Leanest AFR I can get away with?
 

I know you can't tell me exactly, but I was wondering what the leanest people have gotten away with. I searched the forum, but didn't get enough info.
I tuned my engine management system to run lean at cruising scenerio; low manifold pressure at 60-65mph. I use high octane because I have 10.5:1 pistons. At this speed, I was about 17.5:1. That scares me. The car ran fine, and the EMS has a knock detector/ retard. It didn't get hot, according to the coolant temp sensor. However, I don't know if I could be melting the pistons, and not knowing it. When I give it gas, the AFR jumps to 13:1, depending on load.
If it helps to know; it is a short stroke 2.5L V6. There's no way I could hear knock if it happened, because the exhaust is loud.
Any ideas, AFR #s that have been high, but okay?

Well, the civic VX runs 20:1 at times, so I doubt 17.5:1 is that bad, especially if you're not experiencing knock.

From everything I've been told the main times to worry about running really lean it during cold starts and under high load.

There are a few lean-burn Hondas that get away with 25:1 A/F ratios. I've looked into lean burn, but the auto mfrs don't do it because it produces unacceptable NOX pollution. That's why you could get a lean-burn '80s del Sol or Civic, but you can't get one today.

The other thing is the possibility of engine damage. I went through 6 pistons on a single cylinder motorcycle. It would routinely burn a hole in the piston every 2000 miles. I finally thought back to what I'd done 12000 miles earlier. I'd swapped the low mount fender for a high mount fender. The change in cooling air flow was enough to cause the engine to run hot enough to hole the pistons.

If I were you, I wouldn't risk engine damage to save a few pennies with lean burn. Especially knowing you're increasing air pollution while doing it.

Lean burn started in Hondas in the mid 70s with the CVCC engines. Compound Vortex Controlled Combustion (CVCC) used a prechamber and a very small intake valve with its own individual passage in the carburetor.

The mixture in the prechamber was around 12 to 1 while the main mixture was around 18 to 1. Those mid 70s Hondas passed the then current emissions standards without catalytic converters. This was also true of the first fuel injected Nissan Z cars, which came without converters in Federal versions.

As emission regulations became more strict, the difficulty in utilizing lean burn technology became much more difficult.

The Civic VX 92-95 utilized several different strategies to accomplish lean burn. One intake valve barely opened below 2500 RPM, which helped create enough turbulence to better homogenize the mixture. Additional egr helped as well as the 5 wire linear O2 sensor. Combustion chamber design was another factor in promoting the turbulence necessary to get the fuel distributed more evenly in the incoming air charge.

Sadly the more stringent emission regulations made lean burn virtually impossible in later years. I am not anti emission regulations and lean burn should have been supported by looking at the sum of the total emissions instead of the slightly higher NOX emissions by themselves. Lean burn created substantially lower HC and CO emissions, and it would have been (just my opinion) a better long term strategy to allow the manufacturers to continue development of emission strategies that could have possibly addressed the NOX issues.

Better control of the EGR volume would have been one way to satisfy the emission regs while retaining lean burn technology.

The problem as I see it (just my opinion again) is you had the govt determining absolutes in regulations instead of a weighed overall strategy, that encouraged better technology to continue refinement.

California also enacted a 0 emission requirement that was supposed to promote electric vehicle development, but they backed down later.

regards
Gary

Watch for an EGT spike. That's all you really need to worry about. A spike in EGT means that the air/fuel mixture has leaned out to a point that it can't burn completely before the exhaust valve opens.

Usually feels like the engine is missing or stumbling and is where you end up with burnt exhaust valves. A leaner mixture should be less prone to knock as well, not more. Stoich is most prone and anything else becomes less.

Thanks! That's alot of the good info I was hoping for.
I didn't run it with the laptop hooked up, but will get someone to ride with me. I'll look into an Exhaust Temp sensor. I watched my dash Temp sensor, and it didin't move. I know it's not the best way, but if it had moved at all, I would have shut the car off, and reprogrammed the AFR. It did seem to run less than smoothly after staying at 17.5:1 for a while, but I get VERY particular when paying attention to stuff like this (I used to tune V8s by sound, smell, etc.. in the days before ECUs).
20-25:1! That's good to know! I'm not going to try it though!
My Coolant Temp Sensor is very particular in MegaSquirt, so I'll have someone watch it.
The car runs stoich at idle, and slightly leaner while normal driving, except when accelerating, when it goes to 13.5:1.

Oh, and I forgot to mention that I have a Blaster2 coil that's about 5,000v stronger than the stock one, MSD ultra low resistance wires shortened about 6 feet (total) from stock, and indexed plugs, so it should be able to burn the mixture better. Plus the chambers have been polished, so there shouldn't be hot spots.

Thanks again!

Remember when it comes to ignition systems that it will only run the voltage that the plug needs. If you have a 20KV ignition coil and your plug only needs at most 15KV to fire you will see zero improvement going to a 45KV coil.

