S14 SR20 starting tuning after 550ccs, basic questions

[MomentFD]

(tldr & real questions below if you want to skip my intro story)

Hey guys, finally getting my hands dirty with tuning and it's pretty addicting, especially with the LM-1 wideband log going at the same time as making changes. My SR20 is pretty stock, it's got all the bolt ons and a set of 550cc injectors, still on the stock MAF.

Initial logs with the setup were very very rich, getting to AFRs in the 8s. This was with a Nistune-recalculated k of ~22500. Stock was 33100. I tried to verify this by turning off O2 feedback and watch the AFRs while cruising constant throttle at 3k rpm. They seemed to jump a little but it was okay around here. Next I thought the map could be rich, so I compared my map to a 550cc SR20 map posted by PL and it was almost double mine!

Mine:


PLs:


So here is when I figure alright I'll just slowly make mine like his, watch the AFRs to make sure they're sane, and I'll be in the money. When my table was identical, I was still pretty rich. I revisited the K constant and dropped it to about 20200 which kept my AFRs still 14.7-15. Thing is, it was still rich, so I flattened out my table some more. I was able to get AFRs to about 11 range which felt a lot better, bu I decided I need to ask some questions before proceeding.



1 - first of all, which some of you are probably thinking, I need to boost leak test. I've been driving/lightly tracking the car for the last year and I don't THINK I have a leak but I will definitely confirm this asap.

The real questions:
K constant vs fuel map - what are the differences of reducing k constant 5%, or doing a -5% change across the entire fuel map? I ask because in the map I ended up with was getting very low to raw value 0 - the minimum. I'd much rather have a lower k value and richen the table up for more resolution, than using 0-20 points of adjustment. Or am I looking at this wrong?

Verifying K - What's the best way to do this? I turned off O2 feedback and cruised on very flat road ~3k rpm while watching the LM-1 O2 log. Going back to the question before, how do you know what your fuel map values should be at this point? It seems like you could arbitrarily pick these 2 to work together. Now that I asked the question, I'm thinking it should be near 0?

I know the recommended course is doing the Z32 MAF at the same time, which I have on a shelf, but I don't have the ability to fab an intake at the moment, so I'm working on tuning. It shouldn't be a problem though when we get into it...

Anyways thanks for reading and any feedback. Also PL I'd love to hear what the k value was for that tune, although I know there is a huge scale difference in different ECUs K's.

[unex]

1 - first of all, which some of you are probably thinking, I need to boost leak test. I've been driving/lightly tracking the car for the last year and I don't THINK I have a leak but I will definitely confirm this asap.

no, we are not thinking about boost leak, because if you are doing tuning, we hope you already done boost leak. This isnt a question do you have boost leak or not. The question is HOW MUCH boost leak do you have. In most cases
And one day somebody should say this (how to get close to correct K):
1. turn fuel table to show target afrs.
2. make for example 12 afr all the table. I mean ALL.
3. Drive and adjust K till you get close to 12. The more load you can do (if you have safe ign table) the more accurate K will be (but doesnt mean max load). I would say it is nice to make 50-70% load to get correct K.
4. Somewhere in forum PL posted how to fine tune K and/or latency with o2 feedback, but in most cases I am happy with these 3 steps, because you cant get straight afr line because of not accurate MAF table (or differently workign MAFs).
Happy tuning

[MomentFD]

I'll do those steps, I like the setting of the table to a constant. I was wrong that the table/K can be changed similarly, I realize that the values in the fuel map are aiming for a specific AFR and the K needs to be set right for that to happen.

For the load testing, any tricks there? Flat road, constant throttle, and you're saying try to get up to 70% of the load scale?

[unex]

For the load testing, any tricks there? Flat road, constant throttle, and you're saying try to get up to 70% of the load scale?

this is not so much important. The point is to make some load for MAF, to get it more accurate, but it doesnt need to reach MAF limits. Middle or a little bit higher MAF load is OK to get K. Anyway your AFRs will float, its impossible to make it straight line.
After you done this, you can adjust your injector latency on idle, to make AFRs to match fuel table AFRs

[MomentFD]

Gotcha, I can do this. Once I get this dialed in, do we have a sticky for peoples maps/tunes where I can poke around?

