Page 3 of 4
Re: VemsNewb
Posted: Tue Mar 26, 2013 12:34 pm
by chaloux
Word. Sorry, I wasn't trying to come off like a dick. I've read that book so many times I really hope things stick when it comes to tuning my car
always easy to spout off online but actually making use of that info/knowledge is the key thing
This thread is actually a great thread for people to read who are, as the title says, vemsnewbs (myself included)
Re: VemsNewb
Posted: Tue Mar 26, 2013 12:36 pm
by EDIGREG
Hank wrote:Yeah, but decel is decel. That should be killing injector PW all together.
The testing on the FSAE car was NA testing from 35kpa to 100 kpa in 10 kpa increments at .80 lambda-1.15 lamda in .03 increments from 1500-12500 rpm. On the university Land and Sea chassis dyno, we could vary load and did such in 4 different loads simulations. We took a graduated cylinder and made a rig so we could accurately measure out fuel consumption over a 10 minute period at those variables. Ignition timing was also optimized before hand (Thus the 40 hours on the dyno figure) for each lambda and load setting. We also tested on 91 octane and e85 for all variables. While EGT's were obviously much higher at 1 lambda, that is where the most power was produced per BTU expended. In some instances at WOT, slightly rich of around .97 Lambda would make slightly more power with equivalent consumption, but that was only with gasoline.
I'll dig up the charts and data behind the testing. I have it on a thumb drive somewhere in my "school box" I tried to forget about after college
My Turbocharged experience with e85 is that it won't physically light off 380kpa worth of fuel at .75 lambda. Furthermore, I find that I make the most power at .83-.84 levels, but at the expense of high EGTs. .8-.82 is a level I can get clean pulls without rich misfires, and still keep EGT's safe and sound.
Not decel, but everything between full decel (0% throttle) and 100kpa
That is some great data. I won't be hitting 40psi but I'll definitely lean out my entire map when I re-tune after the new fuel system goes in.
Re: VemsNewb
Posted: Tue Mar 26, 2013 12:40 pm
by EDIGREG
Will, be very careful adjusting timing based on EGT. There will be a EGT valley when you hit MBT but it's easy to miss. I would not suggest messing with ignition timing without a proper knock detection setup.
If you think your knock sensor is bad, try listening on the other one. If neither is providing a signal, it's likely the knock listener, cause knock sensors don't go bad too often.
Re: VemsNewb
Posted: Tue Mar 26, 2013 12:47 pm
by loxxrider
I'll chime in here with my experience...
I've now had VEMS on my car for what... at least four years? I have gone to Carlisle and back (over 3k miles round trip each time) with it two times, and since I was broke and in College, I paid a lot of attention to fuel economy. I'd calculate the mileage every single time I filled up. I had it targeting 1.0 lambda on 93 octane for a while and that was netting me about 18 mpg. By the time I got to the car to stumble from being too lean and then backed it off just a hair, the car was getting no less than 26 mpg. This was just fuel tuning, no timing adjustments. I saw this on both trips, but the second one was less pronounced since I kept it fairly lean after the first trip.
Hank, I totally agree with your theory on paper as Ed said, but it doesn't really seem to work in real life. I always thought it would be the case that leaner would mean more throttle position and thus the same or worse efficiency, but that was never true in my testing. MAYBE it is because if you target 1.0, you will dip down below that quite often with small adjustments of the accel pedal (dTPS/dt). At a higher lambda, you will never dip below stoich of course, so that is probably a contributing factor. I have no doubt that your findings were well founded on the dyno, but I don't think the theory carries over to real highway driving.
In the end, I think it is best for anyone who is concerned about this to just go out and test it yourself. My lambda target looks much like Ed's does on the lower end (I may have had a hand in that map looking the way it does?). However, my maps for 93 also look a tad leaner than that until 280 kPa. Anyway, the point is, go test it! My experience has repeatedly shown that I get much better mileage with leaner-than-1.0 lambda while cruising on the highway. Not just in my own car either.
Re: VemsNewb
Posted: Tue Mar 26, 2013 1:26 pm
by Hank
Well, on our FSAE car I did the test/paper on, 35kpa was basically off idle and the least amount of throttle that would actually drive the wheels. Basically 5% TPS, with a Decel clamp at around 30kpa to cut all PW. That test I did was to test all loads and lambda levels to maximize economy at the ragged edge of safety.
When we did the final altitude tuning at EFIexpress in NH before the event, Marc and I settled around .94Lambda at WOT for a safety margin of EGT vs power vs economy. We were both rather surprised that the motor responded that well to power at a relatively lean condition on e85. During the acceleration and road course events that were not measuring fuel consumption, we map switched over to richer map that helped prolong plug life.
Ed, it really surprises me that you can light off .75 lambda mixtures. It will be interesting to see if you can do it in the mid 30s.
