How to Safely adjust timing
 

I was just wondering if anyone can shed some light on how to adjust timing. I think i got the fuel trim adjusting down pretty good. Just kinda confused on how to do the timing. Any input is greatly appreciated.

Thanks in Advance

Freddie T

I doubt any pro tuner will tell you their technique since it is their livelyhood, but Always start conservative and work your way up.

I am not asking for any pro-tuners advice. I would rather have people like malibu jack, warrtalon, or even Alwaysinboost. I am just looking for where you should start. I remember how Alwaysinboost was showing us where to look to know what blocks we were in in ver .7. But with the program not working with version 8 I am just looking to safely increase my timing even a lil bit for a bit more power.....

Thank you everyone.

But if Shiv or Al wants to chime in I wouldn't want to discourage them :D

If you want info like that man you need to do your homework so you can at least understand though. Someone can tell you a basic step, but understanding is necessary also.

ok. where should i go and read up to do my homework. The fuel map was easy but, i figure when i start to mess with the timing i will be redoing my fuel map so its all good. I just really want a good starting point

By the way where in houston are you man. I went to high school in pearland, and went to Montgomery College in the Woodlands??

There's really no big secret to adjusting timing at all.

You'll need:
- lots of time
- an open (err, closed) road. Actually a dyno would be greatly preferred, since it eliminates varying environmental conditions.

Make a couple of 3rd gear pulls with your existing map and log them. You'll at the very least want to log RPM, Timing and Knock Sums, airflow is good to know as well.

Now take a look at your logs.
- Look at the timing values and figure out which load cells you are hitting on your high octane timing map
- If your logs show knock at WOT, then you are done and you should consider backing off your timing by one degree or more (depends on how bad things got).
- If your logs show no knock under WOT, increase your timing by 1 degree for the load cells that you are hitting. Pay attention to the adjacent load cells. For the sake of knowing which load cells you are hitting, you do not want to have two or more cells next to each other with the same value (i.e. something like [3][3][3] ). So, if you have values that are something like [3][2][1] and you want to bump the [2] to a [3], then you'll probably want to change it to look like [4][3][1]. In other words, you are not simply changing one load cell, you are changing the 'curve' at that particular RPM value.

Make another couple of runs and log them ... rinse and repeat.

The key to tuning is taking small steps, logging, and having consistency in your WOT runs.

This method will allow you to tune your car for the maximum amount of timing advance that it will run on that day. If environmental conditions change (e.g. hotter, colder, humid, dry, etc), then your tune may not hold up, which is why you will want to log as much as you can. The more you decide to push the envelope, the more you will have to keep tabs on things...

l8r)

how can you tell which load cells you're hitting? that's the one thing i don't get, i can see what rpm, timing, and knock are, but how do you determine load? do you just match the rpm and timing number up and see which load cell it's in?

Just make sure there is no knock present. And then, that is it....very simple ;) .
knocking will change the timing value from the table. So, you can't follow the load value if you have the knock counts.

i guess i'll just stick to fuel, i don't follow. how do you eliminate the knock from a particular load cell if you don't know what load cell needs timing pulled?

Trial and errror. :D
knocks more than 5 counts will definitley pull the major timing from the table. But, for small knock counts like below 3, ECU doesn't really pull the timing that much...
so, just do the wild guess around that region and try fueling it more or lessening the timing.

you should tune fuel first, then mess with timing, correct?

there is also a point where u can go overboard on timing, without gains in power... just closer to catastrophe when u have a mishap, such as bad fuel, or a significant temperature change....

wideband, close observation of knock voltage etc...

I'm just a beginner like you but one thing I can tell you is if you allready have a custom tune timing map on your car and you need to add timing for say, alcohol, it can be easy to do.

That was what I did, I added just 1 degree to where my load vs rpm, furthermore alcohol was spraying and that 1 degree made ONE BIG difference. I actually posted that map (like an idiot) thinking that it was now my map since I added one degree and got into big trouble with all the EvoM CEOs.

One degree at a time- would be easier if 0.0 able

Do I need to write another how-to? :lol:

Actually it's pretty simple. Log your RPM and your Timing, then take those data points and look at where they fall on your high octane timing map.

How to do this?
- open your high octane timing map
- Hit [ Alt ] + [ PrintScreen ]
- open Paint
- Click on [ Edit ] -> [ Paste ]
- Now take your RPM and Timing numbers and put them on the picture of your map

So, assuming a stock timing map and a log that looks like this:
RPM ... Timing (degrees)
4633 ... 2
5005 ... 2
5406 ... 3
5759 ... 7
6313 ... 9
6791 ... 13
7289 ... 15

The end result will look like this:
https://www.evolutionm.net/forums/at...chmentid=93255

So now you would know roughly which cells the ECU hit.

