Megasquirt thoughts.

[shadow88]

I have a megasquirt controlled 1997 neon dohc turbo 5 speed as a winter car. My goal is better fuel milage, not more power. I was wondering what anyone would suggest for safe lean a/f ratio. I'm running forged 8.6:1 pistons and I have knock detection. I'm only after fuel milage. How lean can you go? 15.9:1? in the 16's?

Those lean a/f ratios are for part throttle, hyway speed cruise. Normally, it's in the mid 15's , but I'm trying to lean out the cruise mixture.

[turbovanman²]

Pics or it didn't happen,

Seriously though, I would say 15.5:1 would probably be the leanest you can go, I know when my van gets into the mid to high 15's and beyond, you can feel it lag so to speak. What are your current AF ratio's at idle, cruise and part throttle boost?

What MPG are you getting now?

[cordes]

In all of my gas mileage tuning I found much more mpgs in the part throttle timing in vac than I ever did in leaning it out. I do agree with Simon that around 16:1 you start to lose power and thus mpgs pretty rapidly.

[shadow88]

Pics or it didn't happen,

Seriously though, I would say 15.5:1 would probably be the leanest you can go, I know when my van gets into the mid to high 15's and beyond, you can feel it lag so to speak. What are your current AF ratio's at idle, cruise and part throttle boost?

What MPG are you getting now?

Pics later. If you really need to see something, then here. It's pretty old, but the same car, just re-purposed.http://www.turbo-mopar.com/forums/sh...ect&highlight=

Idle is difficult to control, but it's around the 14.7 range.
WOT is well into the 12's so no worry there. Along with the 87 octane required, very little ignition timing advance.
Part throttle, I've been as lean as 16.3:1, but it's difficult to keep steady.
Today was about 15.4:1 cruise at 115kph on the hyway

Cordes, you mentioned timing, and I'm running an indicated 44-50 degree advance at part throttle cruise. Running about 15 inches of vacuum. It varies with load, but it's pretty advanced.

Right now, I don;t have an accurate MPG reading because I just got this 2 weeks ago and it ran like poop. I've adjusted something almost every day. So far the range isn't as good as I had hoped, but I wanted to know how lean I could get without damaging anything.

What would fail, or what could I check for if I have been running too lean? I can look at plugs, knock sensor output, a/f ratio. It actualy runs ok in the high 16's a/f ratio.

[GLHNSLHT2]

My 2.5 turbo with a stock head and intake would get 34 around town and 40+ on the freeway running 16.3 a/f's at idle and cruise.

[turbovanman²]

Pics later. If you really need to see something, then here. It's pretty old, but the same car, just re-purposed.http://www.turbo-mopar.com/forums/sh...ect&highlight=

Idle is difficult to control, but it's around the 14.7 range.
WOT is well into the 12's so no worry there. Along with the 87 octane required, very little ignition timing advance.
Part throttle, I've been as lean as 16.3:1, but it's difficult to keep steady.
Today was about 15.4:1 cruise at 115kph on the hyway

Cordes, you mentioned timing, and I'm running an indicated 44-50 degree advance at part throttle cruise. Running about 15 inches of vacuum. It varies with load, but it's pretty advanced.

Right now, I don;t have an accurate MPG reading because I just got this 2 weeks ago and it ran like poop. I've adjusted something almost every day. So far the range isn't as good as I had hoped, but I wanted to know how lean I could get without damaging anything.

What would fail, or what could I check for if I have been running too lean? I can look at plugs, knock sensor output, a/f ratio. It actualy runs ok in the high 16's a/f ratio.

Trouble is, keeping it that lean without better control so to speak, then your risking detonation, damaging valves, melting pistons, popping head gaskets etc. Honda had a lean burn engine years ago but they used a wideband to control it and other gadgets. Honda also has the time and R@D to do it right so it lasts.

If you take out timing to run shitty gas and leaner mixtures, then you go backwards as timing does lead to MPG, hence why vacuum advance's on carb engines add 20-30 degs at cruise.

