conservative aggressive tuning / good bad tune jobs

Hi,

1) what is considered conservative and reliable tune ?how many degree do you retard the timing from knock,is 12.7 afr conservative or aggressive in general.

2) what is aggressive is 2 degree retard from knock considered aggressive?

3) what cause the tune to go out of calibration i see this in cars with no o2 and no maf.

4) usually changing as little as changing muffler can make mess with the tune is this only with piggyback?and why

5) is there a too rich tune i know you can wash the cylinder wall but could it cause pre-ignition or something, if so would it happen more likely with retard timing or advance?

6) can you give me some examples of what you saw from a bad tuner that could cause engine damage?from what i learned about tuning it seems idiot proof as long as there is no knock or heat its a good tune. or do they take short cuts like tuning targets only and not touching maf and ve?

7) dyno and street load difference am i always gonna see the car running leaner on street(like .5 difference?) or is it just because of the wideband location (wind blowing at the wideband) and timing should i be more conservative when tuning on dyno compared street.

8) wideband location tail pipe and headers afr difference? is it gonna be like .5(+ the .5 leaner from street?)should i tune 11.8 from headers wideband on a dyno to get 12.8 from tail pipe on the street?

i have seen so may engines blow up but cant understand what makes it hard it seems straight forward but if i understand those questions maybe i will find the answer and i think those question should be explained in efi fundamental and practical reflash.

thanks

First, there is a range of opinions on a number of these topics. Two people who know what they are doing might disagree, and also sometimes you have to make judgment calls because you don't have enough information (dyno time is expensive for example).

1. Conservative and reliable tune? It's nice to have a point of reference. The first thing I look at, if it's available at all, is what the car does in the 100% stock tune or map (assuming the engine isn't drastically different from stock). For example, if 5000 rpm WOT in the stock tune is at 11.5:1 and 15 degrees of timing, that's a good baseline for "safe" tune. It doesn't mean that 12.0:1 and 18 degrees is automatically unsafe, but it's a baseline.

Generally speaking, n/a can run leaner than turbo, and direct injection can run leaner than port injection, BUT you need some context, such as the compression ratio, cams, amount of boost, what kind of octane.

2. I wouldn't say 2 degrees retard is aggressive, but it depends on whether knock control is on and how aggressively it reacts. If you have no knock control well 2 degrees could be pushing it, but it also depends on compensation tables (high temp compensation for example). The bigger question is how you're finding the knock limit. Severe knock is obvious (you can hear it audibly without any kind of equipment), no knock is often obvious (low speed & low load), but the "borderline" is often opinion. Even the knock sensor on stock cars has to be set according to someone's opinion of the threshold, and is subject to the noise profile of the engine and the location of the sensor. For example, I4 engines with 2 knock sensors (think recent generation BMW engines) are going to read knock more accurately than 1 knock sensor.

3. What you are describing, an engine without an o2 sensor or a MAF sensor, is basically a carburetor in its accuracy. It's hard for a couple tables to cover all areas of engine operation, especially as the engine accumulates running time.

4. A lot of that depends on the specifics of the engine and the control system. Generally speaking a boosted engine running without a MAF sensor is going to be more sensitive, but the newest engines (think Ford Ecoboost) are pretty smart already. Can you give a more specific example? A lot of piggybacks don't work well on modern engines.

5. Too rich is most likely to cause a misfire because the spark can't ignite the mixture.

6. Bad tune depends on context. Is this a stock ECU reflash, is it a standalone, etc. Generally speaking, a bad tune on a stock ECU ignores knock feedback and fuel trims, or unstable boost, i.e. it demonstrates laziness (didn't spend the time to calibrate the MAF or VE) or lack of basic knowledge (lots of knock retard is bad but the tuner didn't understand that maybe). On a standalone it's often not understanding or not spending the time to get a lot of the compensation tables set right (air temp, water temp, etc), or the idle/driveability stuff. Stock ECUs are typically more forgiving for driveability than standalones.

7. Depends on the dyno. An inertia dyno like a basic dynojet doesn't load the engine down so well. A really sophisticated dyno that can do road load will be more representative.

