With all the alcohol tables in a Hellcat tune there has to be a way to make it flex fuel. My "guess" is since it has factory widebands something could be done similar to the Coyotes. I believe the Ford guys use part of a F150 tune in the Mustangs to get it to work.
Has anyone messed with this yet?
So I have some interest in this as well. Though I have some concerns in the factory widebands. We recently did some testing with the factory widebands vs our dyno widebands. Based on the dash in the hellcats most seem to idle at around 13.8:1 AFR. Based on the stoich value in the tune this makes sense due to it being at (and I'm going off memory right now) approx. .072 FA. So we noticed at idle at the tailpipe our dyno was reading 14.7 - 14.9 on average. This held true for 4 different hellcats we tested. At first we assumed this was simply due to Cat converters, but to test our theory we installed a wideband just upstream of the CAT on 2 of these cars and retested. At wide open throttle the tailpipe readings would be approx. 12.4:1, the wideband readings on our installed wideband would match withing .1 AFR of the tailpipe readings, but the dash in the car would read 11.8 - 11.9. At idle our installed wideband would read very similar to the previously stated tailpipe readings (approx. 14.6 - 14.8). So our initial thought was that our dyno wideband was wrong. We installed two brand new sensors and had the same results. So to be sure our entire system was not an issue we installed a wideband gauge using the same upstream bungs we had installed previously, again the readings matched our tailpipe readings from our dyno wideband.
Ultimately we decided to not pay any attention to the factory readings on the dash, and go with our dyno readings. So in doing this we have found that we hit what appears to be a 90% injector duty cycle limit at approx. 6200 rpm when running a 15% reduction pulley on the supercharger with stock injectors. It seems as soon as we see 90% duty cycle it shuts the injectors down, then turns them back on until 90% is reached again.
Seeing as this has held consistent with other Hellcats we have had on the dyno, we feel as though the factory sensors may be widebands, but we believe the dash readings are actually just a calculated reading, or just basically going off of the requested fuel. If this is the case then I would question how the widebands are actually being used by the pcm to do fueling. On the other hand the pcm does use a WOT closed loop strategy for fueling, though in some of our testing we have found that even though the logger is showing a trim for fueling in WOT conditions, it does not seem to actually be changing the duty cycle of the injectors. Meaning if you cause the car to run rich, and the WOT trim value is say -10%, then change the tune to add another 10% of fuel, the total WOT trim now says approx. -20%, but the injectors and the AFR are not doing what the trim is commanding. Instead they are following the fueling curve we set in the tune.
Hope all of the makes sense.
You need to be using lambda not AFR. Your dyno wideband and the widebands in the hellcats are calibrated for a different stoich value. All widebands read in lambda and then use whatever stoich value it's set for to give you AFR. You're causing all kinds of unnecessary headaches by using AFR instead of lambda.Originally Posted by mxatunerjg
There are definitely provisions in the hellcat calibrations for flex fuel. Actually there has been flex fuel stuff in the dodge calibrations for years but no one has made it work. Unfortunately the dodge tuning is still way behind the ford stuff. There is a binary bit in the older dodge stuff to enable sensorless FFV but I have never had an opportunity to try it. I'll email Mark with the address and see if it's there on the hellcats. If so we may be able to make this work.
Yep it seems to have a wide band calibration table that I wish was unlocked as the dash readings are dead on when you set fa stoich to .0682. When I changed it it would idle at the dash readings and show 1.0 commanded and switch at .95 or so as the car would stop pulling fuel to get them back past .95 while logging. Some cars calibrate with the rear 02 so I don't know if this is the case or not.
I was going to post the same thing. Some people say that there is a 6% bias in the ANN which I believe is incorrect. I agree with you it's because wideband AFR is based on a 14.7 scale and the Hellkitty's stoich is 13.85
For information purposes for those who don't know how Dodge does their fueling.
Dodge uses a Fuel/Air ratio instead of a Air/Fuel ratio. To determine what the Air fuel ratio is take 1.0(lambda)and divide it by the PCM's FA stoich ratio.