[shadow88]

The low timing advance is only under boost. As it passes the line from boost to vacuum, the timing increases quite a bit.

I remember Honda's 12 valve being a lean engine (if that's the one you mean) but how lean?

So far, I have seen zero detonation on the knock sensor's output using the drb3.

[turbovanman²]

The low timing advance is only under boost. As it passes the line from boost to vacuum, the timing increases quite a bit.

I remember Honda's 12 valve being a lean engine (if that's the one you mean) but how lean?

So far, I have seen zero detonation on the knock sensor's output using the drb3.

I think your thinking of the old 1500 CVCC engines back in the late 70's.

The Civic I am talking about is mid to late 90's, it ran around 17:1, I've only seen one and I had to fix it for our smog tests, it failed.

I wasn't saying your timing is bad, just saying that if you or whomever retard it to run leaner then you kinda go backwards.

[cordes]

I've run my omni pretty lean (16 or 17:1) for extended periods with no ill effects detected.

[glhs0426]

To help me see it correctly, I printed a copy of both the timing table and the fuel table since they did not use the same x and y steps. The cruise area was monitored between 45-75mph and highlighted on the paper copy. I then would raise the timing values in these areas until the engine would surge or knock then backed down 2 degrees. After the timing was set I would lean out those areas of the fuel table from 14.7 to XX.X until the engine would surge or misfire. On a junk 2.2 the timing was 50+ degrees, but I could only lean it out to 15.2-15.5 before it would surge or misfire with the timing set so high. There may be better methods to find the best combination of spark and fuel for mileage, but I was always taught: set to stoich, run up the timing as high as you can, then lean it out as much as you can.

One area I had great difficulty was below 45 and light throttle. The AFR table did not have enough rpm resolution in this area (I should have changed the x and y values to better match the timing and fuel tables). The AFR required for idle and the AFR required for a lean cruise were very different. I ended up smoothing the map stoich to lean at the 45mph and up area since this was where most of the cruising took place. What I am trying to point out is:

bias your tables for the cruise area; they don't have to be a smoothly stepped rpm scale
make the timing, fuel, and AFR maps mirror each other on the x and y axis as much as possible

Which version of MS are you running? I, II, III? MS&S, HiRes, Extra?

[Ondonti]

I have run 18 and 19:1 by tricking OEM computer (lying about narrowband signal). OEM computer does not like that and if you need to "scoot" then you will have no power available until you get enough TPS signal to switch into open loop fueling mode.

With megasquirt, you don't have that problem. When load is super low, you can go super lean. When your load increases, you can set your fueling to come back on and give you the power you need in that spot. Easy with Megasquirt I never spend any time perfecting it but I cruised above 15 on my MS car and would go to 20 in super low load situations if I could get away with it.

[shadow88]

It's megaquirt 2, pretty basic system. I'm taking Cordes advise and going for much more timing.
So far, 53 degrees advance if it's between 2400 and 3000 rpms and about 17-14 inches of vacuum.
I have only had the car for 2 and a half weeks so I don't have any MPG baseline so far. The first week was a nightmare to get it "close", then the following week is fine tuning the a/f ratio. Now I'm onto the timing advance. This tank looks like it should be in the mid to low 30's with mixed city/hyway.

There's a little something going on at these settings, not sure if it's a miss or what, but I'll keep at it and try not to be as lean as I was.

[glhs0426]

There's a little something going on at these settings, not sure if it's a miss or what, but I'll keep at it and try not to be as lean as I was.

A little like a surge and a little like a miss? If so, it's too lean. On my 2.2 I had add fuel to make the misfire go away. No amount of timing reduction seemed to help, so, I added the timing back and added a little fuel.

You could try adding a little dweel to the coil, but the default setting in the Neon manual seems to be the best compromise between life of the MS coil drivers and spark energy.

