SR20 individual cylinder AFR tuning..

[600supersport]

Just curious if anyone that has actual experience with lambda readings per each cylinder on the SR20 would care to share any info. Ricer math says cylinder #3 could use maybe 3-7% of extra fuel and #2 & #4 a little less, but still extra. Obviously every setup will be different and keeping overall block/head/trans temps under control is important, but EGT sensors are slow to react and reading spark plugs isn't as precisely quantifiable as AFR...maybe you've done a lot of testing with individual cylinder EGTs on the SR and I certainly wouldn't turn down your findings either...just curious

[fatduece]

Mate, you're out of luck. Ain't nobody here can afford such lavish tunes and even if they could, they're too busy trying to achieve credit card fitment.

[600supersport]

I'm sure there's some decently funded teams out there doing it...too bad they don't bother to share with us mere earthlings. Some day I'll bite the bullet on another manifold and the AEM 4-channel setup and contribute something relatively newish here.

[fatduece]

Don't forget to report back!

[Croustibat]

"Ricer math says cylinder #3 could use maybe 3-7% of extra fuel and #2 & #4 a little less, but still extra. "

Why ?

[600supersport]

I don't have personal experience with individual cylinder EGTs (only ran one on #2 personally on my last setup) on the SR and haven't seen a whole bunch of variation in my own spark plugs, but others have observed it running warmer in the middle of the engine with bias toward the warm transmission too, fuel rail pressure drop-off at the rear and I'm just guessing intake manifold design may give a little more air to #3 as well so that's my idea of ricer math in this scenario. I'm just curious if anyone has done some testing with the 4-channel eugo setups yet on this platform.

[ultimateirving]

Do it do it. No cash to spend on car parts right now

[HybridAndy]

If you imaging a log intake manifold

_________

TB ==

|| || || ||

cylinders

You'd expect cylinders closest to the throttle body to get the least air and the cylinders at the end to get the most. Image spraying a water hose into that system, where would most of the water flow out? That's not a completely fair comparison but demonstrates the point. In normal operating conditions, it's a small difference and if your tune is sufficient to keep your "combined cylinder" AFR (IE post turbo O2 sensor) where you need and without detonation, there's no reason to tune individual cylinders.

The only people who would have this needs is if they run in a class where your forced to run certain parts or a restrictor plate and you need to extract the maximum possible from what you have (think a spec series where 2-3hp could make a difference if everyone has the same setup). You'll get much better gains and better return on investment somewhere else (more efficient turbo for your power level, cams, intake manifold ...).

I'd be curious if my hypothesis is correct (water in manifold thing) with some kind of scientific testing but with my experience working on Honda's, when a headgasket or piston failed, it's was usually the next to last or last cylinder from the TB. My assuming was those cylinder were running the leanest and also furthest from the incoming cooled water from the radiator (and a bad tune).

[600supersport]

That all definitely makes sense to me and I'd prioritize the money spent elsewhere to obviously since I haven't done it yet, but as you pointed out with the furthest cylinder results, one thing about seeing the average cylinder lambda being where you want it is that one or two cylinders may be leaner than you want, but the other 2 offset it into the average. It's very easy to account for individual cylinders fuel & timing in the tuning software, but real world lambda data on this platform would shed some more insight as a starting point window on other setups beyond what others have observed with EGTs.

[Kingtal0n]

Originally Posted by HybridAndy

The only people who would have this needs is if they run in a class where your forced to run certain parts or a restrictor plate and you need to extract the maximum possible from what you have (think a spec series where 2-3hp could make a difference if everyone has the same setup). You'll get much better gains and better return on investment somewhere else (more efficient turbo for your power level, cams, intake manifold ...).

yes!

My assuming was those cylinder were running the leanest and also furthest from the incoming cooled water from the radiator (and a bad tune).

sometimes a motor is setup to deliver coolant to the back of the head first, making it less prone to detonation in #4 even though that cyl may run leaner. In other words, Factors you do not see sometimes contribute or reverse the situation from what you would expect.

