360S in Charleston(ish)

[Austin David]

FYI I manually checked timing, the crank sensor appears to be rotated 275*, where the ME map expected 265.  I also saw reference to 255 in other docs, so ... reminder to double-check what you think are sensible defaults.  

[Austin David]

Some minor catch-up to share my joy.

 

First, the good: I've been pulling a lot of logs and I can see (measured) 0.1s between throttle open and RPM leaping.  Throttle response is VERY fast at idle, where the original plenum had a noticeable lag.  The plenum lag is probably fixable with tune, but not doable on the locked factory ECU.

 

When it works (see below) it does work well.  Throttle response is what it should have been from the start, sound is great, and torque is stronger at very high RPM.  The stock 360 kinda started falling off at 6500-ish, this tune pulls through to 7k.  I have not tested, but there is an ignition cut programmed at 7400.

 

I also got a free* knock sensor, which does seem to work well by default.  Very little actual knock, but I've found a couple events including the one I looked for.

 

Learnings:

0) "sharp tools" are important.  In this case it means a laptop, software for log analysis, experts ready to help (remote tuner), and literal tools for repinning cables, properly making terminals, etc etc.

1) the ECU probably doesn't "matter".  If doing this again, I'd pick my tuner and hardware kit (intakes, fuel management, etc) THEN select an ECU recommended by the tuner.  That said, I am getting good support from overseas, though with a lot of overnight-waiting.

2) science is important.  This would be easier if I knew what the "baseline" 360S acted like: fuel pressure, AFR at RPM range, etc.  The "factory tune" might not be perfect, but with all this new control and measurement I am tempted to make the new tune perfect.  Hard to tell how "perfect" it needs to be, to be better than the factory tune

3) there are a lot of variables, and some things that can only be done once (like cutting a hood) and some things that can be re-done without too much cost, like making a custom harness or cutting an air filter backplate.  This would have been easier, and maybe 5% more expensive, if I'd planned some of these ahead.  

 

Some annoyance, which led to delays but probably not show-stoppers:

- alternator electrical noise... and while troubleshooting, seem to have led the ECU to dwelling on plug 2 which fried the driver.  Couple days of troubleshooting + offline exchange, but will be getting a new ECU (see above, excellent support from ME)

- maybe ignition noise, TBD.  infrequent "lost sync" events at ~ 4k RPM.  More data required

- to work around the fried driver, may (temporarily) run a traditional coilpack + wasted spark setup.  I assume I'll eventually go back to COP

- assuming the alternator noise persists, I'll go with a 1-wire "Denso" style option.  Appears to be a 90mm / "mini race" alternator, they run $150(ish) for 55A or ~75A.

[Austin David]

Some investigation about the size of the bonnet cut, which led to some investigation about heat and airflow

 

tl;dr:

• the intake hole may let hot air out of the engine bay, which will increase IAT 15-25C above ambient. 
• the air filter is sized for 300+ HP, and can be blocked off to prevent heat intrusion, without significant impact to intake performance.
• the actual bonnet penetration can be fairly small, provided excess heat from the cooling system has somewhere to go

Some offline investigation into runner length.  This ITB kit offers 70mm, 100mm, and 130mm intake runners, I currently have 100mm.  The manufacturer (or their associated engineering company) won't offer CAD, but do offer 30mm machined extension "tuning rings", with 50mm ID and M59 x 1.5 threads.  The process appears to be something like "add 30mm, see if that's what you want, repeat as necessary", then get custom runners and/or screw-on trumpets to match the required length.   Literature is thin; I THINK the math is linear with distance to the valve top, but the path is not straight and the tube is not a cylinder so changes may not be PERFECTLY linear, but I expect it to be close.  Experimentation underway with printed rings.

 

Peak VE is about 120% at 6500 RPM, with soft cut around 7400.  Shorter = peak at higher RPM, longer = peak at lower RPM.  The literal "tuning" here is setting the length from valve -> trumpet such that a standing wave bounces closed valve -> trumpet -> open valve, and forces its way in... hence the > 100% efficiency in a specific range.

 

Butt Dyno indicates very linear performance to soft cut, but I'd like to see if I can shift the curve down a little.  The attached model fits snug without trashing threads on my intake runner, and 20mm fits inside the air filter so I can run with the bonnet in place.  I also added +15mm outside the filter, which is about the minimum extension because of the thread pattern, and happens to just barely fit under my bonnet. 

tuning ring 15mm.stl

[Austin David]

Analyzing some data, the trumpet lengths appear to make a difference but it's subtle, and really only relevant at WOT.  I assume this would be a lot more effective when paired with cam changes, which I'm avoiding.

 

White = RPM

Yellow = "airflow est" from the ECU

Red, Green are AFR & AFR err.

 

Hypothesis is that length changes would shift the shape of the airflow graph.  I've reviewed about a half-dozen pulls from each and these are representative.

 

normal 100mm extension:

 

+20mm inside the filter:

 

+35mm total, with 15mm outside + 20mm inside the air filter:

 

note that the extensions are cylinders so I expect they would have a small net dampening of total airflow at WOT, so the before/after aren't perfect comparisons, but I do expect the 20 vs 35mm measures to be really good internal comparisons.

 

net/net: I think the graphs DO show a subtle lowering of the airflow / torque curve.  Butt dyno confirms, but the effect is SUBTLE.  Mostly at WOT I can feel it tapering off around 7k RPM, where the standard pulls smooth to the soft cut.  In midrange it still feels great both ways.  The airflow is also 35mm further out from the bonnet and the bonnet-hole would need to get a lot bigger.

[sltous]

I had some fun with the extreme version of your tape experiments if you want to take a look at the RAM air scoop data I had

 

 

[Austin David]

On 10/27/2024 at 3:38 PM, sltous said:

I had some fun with the extreme version of your tape experiments if you want to take a look at the RAM air scoop data I had

 

 

 

 

Great writeup, thanks!  There's definitely a lot more auditory feedback with so little between the intake valves.

[Austin David]

ITB update: heat soak in the injectors.  Symptom is a rising AFR (lean) at idle or after resting when hot.  Does not correspond to IAT, and drops with a few minutes running.  These are different injectors + fuel rail, positioned in the TBs, further out from the stock locations.  Heat is most likely conducting through the ITBs and into the injectors, and they're warming enough that they open slower than expected... less fuel -> higher AFR, enough that the ECU cannot compensate (it's limited to 20% correction).  When the throttle picks up and more fuel flows, the injectors cool off and the effect goes away (but returns with a few minutes idling at temp).

 

This appears to be due to the "dead head" fuel rail, which doesn't move any fuel (or soaked heat) at low duty.  I assume the stock rail also has this problem, and is somehow managed by the stock ECU.

 

The simplest solution is to bump VE around idle, which gives the ECU a little more room to compensate for the changing flow rate when hot or cold.  +20% VE and -20% startup tuning leaves the cold starts still correct, and the ECU then has plenty of room to trim out any changes with injector behavior.

 

A conventional return-style fuel rail would mostly hide this problem, as the fuel gets a chance to cool off on the way to/from the tank.  I'll put this near the bottom of my list for next time I get bored, below "rewire the whole engine".