However that is assuming that no timing was being pulled due to knock. Therefore you also need to look at your knock counts during the run and you need to take a look where the timing would have fallen on the low octane timing map. If you map the same hypothetical timing values out on the stock low octane timing map, you'll see that they would have hovered around the 160-180% Load columns. It would be very unlikely that your Evo was only able to generate 180% load at WOT.

l8r)

so usually at WOT you'd be above 180 load? thanks for the little how-to, it definitely helped me understand, thanks! i hear of some people making the high/low octane maps the same, this isn't reccommended then?

Welcome.

As far as the setting of high/low octane maps equal ... It's not something that I would ever do as part of my daily driven (or race) tune.

The only time that I've had my low octane timing map set equal to the high octane timing map was to troubleshoot why my car was pulling a couple of degrees of timing at WOT with no knock. However the low octane map was set back to its original state before the car was even pulled off the dyno.

l8r)

Thanks for this, before I had to copy all the cells and paste into excel when all I needed to do was [ Alt ] + [ PrintScreen ]!

hahah yea man I am.

because the logging program Second Chance created is so useful in tuning I'm still using the v.7 EVOScan program. When something comes along that is better I will reinstall the v.8. But for now I'm extreamly happy with how my car is running.

If you want to come down this weekend I'll help you tune your car on the street. Your going to need a WB though.

Datlogging your run from the diag port will also tell you if you are pulling timing in a range; you want to make sure that your not using your low timing map in that range, it will tell you if you are knocking. If your RPM doesnt match your cells, look in your low timing map and see what the cells are, then adjust the corresponding cells in the range on your high timing map down a little and see if the timing changes back to the high map in that range.

this clarifies even further for me, thanks. also, what's [printscreen]?

[ PrintScreen ] is the key on your keyboard that has the words Print Screen on it. Sometimes also labeled as [ PrtSc ] ;)

l8r)

Look on your keyboard above the insert button.

I would be very careful of what you read online and what you actually follow. Tuning is not something everyone can just pickup on. You have to put time and effort into it. Most of all understand why you're doing what you are doing.
I never tune with the knock sensor, meaning waiting for it to have counts and back it down.
Good luck.

When setting up base timing maps a dyno is helpful as you dont want more timing than is needed to achive MBT

Usually a good starting point is to dial in MBT on a dyno or with an accuarte G force meter and then back off 10% for saftey (with ecu flash you can only go a full degree at a time)

With good timing base map in hand then you want to get on a road and test it in all gears and varying normal operating conditions to assure that knock activity is not present with the fuel normally used


On the IX's timing is more compicated as it works hand and hand with MIVEC maps

Remember the stoc ecu uses a average on 4 boxes around the targeted map cell


Finaly, for reference the Evo ecu is reactive in terms of knock control not adaptive as in the case of a Subaru. In short in an Evo you start off with your desired Ecu timing profile and the ecu pulls away timing as the ecu percieves defined knock events based upon what the knock sensor hears. In a Subaru the ecu can add in timing in the absense of knock - (not the case on evos). For more details review my article in Turbo magzine on Evo ecus.

would you believe i've never used or noticed that button there before? :lol:

also, thanks for chiming in Al!

^^ Shocking! {pick}

:lol:

l8r)

Have you ever gotten an ID 10t error??

Here's a nice 3D representation of the stock timing table... notice how as load and RPM are compared, the stock ECU timing table follows a smooth ramp up of timing as RPM increases... the 3D representation makes this eas to see...

https://www.evolutionm.net/forums/at...id=93289&stc=1

When you adjust your timing cells you want to make sure that the ramp rates between cells aren't too drastic. This could lead to all sorts of problems... a sudden "safe to knock" transition and a rather non-smooth seat of the pants feeling as you traverse different timing cells (rolling on and off the throttle rather than treating the gas pedal as an on-off switch).

Here's an example of a ignition table that wouldn't feel very smooth when jumping from the 180 to 200 load regions at 1250rpm (sudden increase in timing that could result in knock) and at 5000rpm (sudden drop in timing that could feel like a rapid loss in torque).

https://www.evolutionm.net/forums/at...id=93292&stc=1

I purposefully left the timing values out so you couldn't recreate this table yourself.

HAHA I wonder what that is susposed to represent...

Nice post as usual.

What is that done in?

MS Excel

Im not an expert but I personally don't feel that this graph is a good method to judge the quality of a timing map. Dont ask me why - I dont know, but I do know there's so many more important factors at play than load vs rpm

i see how you reference the timing load cell by referencing the datalogged timing and rpm and looking at the map to find the load cell, but once your car isn't stock, how do you follow the fuel being that the AFR you datalog doesn't correspond to the AFR target in the map for a specific RPM?

Once you chart your timing through the load cells in the timing map, you should be able to use the same pattern on your fuel map. I.e. if the ECU was hitting 260% load at 5000 RPM, then look at the 260%, 5000 RPM fuel cell. Your logged AFR will rarely correspond exactly to the AFR target value in the fuel map, especially once you start modding.

l8r)

That type of map is not optimal for several reasons, IMHO:

1.) The flat timing ranges basically assume that at 200% load you can run the same amount of timing as you can at 260% load.