Chris, my testing negated wind resistance as an equation of load, but it was taken into account by applying more load to the dyno. This dyno session was as real life as you can get, not a paper thing! Since this test I have done testing on my car in "real life". I urge you to take your car to 75mph, set the cruise on your favorite stretch of highway(doens't have to be flat, can be "hilly") and log TPS relative to lambda at 1.0Lambda targets. The Cruise control will modulate the pedal for you and make a very smooth attempt at maintaining 75mph. I then challenge you to do the same at 1.15 lambda targets. The car will run just fine at 1.15 lambda, but your TPS will increase the same ratio'ed amount. Your manifold pressure will increase as well, from an arbitrary 50kpa to say 65kpa. This is another tell-tale that you are ingesting more air that needs to be matched with more fuel, regardless of how lean the final chemical reaction is. Injector PW will go up to match the extra airflow.
Search the geek for me on the other side of the fence with this debate. Embarrassingly enough, I advocated the lean out theory to my FSAE team and coach (a ICE professor). They shook their heads and told me to test it. I did a few basic tests on the dyno and realized my error. After I had done the test, the professor called me into his office and he had done the math(the "on paper") for the chemical reactions and molecular weights per BTU expended. It isn't unreasonable to expect that X amount of load requires Y amount of power and Y amount of power requires Z amount of energy(fuel). No magic here.
The only thing you can fudge is Thermal efficiency, and that does change with RPM. You can convert more chemical energy to mechanical energy if you are in certain RPM ranges of a motor(no really lower). Using the least amount of TPS or kpa is still the best way to monitor if you are in the best gear or not. I have TPS on call with my AiM MXL dash for commuting in the URQ.
What do I know though?
That is me winking through my block.
Re: VemsNewb
Posted: Tue Mar 26, 2013 1:35 pm
by EDIGREG
LOL epic photo
Hank wrote:Ed, it really surprises me that you can light off .75 lambda mixtures. It will be interesting to see if you can do it in the mid 30s.
I'm limited to about 25psi with my fuel pressure issue so I am not hitting those cells currently.
Re: VemsNewb
Posted: Tue Mar 26, 2013 1:46 pm
by loxxrider
Haha, awesome picture.
I do follow the math and the equations, and the theories, and your dyno testing too. The engineer in me wants to agree with you, but it also wants to question the results
There is something else coming into play here. When I was testing this, it was 80 mph the whole way as I don't like to push it much past that for ticketing purposes, and going slower is just too slow on a long trip. Believe me I get the whole, "you need a certain BTU to output a given amount of work" thing, but it just isn't working in practice for me.
I really think it is something more along the lines of not actually being able to hold 1.0 lambda and sometimes dipping below that while cruising (constant modulation of the throttle required, and bad dTPS/dt settings could really make for some bad dips in lambda) or a margin of error in the O2 sensor... something along those lines and likely a combination of things. This only gets amplified the more you lose control of the other variables. Obviously the physical principles hold up, its just that in the cases where people are getting more economy leaner than 1.0, something else is coming into play that I/we aren't seeing. It may be that at a gauge reading of 1.0 lambda, the average value for lambda I was seeing really was more like 0.95 (arbitrary). All I know is that I saw a marked increase in fuel economy when the car was reading much leaner values of lambda than 1. I don't have cruise control though, so that may have something to do with it.
It'd be fun if people would go test this and report their findings. I have my findings, but no proof of them since it was quite a while ago.
Re: VemsNewb
Posted: Tue Mar 26, 2013 1:52 pm
by chaloux
I will. Way to and back from Carlisle how about
Re: VemsNewb
Posted: Tue Mar 26, 2013 1:53 pm
by loxxrider
chaloux wrote:I will. Way to and back from Carlisle how about
That'd be great!
Re: VemsNewb
Posted: Tue Mar 26, 2013 2:02 pm
by DuBistS4
Me too.
Re: VemsNewb
Posted: Tue Mar 26, 2013 2:05 pm
by Hank
chaloux wrote:I will. Way to and back from Carlisle how about
Test 1.0 Lambda on the way "down", and 1.15 lambda on the way back "up"
I swear it is a Placebo effect Chris. When I am leaning out to 1.10 in the past for economy, it is usually on long road trips where I am bored. I feather the throttle, and plan ahead to make sure I can pass that semi on the downhill section, coast to a stop on decel, ect. Those were just great practices to getting great economy, not a bi-product of leaner mixtures.
I think the "hidden" variable you can't put your finger on is ignition tuning. If you are just leaning out a mixture while keeping ignition values the same, one of the two calibrations is incorrect. Most likely, your 1.0 lambda tune is conservative, and 1.10 lambda is getting on the ragged edge and thus returns a better economy. Being more agressive on timing would probably yield the same economy or better with 1.0 lambda
Monitor your EGTs too, especially those with warranty left on their headers
Re: VemsNewb
Posted: Tue Mar 26, 2013 2:33 pm
by S4Pirate
chaloux wrote:Word. Sorry, I wasn't trying to come off like a dick. I've read that book so many times I really hope things stick when it comes to tuning my car
always easy to spout off online but actually making use of that info/knowledge is the key thing
This thread is actually a great thread for people to read who are, as the title says, vemsnewbs (myself included)
Don't you know Will is the Stig's brother from another mother?