Also, acoustic tuning of intake runner length often has much more of an effect 5-10% of air mass ( the manufacturer knows that for a given range VE will be higher than it should be ) than a discrepancy in the design ( i.e. an extra 3% at the rear - you never see an OEM sr20det runner smaller or differently shaped in the rear, despite all the manufacturers testing/development which would show any "problem area")


Once you step up to "greddy style" intake manifolds the loss of traditional long-runner acoustic tuning leads to different designs, i.e. the funnel shape as the plenum approaches #4 with its really short runner design must have been what they found to be most effective (i.e. there would be some kind of testing/development before you buy a part like that, otherwise you should not buy it or trust it )

[600supersport]

If we keep talking about this long enough, someone will chime in with actual experience lol

[LoSt180]

This sounds similar to a problem with the G35/350Z where the first two cylinders got less air due to the shape of the intake manifold and ran rich. Most fixes involved changes to the manifold to balance out and provide better airflow.

Interesting stuff.

[Croustibat]

Originally Posted by 600supersport

If we keep talking about this long enough, someone will chime in with actual experience lol

We are on zilvia, the thread is about expensive and highly advanced tuning.

People who do that won't share the info ( first, because they don't post here, second because they don't want to give precious information for free).

[brndck]

I thought Bill Washburn from DentSport ran a setup for individual cylinder tuning. He's into some pretty advanced shit.

[600supersport]

I know most don't bother sharing info like this here, but it's not THAT expensive, not THAT advanced and not the secret key to a professional tune...

[brndck]

there was an article a while ago, where iirc bisimoto tuned a turbo s2000 with aem individual wideband sensors, and they picked up like 40whp just by balancing out each cylinder; leaning out the overly rich ones and richening up the lean one(s). I'd say that's worth the cost right there.

[600supersport]

Cool, I'll have to look that one up, though I doubt most setups will gain that much power if any at all, just better overall calibration since we're making a lot more than stock figures.

[brndck]

this is not the article I was thinking of, but similar. they picked up 17hp/16ftlb on an NA s2000
http://www.superstreetonline.com/how...-series-2-ems/

[brndck]

this is not the article I was thinking of, but similar. they picked up 17hp/16ftlb on an NA s2000
http://www.superstreetonline.com/how...-series-2-ems/

[Croustibat]

Originally Posted by 600supersport

I know most don't bother sharing info like this here, but it's not THAT expensive, not THAT advanced and not the secret key to a professional tune...

4 wide afr sensors, 4 EGT sensors, harness, an exhaust header custom made for the sensors, ECU capable of commanding injectors separately, software to datalog all that, countless hours for tuning, YES that is way more expensive than an AFR sensor in the elbow and a nistune. I'd say that setup is easily worth 4000$ more.

On a similar budget, you'd either get a perfectly tuned T28 based engine with like 300WHP, or you'd get a twinscroll EFR turbo with a standard custom tune and 450WHP.
Sure, it might be less efficient. But it will have both less lag and more HP, even if a 4 AFR/EGT could net more power from the same setup.

I know which one i'd get.

It is not much more complicated to tune, sure, but it really is way more expensive and longer to do, and unless you NEED these last HPs in a competitive class, it is absolutely useless. My tuner is getting around 700HP out of a ca18det with e85, and he is doing that with a simple nistune, 1 AFR and 1 EGT sensor (he just blew his z32 gearbox and went Z33)


BTW that "S2000 example" is worth crap, the HP gain has nothing to do with balancing each cylinder, it has to do with going from a mail order tune to a custom made tune ...

[600supersport]

Croustibat, this isn't a discussion about why and who should use this lol especially if your tuning setup can't already handle analog inputs (or CAN) for datalogging like this, but I guess it's just another reason to use Nismotronic if nistune can't, but many others can too.

You absolutely do not need EGT probes to go with this. All of the 4 channel hardware is ~$1k plus 4 bungs welded into an existing manifold (if you can't weld) so maybe $300 more and an extra $100 hour on the dyno (not a 20 hours) so an absolute max of $2k because you already have a capable tuning setup. I'd get it done for about $1,300-1,500 max...people spend more on head work and wheels all the time lol.

Again this isn't to discuss who and why, I'm just curious if someone has actual usable input to share.

