• Hello Guest, welcome to the initial stages of our new platform!
    You can find some additional information about where we are in the process of migrating the board and setting up our new software here

    Thank you for being a part of our community!

Megasquirt flex fuel tuning E85 with ethanol sensor.

qwkswede

Old and Slow
300+ Club
Joined
Mar 19, 2003
Location
Denver, CO
I am trying to work through a decent flex fuel tune on my daily driver 240. I have done plenty of E85 tunes on red blocks over the years. But on my daily driver car I wanted to experiment with true flex fuel automatic tuning. I bought the sensor and plumbed it into the return fuel line near the fuel rail and wired it up. I get realistic readings from the %ethanol gauge in Tunerstudio. Everything seems to be working OK. I was getting 12% reading on my standard pump gas 91 octane. I have heard and read the signs at gas stations in Colorado that standard fuel blends contain about 10% ethanol. So that 12% reading seems about right. I added a couple gallons of E85 to my 3/4 full tank of gasoline and the reading climbed to 21% ethanol. Good enough. The sensor reads in a somewhat useful way matching the fuel I added. I think I can tune with that.

I open up Tunerstudio and find the screen for the flex fuel sensor and activate it. And I have to say that I'm somewhat surprised with the basic settings available. I was hoping I could have a totally different set of maps for fuel, ignition, and boost for e85 and gasoline. And some sort of blending between the pairs depending on the ethanol percentage. But below is what I see. I guess it could do the trick buuuuut.

Is a simple linear multiplier for the fuel map and spark map really good enough? It seems overly simplified. My motor is a high compression NA motor with a turbo. The gasoline tune is really tricky and conservative in the timing to make it survive with 12psi boost. It's a funky set of maps especially in the positive pressure parts of the maps. High octane E85 should allow me to set up a more optimum spark map. But all I get is a "distributor twistin" adder of X degrees? Really?

Am I missing something here? Am I the first person to want the flexibility to drive the car around in the wild and fill up anywhere I can?

attachment.php
 

Attachments

  • flex_fuel_ms2_screen.JPG
    flex_fuel_ms2_screen.JPG
    124 KB · Views: 249
It's been a little bit since I've messed with it, but that's not what it looks like in my MS3X setup. I had separate fuel/spark/boost maps, and separate things like cranking pulsewidth, etc. And it interpolates between the two sets based on the sensor reading. And you could even set up an interpolation curve instead of a flat interpolation because a little E85 goes a long way.
 
Thanks John. What you described is what I thought i was going to be doing. This is an MS2Extra setup running on microsquirt hardware. Let me hunt around a bit. Maybe there is a better version of code to run. I can't easily go to MS3. Maybe this is the first real reason I have come across to upgrade though.
 
I know My old MS1 setup wouldn't have done anything at all like that, or even what you had. All I could do with it was (mis)use the Nitrous switching functionality to swap spark and fuel tables from gas to E85 with a dash switch. You just had to run it really low on one fuel before refueling on the other and flipping the switch. And cold starts were a bit iffy because it didn't change any of those settings.

On the MS3X system the only trick I ever needed was to turn all the flex fuel stuff off when doing tuning/autotuning. You obviously can't tune two tables at once, so I'd just do a similar thing to the old (run it super low on one fuel, switch to the other) and then do some tuning. And even then, only really do E85 tuning if the tank read as close to 70% or more Ethanol. I would occasionally get some very weak E85 fuel out of pumps. It's 'up to' 85%, and at times (especially in the winter, presumably for cold starts) they drop the percentages way down. I've never seen it, but I've heard that occasionally they'll just go ahead and put normal E10 gas in the E85 tanks.

That all helps explain why I was constantly chasing my tail tuning the E85 on the old MS1 setup, the fuel I was tuning on was probably constantly changing tank to tank, without a sensor I never knew.