Re: VemsNewb
Posted: Tue Mar 26, 2013 2:52 pm
by loxxrider
Hank wrote:chaloux wrote:I will. Way to and back from Carlisle how about
Test 1.0 Lambda on the way "down", and 1.15 lambda on the way back "up"
I swear it is a Placebo effect Chris. When I am leaning out to 1.10 in the past for economy, it is usually on long road trips where I am bored. I feather the throttle, and plan ahead to make sure I can pass that semi on the downhill section, coast to a stop on decel, ect. Those were just great practices to getting great economy, not a bi-product of leaner mixtures.
I think the "hidden" variable you can't put your finger on is ignition tuning. If you are just leaning out a mixture while keeping ignition values the same, one of the two calibrations is incorrect. Most likely, your 1.0 lambda tune is conservative, and 1.10 lambda is getting on the ragged edge and thus returns a better economy. Being more agressive on timing would probably yield the same economy or better with 1.0 lambda
Monitor your EGTs too, especially those with warranty left on their headers
You may well have a point there which was something I had considered. I don't think the placebo effect came into play as I couldn't really drive conservatively if I tried (gotta make sure the power is all there, ya know?). I do think timing may be part of the answer though. That is the one thing that makes sense here and I'd bet that most will have relatively conservative timing tuning at low load.
Re: VemsNewb
Posted: Tue Mar 26, 2013 2:52 pm
by EDIGREG
Were you guys pumping some iron in the highschool music room?
Re: VemsNewb
Posted: Tue Mar 26, 2013 2:57 pm
by DuBistS4
chaloux wrote:Word. Sorry, I wasn't trying to come off like a dick. I've read that book so many times I really hope things stick when it comes to tuning my car
always easy to spout off online but actually making use of that info/knowledge is the key thing
This thread is actually a great thread for people to read who are, as the title says, vemsnewbs (myself included)
Lol no worries. Over the keyboard I sound rather uneducated on the subject.
Re: VemsNewb
Posted: Tue Mar 26, 2013 3:01 pm
by Hank
EDIGREG wrote:Were you guys pumping some iron in the highschool music room?
lol
Re: VemsNewb
Posted: Tue Mar 26, 2013 3:06 pm
by Hank
Flame front propagation varies a lot with different fuel densities within the combustion chamber. Ignition values can not be the same for 2 vastly different densities. If you were on the raged edge of peak timing on 91 octane, leaning it out to 1.15 would likely induce light pinging even at light load. The fact that it wasn't leads me to believe it wasn't optimized at cruise, and probably more optimized in a leaner situation. This could explain validated results on economy when leaning out.
Re: VemsNewb
Posted: Tue Mar 26, 2013 3:14 pm
by loxxrider
Exactly ^^^ That is how I'll justify it for now at least.
Re: VemsNewb
Posted: Tue Mar 26, 2013 3:18 pm
by Hank
GLad you can finally sleep at night.
Re: VemsNewb
Posted: Tue Mar 26, 2013 5:45 pm
by DuBistS4
What tupe of injector settings are you guys running? The simplified or traditional? I've been trying to get my VE close enough to where I dont need to retune it when I change Lambda targets and I think the fact that I am running the Traditional style strat is screwing it up. Thoughts?
Re: VemsNewb
Posted: Tue Mar 26, 2013 6:14 pm
by S4Pirate
EDIGREG wrote:Were you guys pumping some iron in the highschool music room?
Haha. That's our friends warehouse. Serves as storage/movie theatre/motorcycle shop/music jam out spot/gym/etc.
Re: VemsNewb
Posted: Tue Mar 26, 2013 8:31 pm
by DuBistS4
DuBistS4 wrote:What tupe of injector settings are you guys running? The simplified or traditional? I've been trying to get my VE close enough to where I dont need to retune it when I change Lambda targets and I think the fact that I am running the Traditional style strat is screwing it up. Thoughts?
Re: VemsNewb
Posted: Tue Mar 26, 2013 10:20 pm
by amd is the best
My best MPG in the 200 was 29.8 on the way to Carlisle at an average speed of 70mph through the hilly terrain of NY and PA. Target for cruise is 1.02-1.03. On the way home from H2O, four people and a full trunk I got 27mpg. I am astounded at how well the 200 does on fuel. That was on 93oct pump fuel. E85 my best was 19mpg up to Burlington, VT to visit Casey (so there was a bit of spirited driving involved while up there).
Re: VemsNewb
Posted: Tue Mar 26, 2013 10:43 pm
by DuBistS4
What type of injector settings are you running Nick? Traditional or Simplified?
Re: VemsNewb
Posted: Tue Mar 26, 2013 11:44 pm
by amd is the best
DuBistS4 wrote:What type of injector settings are you running Nick? Traditional or Simplified?
- injector.JPG (53.68 KiB) Viewed 33597 times
This is how it came from Marc. I have made no changes to them and have not had any issue.