If you take a moment to think about that, it doesn't make sense. Simply put, the ECU calculates the load value based on the mass of air being pushed into the engine. At 260% load value you will be pushing more air into the engine and (hopefully) mixing it with more fuel, resulting in a more powerful combustion (more fuel + more air = bigger release of energy), which means that you should be able to run more timing at lower load values.

2.) Sudden changes in timing, no smooth transitions. This can cause detonation, both from a sudden increase or decrease in timing.

3.) It is very difficult to log against flat timing maps. If you logged 5 degrees at 4000 RPM and your timing table shows 5 degrees from 180% to 260%, then it makes it very difficult to pin down exactly where your ECU is running through its maps

l8r)

Timing curve is suppose to be a smooth transition as load increases and with that graph he is showing the difference between a well adjusted timing map and poorly adjusted timing map. If you have seen the aem software, once your timing map is adjusted properly and you look at in 3d you will understand as the graph will be smooth as butter.

A graphical map is highly useful to make sure that you do not miss any cells and also to help shape the map

The taget numbers under load should be determined through dyno and road logging and then you can use the graph to smooth things out

Flash Evo by Ecutek has some great 3 d maps you can spin about in all directions and its very helpful

Aem is the same

gotcha! makes sense, thanks alot, you should make a how-to!

Thank you very much Al for the responce, also everyone else who chimed in. I think i get it a little more now. I still have to pick up a wideband. I figured i would do that before i go and try to sign up for the dyno day with you guys alwaysinboost. I don't want to look unprepared. I just want to try to understand it and the info that I recieved in this thread has helped alot..

You really should have a WB but if your only going to tune your car on the dyno then you don't 'need' one because they have one there. However there is a big difference between tuning 3rd gear on the dyno and tuning 1st-5th on the street with real load in real driving conditions.

The map I made on the dyno is childs-play compared to the one I made on the street.

I am so jealous of you alwaysinboost being that we have the same mods. You have less miles than me, I have thought about loading the maps that you have posted being that my car has been across the country 2 times and 12000 of my 50000 miles have been highway so all i can say about my evo is 18000 of my 50000 miles ok so 32000 are city. I can't be more happy even with the gas mileage increase that I have gotten thanks to evoscan and evoflash. To even say that it is a great time to have a evo. Is a total understatement. No dissrespect to you Alwaysinboost in what you said. You sold me great exhaust and I can do nothing but brag on megan racing...... Thank you again Al for posting again in this forum.

By the way Alwaysinboost how long you think it would take to do street tunes if I got a wideband...(correction when i get a wideband)

I actually have just under 70k on my car so you have nothing to be jealious about... just enjoy your car, thats all that matters.

I'm glad you like the exhaust, its a nice system. I would have purchased the same thing again but I wanted to try the EBAY unit out.

As far as tuning your car goes will probably only take a few hours and about 1/2 a tank of gas, lol.

So [XXXXRPM] - [14], [13], [12], [11] is the same as [12], [13], [12], [13] and so forth?

Hey Al. I would be very interested in reading the above article. Which issue was it? Is there an online version of the article???

BTW I went to Dynoflash.com and clicked the "tech articles" link and got an error. Don't know if this is a temporary glitch or permanent. Sorry for the off topic...

The page cannot connect to the MySQL database.

Miraserver Error: There has been an invalid database query. MySQL said:


Host 'phre2.phre.net' is not allowed to connect to this MySQL server Query:
SELECT name,id,short,parentlist FROM sections WHERE status=1 AND parent=0 ORDER BY name

No, I think that needs to be re-worded a bit. At least from the logs that I've looked at, I've noticed the following:

Example:

sample timing table (240 and 260 Load columns):

5500 RPM ... [4][2]
6000 RPM ... [6][4]

Now, let's say that the ECU has calculated a load of 250%, the timing (barring any other condition that pulls timing) should look as follows:

5400 RPM ... 3
5500 RPM ... 3
5600 RPM ... 3
5700 RPM ... 4
5800 RPM ... 4
5900 RPM ... 5
6000 RPM ... 5
6100 RPM ... 5

If your Load % varies slightly, you may end up with something like this (given the same timing table):

5400 RPM, 260%, ... 2
5500 RPM, 260%, ... 2
5600 RPM, 250%, ... 3
5700 RPM, 250%, ... 4
5800 RPM, 250%, ... 4
5900 RPM, 240%, ... 6
6000 RPM, 240%, ... 6
6100 RPM, 250%, ... 5

If the calculated Load % and RPM value do not squarely fall onto a pre-defined timing cell, then the ECU will average the nearest timing cells to get its timing value. Sometimes this is two cells (row or column), sometimes it is four cells (in a square), but I've not seen the ECU use 4 cells in the same row to get its average.

l8r)

You got to be kidding, I'm of the working class - Im like duuuuh.

Cute