than 10% Ethanol in E10, in fact, what I have tested is they are generally pretty close to 5-6%, definitely not something to think about being more than 10% as a general rule.

Second, the reason to buy a test bottle, other than for testing when something is wrong with your flex fuel vehicle and you have ruled out a lot of more obvious candidates, is to check how much gasoline is in your e85.

The variability of ethanol content in e85 seems far greater than the variability of ethanol in e10 (by geometric proportions), so a test bottle might actually tell you something meaningful testing e85.

One thing I plan next is to try and quit focusing on how far off my virtual sensor is measuring e10 and start seeing how off it is measuring e85. I will be filling up with e85 for a few tanks and then also doing the bottle test. I am guessing, given my primitive fuel pressure management setup in my 2007, that the algorithms are optimized to guess more accurately at ethanol content of e85 versus e10. My theory will be confirmed if it gets a lot closer in percentage terms versus measuring e10.

 

[wsteele]

So I haven't yet started my testing of the virtual sensor in my truck, versus E85. It is a little more challenging to get E85 where I am, so haven't done the fill and see what the sensor says I have versus the bottle test yet.

The last fill of E10 I did a few days ago settled at 28.6% Ethanol in gas that tested at 6% with the bottle, so officially, my virtual sensor is out of spec according to the shop manual.

I know I will end up having to change my fuel pump in my last (likely in vain) attempt to get my virtual sensor reading correctly for E10. I did just reset the Ethanol content in the computer with my Tech 2, so it isn't killing me to have the sensor be wrong.

I did figure out how I can run my analog pressure gauge to a spot on the windscreen to see what the actual rail pressures are under heavy load. I am still a little wary of doing this (so keep putting it off) because if one of the joints in the hose starts leaking while driving, I run the risk of burning the old gal down.

In addition, the only thing holding the hood in place will be the safety catch (it will be unlatched from the normal hood latch to create enough space at the rear of the hood to snake the hose through). I will try and record the actual test when I do it, so if I do burn it down or loose the hood at speed and crash the thing before getting it stopped, maybe I can enter a Darwin Award contest and recoup some of my losses... .

What we are willing to do for the sake of knowledge... .

 

[wsteele]

OK, so I got my first pretty full tank of E85 at the local dealer in a very long time.

I was hoping that the error percentage would be less on the high side (close to 100% E85), thinking maybe they set up the virtual sensor such that if it was going to err, it would be on the E10 and below calculations.

The tank before filling was down to about 2 gals of 6% E10, so a very small percentage of Ethanol.

After filling up with E85 and driving for about 25 miles, the virtual sensor settled at 56% Ethanol. I took a sample of the E85 as I pumped it into my tank and when I got home did the water in bottle test and it showed 80% Ethanol, so a pretty good percentage.

I am disappointed that my virtual sensor is so far off on the high end, maybe more than it is when calculating E10 percentages. I will wait until tomorrow when I do a cold start to see if the calculation jumps up and more (I have noticed that happening when running E10). Maybe it will get a little closer.

I do plan to do some Tech 2 runs and see what trims and knock activity seems like with the E85, I am guessing the trims will look pretty rich and the knock sensors may be asleep.

 

[wsteele]

OK, it settled at 59% content on the virtual sensor. Better than the 56% it settled at yesterday after the fill up, but still way off the 80% I have in the tank. I have to say, it started right up and idled smoothly first start this morning, the trims looked good (it was already in closed loop by the time I got the scanner reporting). Obviously, I didn't have the overly rich smelly cold start I get when it is way off the other way when running E10.

Looks like all I have left is to hook up that analog fuel pressure gauge, snake it through the rear of the hood opening, tape it to the windscreen and see what the pump is doing under heavy load. If it stays at 50 psi the whole time, I will be stumped.