When your engine is running, you would want to have the smallest ground loop to the power source -- which is the alternator (unless your engine runs without one), which is why it is recommended to have the grounds connected on the engine block/cylinder head.
Smaller ground loop means minimal chances that other accessory loads would be having to share the same ground (having more resistance) to the power source. More grounds, less resistance. Then there is less chance the ECU would be picking electromagnetic interference from ignition system.
If you have the ECU grounds connected to battery negative, it would still have to share the ground wire from the battery negative to engine block with other accessories.
Note as others have said above, in modeled mode the fuel table has nothing to do with injector PW or duty cycle, it is the volumetric efficiency, or in other words it is a measure of the air flow.
So if you have a value of 100% in the fuel table it means your 2320cc engine will inhale 2320cc of air in one engine cycle. If you have a VE of 150% it means your 2320cc engine inhaled 3480cc of air in one cycle (obviously impossible).
Typically in modeled mode you will have a range of numbers from about 50-60% at idle to 100-120 at peak efficiency. The areas in your VE table where the numbers are 20-40 is probably just because your injectors are working in the non-linear region and you dont have proper characterization data to correct it. The areas where the VE is un-realistically high is possibly low fuel pressure or incorrect injector flow rate.
It would be nice to be able to bound the max and min duty for idle control (3 wire pwm in particular) to vary either by some amount above and below the base position, or by a coolant or possibly other axis variable. It sometimes has been a bit of a hassle to get the idle, return to idle, startup, etc. all to work as smooth as I would like (granted most of the cars I'm using these on are 30 +/- years old now), and it just seems like this would aid in dialing this in a little quicker.
My other car has been off the road for ages, and life has been in the way.
So I thought it might be fun to just wire up my ECU into the Echo so I could have a play around with some tuning stuff again.
It's a 1300cc economy engine so not going to set the world on fire. But a bit of fun.
I need/want to revert it all to factory though, so I bought a spare ECU from wreckers and cut the plugs out to make a patch loom.
This car has VVTI on the intake side, which means I could do my usual sweep through the vvti angles and see which delivers the most airflow into the engine.
Results were unexpected:
(Colour scale is cam angle, blue is 0 degrees advance)
The engine pretty much hated having any cam advance at all over say 3000rpm.
As opposed to my other engine where it likes lots of advance right to say 6500rpm then tapers off.
I also noticed that the mass air flow rate just absolutely flatlines when you got to around 4000rpm.
I didnt have a map sensor connected, but my suspicions were that it relates to the absolutely tiny throttle body, and the super long skinny runner on the intake manifold.
There didnt look to be a particularly easy way to fit a bigger throttle body, as the whole manifold is plastic and there's not much scope for boring it out.
So looking in the parts bin, I had some silvertop 4AGE throttles in the garage.
I have a 3D printer that could bang together a usable prototype so after a few adjustments ended up with this.
Then printed some trumpets to available clearance.
Everything fit up surprisingly easily, the most expensive part of the swap (of parts I didnt already have) was just buying a radiator hose so I could move the filler cap a little further over.
Results - were amazing!
Instead of flatlining at 4000rpm, it just kept going up. I'd keep revving it a bit higher to help mitigate the crummy gearing from 1st to 2nd, but it starts getting valve bounce at around 7200rpm.
Also, now since the pressure differential from intake side to exhaust side has lessened (less intake vacuum at WOT) the engine now likes having more cam advance to make use of some scavenging.
Virtual dyno showing really good gains, and the difference in the fuel map backed this up too.
I took this car to the drags last year for a laugh, and when standard it ran a 17.6 @ 77mph with a 2.5 sec 60foot.
So I wanted to try beat this as best I could, so setup a basic launch control using just rpm limiter with vehicle speed as a load axis
I figure I can tweak the 0kph limit up or down to set how much initial wheelspin it has, and then tweak the decay rate to calm down any excessive wheelspin as it gains speed.
Some testing was looking promising but I also havent really been in the mood to deal with a blown up clutch so I didnt spend too much time on it.
The printed manifold got some reinforcing just in case.
Come to drag day.
I think this launch control method would have worked well if I was doing a burnout to scuff the tyres. But since I was clutch concious I was just rolling through. So the tyres would get wet to a varying degree and also the rubber debris you pick up varies from run to run. So I found it was a bit inconsistent.
None the less, managed to improve the PB significantly to 16.4 @ 83mph. With a slower 60ft at 2.6 seconds
So possibly even quicker however I only completed a minimal amount of runs for sake of clutch preservation.
(I've owned this car since 30,000km, it's now on 195,000km with original clutch)
It's actually been heaps of fun, and it's a completely different car to drive with the better intake.
I've got a gearbox ready to fit with better final drive ratio and gearing from the RS model, and a torsen LSD to fit in it.
I'll also chuck the 1500cc motor in at some point, I'd imagine it would dip into the 15 second bracket pretty easy with that combo.
And it's great because all of the parts are hilariously cheap. If I could wring 100hp per litre out of the 1500cc setup it would be pretty hilarious to drive, as it's only around 800kg. Still not the same power to weight ratio as my other car, however, would be a heap of fun to do some autocross/hillclimbs/small track stuff with.
It seems this is quite variable in my experience, possibly depending on size of throttle and plenum volume etc. I will attach a few example settings from cars I have tuned but you will notice they are hugely different. The LS4 needs 7% dashpot throttle, whereas my evo needs no dash pot at all...
To tune it set up a time plot with RPM and idle target one the same plot like pic below. The idea is you want the RPM to drop down to target as fast as possible without significant undershoot.
Here is our LS4 race car with our default base map dash pot settings, notice the big dip under target then a big oscillation after the near stall:
I finally got to the bottom of the issue. The meter movement was the cause of the needle rising when power would be applied to the tach controller. I think in the process of me replacing the signal resistor I bumped meter movement out of calibration. I made a complete video on how to fix it and how to adjust the tachometer range:
https://youtu.be/y2TRMWldaQA
IAT Fuel correction should be turned off for modelled fuel equation. The air density change due to charge temperature is already taken care of in the fuel model.
essb00 got a reaction from JeremiahJ in Is it necessary to connect all GND (black) grounds?
