Temp related knock
 

It's not news that our FA20 engines are sensitive to high temps and seem to be a bit knock prone. I wasn't sure where to post this so I started a new thread. Hopefully this will result in some good solutions in tuning software and/or hardware that will not cut into already lowish power and prevent loss of power due to high temps.

I have already taken some precautions to try to lessen knock in my own modified tune using the OFT and Rom Raider. Here are a couple of logs taken at the same road in different temperatures with the same exact setup. Logs start with a 3rd to 4th to 5th WOT pulls with FLKC and Knock Correction being logged.

In 75*f ambient temps:
http://datazap.me/u/solidone/stg1-20...2&zoom=268-590

As you can see, not a lot of corrections going on here. I'd actually consider this good, since the minor corrections is telling me I got the maximum amount of ignition advance in the tune where the computer is only dialing back a little bit in a small area. Not enough for the IAM to be cut back.

IN 100*f ambient temps:
http://datazap.me/u/solidone/stg-1-f...12&zoom=63-368

Here, with only the difference being the higher ambient and intake temps, you see the computer pulling more and more timing as I continue to up shift. I'm sure less power is being put down than the earlier log in cooler temps.

What this data tells me is that under conditions that we see at the track, in hotter than 100*f temps and more continuous aggressive up and down shifting the engine is likely putting down less and less power as the day and on track sessions go on. My butt dyno concurs. And the resulting IAM once I check the status after a few sessions on track also concur. Where do I go from here to try to reduce this degradation in performance without cutting into the already low output of this engine?

Edit: I'm just going to add a few more logs as the earlier 2 at different temps for my own record. :)

84*f ambient:
http://datazap.me/u/solidone/stg1-20...2&zoom=121-386

[*after uninstalling oil catch can from PCV vacuum line, fuel trims and afr's changed and more knock corrections logged without other changes*]
84*f ambient w/high IAT's (+100*f):
http://datazap.me/u/solidone/stg1-20...-12&zoom=8-308

86*f ambient: (12.6:1 afr)
http://datazap.me/u/solidone/stg1-20...12&zoom=52-347

73*f ambient:
http://datazap.me/u/solidone/stg1-20...12&zoom=27-299

You can alter the IAT/Temp compensation tables to pull some timing as temp increases, your still pulling timing but you generally come out ahead if you pull the timing and avoid the knock as the ECU is pretty aggressive in pulling timing once the knock is detected. If you preemptively pull say 1 degree via temp compensation it might avoid the ECU pulling 2 degrees when the ECU is left to detect and correct the knock.


As your on 91 octane for track days it would probably be worth paying some extra dollars for some octane booster or a drum of better fuel for the odd track day its cheap insurance.


below is the IAT/Temp tables you could just make the compensation more agressive by making it start earlier or increasing the values.

The second table I originally interpreted incorrectly. -100% means full compensation applied and 0% means no compensation applied from the prevoius table.

If I was going to the track i'd throw in some octane booster as well just for some added protection.

http://www.ft86club.com/forums/attac...1&d=1408588907

I'm actually considering this. Although I've asked some more experienced track folk like @CSGDavid about this. He seems to think octane boosters don't do shit. Not sure what his experience is with different octane boosters. While I recall reading some material showing certain octane boosters do actually do as advertised, I'm not one just take peoples word on some of these things. There's one sure way of finding out, I always say. :)

I would totally agree that better quality fuel is always better than octane booster, but its better than nothing.

Another alternative is to have a track map and pull some timing out at track loads and RPM. It's a safe option and then add in some protection with the IAT timing comp table. Better fuel is definitely the 1st option though.

On a side note, Steve I can't work out the RR 3D map above. I assume the -100% means 0% is applied (-70% is 30% etc), or is it the other way around?

Tdd conformed it works oposite to what I thought by bad

-100% in the compensation table means no compensation applied
0% means full compensation applied from activation table

The only reason I ask is that ECUtek table is basically a multiplier. This is the stock table:
https://dl.dropboxusercontent.com/u/...20Ignition.jpg


Makes it a little easier to understand IMO. Either that or the numbers in the RR table should be 100% and not -100%

hmmm so are we able to increase ignition advance in cooler temps (below 70*f) on these tables? Winter is right around the corner, maybe we can cash in some ponies with the lower temps while at the same time tune for knock resistance when it gets hot by tweaking these tables.

I think the answer to your question is yes, however do you really want to test it and how would you set it up? Where would your standard temp be and will the lowering of the IAT bring you any further knock resistance? You would need a totally knock free base to test it out on and be willing to induce knock across the whole range. Plus, you'd need to do multiple tests to ensure that no results are "one offs". I don't think the time/effort would be worth the extra degree in timing you may be able to get in there.....

FWIW, on our visit to Cosworth all the dyno figures were stated as at 55C..... potentially this is where most would set their tables up at and work from there.

@Shiv@Openflash can we get some input here? Im wondering this as well

Well, I don't have the luxury and access to a dyno located in a climate controlled temperature adjustable room. I was thinking of just making conservative changes to the existing tables and log what happens once the temperature cools. It still pretty hot in cali so I'll just be working on the hot side of the table and retarding the ignition. While I do that I will apply a conservative amount, like half a degree of advance to the cooler side at 68*f and below and see what happens when it starts to cool down. I know the existing ignition table works pretty well in 80*f and cooler temps, so that will be my starting point with the OTS tune provided by vishnu.

