E85 Discussion - HPFP lubrication and Flow issues

[Realgib3]

Originally Posted by superskaterxes so just to make sure i have this straight, people arent seeing a change in WOT targets until greater then E40? i was running the same tank of E20 for the last week and i originally had my targets set to 12.2, then i bumped them to 12.7 and all of a sudden im seeing mid 13's on my log tapering to 12.7 near RL. not sure what happened but i just added some more E85 to make E30 total and im gona keep playing with it.

Yeah my target vs logged changed with 1.9 and 1.8g by ~.2-.3 so 12.8 became 13+. Not sure why people wouldn't see any difference unless they're stuck in CL or something.

[RichieRichness]

fuel pump arrives today...time to increase timing and boost.

e85 impressions so far. (first tank was e20 now I'm on e22)

-Cold starts are a little odd/rough
-I've been able to add 4* of timing in the mid range and 2* up top w/out KR
-Car pulls harder!
-I've seen about 15 awhp and 30 awtq increase according to VD
-MPG has increased according to DH but in reality, It's about the same...(20-25mpg)

Future Plans
-Install fuel pump this week
-increase timing 1*
-increase boost to 20 psi

Questions
-Should I lean out AFR a little? I'm currently at 11.8

Concerns
-Whiplash
-Money pit
-Boom?!

[mrmonk7663]

Lean out the AFRs 12.0 This will be my base target on my 3 gallons E85

[Ziggo]

Originally Posted by superskaterxes so just to make sure i have this straight, people arent seeing a change in WOT targets until greater then E40? i was running the same tank of E20 for the last week and i originally had my targets set to 12.2, then i bumped them to 12.7 and all of a sudden im seeing mid 13's on my log tapering to 12.7 near RL. not sure what happened but i just added some more E85 to make E30 total and im gona keep playing with it.

I don't know why, but I have not seen WOT changes for 3/9. I secretly thought Dano's MAFCAL might have been slightly off before, but I doubt that now. My MAFCAL was spot on up top on pump, and still spot on. I am targeting 12.0 and getting 12.0


Zigatapatalka

[cld12pk2go]

Originally Posted by Ziggo I don't know why, but I have not seen WOT changes for 3/9. I secretly thought Dano's MAFCAL might have been slightly off before, but I doubt that now. My MAFCAL was spot on up top on pump, and still spot on. I am targeting 12.0 and getting 12.0 Zigatapatalka

I had the same behavior WOT OL...but my CL LTFT shifted by ~8% on that much E. Have yours shifted?

Odd indeed.

[silvapain]

Originally Posted by RichieRichness fuel pump arrives today...time to increase timing and boost. e85 impressions so far. (first tank was e20 now I'm on e22) -Cold starts are a little odd/rough -I've been able to add 4* of timing in the mid range and 2* up top w/out KR -Car pulls harder! -I've seen about 15 awhp and 30 awtq increase according to VD -MPG has increased according to DH but in reality, It's about the same...(20-25mpg) Future Plans -Install fuel pump this week -increase timing 1* -increase boost to 20 psi Questions -Should I lean out AFR a little? I'm currently at 11.8 Concerns -Whiplash -Money pit -Boom?!

You can certainly go leaner than 11.8. I run 12.1-12.2 up until 5K+ then I taper to 11.8 on the K04 with pump gas.


Tapadatass

[superskaterxes]

i want to bring up the topic of CL fueling vs OL.

to my knowledge no one has ever INCREASED the time we stay in CL vs setting the ~1.2L cap then exiting into OL. i know the BMW's dont even have OL fueling at all so my question is, why not run CL for most if not all the time?

this would eliminate the need to ever change your tune (within reason) when switching between E85 blends. if you decided on E30 you wouldent have to go nuts at the pump wondering if you put E25 or E40 in because the trims would take care of any inconsistencies. total CL fueling has been something dustin and i have wanted to try forever and i might just take the leap.

does anyone have any negative effects they can think of? more wear on the O2 sensors? missing tables needed for complete CL control?