[black86glhs]

To offset the lean surge, you might be able to put on a high output coil. Might ignite the leaner mixture and take some of the misfire away. You could also run a bit more plug gap, but I can't remember if you want more gap or less.

[zin]

Remember the goal is optimizing ignition timing or, more specifically, where peak cylinder pressure occurs, not just a big number. More efficient burning requires less timing for optimum performance (HP or MPGs).

During EGR (charge dilution) and/or when Lean Mixtures are being burnt, it will be more difficult to ignite the mixture (and keep it lit) and require more time for the flame front to travel across the chamber. This is mostly because the fuel molecules are further apart, kind of like having to hop from one stepping stone to the next (lean/diluted mix), rather than just walking on a sidewalk (rich mix). Adding a small amount of gaseous fuel (propane/hydrogen) will help speed things up by helping to connect the dots so to speak. It can also help prevent misfires, a common problem with lean mixtures. Another way to help ensure ignition is multiple sparks, incidentally this is why MSD (Multiple Spark Discharge) Ignitions came to be, they (the engineers that started the company) needed an ignition that would reliably ignite lean burn mixtures.

Typically 17:1 yields best mileage, but leaner mixtures can/have been run, but the optimum mix will vary wildly with all the variables.

I think a turbo engine is a great foundation for a lean burn engine, they are generally stronger and so will deal with detonation better, which will be fairly likely, though not nearly as destructive as it would be happening at part throttle and with much less energy/fuel involved. They also have exhaust valves made to deal with high heat, another likely side-effect of a lean burn...

I really can't wait till I'm at the point where I can program my own cals and explore this area myself... I think there will be quite a bit that can be done once we unburden the calibration of all the emissions BS and possibly add the gaseous fuel/accelerator and possibly adding full-time water injection, although I'm think it may be a fine line on the water injection, it may increase the chance of misfire, especially if no gaseous fuel is added...

Sorry for the book, hope it's more helpful than clutter!

Mike

[shadow88]

Mike, I always like your answers. There's something good in every one.
This engine is egr delete. It's part of its past as a part time drag racer. The car's best was a soft launch 14.1@104, but that's no longer my intentions.

I'm Pretty sure the slight misfire was lean, in the 17's, corrected to the 16's and all is good. Still adding timing and removing fuel to try to get the best of my fuel energy. I should have a video and pics this weekend.

[shadow88]

http://www.youtube.com/watch?v=mvtbT-Kn5lg

A little drive up the hyway last weekend. So far it's looking like Cordes was absolutely right, ignition timing will play a bigger role in improving MPG's than leaning the mixture.

So far, at cruise of 70 mph I'm showing around 15.3:1 a/f ratio and 58 degrees ignition advance and it's looking like a good tank of gas so far. Thanks for your ideas and I'll update once I get a good hyway run this upcoming weekend.

I'm sure things would improve even more if I had a 3.55:1 diff transaxle, but I'm stuck with the 3.94 non r/t trans.

[glhs0426]

Success is sweet! Where are you going to post the msq. when you are comfortable with the tune?

I don't remember which tune is the good one for the 2.2 or I would have posted it for others. It's easy to get wrapped up in the moment and forget to document success. Failure seems to leave evidence.

IIRC the R/T and DOHC cars used 3.94 and the SOHC used 3.55.

[shadow88]

What's an msq? The actual timing chart and fuel map? Any time is good with me. I'll do minor corrections above 75mph due to a sudden rich condition and I'll probably add a little more timing to the boost area but I'll keep it slow with the 87 octane gas. I'm pretty bad with computers and I use the laptop only for the car, so if you want to see the mpas, I'll have to take a picture and post it?

I had a 1995 single cam neon and I miss those MPG's. I'm on the lookout for the transmission.

I don't mind making my failures known because it might help someone out who's trying the same thing.

[glhs0426]

The msq is the tune file. I think anytime you make a change to the tune the program automatically saves the changes as a new file unless you manually overwrite the current tune.

Start a new thread for Megasquirt msq's and use the basic idea behind the MS forums.