Smoothing timing tables

[superskaterxes]

so i recently saw a post from @djuosnteisn where he posted one of his timing maps and it looked like a cheese cloth draped over a bar stool. shit was smother as butter. i decided to take a stab at smoothing my timing maps as the stock ones are rocky mountain status. i got to thinking about the possible ramifications of this but couldent think of much. i flashed the map today and only put a few miles on it but it seemed to feel smoother. thoughts?

before:




after:




@rfinkle2
@Dano
@Lex
@socks

[rfinkle2]

LOL. I saw that same post.

I don't know what the other guys will say concerning the 3 dimensional graph, but I'm primarily concerned with graphing spark advance vs. rpm in my logs.

pming you.

[superskaterxes]

lol MSF encyclopedia finkle sent me this.

im pretty sure this is the one i was talking about but im not sure. what kind of problems could this much smoothing cause?

@David@COBB
@Cobb Tuning

My timing vs OTS

[SSinstaller]

Originally Posted by superskaterxes lol MSF encyclopedia finkle sent me this. im pretty sure this is the one i was talking about but im not sure. what kind of problems could this much smoothing cause? @David@COBB @Cobb Tuning My timing vs OTS

I've been running a smoothed out timing table for a while now.... As long as your EGT's are good and your not seeing KR your not going to have any issues.

Here is my timing table:

[superskaterxes]

Originally Posted by SSinstaller I've been running a smoothed out timing table for a while now.... As long as your EGT's are good and your not seeing KR your not going to have any issues. Here is my timing table:

you sir need to add some series timing, you have no idea what your missing out on..........

[SSinstaller]

Originally Posted by superskaterxes you sir need to add some series timing, you have no idea what your missing out on..........

LOL... I know what I'm missing, just can't do anything about it right now...

I'm hitting 495g/s and 120% InjDC at 3.3 load, on pump gas with no meth.. More timing in the 2.0 row = KR for me...

[Lex]

Remember that the ECU smooths timing transitions for you anyways. And that the main point of the timing table is not to look prettier than your girlfriend's ass but rather to extract the most efficiency from the motor without knock.

Sooo the proper way to do this is on a load based dyno where you take the engine to each load cell and target efficiency. This is hard to do on the street.

I tune by changing timing along the logged RPM and load curve. I then smooth cells around that area such that timing goes up with RPM and down with increasing load but I avoid going too far beyond what the logs show simply because I don't know how the motor responds in those areas without more extensive logging and dyno time.

[superskaterxes]

Originally Posted by Lex Remember that the ECU smooths timing transitions for you anyways. And that the main point of the timing table is not to look prettier than your girlfriend but rather to extract the most efficiency from the motor without knock. Sooo the proper way to do this is on a load based dyno where you take the engine to each load cell and target efficiency. This is hard to do on the street. I tune by changing timing along the logged RPM and load curve. I then smooth cells around that area such that timing goes up with RPM and down with increasing load but I avoid going too far beyond what the logs show simply because I don't know how the motor responds in those areas without more extensive logging and dyno time.

i basically already did this but i couldent see any positive aspects to all those sharp peaks and valleys in the graph so i did my best to smooth them without changing the base values too much.

something doesent sit right knowing im only tuning timing along the 3k column and 2.0L row lol

[NoLeftTurn]

I'm working on a spreadsheet that will assist in adjusting timing. It is based off of a 3rd order polynomial equation for the most part. Stay tuned.

[rfinkle2]

Originally Posted by NoLeftTurn I'm working on a spreadsheet that will assist in adjusting timing. It is based off of a 3rd order polynomial equation for the most part. Stay tuned.

Hah! I was working on a 4th order polynomial equation. LOL. J/K

[RichieRichness]

Holly hell great minds think alike. I too smoothed out my timing tables today, and tested on my way to school. I honestly couldn't tell a difference with only putting 15 miles on the map. I want to say it reduced some part throttle kr but will report back tomorrow.


Sent from my iPhone using Tapatalk

[socks]

the smooth table we use was rendered in CAD and then transformed into a table. there wasnt much to it, as it is just a starting point anyways.

[Tomas]

^ That looks like my psychiatrist's couch.
Or was it my gynecologist?

[RichieRichness]

Are you guys smoothing all Ign Tables? BTW Max A & B tables seem to be very mysterious...any further info on them?

[superskaterxes]

all your timing tables should be identical

i just made my max tables 5 deg higher then my normal ign table, both are identical otherwise.

[RichieRichness]

Originally Posted by superskaterxes all your timing tables should be identical i just made my max tables 5 deg higher then my normal ign table, both are identical otherwise.

I know that high throttle Ign tables should be the same, while low throttle/cl Ign tables should reflect eachother. Are you saying that the high throttle and low throttle Ign tables are all using the same values? If so why? I haven't been doing this.

[superskaterxes]

Originally Posted by RichieRichness I know that high throttle Ign tables should be the same, while low throttle/cl Ign tables should reflect eachother. Are you saying that the high throttle and low throttle Ign tables are all using the same values? If so why? I haven't been doing this.

every timing table should be the same (ie all 4)

theres no reason not to make them the same because your never gona be low throttle 5500, 1.85L or high throttle 2500 .54L


same with fueling, every single table, when overlapped on each other should be the same. this prevents any confusion for the ECU and helps keeps things consistent.