[Ziggo]

Originally Posted by cld12pk2go I had the same behavior WOT OL...but my CL LTFT shifted by ~8% on that much E. Have yours shifted? Odd indeed.

I have not left a map on there long enough yet for the LTFTs to settle other than idle. The idle breakpoint is 8%, and the others are floating around the same, but I never lost much sleep over my CL trims, they were not 0 before.

[Dano]

Originally Posted by superskaterxes i want to bring up the topic of CL fueling vs OL. to my knowledge no one has ever INCREASED the time we stay in CL vs setting the ~1.2L cap then exiting into OL. i know the BMW's dont even have OL fueling at all so my question is, why not run CL for most if not all the time? this would eliminate the need to ever change your tune (within reason) when switching between E85 blends. if you decided on E30 you wouldent have to go nuts at the pump wondering if you put E25 or E40 in because the trims would take care of any inconsistencies. total CL fueling has been something dustin and i have wanted to try forever and i might just take the leap. does anyone have any negative effects they can think of? more wear on the O2 sensors? missing tables needed for complete CL control?

I have threatened to do this very thing more than once but never did it... I think the idea is in my load cap thread and my boost tuning thread LOL.

it would be interesting but perhaps the ECU isn't fast enough to process all that data. IDK

4500 was the OE transition point so its worth a try.

doooo eeeeeet

[cld12pk2go]

Originally Posted by superskaterxes i want to bring up the topic of CL fueling vs OL. to my knowledge no one has ever INCREASED the time we stay in CL vs setting the ~1.2L cap then exiting into OL. i know the BMW's dont even have OL fueling at all so my question is, why not run CL for most if not all the time? this would eliminate the need to ever change your tune (within reason) when switching between E85 blends. if you decided on E30 you wouldent have to go nuts at the pump wondering if you put E25 or E40 in because the trims would take care of any inconsistencies. total CL fueling has been something dustin and i have wanted to try forever and i might just take the leap. does anyone have any negative effects they can think of? more wear on the O2 sensors? missing tables needed for complete CL control?

I did this about 8 months ago briefly. Basically forcing CL by taking the load transition points above my max load and the throttles to 100%.

I noted that the computer did some unusual things for me (like actual restricting timing to the Max timing tables, which I had never seen before). It was working for the most part though.

I chickened out after blowing my turbo discharge coupler off and having to limp home and haven't tried it again.

[Dano]

Originally Posted by Ziggo I have not left a map on there long enough yet for the LTFTs to settle other than idle. The idle breakpoint is 8%, and the others are floating around the same, but I never lost much sleep over my CL trims, they were not 0 before.

my idle LTFT settled in at between 11-15 but cruise ended up at around 5-7 and as you have observed, WOT AF did not shift at all.

[Lex]

Only issue is it would take some time for LTFTs to be learned for the new fuel - so you'd have to spend significant time in the WOT region before trims settle.

[silvapain]

Originally Posted by Lex Only issue is it would take some time for LTFTs to be learned for the new fuel - so you'd have to spend significant time in the WOT region before trims settle.

I see no problem with that


Tapadatass

[djuosnteisn]

Lots of people saying that there has been little to no shift in wot AFRs with mixes.

Is it possible there is indeed a sensor that measures the ethanol content and adjusts accordingly for OL? But not necessarily for CL cause the trims handle it there?


I have one guy i've been helping with a tune on an e85 mix. And on some logs, his AFRs are spot on.... others.... quite lean. There is no problem with the MAF, calc load, etc... just on the lean logs... the ecu is actually commanding less fuel (as observed by IPW).

I really really really want to chalk it up to a sensor issue, like his ecu knows there's e in some situations, and not in others.


Curious on people's thoughts.

[phate]

I really don't think there's a sensor for this. I've looked through the fuel system diagrams a number of times and haven't seen anything which would indicate this.

