limitations of engine tuning... timing vs AFR

[newaccorddriver]

i was reading on a website that when you are street tuning, you should try and flatline the AFR at say 12:1 on any given car, annd slowly advance the timing to lean it out. in theory this makes sense because you are extracting more energy with the advanced timing and creating more torque in the process. at the same time, when you advance the timing, you should also add a bit more fuel the richen up the AFR for safety measures.

altogether it makes sense because extracting more power means more heat is created in the process, and you should try and control temperature at which things start to melt away. one of the things that i actually question is when your advancing the timing, and adding in more fuel to the equation. by adding more fuel, you keep richening the mixture, and by advancing, your leaning out the mixture. is there really a limit on how much of this can be done? i know that you should back off the timing when knock is heard and detected, does this register on the wideband or an EGT gauge? also, on a few JDM skylines ive seen with MINE's ECU's, they tuned it so on the higher load areas, they kept advancing the timing beyond what is reasonable, and kept adding fuel in the process. im sure it makes alot more power compared to the stock ECU, but at the same time, they advanced the timing by lots, and pours fuel in by the gallon when on high load. basically the fuel cools down the pistons and the air and prevents detonation at the same time while also being able to advance the timing. im sure theres a limitation to it, but what is the limit on how much advance you can have along with a richened AFR before it is impossible to prevent detonation?

is the maximum amount of torque that can be achieved through the tuning process based on how much advance you will be able to run alot with the AFR while staying out of the detonation territory?

i will be tuning my car when its done on a chassis dyno so i can tune each cell to perfection, but in the meantime, id like to know the process of tuning..



i might have repeated myself a few times over, but ive had a long day, and its late at night.

[Hauntd ca3]

that 12:1 afr is considered a fairly normal ratio for full throttle/load
you have to advance the timing as revs increase since the af mix takes the same amount of time to burn regardless of revs.
the idea is that you advance the timing as revs increase to get the time of max cylinder pressure at about 10 to 20 degrees atdc so that the piston/rod get the best push it can on the crank.
with the good old dizzy you have a limit on how far you can advance the spark
if you have a timing light that you can adv/retard the flash with, you'd prob find that max adv would be 30/35 degrees.it could be a bit more or less depending on the dizzy but as said before since it takes the same time to burn the 12:1 afr regardless as revs rise and dizzy gets to max advance you might find that you need more advance than the dizzy can provide to get the point of max pressure where you want it.thats where aftermarket ecu's and dizzyless ign come into play..

adding fuel will hold off knock to a point but they'll just be adding it as revs increase aswell.
same as any engine . It all depends on how much air your motor can suck in to how much fuel you can add and keep the afr in the safe range.

[AccordEpicenter]

it depends on if your tuning with boost or na, Na you can get away with rediculous amounts of advance somtimes, but tuning for boost, if you are agressive with the timing, all the fuel in the world cant prevent you from detonating, and if you are even a little agressive with timing, adding fuel like crazy may or may not help if you are already at somthing rich, like 12:1 afr. There are alot of reasons why this happens, and there is no set in stone limit, it depends on the engine, setup, fuel, iats etc there are alot of factors that contribute to detonation. I would say for boost, id shoot for at least 12:1 afr or richer (even 10.5:1 should be fine) and be VERY conservative with timing, then once you have your target AFR across the board, start adding advance slowly, make sure your making more power, and once you get to the point where you lose power or gains start to level off, id back it down a few degrees to be safe. Its very common with pump gas turbo cars that the peak advance under boost is less than 18-20 degrees btdc, which is nothing compared to na engines that could have 35-40 degrees+ btdc timing stock. Generally, with more advance and leaner mixture on boost youll usually make the most power, but this can lead to severe detonation if you overdo it, lean is mean. Even with a built engine, if the tune is wrong, youll still blow it. Youre better off adding a couple psi of boost and tuning accordingly than tuning very agressively and running the engine ragged. You cant always hear knock, and even an inaudible ping on a turbo motor will melt pistons, or crack the lands and rings, if you do hear it, severe engine damage is imminent. On pump gas, tuning is the difference between a setup that is going to last or blow up almost immediatley. Ive seen guys put together turbo kits on their civics, and went out for a few rippers, or just revved the engine in their driveways building boost, and the tune was soo bad the engine was severely damaged in virtually no time. Tuning with an egt gauge is nice, but it should also be used with a wideband.

[newaccorddriver]

thanks for the explanation on the entire topic. the entire MINE's ECU talk on a skyline forum got me thinking as to why they richen up the mixture and kept advancing the timing.