[HybridAndy]

The other issue is, anyone who's done this testing, probably isn't running the stock intake manifold or using an unmodified head so the values won't have much meaning for your package.

Honestly, this is the kind of tuning someone like Nissan would do when it's invested hundreds of thousands of dollars into a world challenge or touring car program and needs that extra 2-3hp and 1-2 mph on the straight so they can shove it in Honda or Volvo's face. Your probably talking about going from 88% to 89% duty cycle on individual injectors and making very little difference. They would also probably share this information with other factory backed teams running the same engine but wouldn't ever release it publicly.

That said, there are people who are more interested in efficiency than ultimate output and it's possible someone has done this testing. I have a Greddy EGT sensor tapped into one of the head shield bungs on my #2 cylinder but I just like having it to monitor the overall health of the engine (figured it would help if I ever had a failing fuel pump or something to warn me of climbing EGTs before the engine pops). I've worked on a few equal length Subaru manifolds before that have existing provisions for individual EGT sensors but I've never seen someone with them hooked up.

[Kingtal0n]

You have to be joking with these figures.

If I can make 500-600bhp with an $800 OEM longblock on an $140 single wideband sensor safely,

and others are producing 2000+WHP with the same $140 single wideband sensor,

I don't see any demand for such a product/project and infer that anyone using such a complicated mess is either rich and bored or just foolish, disgusting thread is disgusting, flow match a set of high quality injectors and call it a day

[fatduece]

I think the advantage of these tunes are in the added reliability. It's probably the reason why "built and tuned" motors don't last as long as OEM. The power gains are just a bonus.

[600supersport]

I'm glad AEM does "see any demand for such a product". The trick is compensating for the backpressure present pre-turbo or else you'd be able to just route 4 independent O2 sensors into your analog inputs and log that and tune according quite easily. It is nowhere near a necessity to making power and even reliably as I have and still haven't invested in one yet beyond my single O2 and EGT sensors, but I'd say it is the best way to maximize your true lambda since there's 4 combustion chambers, not one. Every mod that everyone does to every car obviously affects someone else's differently, but findings on my engine are something I'll take with a grain of salt...again, not here for the why and who lol but if you guys still wanna theorize, at least it hasn't been beaten to death a million times yet...

[RedtopTech]

Weld an o2 bung to each runner. Wear a leather glove and swap your sensor between runs. This is a cheap and easy way to ensure that your money and hardwork doesnt "go up in smoke" by tuning to the leanest cylinder.

[600supersport]

I'd be interested in the steady state too, but give it a rip and I'll read the shit out of your findings

[Croustibat]

using 4 AFR probes but no EGT is so stupid ... i can't ... even ...

Anyway, i got my ass in an S13 yesterday. Has only 1 AFR/EGT. 9000rpm, 2.2bar boost ... with the boost controller turned off. Can do up to 2.8bar boost. That thing hauls, you have no idea.

Guess what ? He went the "i'd rather get a bigger turbo than a fancy tuning system". He has been drag racing this engine for years now, he has blown turbos, transmissions, but the engine block ? Nope. Still there, no piston, rings, or rod change. Nothing.

So yes, this IS a discussion about when you should spent money and time on an overkill tuning setup, and when it is not worth the trouble, because you may have unlimited funds but most of us have to make a choice.

And frankly, you should not even talk about tuning an engine, let alone individual cylinder tuning, when you have no idea why an EGT probe is needed.

[600supersport]

You're still missing the point, but keep trying

[HybridAndy]

Just to throw a little fuel on the fire, I saw the following article on MotoIQ regarding their MkIV Supra with the following line.

The car ran great for several years until multiple fifth-gear passes one day caused cylinder #1 to eventually melt down at the piston. The culprit was determined to be not insufficient methanol reaching cylinder #1, due to the sudden bend of the factory lower intake runner. Therefore, the ignition timing was too high in that cylinder, given the available octane. The rest of the cylinders, which were getting methanol, were perfect, but they also had a much straighter shot through the factory intake runners into the intake valves.

http://www.motoiq.com/MagazineArticl...CEA-turbo.aspx

Just thought it was interesting.