8. Wideband location - before the cat is going to be more accurate than after the cat, but the location of the sensor in the exhaust stream does matter if it's not mixing all the cylinders (for the whole engine or for a given bank of cylinders). You also have to be careful about exhaust pressure and temperature affecting the signal.

thanks raymond for taking your time to answer the questions,

1)For example, if 5000 rpm WOT in the stock tune is at 11.5:1 and 15 degrees of timing, that's a good baseline for "safe" tune. It doesn't mean that 12.0:1 and 18 degrees is automatically unsafe, but it's a baseline.

lets say for gm engines baseline is 11.5:1 or richer with cat protection and 16 degree of timing, first tune is 20-23 degree of timing and 12-12.5:1 and the second is 25 degree of timing and 12.6-12.9:1 the knock limit is 27 and 13.2:1 ,how much more aggressive is the second tune is it still conservative ? i mean if i tune for business is it worth the extra 5-10 horsepower or should i go with the first one for a good reputation.

2)lets say gm engines and toyota cars specifically gt86 ,land cruisers and supra, lets say i have 2 degree of timing retard and the tune lean out for a problem in the engine like fuel pump and the driver doesnt have afr gauge is he gonna blow up the engine after couple of wots from afr 12.7:1 to 13.5:1

3)the engine is 1fz-fe with unichip, im not sure if unichip have a setting for knock control and have to be set first (most of the tuners upload a file and make few changes for ve and timing)

4)1fz-fe with unichip i find that changing as little as a muffler would leanout the car does this happen with other cars or ecu is it the tuner fault or is it common with speed density cars.

1) Well for the first example consider that the stock tune is trying to protect the life of the cat over many thousands of miles of driving (including towing or other harsh conditions) to meet government regulations. The government checks the emissions after 4 years and if the cat cycles too hot enough times it will deteriorate the emissions and could also cause a catalyst efficiency code sooner than it would with a richer tune. So that's one of the major reasons why the stock tunes are so rich. GM uses a model of the catalyst temperature for catalyst protection; a lot of ECUs don't. Often that GM vehicle's cataylst temperature model feedback is turned off in an aftermarket tune, and only the power enrichment is used.

Also, there is a trade off between spark and AFR, and you get more margin when you have knock control working. See attached article I wrote a while back. Another important things to understand is what is the Lean Best Torque Air fuel ratio. Generally speaking it's low to high 12's in terms of AFR across a lot of engines but it can vary a little bit. So leaning it out to 13.3:1 might actually not buy you anything. See attached image showing relationship between AFR and torque.

2) I've seen engine survive a quick lean out. I'll put it to you this way, you've got a lot more margin for things going wrong on a non turbo. You might overheat a cat or hear some knock, maybe have some damage to the exhaust valve. On a turbo car the risk is much higher of throwing a rod or something.

3 & 4) I don't know much about Unichip but I think it is a piggyback and most don't work very well. A JB4 by BMS (most commonly on BMWs) are probably one of the only piggybacks on modern cars that work ok.

Sweet article Raymond, thanks for sharing!

Interesting post, thanks for sharing the article Raymond.

thanks Raymond for the article this actually cleared some of the confusion in my head so if you know more articles i would be interested to read them,

a week ago i took hand on dyno lessons and i think it kinda answered some of my questions,

on the 1zz-fe engine the peak torque was on .8lambda and 16 degree of timing, we started at .93 and it would knock past 16 degree and with .8lambda we got 22 degree but no gain after 16 degree,

1)now my question is would a conservative tune be the tune that wont knock at .9-.93 and tuned at peak lambda .8 for example? lets say 22 degree with .8 would gain 10 more hp is that aggressive and tuned on 16 with .8 lambda be more conservative (my reasoning is if the engine ran lean for example because of mechanical issue it wont detonate but then would it be an issue of too little timing if it away from mbt or maybe even pointless) can i use this approach for all engines, let say im tuning someone else engine and they want to be sure it would never blow up because of the tune is this the best approach detune timing and keep peak lambda even if it .9 with consideration of egt, or still detune lambda too to .8-.85 just to be extra conservative.