[AccordEpicenter]

this is also why you cant take a tune from another engine and or setup and use the same exact tune for your car, it might work, it might not, different setups and engines tune differently, it all depends

[Accordtheory]

you should try and flatline the AFR at say 12:1 on any given car, annd slowly advance the timing to lean it out.

by adding more fuel, you keep richening the mixture, and by advancing, your leaning out the mixture.

Rich/lean is a/f ratio. That has nothing do do with timing. Advancing the timing does not "lean out" a motor.
What program are you using? Are you talking about your 3g? I assume you have obd-1.
Do you know what mbt, or "maximum brake torque" is? You don't just keep adding timing and fuel. Maximum brake torque is the most torque the engine will produce for a certain map/a/f. So you set your target a/f, and then start out with a certain low amount of timing that you know is safe, and advance the timing until the torque stops increasing, for that load point. Once you reach that, your timing is at mbt. If your torque is still increasing, but you run into knock before mbt, Then you add fuel, or change to a higher octane. Adding fuel will drop the power output though, just like running less timing, so it's all a series of tradeoffs at that point. Those tradeoffs are what tuning is all about.
11.8:1 is good for most turbo hondas, if you're not octane limited, and can hit mbt with it, something like 13.4:1 is good for n/a.

[Accordtheory]

[QUOTE=Hauntd ca3;861975]you have to advance the timing as revs increase since the af mix takes the same amount of time to burn regardless of revs.
QUOTE]

That is not always correct, many factors come into play here simultaneously that affect this in different ways. I have seen timing graphs that have less timing at higher rpm. Just looking at a stock p72 high cam timing map on the hondata s300, the timing from 6500-8950 is exactly the same, 30.5 degrees. But from 2000-4000rpm, it does increase in a fairly linear fashion, from 10-22.5 degrees, on the low cam map.

[AccordEpicenter]

also, when tuning, try to use a SuperFlow or Mustang dyno, a dynojet tends to be a bit innacurate and can skew results and thus tuning a bit. Its due to the design of it, you can actually put a precise load on an engine with a mustang, but a dynojet just spins a weighted drum and determines power and torque differently

[Accordtheory]

What I want to know is if you can find mbt with the gtech pro accelerometer. Imagine, a couple hundred bucks for all the tuning you'll ever want..

[newaccorddriver]

What I want to know is if you can find mbt with the gtech pro accelerometer. Imagine, a couple hundred bucks for all the tuning you'll ever want..

although that idea would be really nice, i dont think that would be an ideal way to do tuning. at $150ish an hour for dyno time, i really probably only need to book in 2 maybe 3 hours to get a precise tune on my car. alot of the time is spent letting the car cool down anyways, so i figured 2-3 hours should be good enough to get a good tune.

[newaccorddriver]

Rich/lean is a/f ratio. That has nothing do do with timing. Advancing the timing does not "lean out" a motor.
What program are you using? Are you talking about your 3g? I assume you have obd-1.
Do you know what mbt, or "maximum brake torque" is? You don't just keep adding timing and fuel. Maximum brake torque is the most torque the engine will produce for a certain map/a/f. So you set your target a/f, and then start out with a certain low amount of timing that you know is safe, and advance the timing until the torque stops increasing, for that load point. Once you reach that, your timing is at mbt. If your torque is still increasing, but you run into knock before mbt, Then you add fuel, or change to a higher octane. Adding fuel will drop the power output though, just like running less timing, so it's all a series of tradeoffs at that point. Those tradeoffs are what tuning is all about.
11.8:1 is good for most turbo hondas, if you're not octane limited, and can hit mbt with it, something like 13.4:1 is good for n/a.


i was refering to using megasquirt for my car. the reason why i said timing changes your AFR is mainly because of the amount of time it would take to burn the fuel. the more timing you give it, the more fuel gets burnt, and less oxygen will be present in the exhaust.


im going to be tuning to use 87 octane at first since this is mainly an economy car anyways, but theres a chance ill put premium in it and tune that instead.


im just starting to grasp the concept of tuning each individual cell on the tuning table, so i guess any reading i do will be considered during the actual tuning portion

[newaccorddriver]

this is also why you cant take a tune from another engine and or setup and use the same exact tune for your car, it might work, it might not, different setups and engines tune differently, it all depends

i wasnt taking a tune from another engine, i was wondering why MINE's did what they did with their ECU. it seemed quite simple to counter the knock with more fuel, but had no idea why they added timing

[Hauntd ca3]

[QUOTE=Hauntd ca3;861975]you have to advance the timing as revs increase since the af mix takes the same amount of time to burn regardless of revs.
QUOTE]

That is not always correct, many factors come into play here simultaneously that affect this in different ways. I have seen timing graphs that have less timing at higher rpm. Just looking at a stock p72 high cam timing map on the hondata s300, the timing from 6500-8950 is exactly the same, 30.5 degrees. But from 2000-4000rpm, it does increase in a fairly linear fashion, from 10-22.5 degrees, on the low cam map.

