TD_D mod journey. From stock to the 'bastardmobile'

[Td_d]

Wow, very nice car and incredible progression. :shock:

:tup:

Thanks man! Yeah, I've totally geeked out on the tuning side of this since I started this journey with Holy helping me out. My STI owns my soul now :tard:

 

[R3V]

Thanks man! Yeah, I've totally geeked out on the tuning side of this since I started this journey with Holy helping me out. My STI owns my soul now :tard:

In other words, living the dream eh? :lol:

I'm hoping to be able to tune my own car within the next year, but even then it will be with heavy help from various people.

 

[Td_d]

In other words, living the dream eh? :lol:

Chuckle... there are days when I think it would be nice to step in my car, and just drive it without instantly thinking of 101 variables And then there are days when I feel like driving it off a cliff...

And then of course there are those days when I give Porsche's and M series' unpleasant surprises

I'm hoping to be able to tune my own car within the next year, but even then it will be with heavy help from various people.

It's a steep learning curve, but I found beyond a point, you start seeing the bigger picture, and the less overwhelmed you feel, the quicker you learn.

 

[R3V]

Chuckle... there are days when I think it would be nice to step in my car, and just drive it without instantly thinking of 101 variables And then there are days when I feel like driving it off a cliff...

And then of course there are those days when I give Porsche's and M series' unpleasant surprises

It's a steep learning curve, but I found beyond a point, you start seeing the bigger picture, and the less overwhelmed you feel, the quicker you learn.

I've already had the first two (suspension issues :lol, I'm eagerly awaiting that 3rd option though! When I first got my STI, I went in assuming I knew absolutely nothing. That tactic worked well since it allowed me to actually listen/read without my own BS holding me back. I've since applied that approach to every aspect of my life and have found enormous improvement. :tup: I've learned so much so far, I'm not at all nervous for learning more. I'm more excited and eager than anything, that being said the only reason I even set time goals is to push myself. I know there are a million and one mistakes in my future, but that's all part of the process. Once I'm over the major tuning "hump" I'll be extremely proud of myself, which is all that really matters anyway.

Expect to see me poking around :tard:

 

[Td_d]

Ask, ask away! There's some good heads on here - and you've definitely got the right approach - I'd like to think I followed a similar path.

I guess as a general rule in life, if you don't let your ego get in the way, you're more likely to actually progress :tup:

 

[R3V]

Ask, ask away! There's some good heads on here - and you've definitely got the right approach - I'd like to think I followed a similar path.

I guess as a general rule in life, if you don't let your ego get in the way, you're more likely to actually progress :tup:

Totally agree, I learned that bit the hard way.

 

[HolyCrapItsFast]

Hehehe... [Drumroll]!



Alright - I am now officially running a speed density tune

Using the spreadsheet tool that's up on the Romraider forums, with a couple of tweaks of my own, managed to get a workable (well, actually a pretty decent) SD map to load on and ran it, with a couple of iterations and edits today.

I left closed loop switched on (Cobb also recommends doing this) as it leaves you with a bit of a safety net in terms of heat soak with the IAT in the cold air intake. Zeroed the load compensation table completely (which I am loath to do, as I have major resonance issues) and ran it. Open loop is remarkably close in terms of tracking the fueling tables.

Closed loop is going to need a lot of attention. Trims are quite erratic - although I believe that also has to do with the zeroed load comp tables. My thinking is that the VE map is in a better position to deal with resonance, given that it is using the same input parameters as the problem - RPMs and MAP. And you can see it in the map as well - the areas where the load comp tables previously compensated have similar patterns. I'll make up my mind as I go along as to whether I revert to using the load comp tables strongly or only to smooth things out.

Idle was actually relatively stable, although there is a specific situation where I managed to get it to basically create a circular feedback loop - it swung from -7 to 7 and so on corrections. High IAT after heat soak, and a tap to accelerator and letting go (creating a peak, and then a trough). My gut feel is that the latencies are still not right. I'm running 1.05ms now, as opposed to 1ms which is the advertised deadtime at 14V. Looks like I'm going to have to test different values - but I'm thinking that the pulsing (which I can see in the AFR's on the wideband swinging wildly) is likely due to the injectors being slower than the advertised rate. We'll see. Frankly, I wish I had some input from other people running 1200cc five-O injectors... If any of the mechanics on here have encountered them, would greatly appreciate feedback.

Once I have a relatively stable VE map, which I need to make sure is done at a relatively similar atmospheric temprature (it's been damn hot lately...), I will need to proceed to the g/s vs IAT compensation tables.

This is going to be a lot of work - but I suspect worth it.

Allow me to interject an alternate reality :lol:

You shouldn't have to worry about resonances and compensations for load with speed density. At least not as much as you would for a MAF. SD relies on temp and pressure, not flow... so resonances don't come into play with this system. Pressure is pressure and it is not effected by things like flow profile. Additionally the compensations are meant to correct for g/s relative to RPM and MAP and since we are SD we don't care what our flow rate is. So what are we really correcting for.

