mrfred

iTrader: (50)

EDIT: My current latency values are in this post:

https://www.evolutionm.net/forums/sh...76#post5806676

For E85, I'm now using an injector scaling of 731 cc/min, and for pump gas, I'm using 1008 cc/min. Trims are typically within +/- 2% for either E85 or pump gas with these values.

----

I finally installed my FIC 1050s today. Had to do it twice. What a PITA. Had to do it twice because one of the injectors was frozen shut (it was the one that I ran water through to measure latency under pressure). Anyhow, I happened to have some extra FIC 1050s on-hand, so I switched out the bad one for a good one, and I was on my way.

- Here's what I found:

Dialing the scaling and latency was pretty easy. Within a half an hour of tinkering, I had the fuel trims easily within +/- 7%. I think now with a little more tweaking they are within +/- 4%. Here are the values I'm using for now:

When I made a change to the latency, I usually changed all the values from 11.72 to 18.68 by the same amount. The latency values are quite different than what I measured on the benchtop (the road tuned values are about 0.25 ms larger), but its what it took to get the fuel trims near zero.

Driveability with the FIC 1050s is amazing. Its easily as smooth and as responsive as stock at light throttle. Might even be a little better. Idle is just a tad bit more wavy and choppier than stock, but not much. Perhaps with a little more tweaking of the latencies, I can get it a bit smoother at idle.

The one annoying that that's happening is that its taking quite a few cranks of the starter motor to get the engine to light up after its been sitting for more than 10 minutes. Gets worse the longer it sits. Is this common with big injectors on the stock ECU, or do I have a leaky injector that I need to replace (yet again, ugh)?

dude

iTrader: (12)

Try logging startup. I found that I was seeing about 85-90 Load (2 byte) on initial crank, and there wasn't nearly enough timing advance (rescaled load axis). I was having the same symptoms as you, and after bumping the timing from 0-1000rpm in the load ranges under 100 it helped greatly.

Another interesting note, it seems that lowering the 16 and 18 volt latency values to near stock values also helped startup, although I'm not quite sure as to why. I logged batt voltage and it never got anywhere near that high on startup, but for some odd reason it does help the car get started faster. Maybe it has something to do with the cold start enrichment tables recently found but not yet defined for my ecu (947100015)

RazorLab

iTrader: (8)

If it gets worse than longer it sits after being hot I would lean towards a leaky injector.

logic

iTrader: (2)

I had the exact same behavior with a fuel pressure regulator that wouldn't fully seal, so I'll ditto the other responses: leaking injector or seal (look for puddling fuel on the intake manifold, not all leaking is internal).

mrfred

iTrader: (50)

Definitely no external leaks because no gas smell or signs of puddling anywhere. I haven't put the clips on the injector plugs yet, so I think I'll make sure the problem still exists today, and then after it sits for a while, pull the injector plugs, crank the motor, reinstall the injector plugs, and then try again. It if fires right up, then I'd say its a leaky injector (or possibly fpr).

mrfred

iTrader: (50)

So, the engine lit up on first crank of the starter this morning on a stone cold motor. Perhaps it is a latency thing?

Pd1

iTrader: (38)

You may be flooding the motor during startup. Check to see what your voltage is during cranking when you're having the problem. I fixed one a few weeks ago with the same symptom by increasing the latency values around 9V -- which was where it was cranking.

I'm also a believer in increasing the resolution of the latency table. Feel free to add points of resolution and ditch the useless 4v and 18v entries.

dan l

iTrader: (3)

I find out what % I am off on my LOW fuel trim. Say I'm 10% lean. I will then add 10% to all my STOCK latency values. The theory is that since the stock injectors use a coil just like after market ones the main difference is that you need more latency everywhere. If you just add a static ms of latency you won't have enough latency at 10v which is about what the battery voltage is when starting. The car will be lean and doesen't start cleanly. If you add a % amount you'll notice that the bigger latencies (low voltage) are growing larger and the ones at 14v (where your trims are usually updating at) are changed by only the desired amount.

So basically, I can't describe it very well, but just add latency to the STOCK values by a multiplying it by a certain percentage not a base ms value.

Pd1

iTrader: (38)

It's possible that the problem is a leaky injector, because it would exhibit the same symptom. Imagine the pressurized fuel system leaking just a tiny bit of fuel into the combustion chamber every few seconds after the engine is shut off. Burning it off will take a few cranks, and leaving it sit for hours will allow it to evaporate and start as if there's no problem.

As I said a moment ago, this could also be the result of imperfect latency values at the crank voltage level. You may find that the crank voltage 10 minutes after shut-off is different from the crank voltage first thing in the morning.

mrfred

iTrader: (50)

Thanks guys, I think I'll try tinkering with the latencies a bit before doing another R&R. I'm running pretty high values at the lower voltages, so I may try lowering the values a bit.

scheides

iTrader: (10)

Good deal mrfred! I'm about to put mine in as well, I'll post the values once they're dialed in

tkklemann

iTrader: (6)

I had this problem on a car last Sunday, and have been trying to tune it ever since. He just told me this past Sunday that his Turbo outlet gasket was half MIA. When I looked down into the engine bay, sure enough, it was half sticking out and half torn apart. Damn Mitsu, they were the last to touch it. I just installed RC1000's on this car, and it seemed, just like mine did, that even the smallest boost leaks are amplified with big injectors. I had a small leak on mine when I originally installed my FIC1050's, and couldn't figure out why I couldn't get the scaling nailed down. Eventually I found that I had a small boost leak, fixed it, and after that all was good.

Your Latencies are a bit different than mine...

93 Scaling is 1083, E85 is 713

Latencies are:

3.912
2.304
1.512
1.152
0.816
0.6
0.456



I have noticed that on 93, the fuel trim "swings" are worse on 93, but they always center around 0, it just depends on the weather. On E85, they don't move around nearly as much.

deanner

Put a fuel pressure guage on it and see how long it takes to bleed off, thats how we check hondas, should hold....

mrfred

iTrader: (50)

I may have to do that if I can't fix it with latency adjustments.

mrfred

iTrader: (50)

Your latency values were pretty close to what I measured in the lab, so I started with yours. However, I found that the STFT would max out (24.80%) almost immediately (like within 15 seconds) during idle after startup (whether hot or cold motor). Increasing latencies was the only way to bring it down. Then I had to bring down the scaling to get a proper trim during cruise. I've checked the LTFTs twice in the last two day daily driving, and the idle trim was exactly 0 both times, and the cruise trim was about -0.2. Pretty amazingly close to zero, but as you said, I'm sure there will be some variability with weather.

I must say that I am surprised at the variability of results that people are getting for the same injector.