The issue of going lean sometimes and other times not is more than likely mechanical - pre-turbo leak. I've dealt with them on my car, and I'm dealing with it on another car I'm tuning right now. Leaks, unfortunately, can be very inconsistent.

As for the AFR shift/no-shift: Could it have something to do with the density of the two fuels? Meaning when you mix 1.00 gallon 93 and 1.00 gallon E85, you don't actually come up with 2.00 gallons? Sort of like taking a glass of sand, then filling it with some water....The two can occupy the same space until all of the space between the sand particles are filled, only then does it overflow.

[silvapain]

That might explain why there are duplicate fueling tables. One set for pump gas; another for E85.

I find it hard to believe any auto manufacturer would spend money putting in a sensor that wasn't needed though.


Tapadatass

[djuosnteisn]

Originally Posted by phate I really don't think there's a sensor for this. I've looked through the fuel system diagrams a number of times and haven't seen anything which would indicate this. The issue of going lean sometimes and other times not is more than likely mechanical - pre-turbo leak. I've dealt with them on my car, and I'm dealing with it on another car I'm tuning right now. Leaks, unfortunately, can be very inconsistent. As for the AFR shift/no-shift: Could it have something to do with the density of the two fuels? Meaning when you mix 1.00 gallon 93 and 1.00 gallon E85, you don't actually come up with 2.00 gallons? Sort of like taking a glass of sand, then filling it with some water....The two can occupy the same space until all of the space between the sand particles are filled, only then does it overflow.

We've beat the leak horse to death several times. And when comparing logs... there is no indication of a leak, e.g. the maf g/s is as expected between the good and bad logs (meaning the air is properly metered), calc load is correct (meaning load cap isn't an issue)......, but for some reason, the IPW is simply lower on the lean log, meaning the ecu actually intentionally sprayed less fuel.


I don't think the ethanol sensors are too terribly complicated, possibly just measuring conductivity of the fuel or who knows... but it's something i've been thinking about. Cause this guy's issues have me pretty puzzled right now. The ecu seems to be intentionally changing it's fueling strategy intermittently.

[superskaterxes]

Originally Posted by djuosnteisn Lots of people saying that there has been little to no shift in wot AFRs with mixes. Is it possible there is indeed a sensor that measures the ethanol content and adjusts accordingly for OL? But not necessarily for CL cause the trims handle it there? I have one guy i've been helping with a tune on an e85 mix. And on some logs, his AFRs are spot on.... others.... quite lean. There is no problem with the MAF, calc load, etc... just on the lean logs... the ecu is actually commanding less fuel (as observed by IPW). I really really really want to chalk it up to a sensor issue, like his ecu knows there's e in some situations, and not in others. Curious on people's thoughts.

im seeing this same exact thing on my logs. and im 100% positive i dont have a leak. the only thing that changed between the logs was some timing and VVT adjustments.

i dont think that theres a sensor but we do have a table for "low octane" ign timing. is that table maby based on a DAM like for the subies?


@Lex

why do trims need to settle at all ever? isent that the point of having them?

[djuosnteisn]

Originally Posted by silvapain That might explain why there are duplicate fueling tables. One set for pump gas; another for E85. I find it hard to believe any auto manufacturer would spend money putting in a sensor that wasn't needed though. Tapadatass

I agree, but just to play devils advocate... I'm sure it would be in the auto manufacturer's best interest to protect against a blown motor due to clueless Aunt Susie borrowing her son's speed 3 for the weekend and accidentally filling up with e40 or something crazy.



Plus... having no shift in AFRs until a certain mix ratio threshold sure sounds to me like a sensor is simply hitting max. Like an ethanol sensor that only reads up to like e20 or something... then suddenly adding more e above that actually start leaning out.


All speculation at this point.

[superskaterxes]

if you hook the DH up to a ford, there is a PID that says "alcohol content" or something like that. this isent available on the mazda section of the PID's though.