[SRTie4k]

Originally Posted by superskaterxes every timing table should be the same (ie all 4) theres no reason not to make them the same because your never gona be low throttle 5500, 1.85L or high throttle 2500 .54L same with fueling, every single table, when overlapped on each other should be the same. this prevents any confusion for the ECU and helps keeps things consistent.

It doesn't seem wise to make the knock tables the same as the no knock tables. Doesn't that mean that if you start receiving knock, the ECU won't remove timing?

[RichieRichness]

Originally Posted by SRTie4k It doesn't seem wise to make the knock tables the same as the no knock tables. Doesn't that mean that if you start receiving knock, the ECU won't remove timing?

It doesn't matter...ecu still pulls timing if it sees kr

[RichieRichness]

cool, makes sense. But what about the Max Ign tables? Has anyone smoothed it?

[RichieRichness]

I haven't seen a kr event above 1 in 3 months


Sent from my iPhone using Tapatalk

[RichieRichness]

Originally Posted by RichieRichness I haven't seen a kr event above 1 in 3 months Your tune needs a little work if you see that kind of kr. Sent from my iPhone using Tapatalk

Sent from my iPhone using Tapatalk

[silvapain]

I only worry about the timing values I hit when going WOT or in CL I don't waste time smoothing out parts of the tables I never use.


Tapadatass

[RichieRichness]

495 g/s... jeasus

[RichieRichness]

So I just loaded the following timing table and went out for a lunch drive. Feels a little bit smoother.
RichieRichness - Timing.PNG

I wish we had more info on the Max a,b tables.

[superskaterxes]

max a/b just needs to be higher in every area then your normal timing tables. that way u only have 1 table controlling ign

[RichieRichness]

Originally Posted by superskaterxes max a/b just needs to be higher in every area then your normal timing tables. that way u only have 1 table controlling ign

Therefore, we can use the same timing table for all 6 tables except bump up the max A/B tables with a few extra degrees ? (see pic)

BTW - have you guys noticed the MAX A/B Ign tables have an extra 3 rows of cells? (see pic) Maybe this is where the mystery timing values are coming from for those e85 people trying to increase gas milage. FYI -The Gen 2 ATR doesn't have these extra rows.

Max A&B - Smooth Ign Timing.PNG

[rfinkle2]

Originally Posted by RichieRichness Therefore, we can use the same timing table for all 6 tables except bump up the max A/B tables with a few extra degrees ? (see pic) BTW - have you guys noticed the MAX A/B Ign tables have an extra 3 rows of cells? (see pic) Maybe this is where the mystery timing values are coming from for those e85 people trying to increase gas milage. FYI -The Gen 2 ATR doesn't have these extra rows. Attachment 47108

That is an interesting point RichieRichness.

We have 3 rows more resolution, but in high load areas vs. low load areas.

http://www.mazdaspeedforum.org/forum/foru...d-3-atr-99304/

[RichieRichness]

I see my max timing hits 60 according to DH. However, the highest timing I'm targeting is 58. My best guess is that Cobb has not found some of the timing tables/rows...like PT Ignition?

[rfinkle2]

Originally Posted by RichieRichness I see my max timing hits 60 according to DH. However, the highest timing I'm targeting is 58. My best guess is that Cobb has not found some of the timing tables/rows...like PT Ignition?

They have found em' just not sharing em' LOL.

[djuosnteisn]

Originally Posted by silvapain I only worry about the timing values I hit when going WOT or in CL I don't waste time smoothing out parts of the tables I never use. Tapadatass

One of the biggest benefits of a smoother curve is it's easier to make adjustments. For example, when your smoothing out to higher and lower load cells surrounding the cell of interest, the stock values are kinda all over the place. It makes it more difficult the interpolate up and down and you wind up spending more time making it look some what reasonable. But when the curve is already smooth to start with, making tweaks and smoothing them around is super simple.

I've been running a brillo curve for quite a while now, on both cars, and have experienced no ill effects. Gas mileage is the same, "feel" is the same, and performance is the same. It's just easier to tune with.


I know that some peaks and valleys are to be expected when doing a thorough tune on an engine dyno, like mazda probably did, but those same peaks and valleys are contingent on the entire motor and tune as a complete system. We can't just change the fuel a lil here, and a lil there... and maybe "dink" with VVT a tad, and throttle tables... etc etc etc.... and expect that same timing table to be applicable anymore. IMO, a smoother table is probably a better approach, or at least an equally harmless one as no change at all. And for me, the benefits in "ease of tuning" is worth it.

[RichieRichness]

i concur

[mrQQ]

can you post that smooth table in numbers format?

ty

[BigjohnB20]

Originally Posted by mrQQ can you post that smooth table in numbers format? ty

At the top of the thread (post 3 from superskater) there is a link to another thread on this same topic. In the linked thread member noleftturn posts an excel file with the smoothed curve values. That is where I picked it up a long time ago.

Not sure why you PMd me yesterday asking for it. I don't care, just wondering how or why you got my username. Anyway, that is where I picked it up from

[ASHMS3]

Zombie thread rises from the dead.

[mrQQ]

Found your name on one of the similar threads

Thanks, I'll look for it.