Hellcat FA stoich is .0722
Divide 1.0(lambda)/.0722=13.85
Let's say you are commanding a WOT FA at .01551
.0722(stoich) +.01551(WOT FA multiplier)=.08771
1.0 /.8771(commanded WOT FA)= 11.40, the PCM will show this as commanded EQ of .82
13.85(Hellcat stoich) x .82(commanded EQ)= 11.36(roughly)
14.7(gas stoich) x .82(commanded EQ)= 12.05
11.36(Hellcat WOT FA) x 1.06(offset)= 12.04(AFR)
13.85(Hellcat stoich) x 1.06(offset)=14.7(AFR)
If you tune with AFR you have to take into account there is a 6% difference in the stoich value that the PCM is using and what the AFR wideband is using.
Last edited by Jay@HAP; 12-17-2016 at 10:04 AM.
Please let me know what you find out about this.
Will do.
It looks as though the switch is there. Mark is going to add it for me to test. I have an e85 car at the shop now so well begin going through the process to see if it works and what else is needed.
Curt and Jay, I apologize for sounding dumb in all of this. I had to think pretty hard as to what you two were saying about the air fuel ratio, lambda, and the way Dodge does fueling. First off I am aware of the way the fueling is done with the Hemi's. The problem is the Stoich setting. I was under the impression that was used as a fueling target, rather than an actual stoich value. This came from Diablo. I was talking with one of the people there at one point. We were discussing setting a richer fuel target for closed loop fueling. They said to change that stoich value and that would do it. Problem is it never did work, but due to that I was always under the impression that was what the stoich value represented.
After thinking about what you were saying, and thinking about my past experience of how changing that value affects the fueling, I now realize that value is simply giving the computer a fuel/air value for the stoich set point, aka what fuel/air ratio equals a measured value of 1 lambda. This explains perfectly everything about why the Hellcat is showing 14.7:1 AFR on our dyno or a gauge but 13.8 on the dash. As you two stated it is just simply the calculation for the AFR based off the stoich value setpoint.
Just wanted to say thank you for pointing it out. It is something I guess I halfway understood, but due to some mis-information, did not completely understand. Now that I know what that value actually is doing, it explains so many things I could not seem to explain about some of the fueling issues I've run into, and why my calculations did not always work out the way I thought they should.
As for the E85 testing, I think the big thing to really make it work properly is to have a flex fuel sensor. I'm not sure that it would work, but I think for consistency it would be best. Either way I'm very interested in what you find Curt.
the dash always seems to read 13.8 at idle on dash. I might need to scale the injectors as it was adding a lot of fuel at idle and it looked like the widebands switched at .95 when set it to .068. I think they meant it's for base calculations before trims are taken into account. It inturn offsets all the lambda targets which inturn makes me think. Setup a tune for gas get trims inline and just change stoichiometric to the fuel you use and trims should still be good. Personally I don't know why my idle was rich other then the trims having to add 10 plus percent at idle and the computer wanting to overshoot more because of the big offset. I'll try again when I get a chance to drive my car more.
For tuning with the NN on it seems leaving the stoich value stock works just fine. I agree in "theory" it should be changed but it doesn't seem necessary.
I mean Dodge didn't even use the correct stoich value for pump gas. Why didn't they? Who the hell knows, but it works. Dodge does some crazy stuff.
FA stoich for 100% is .0680. If you change the stoich, IMO you will have to change every FA table in the tune to compensate for the difference.
Not to mention that the factory air model and fueling is way off from the factory. What the ANN "fixes" is amazing. It's "hacked" from the factory. That's one of the reasons when you try to tune it "the right way with VE" things are so complicated. If ANN did not do so much virtual calculations IMO when you turn it off it wouldn't be so far out of whack. I think it's similar to the way GM does their virtual VE that is MAF based(even though I do not have any first hand tuning those)
The more tables, PIDS, switches, etc, takes more RAM and more processors which equals a more expensive PCM. Having a "virtual network" greatly reduces that. On the flip side of that look at the new Copperhead PCM's. The 15's have like 50+ spark tables? Not to mention all the VCT stuff and torque tables. Dodge PCM's don't have that.
Sorry I got side tracked. My point is Lambda.