[Realgib3]

Originally Posted by djuosnteisn I agree, but just to play devils advocate... I'm sure it would be in the auto manufacturer's best interest to protect against a blown motor due to clueless Aunt Susie borrowing her son's speed 3 for the weekend and accidentally filling up with e40 or something crazy. Plus... having no shift in AFRs until a certain mix ratio threshold sure sounds to me like a sensor is simply hitting max. Like an ethanol sensor that only reads up to like e20 or something... then suddenly adding more e above that actually start leaning out. All speculation at this point.

So I looked at a couple more logs since the E started and now I'm intrigued... Seems to have learned something at least in my WOT region cuz afr is back down to my original gas commanded... hmmm but right after the fill up it was leaner... hmmm

Also, considering the fact the gas stations mix different % of ethanol in their gas would lead me to believe we may have something in the ecu to account for it, since even without mixing like we do, ethanol is something the cars are guaranteed to have to handle, to some degree, in normal daily driving.

[Ziggo]

Could be a sensor, but my guess is it has something to do with density as well.

Here is the logic:
AFR is mass based. Per my sheet a 3/9 mixture requires ~8% more fuel by mass.

In the extreme, the specific gravity of gasoline is .7, for E85 it's .81, an E30 mixture would be ~.73, an increase of 4%, basically cutting the required correction in half.

Where this breaks down is while E85 is pretty consistent at .81 gasoline can vary from .7 to .77 so the amount of density related correction you get varies wildly.


Zigatapatalka

[superskaterxes]

heres a nifty little article dustin sent me:

Auto Tech: Chevrolet Impala E85 Flex Fuel System - Autos.ca

seems that 06 Chevy Impala can CALCULATE the % ethanol just using the O2 sensors. this would be my best guess at why our ECU's are freaking out but ive put a call into cobb to see what they have to say.

[Nataphen]

I'm on the edge of my seat with all these new thoughts flying around.

[Pedal-Force]

I just wanted to thank you guys for your work. I love this platform and want to see it go far. I'm still only dipping my toes, but eventually, when I have money and a spare car, I'd love to see how far it can go.

Sent from my DROIDX using Tapatalk

[Realgib3]

Originally Posted by superskaterxes heres a nifty little article dustin sent me: Auto Tech: Chevrolet Impala E85 Flex Fuel System - Autos.ca seems that 06 Chevy Impala can CALCULATE the % ethanol just using the O2 sensors. this would be my best guess at why our ECU's are freaking out but ive put a call into cobb to see what they have to say.

^^^ I suggest everyone read that article. Here is a quote that really caught my eye...

GM has developed a Virtual Flex Fuel Sensor (V-FFS) software program that calculates the ethanol content in the fuel, instead of using a sensor to measure it. When the fuel level in the tank increases as the vehicle is refueled, the computer recalculates the percentage of ethanol in the fuel and automatically changes the air/fuel ratio. To do this, the computer temporarily stops the operation of other emission systems and monitors the oxygen sensors to determine ethanol content. The test is done several times until calculations remain stable. This can take several minutes when the engine is idling, but much less time at higher fuel flow rates.

Seems like exactly what happens when I fill up. I'll try to get logs today, but basically after I fill up, I seem to be leaner for a little while, then after some driving, it settles back down, even in OL situations.

[Ziggo]

Things that make you go hmm.

I have yet to make a "hot" map change on my car since switching to the blend. I always drive for at least 5-10 min to let the car warm up. Assuming the compensation were to get wiped out when re-flashing. If I were to reflash the map on a hot engine, and then immediately make a WOT pull, I should see it leaned out ~8%


Zigatapatalka

[phate]

I just checked the fuel system diagrams and fuel systems parts lists again. No sensors which are out of the ordinary. There are a couple sensors in the tank, but they are related to the evap system/leakage detection sensor.

[djuosnteisn]

The link Super posted proposes an o2 sensor based system... which is kinda like an invisible fuel trim.... So no direct application specific sensor would be necessary.