@steve99 @Shiv@Openflash Steve, this chart makes me think that the alterations you've made to the percentages are actually decreasing the amount of timing being pulled.

Correct I had it arround the wrong way :-(

It's definitely confusing though. We just need @Shiv@Openflash to chime in here so we don't blow anything up :bonk:

I will have a look in the next few days the amount of timing adjustment is very small and may be difficult to actually see in the logs

I don't think he is down with that. If you want to mess with the tunes you'd be doing it at your own risk. People get paid big money to mess with this stuff, if you catch my drift. Just try to understand the 'what why hows' before you start changing stuff. Same goes with anything else. I do this because I like tinkering with this stuff and I have at least a basic understanding of how a car and a car's engine works. I'm learning a lot of this as I go and I fully understand/accept the risks. If you don't, then don't mess with it.

+1 on that were all learning and using open source definitions and software ie romraider, their are no guarantees here, i dont mind sharing what i discover but if you choose to use stuff guys post up here it at your own risk, like SolidONE im a tinkerer not a tuner some of these tables we alter have not been altered before ,dont expect shiv to help out when we are messing with stuff that he has not touched, just make small changes log it check the results and try again its all part of the fun and learning experience. :D

I understand the "what why hows", all I'm wondering is whether the percentages Steve changed in one romraider table are adding or subtracting timing, because it's not clear. Not asking for all of the secrets of the trade here. I do respect what he does though
and understand if he doesn't want to help.

the more i look at it after seeing the ecutek version i have the % table arse about,

That's exactly my plan, after I do some data logging in the next 48 hours to see if it is indeed temp related knock. I've already pulled .35 in quite a big chunk of places, if it's not temp related I'll be pulling .35 more where it needs it. Will share my findings, hopefully it will guide some people using socal 91 gas.

Mate that was a stock A01G ROM in ECUtek, open the same in RomRaider and post the table. That should be an easy confirmation ;)


*EDIT* - Just checked a B01C in RR and it has -100% where ECUtek is 0 and 0% where ECUtek is 1.

This is stock tables in A01G

http://www.ft86club.com/forums/attac...1&d=1408588907

So RED would mean more compensation (more timing pulled)?

Nicely confirms my idea, RR is a bit arse about face. -100% is 0 correction applied.

Yes, spot on. 0% is full correction applied.

Okay thats what I thought. I was looking at the -100% thinking hmm..why would there be more timing pulled at much lower engine load.

Looks pretty convincing , @Td-d could you look at the scaling on that table for temp compensation please

Guys - I know Ecutek is awesome and all, but this scaling has been on the IAT activation tables used on Subarus for time immemorial. Maybe I'm just used to reading it, but -100% makes sense to me - 100% of the compensation is removed, in other words no compensation is applied. 0 means none of the compensation is removed.

Tomato / tomato.

P.S. I'm tired and cranky ;)

They both make sense if you know how to read them ;) Even if the scaling isn't changed, at least some might now realise. I didn't look into the RR definition to see whether it was made clear.

Yes had it round the wrong way corrected my stuffups thanks guys.

Interesting their is no timing compensation applied for Engine Coolant temperature.

Might be a good failsafe to pull a bunch of timing if the coolant gets too hot at least you would feel the car loose power and know something was up.

Hey @steve99 I'm curious if you've done any changes to the tip in retard and tip in enrichment tables. From the looks of my 3rd-4th-5th logs it really look like the amount of knock correction applied by the computer is effected by both temperatures AND tip in knock.

The tune that I'm on in these logs on the first post already has a bit of tip in retard and enrichment added. I'm thinking maybe the tip in retard needs to be increased in relation to the IAT as well.

Have not adjusted tip in retard or enrichment. by running our 100Ron fuel and a couple of tweeks to temp and Knock correction timing I have almost zero knock now. Just waiting for the hot weather to return to check at high temps.

Has someone been able to confirm that you can put positive numbers in the IAT compensation table and have a resulting advance in timing? I'm hoping there is no absolute value computation in the formula which would result in less timing regardless of sign.

With cooler temperatures arriving now that it is fall, I am playing around with increases in base timing. I'm hoping to see a knock free increase in timing develop below 10C. I can see this is likely a tough balance given MAF loads increase with cooler denser air as well.

Haven 't had a chance to test if it will add advance yet. It's still hot like the devil's butt crack around these parts.

I wish, take a look at my forecast, it's colder than a nun's **** here this week. http://weather.gc.ca/city/pages/sk-32_metric_e.html

Looks like I'll have that answer soon.

I accidently put a positive value in the table iat vs temp and it did not upset anything, unfortunatly it was winter and iat did not get to the trigger temp so i carnt tell you if it worked but it certianlty did not cause any problem.

It is odd, all the other roms start pulling timing heavily beyond 230F.

thanks
might check out a wrx rom and see what they do seems like a worthwhile addition.