EDIT: Googled up a pic of what the flex fuel setup on an MS3 looks like in Tunerstudio:
169645-my-quest-running-flex-fuel-ms3-ms3x-80-flexblend_163948e8b2a154a4fbf34b272725ffdc225842b1.png
 
Last edited:
I should probably do that for the 940, I'm using the basic flex setup and it's close enough for me not to fool with it. I think I have it add 2 degrees of timing for full e85 or something like that. the basic stuff can work, and if you're running microsquirt you may be limited to just that. Don't go crazy with the timing on e85... just cause it's not knockin' doesn't mean it's not bendin'
 
Wow,
That ethanol % changing under load is a bit alarming. I'll defintely datalog the sensor values and keep an eye on them. It seems stable just driving around town.


Make sure you check you log first. There is no smoothing feature on the flex fuel sensing on current MS2 extra firmware, which makes flex fuel auto tune useless and can blow up your engine.

http://www.msextra.com/forums/viewtopic.php?f=101&t=67564
 
What you posted is all I could find for flex blending when I was using microsquirt. It's really basic and not that useful. Once you go to ms3, you can do different fuel, spark and boost tables and blend between them based on flex %. It's much more advanced and more useful.
 
Yea on my ms3x I just ran power and ground to the sensor and ran the signal wire and turned flex on. Works really well so far and I haven't messed with it at all
 
Well,despite the simple interface. The fueling seems to be somewhat usable. I added another couple gallons ethanol and brought the mixture to 32% today. I have the multiplier set up to keep the AFRs where I want at WOT. Mid 11s on the wideband. It is doing OK at idle and better at cruise in closed loop. Idle seems to be richening a little too much because I don't run closed loop there. But it is workable so far. I'm getting enough octane boost from even 30% ethanol to really help under boost. I'm still running very conservative timing. I'll see what I can do with it. Running some mixtures in the 20-50% range would be good in the summer when I have trouble detonating.

I'll keep upping the ethanol and tuning the flex table. I haven't touched the base maps. I want to have those set up the way they were and be able to get back there with just a straight 91 octane fill up.

Not ideal, but really simple to use. It does work reasonably well with the default settings that were there in the stock tune file.
 
Last edited:
that's the problem with flex and uS... and to a certain extent ms3, but it's better there... you really need table blending to take care of some of the possible transient issues (the fallback is to make things richer than really needed), along with accel blending and warmup/cranking blending to do things really well.
 
Yep, now I know. I'm surprised there isn't more pressure on the MS development crew to add some more sophisticated tuning for flex fuel. Straight E85 tunes are good for weekend fun cars. But I need the easy flex fuel switcharoo.

It will be intersting to see if I'm going to hose up my emissions enough to fail the roadside emissions test I drive past every morning. :( I don't need another visit to the State Inspection Center.

that's the problem with flex and uS... and to a certain extent ms3, but it's better there... you really need table blending to take care of some of the possible transient issues (the fallback is to make things richer than really needed), along with accel blending and warmup/cranking blending to do things really well.
 
Yeah, after your last comment I did take a look at the fuel % log. Under all kinds of loads, it looks stable. It fluctuates +/- .1 or .2 at the most.

Have you check you logs and see if your flex fuel sensing is not jumping all over the place?
 
Yep, that is the conversation we were having the last couple posts. It isn't exactly borked. And it is working just fine for me. I'm not seeing any unexpected sudden changes in the fuel % reading that would require a lagging time based average. It makes sense though. The only time that % should be changing is when you add fresh fuel at the pump.

FYI - - Flex fueling on a MS2 is borked

http://www.msextra.com/forums/viewtopic.php?f=101&t=67564

James needs feedback on the 1.5.1 up MS3 firmware where a test fix was implemented.
If he gets good feedback he says he may be able to backport it to the MS2 firmware.
 
MS2 or MS3?

I posted in hopes to stir up some feedback for James from guys running newer code on MS3 setups.

MS2 guys are complaining that the % changes for no reason and under load, YIKES!
as you said the only time the % should change is when different fuel is added to the tank.
 
Back
Top