To take advantage of the higher voltage coil you need to widen your plug gap. And even then the voltage required to fire the plug varies depending on load and how advanced your timing is. More advance means less voltage, less gap mean less voltage, less throttle means less voltage.

An article on lean burn. You can also read my experience with lean AFR on my megasquirted Mercedes near the bottom of the first post.
https://ecomodder.com/forum/showthrea...ura-11717.html
It wasn't encouraging.

Thanks! I did increase the gap a little. I got the coil mostly because the ones for this engine go bad. I also have a MSD 6A CD box, but am still figuring out how to wire it.
Thanks for the link, tjts1. If it's not encouraging, maybe that means I can be OK with not get obsessive about this :)

I think there is some FE to be gained here but not without spark control. At this point I'm using MS1 extra HR but ignition timing is being done by the stock mercedes EZL computer. The plan next summer is to add a 32 tooth wheel and ford EDIS. Until then I'm going to stick with 14.7 afr at low load.

Back in the old days (70's) the best A/F ratio for economy was 18:1 so you are close. Having run some pretty lean mixtures in my old 65 Rambler American flat head 6 with Polished heads and pistons and rather high compression (about 180psi) I found that when you lean it out enough you start to run less vacuum for the same speed and at highway speeds the coolent temps would start to rise a lot more than normal on warm days. Running water injection helps cool it a bit and having variable mixture control really helped a lot too. I was running a single SU carburator - single modified needle variable venturi carburator. Mileage was in the high 30's for a 6 passenger car in those days was very good.

Here, we have automatic water injection from the sky. Hooray?
In the future, I also want to do either the Ford EDIS, or a coil over plug from a Honda CBR Motorcycle. I'm also looking into Independant throttle bodies from a M/C. There is a guy who drag races with thre same engine as mine, and does 11's in the ? mile. I'm concerned about the torque curve, though.

I don't know if it makes a difference but my engine has a relatively low 9:1 CR. That might be part of the reason why I actually INCREASED my fuel consumption by running leaner than 14.7. It definitely needed higher KPA (manifold pressure) to maintain the same speed and lost some throttle response but no FE gain for me. I'll try again later this summer.

Aha.
I have a 450SL that I was restoring before the economy took a dive. I never dreamed of getting good FE out of it.

Mine is a 93 190e 2.3. The last of the Kjet cars. Even through I'm running the same 14.7 afr as Kjet did before, it still managed to improve FE by 10-20% depending on the driving condition. It did a 30mpg tank last week with the non lock up TC automatic transmission. I'm hoping to get it into the mid 30s once I install a 5 speed manual.
https://190rev.net/forums/showthread.php?t=35155
Don't get too hung up on lean burn. With MS and a wide band O2 you should be way ahead of the stock setup.

I've been using lean burn on my car. I found that the engine starts to misfire when it's too lean. Sometimes it's hard to tell when it's misfiring, so try running too lean, then richen it up about .5 AFR. I also learned that lean AF ratios have a slower burn rate, so you should add 1 to 3 degrees of timing to your maps in the same areas where it's most lean.

It's a good idea to gradually richen the AF ratio as more power is needed. AF ratio should be at it's leanest during steady light cruise. On my car I found that using less throttle than steady cruise caused some misfires, so my fuel maps get richer around 2-5% throttle. (I cruise around 7-9% throttle, for example).

I also found that opening the plug gap helped reduce misfires slightly.
FYI my car has 7.9:1 compression, stock.
HTH

I read your ecomodder link, tjts, and will check out the 190rev.

DRW, I got scared, and richened it up to about 16. I'm going to screw with it more when I get an exhaust temp sensor, and move the O2 sensor to the collector, since it's in 1 runner. It richens up immediately when there's lower pressure in the manifold.
Good news; I went through 2 tanks of gas (22 gallons) in about 3 days (677 miles). That's about 30mpg, which doesn't sound great, but considering it was about ? rush hour & downtown traffic, and ? having alot of fun in the twisty mountain roads, it's great! I would expect more than 35mpg FWY.

Are you using a wide band O2? Bosch LSU with a 14point7.com controller has a built in EGT output from the O2 sensor.

Sounds like you're making good progress. Keep tweaking the settings while observing the car and it'll get better.

From my notes, 16:1 is the AF ratio with the highest EGT's. As you go leaner EGT's, NOx, knock propensity, and burn rate go down. Of course all cars are different, so what works for me might not work on your car, so be alert.

Cool. Good to know!
I have a DynoJet wideband controller with a needle guage, and datalogging (which is redundant, but it was a good deal). It's consistant with the MegaSquirt readout. I'm now running up to 16.5:1 at the leanest, but mostly around 16.
With the power, and low resistence of the car, the manifold pressure is very low at cruising.