but wouldnt the bigger lobes on the vtec cams be effectively similar to bumping the comp ratio.
thats effectively what turbocharging does really and you have to be conservitive on the timig with those engines to avoid knock and melt down.
i'd say that the hondata motor puling such high revs is real high in the comp ratio and when you do that you end up with higher combustion temps which would bring you close to those temps they try to avoid to stave off knock.
keeping the advance back abit would help keep the pressures and temps down in the safe range.
mindyou what fuel were they running.
were they using that 80 something octane you have there or was it in japan where 100 octane is availible at the pump or even race gas

[Accordtheory]

the reason why i said timing changes your AFR is mainly because of the amount of time it would take to burn the fuel. the more timing you give it, the more fuel gets burnt, and less oxygen will be present in the exhaust.


That is totally incorrect. Changing your timing does not affect your afr. It might slightly change the results of a smog test, but not how much fuel is burned. That is not possible. I wouldn't attempt tuning yourself if that is the level of your knowledge, no offense intended.

[newaccorddriver]

That is totally incorrect. Changing your timing does not affect your afr. It might slightly change the results of a smog test, but not how much fuel is burned. That is not possible. I wouldn't attempt tuning yourself if that is the level of your knowledge, no offense intended.

doesnt timing dictate how much fuel gets burned? i mean, if the plug fires at TDC, then it isnt utilizing all the potential of the fuel in the cylinder, and not all of it will get burned cause the plug fired too late?

any criticism is more then welcome. i know what to do what it comes to different loads and different RPMs and such when it comes to fueling a car, but dont know much about ignition timing

[Accordtheory]

[QUOTE=Accordtheory;861994]

but wouldnt the bigger lobes on the vtec cams be effectively similar to bumping the comp ratio.
thats effectively what turbocharging does really and you have to be conservitive on the timig with those engines to avoid knock and melt down.
i'd say that the hondata motor puling such high revs is real high in the comp ratio and when you do that you end up with higher combustion temps which would bring you close to those temps they try to avoid to stave off knock.
keeping the advance back abit would help keep the pressures and temps down in the safe range.
mindyou what fuel were they running.
were they using that 80 something octane you have there or was it in japan where 100 octane is availible at the pump or even race gas

You said the amount of advance increases with rpm, I'm explaining that is not always the case. The hondata map I referenced is a factory map for a factory usdm motor, tuned to mbt, actually slightly less for safety. The fact that the motor carries the torque so well as it revs with the same advance for the last 2500rpm or so of the powerband illustrates that the burn speed of the a/f actually increases with rpm, contrary to what you wrote ealier. Kind of weird to think about it that way, but it's true.

[newaccorddriver]

so how would competent tuners actually tune a car? nobody around the edmonton area has every tuned a car that is running on megasquirt, and alot of the tuners arent willing to tune a system because of the fact taht they never heard of it. i can rent their dynos, but that is about all i can really do since i have to tune at my own risk.

anyways, what do tuners usually do when their on the dyno since apparently i got the wrong view of things here. would they do AFR first and then timing, or do they do both of them at once?

[Accordtheory]

doesnt timing dictate how much fuel gets burned? i mean, if the plug fires at TDC, then it isnt utilizing all the potential of the fuel in the cylinder, and not all of it will get burned cause the plug fired too late?

any criticism is more then welcome. i know what to do what it comes to different loads and different RPMs and such when it comes to fueling a car, but dont know much about ignition timing

Okay, this is probably my last post in this thread, so I'll try to make it worthwhile..lol.
Ignition timing does not have anything do to with how much fuel is burned. Once it's lit, it's going to burn, what does it matter when it's lit? If you fire the plug at tdc, the motor will lose power due to a very non optimal cylinder press vs crank angle, and the fuel will still be burning as the exhaust valve opens, further wasting energy.
The correct timing for every load point will optimize the cylinder press vs crankshaft angle, resulting in the maximum torque (power) for that point. If you run into knock before mbt, you're octane limited.
As far as fighting knock, is it better to add fuel and keep the timing up, or keep the fuel the same and take the timing out, who knows. Some people even say running higher boost with less timing is easier on the motor for a given hp level. But most people agree that for a built motor, tune the ignition to mbt, and keep a reasonable a/f ratio. The idea with the former method is to reduce the peak cylinder press for a given hp level, (trying not to crack those fragile sleeves) but it is less fuel efficient, and puts more thermal energy through the motor. more blowby too.
Good luck with your goals..