[phate]

Yeah, I'm aware. Just wanted to make sure we ruled out possibilities

[Ziggo]

We may or may not have it, but I hope the guy that came up that algorithm got a nice bonus, I can appreciate fancy engineering and no doubt there is a valuable patent on it.


Zigatapatalka

[phate]

Guys, that algorithm doesn't really mean a whole lot to us.....because it is essentially the same algorithm as closed loop tuning, then it probably just takes the average adjustment and compares it to the original gas curve....Sounds a lot like normal closed loops trims settling, just with an allowance to run any ethanol content and then assume the fuel has changed.

Interesting, I could see how the programming and validation would work for one of those systems. Could be interesting to see what happens when you introduce a vacuum leak into one of those systems, lol.

[Lex]

If there is a trim it should be captured by a PID. I'd think LTFT is a good place for it but it always goes to 0 at WOT. If it could compensate at WOT would it not compensate in other instances? Such as when the MAF cal was slightly off?

Perhaps there is a WOT compensation that is not called LTFT and we're actually not seeing the PID.

Other things to think about - how well is the fuel mixing in blends? Also what Ziggo is getting at - our fuel is compressed. Does that affect anything?

[Fatguy729]

Perhaps this is the reason (other than cat efficiency) for the 2nd DP O2 sensor that seemingly does nothing?

[phate]

@Ziggo @Lex

Agreed: 14.68/9.76 = 1.50

Which means anyone running E85 should be running 50% more fuel at the same airflow, but we know this isn't true. I scaled my MAF somewhere in the 1.35-1.40 range (going to check this tonight). Every report/article I have seen shows an increase of ~35% fuel, which leads me to believe density is playing a significant role, here.

[Realgib3]

Originally Posted by phate @Ziggo @Lex Agreed: 14.68/9.76 = 1.50 Which means anyone running E85 should be running 50% more fuel at the same airflow, but we know this isn't true. I scaled my MAF somewhere in the 1.35-1.40 range (going to check this tonight). Every report/article I have seen shows an increase of ~35% fuel, which leads me to believe density is playing a significant role, here.

The equations I used for the chart I posted showed 33.x% from gas to e85 so the ~35% is about right

[Ziggo]

It's not really the same thing as the LTFTs. It's actually very inventive. If you hold MAF and thus its current active trim constant, and wiggle the injector pulsewidth, you can measure the delta on the 02 sensor. It's alot more complicated than that of course, but it's possible to isolate the fuel as a source of 02, and thus compute the alcohol content (but what you are really doing is determining the actual AFR of the fuel, the source of the extra 02 doesn't really matter)


Zigatapatalka

[Lex]

Originally Posted by Ziggo It's not really the same thing as the LTFTs. It's actually very inventive. If you hold MAF and thus its current active trim constant, and wiggle the injector pulsewidth, you can measure the delta on the 02 sensor. It's alot more complicated than that of course, but it's possible to isolate the fuel as a source of 02, and thus compute the alcohol content (but what you are really doing is determining the actual AFR of the fuel, the source of the extra 02 doesn't really matter) Zigatapatalka

I wouldn't be surprised to see this implemented in many new cars - just not sure if something similar was done to the MS3.

[djuosnteisn]

I imagine the narrow band would be used for this as well... considering it's used to calibrate the wideband.


And if my guy's narrow band is acting up... perhaps it's affecting this measurement some how.




OR..... perhaps i'm just grasping at straws hahaha.

[event]

Originally Posted by superskaterxes so just to make sure i have this straight, people arent seeing a change in WOT targets until greater then E40? i was running the same tank of E20 for the last week and i originally had my targets set to 12.2, then i bumped them to 12.7 and all of a sudden im seeing mid 13's on my log tapering to 12.7 near RL. not sure what happened but i just added some more E85 to make E30 total and im gona keep playing with it.

Targeting 11.8 on E28 and my WOT AFRs are 12.0 ish