[PL]

Just had a look at about 6 different S14/Z32/550 tunes I've done recently and K constant varies between 32,000 and 36,000. I use 34,000 for the starting point in my S14/Z32/550 base tune.

If you had to adjust K outside of this range I'd be looking for a problem. Check fuel pressure first.

Values in the fuel table are all in the range of 35 to 45 in the high load area.

Regards
PL

[MomentFD]

Good info PL, my K isn't there yet because I'm on 550s but not the Z32 yet. I should be able to tune the fuel map and just adjust K as needed when I get the Z32 in, correct? I know putting injectors/MAF in at the same time is recommended, I'm just trying to get out of the 8s for AFRs though until I can get my Z32 in.

[PL]

Yeah you definitely wanna be doing AFM and injectors at the same time otherwise you're doubling your tuning effort. But if you're just starting out then it's all good practice!

PL

[MomentFD]

All the double work will be dialing in K twice, right? If the motor is setup well, then a proper change in K after a hardware change should result in similar AFR behavior if the fuel table is held constant? Does that happen much in practice?

happy weekend chaps!

[PL]

Never seen it happen when changing AFM. Injectors are usually fine - if you've tuned for 550's then fitting 740's is usually pretty
straightforward. Often no changes to fuel/IGN maps.

If it was tuned properly to start with then to change injectors is often little more than:

- Use Injector resize dialog (tick boxes to auto adjust load scales etc.)
- Tweak K and latency as reqd to get things spot on again
- Tweak load scales to get things tracing exactly where you want (most important for IGN map)

But changing AFM tends to take more effort - requiring changes to fuel map values and often messing about with other secondary maps/tables.

But the main reason to do both AFM and injectors at the same time is that they tend to cancel each other out. A larger AFM needs a higher K constant. Larger injectors need a lower K constant. Certain combinations (S14 SR20DET - fitting Z32/550's) almost cancel each other perfectly and K finishes up back close to standard. This is the best possible result as everything then falls into place with minimal effort. The amount of tuning required is then similar to doing an engine with stock injectors and AFM - largely just tuning in fuel/IGN maps and adjusting load scales to allow for whatever boost you're running.

Any combination that results in K constant being thrown out significantly will need more effort to tune. Things to avoid are fitting Z32 AFM's to otherwise stock engines (it's usually a waste of time anyway because injectors/turbo will max out well before the Z32 AFM). Same with injectors - fitting 740's to an otherwise stock engine will usually throw K out by a factor of around two. Which will take considerable effort to tune.

And stuff like using the really big HP slot-style AFM's in 4" tubes and the like with stock injectors is just silly. But balance your huge AFM with big injectors and suddenly you'll find the world is a wonderful place! Combinations like R35 GTR/HPX AFM in 3.5 to 4" tube and 1000cc injectors on E85 can be surprisingly easy to tune. Just see where your K constant ends up - if it's getting close to standard then you can sit back and enjoy your tuning. If it's miles out then it may be worth re-considering your AFM choice. For slot style AFM's this can sometimes be achieved by changing the tube size. But don't get too carried away as editing the VQ map as part of the tune tends to take a long time. We're trying to have drop-down VQ maps available for a series of tube sizes but it's proving to be a challenge.

PL

[MomentFD]

PL, you're awesome, all that's crystal clear. It's easy to forget that while it should all work smooth any configuration, it doesn't always happen that way.

I'm gonna work on getting the Z32 MAF in and the car boost certified before digging in much more.

Is there a thread or page with people's tunes or do I just need to look around? Seeing some example maps has helped a lot, really explains why my 'stock' AFRs are so high.

[PL]

No worries. Glad to be of assistance.

Just drop me a line at home pl <at> plmsdevelopments.com if you need an example tune. If I have something suitable then I'll try to help. I mainly do SR stuff so I have some for them. Got a few decent RB tunes but not a lot.

Careful with Z32 AFM's - lots of China fakes out there. There's a thread somewhere with details. They're getting very good - from the outside it's hard to pick. But if you look down the middle the sensing element is quite different.

PL