[Hauntd ca3]

i see where you comin from now.
it may not always be the case but is a fair chunk of the time
most tuners down here have a special knock sensor and headphone set that they can listen for knock with and set the timing from there at each load point in the aftermarket ecu.
seems to be pretty effective.
the only time i,ve seen it fail is a dick who tuned on 96 octane with octane booster then ran the thing on 91 and cracked all his pistons
typically it was a highschool kid with rich parents( dont you just hate them) who watched fast and furious to many times and took everything they said as law
only cost him 10k for a rebuild

[newaccorddriver]

if i was to run something like chrome or emanage, then the tuners here would know exactly what to do, and how to do it, but since im running something they never heard of, then they dont want to handle the liability. also, more often then not, when it comes to tuning with an unknown system, they dont really give any advice with it either which is why i turn to here for some advice

[Accordtheory]

Okay, This will be my last post in this thread..I need to stop posting and do something productive! Or maybe just take less than 20 minutes just to write one post..ugghh.


most tuners down here have a special knock sensor and headphone set that they can listen for knock with and set the timing from there at each load point in the aftermarket ecu.
Okay, hopefully they aren't actually using the knock sensor to determine the timing, they're just using it to make sure they aren't hitting the knock threshold in the quest to find mbt. If they're doing street tuning, off the dyno, and actually hit the knock threshold and then back it off, that is not the ideal way to do it, the timing could actually be ahead of mbt, (!) pointlessly making less power and putting the motor at an extremely high risk of damage. If they're approaching mbt on the dyno and then hearing knock and backing it off, that's correct. I have the j&s safeguard on my car, for that added degree of protection..

newaccorddriver, what I'd do I guess is try to find a stock map that sort of matches your motor, and go from there. I'd start out with conservative timing, tune the part throttle to 14.7, richer, like 13:1 or so at the higher load points, and then start playing around with the timing. I don't know if you're doing a turbo build, but if you are, most tuners who are actually willing to talk agree that 11.8:1 is a good ratio for safety and power.
There are also forums that are dedicated to the subject of tuning, I bet you could learn a lot more there than by reading the bs I post up on 3geez..:)

[Hauntd ca3]

some use it to get a good idea of new ign mapping and others use it to findthe limit that they can advance it then set max adv slightly less than that

[cygnus x-1]

AccordTheory is right on here. The common method for street tuning is to bump up the timing until it knocks and then back off a bit. That will get you close but is not always optimal. The only true way to optimize timing is to go for maximum torque. To do this you need a way to hold the engine at a particular MAP/RPM point while adjusting the timing for maximum torque. Loading the engine is the only way to do this. Which leads to AccordTheory's question:

"What I want to know is if you can find mbt with the gtech pro accelerometer. Imagine, a couple hundred bucks for all the tuning you'll ever want.."

I don't have a Gtech but I do have a similar product from Innovative Motorsports. It's a sensor box with a 2 axis accelerometer that you can use to figure out torque and HP based on the weight of the car, gear ratios, tires size, and wind/frictional drag coefficients on the car. This is all fantastic information but there is still the fundamental problem of keeping the engine at a fixed RPM/MAP point. This is impossible on the street. If you could find a perfectly flat and straight section of road you might be able to do something where you take multiple runs while logging RPM, MAP, and HP/Torque, and then start changing the timing in certain places and seeing what affect that has from the logs. This is probably easier out west but around me it's impossible to find roads that are consistent enough. A drag strip is a possibility but you're going to have to make lots of runs so it would take forever.

One thing I've been thinking about lately is if it might be useful to be able to measure engine torque directly, instead of having to calculate it based on vehicle speed/weight or G forces. Those methods aren't so accurate because you don't know if the road is level or what the wind effects are (the car will accelerate faster downhill than uphill with a given amount of torque, same with a tailwind). But what if you built (essentially) an engine dyno right into the car itself? How? Well imagine if you could measure the force exerted on the torque rod that keeps the engine from twisting when you accelerate. That would give you some measure of the torque output of the engine, and is in fact how engine dynos measure torque. You probably wouldn't be able to equate it back to *actual* torque output, but as long as it's reapeatable it doesn't matter. All you need to know is that whatever value you measured this time is higher or lower than what you measured last time.
Now how well this would work in real life I have no idea. It's just something I've been thinking about. Probably for all the effort it would be cheaper and easier just to go do it on a loaded dyno.

C|

[Accordtheory]

Conceptually a good idea, but repeatablity/accuracy/loggability would definitely be a challenge. Check out the software with the road race gtech.
http://www.gtechpro.com/images/bkgd-pass_dr.jpg

[cygnus x-1]

I was searching through old posts and came across this one. Just wondering if you (Newaccorddriver) ever got your Megasquirt running?

C|