You may still need to utilize the load comp tables but not for the reason you normally think. I would emphasize my focus on the VE table and adjust as best I can first. you might find that you need not touch the load comps at all once you have done it correctly.

Of course I'm totally speaking out of my ass right now because I have ever only done this on the AEM and the AEM does not have load comps... well they do... it is just different.

 

[Td_d]

^ that's what I was hoping - hence the focus on the VE table. I think the latencies, and tightening up the VE table is the way to go.

Also, I think I might need to watch the weather quite carefully, and choose a couple of days where it's cooler, and similar temp wise. The Iat's can really throw things out, and I need to get a good baseline to deal with IAT compensations.

I'll be getting a more recent version if the patch from Merp, with MAF blending, soon. So I'll have a backup plan if I really continue to have problems with CL.

 

[Td_d]

Thankfully, I actually saw a screen grab of the full table, by Cobb themselves actually. They've taken a linear approach - ie it's a g/s by IAT tables, so x amount in each column. I guess it's a good starting point...

 

[Td_d]

Yeeha! Car's loving SD at the top end, getting there on most of the map.

 

[orestes85]

some sort of pressure gauge? idk...never seen anything in a car that's measured in kilopascals. about 100-105 kpa = ~15psi though.

 

[Td_d]

1.91bar boosting

For whatever reason, boost also seems to hitting target easier. That's my EVC 6 boost controller.

 

[Td_d]

I'll be point blank: AFRs affect your boost greatly.

I should know that

I've generally been slightly under target (there's an offset feature on the EVC that allows you to adjust for environmental factors that affect target versus what you're actually hitting). I've mostly hit 186 top, never bothered to adjust further. Running SD, yesterday, it just climbed straight up to 1.91 bar (target is 1.9). And my top end is still slightly rich compared to target of 11.5~11.7 AFR.

 

[Td_d]

Think about it: you have to have the correct drive pressure to build boost. There is always a trade off with these cars when it comes to certain variables. It's why I'll stick in a general area when I touch mine, but I just give her what she likes.

Yeah - that's the approach I've been taking tuning speed density - blocking off load / rpm / map areas and fine tuning them. All I can say is on open loop, speed density is the bomb. The delivery is so linear, it's beautiful. I'm very slowly 'deriching' the top end, as I've had some rather wild swings before with serious temperature changes - it just feels awesome.

I'm making good headway in the closed loop, and the map is also beginning to look more smooth and consistent, so I believe I'm moving in the right direction. And I think I've traced the source of the pulsing.

It's the injector latencies - but not where I was looking. If you remember, I had low voltage some time back - which I thought I sorted out cleaning up the grounds. Well, it looks like my battery is taking a crap (it actually drained last weekend because I left the tactrix in for 2 days without driving) - so on a hunch, since I've had some issues with a lopey idle on SD, I logged battery voltage. Dammit - down to 12 ~ 11.9 volts again. So I tracked it closely with what was happening with fueling and map - and there's a direct relationship, with the latency tables as the link. The voltage is swinging from the 13's to 12V, moving the commanded latency a whole cell. Now - five-O only provides the 14V latencies - so the rest I extrapolated - and the 11.5 V cell is off. Since I had quite a bit of idle data in terms of AF learning at 12V, so I shifted that cell by the relevant %, and the pulsing mostly disappeared.

Swung past a battery place, tested did a quick voltage test on it, looks like it's borked, so going back tomorrow to have it replaced (they'll also test it properly, it was closing time).

Amazing how much crap a low battery can cause...

 

[HolyCrapItsFast]

TD... You can write the SD section in the next release of the Tuning Guide :lol:

 

[Td_d]

TD... You can write the SD section in the next release of the Tuning Guide :lol:

Deal! It's the least I can do...

 

[Td_d]

Replaced the battery this morning - it was indeed shot. I now have a gorgeous 14 - 14.3v, nice and steady.

Aluta continua! Will be interesting to see what changes with fresh battery - I suspect a lot of the erratic readings in terms of AFRs.

What I'm finding interesting is that at the same place - roughly 2800~2900 RPM where the load compensations needed to be bumped up using MAF scaling, has the same leaning out using SD.

 

[Td_d]

Interesting...maybe it has to do with valve overlap due to avcs...

That's what I'm thinking too - AVCS setting around that area in particular really messed with trims and knock in the past - which is why I have it very close to stock. Wondering if I should start playing with it a little, see the results.

 

[Td_d]

Amazing what a difference the new battery makes. My idle would randomly go lopey - completely gone with a steady voltage at ~13.7v. I have got to do a grounding kit after this experience...

My map is increasingly looking very smooth, as it should - looks like the battery was causing a lot of random noise with the AFRs. The only oddity is still the 2800 to 3000 mid load 'lean bump'. I'm going to reduce intake AVCS a little, and make it more linear to see the effect.

 

[Td_d]

Yeesh - just did a quick check under the bonnet, see what my meth levels were sitting at, and something caught my eye by the MAF housing - it's frigging cracked on both sides due to the extreme heat over the past couple of weeks! Good I'm on SD now...