I was slightly sad to hear that but the stock stuff does look rushed at least hellcat wise. If it was not for the widebands the fuel would kind of be lost. Scaling the injectors pulled all the trims in kind of nicely. I hope this sensor less stuff pans out ether way e85 is all over around here.
You can still scale the injectors on a car without factory widebands. Obviously you would need a wideband to do the WOT tuning. IMO getting the histrogram setup correctly is the toughest part.
You are correct about changing all FA tables. If stoich value is changed then all FA tables will need adjusted. For example...going from .0722 to .068 will reqire PE table to be increased by the difference of .0722 and .068 to keep PE fueling the same since PE is just added to stoich.For tuning with the NN on it seems leaving the stoich value stock works just fine. I agree in "theory" it should be changed but it doesn't seem necessary.
I mean Dodge didn't even use the correct stoich value for pump gas. Why didn't they? Who the hell knows, but it works. Dodge does some crazy stuff.
FA stoich for 100% is .0680. If you change the stoich, IMO you will have to change every FA table in the tune to compensate for the difference.
Not to mention that the factory air model and fueling is way off from the factory. What the ANN "fixes" is amazing. It's "hacked" from the factory. That's one of the reasons when you try to tune it "the right way with VE" things are so complicated. If ANN did not do so much virtual calculations IMO when you turn it off it wouldn't be so far out of whack. I think it's similar to the way GM does their virtual VE that is MAF based(even though I do not have any first hand tuning those)
The more tables, PIDS, switches, etc, takes more RAM and more processors which equals a more expensive PCM. Having a "virtual network" greatly reduces that. On the flip side of that look at the new Copperhead PCM's. The 15's have like 50+ spark tables? Not to mention all the VCT stuff and torque tables. Dodge PCM's don't have that.
Sorry I got side tracked. My point is Lambda.
Yes, Lambda does help to simplify, and while i have not tested it yet personally, if what Josh said is true then changing the stoich setting along with all PE tables would give an AFR dash reading for the fuel being used. Though using Lambda makes this a non issue, other than just to make the dash read a specific value.
One more thought, a stoich value that would give dash readings in E85 AFR values would require the max enrichment value to be adjusted significantly. If it were not the PE enrichment would be limited to the max enrichment value, and if my memory serves me correctly, this would be awfully lean for even an N/A engine on E85.
My only issue with the dash though is the idle or what i would consider the old cl or the none pe part the af gauge was still showing 13.8 but the rest of it was spot on in the pe part when I switched to .0680. It was weird and I pulled off at a gas station when I saw it not switching at 1 but around .95 at idle. Think of the dash like any other wideband the display is an interpretation of lambda blah I say straight gas the way the computer is is pretty much just like that with how you set stoich. So I leave my widebands in gas mainly because I dont want to have to serial to them to change it. The fa pe table you have to look at total added. http://www.wallaceracing.com/air-fuel-lambda.php
Lambda 1 is 14.7 gas and 9.77 e85 in the computer stoich is .068 gas and .102 e85 so then you get power enrichment to say 11.5 gas the totals would be .0870 and pe is the difference or .0190 gas and e85 total is .7.64 or .131 and pe would be .029. but the lambda in pe is still .78 on the widebands.
Last edited by JoshCP527; 12-19-2016 at 10:18 PM.
Yes, no matter what fuel is used measured Lambda will not change. The only thing changing is the Stoich value for the AFR or FAR of the fuel being used, which is simply just a multiplier. Lambda eliminates that and makes things simple.
Ok so enough of this talk. Let's get back to the main issue at hand, what do we need to do to make these ECU's do flex fuel? Curt, if there is anything I can do to help please let me know. I have access to a few Hellcats that we have been doing a lot of testing with.
I compiled a list of parameters to be added to try and make it work. Just sent Mark an email. We're going to give it a shot but I have a feeling it may take some time. There are going to be a bunch of tables to populate to make it work. It would probably help having a read from a dodge that already has flex fuel enabled.
Ok by defined is it the area that calculates the alcohol percentage out that's the main issue now that it can be enabled?
Reply With Quote