2)which is better lean with less timing or rich with more timing if hp is equal let say .8 to.9 lambda maybe .9 is too hot let say .87, i noticed most of ls based engine tuners use lean with less timing .85-.9 and use race gas for more timing but i always heard lean is mean and the tuner that tuned my car used .8 afr i dont know if it because he is running more timing or because its a drift car so idk cooling or less timing with consideration of the engine application and duration of wot

3) this is kinda the same question but it involve nitrous, Monte Smith says to run the afr same as n/a peak hp but here its recommended as turbo afr for cooling, his argument that engine are more likely to blow up from rich than lean now he doesnt mention how rich is bad but insist on reading the plugs so idk how bad is it to run .87 on nitrous compared to .7-.8 with large shot 200+ if plug doesnt read lean

4)at which lambda would cylinder wash happen? and does type of fuel effect have different lambda? from what i learned so far i personally would never run rich than .75 i would target .78-.8 for all turbo engine unless i make more power with .75 i might consider running peak n/a lambda .85 for example if i know the cylinder temp isnt too high.

5)is there a way to tell cylinder temp without egt other than knock, im guessing its based on timing

6)more boost or timing im starting believe that lean and more boost is always better than more timing as long as it wont melt the piston or that more power can be made with more timing than boost

7) how to reduce injector duty cycle if maxed out is it basicly running leaner or is there anything else without hurting an engine, and does reducing timing only have an effect on duty cycle.

thanks

1) I can only speak in generalities because it depends on the engine. Generally speaking, more spark + richer is less risky than less spark & leaner. Typically you have lower exhaust temperatures when you go the richer + more spark route, but it's worse for emission (which isn't a concern area here but when looking at stock tunes it is). Each parameter you are optimizing has its own sensitivity curves and you have to find the sweet spot for that parameter depending on your goals. A 1ZZ engine is a pretty simple engine to tune. A lot of them didn't even have electronic throttle or variable valve timing which are standard on modern engines now.

2) See answer to #1. The other thing to keep in mind is that every platform has its own set of "groupthink" and common expectations of what a good tune consists of. Sometimes there are good reasons for that and other times it's because so-and-so started doing it years ago and it just became conventional wisdom. Turbocharger choice is one example. On say an Evo (8 through 10 at least) it's common to wind up turbos to very high boost and taper them down at high rpm, similar to stock. Other cars it may be more common to put in large turbos and hold a flat boost curve to redline.

3) You would have to get into the exact details of how the nitrous system is plumbed and how it's tuned. Is it a single fogger? Direct port? What ECU are you tuning it with, what safeties do you have, how is the engine built, etc.

4) Cylinder wash is going to depend on the engine. If you have access to stock fuel AFR maps or datalogs it can give you an idea of the lower boundary. If it runs 10.5:1 stock (assume E0 for simplicity's sake) well then it's probably ok to 10.5:1 . On direct injection engines, the injection timing is a big concern for cylinder washing. Also, most of the time you are ok if you change the oil more frequently. When you are modding a car it's understood that you aren't constrained by wanting to have long oil change intervals.

5) You can't tell temperature or pressure without measuring. You can use basic principles to help guess/predict what would happen "directionally"--for example, richer is usually cooler, advanced spark is usually cooler, more boost is usually hotter etc.

6) Some of this is going to depend on the specifics of the engine and the turbo. If the turbo is already operating near its limits (high altitude, or already turned up the boost compared to its efficiency range, or very restrictive intake and exhaust). As far as spark timing goes, it depends on the fuel and depends on what your engine can handle. If you are putting a turbo on an n/a engine and keeping the stock pistons for example, it's not built for boost and it will likely not tolerate as much timing (due to either knock or cylinder pressure concern). More boost/less timing is definitely not fuel efficient, and there are limits to how lean you can run it. On a modern turbo engine you use variable valve timing to help spool and peak power.

7) The way to reduce injector duty cycle without changing the injector or fuel pressure is to inject less fuel. So either you run the engine leaner or you make the engine somehow more efficient so it takes less fuel to make power (better brake specific fuel consumption at any given lambda).

one of my previous question became my worse nightmare, what causes the tune to mess up on it own,

on my ls2 engine i have an issue that i cant trace it, i think it might be small headers leak but i fixed it over 6 times i cant see it on the 02 mv datalog, my fuel trim stays positive no matter what i try,

my car runs at it best on closed loop, on open loop with short trim it would lean out over 40mph at any gear/load start to stumble hard on the highway

open loop is fine just too lean, i dont think its fuel pressure issue as the engine runs richer than 10 on wot.

interesting post, thank you