Tuning: 3500, 5-speed, 1227730, nAst1, Fiero GT

[LZeppelin513]

So like I said above, I am still not seeing the variable that I need to change for cold starting compensation. If anyone could point me in the right direction I would appreciate it.

[Google Adsense]

[LZeppelin513]

The new narrowband I had bought and installed was faulty so I have been tuning only in open loop. I got a new narrowband yesterday and discovered that my wideband is reading at least a full point or two lean. I checked for exhaust leaks and found a small one around the flange so I adjusted it and fixed it but it did not affect the O2 readings. They are both on the same side but I am not sure what to trust. What is most likely the problem here?

The car was running extremely rich at first due to a wiring problem. Could this foul the wideband sensor and cause it to read way lean?

[LZeppelin513]

Turns out the wideband sensor was fouled. It didn't look too bad but I decided to bite the bullet and buy a new sensor anyway. I replaced the wideband O2 sensor and I think it is now reading correctly.

Roberts 3500 VE table for Nast1 is close so it runs decent.

I have spent a few days now getting my part throttle VE tables closer but I am having a weird idle problem. In open loop it idles nearly perfect. It does not idle correctly after it switches to closed loop AND warms up past around 180* (in closed loop below ~180* it seems to idle fine.)

How it acts when not idling properly: It idles perfect for a bit and then leans WAY out until it either stalls or hunts and recovers... then idles perfect for a bit more until again leaning way out.... repeats this. It sometimes (much more rare) does this when I am coasting as well.

I have tried searching through the parameters but am not sure which one is the culprit.

I will pick up my camera this week to get a vid and a couple pics of the engine bay now that it is starting to look and act like a normal car.

[LZeppelin513]

I fixed the idle problem. It was going into single fire mode each time it would go lean and stall so I slid KAPLL over to zero and now it does not ever enter single fire mode and the idle is perfect.

[LZeppelin513]

Time for an update:

A couple months ago the fuel pump went out on me and have been borrowing my dad's car since. I finally got a chance to do some work on mine. I put in a TRE fuel pump and then discovered that the larger current draw was causing a large voltage drop on the stock wires. So I upgraded the wire system for the fuel pump. Also, the exhaust work done from a shop was pretty bad. It was leaking at the flange and very uneven at the tips. I welded on some 3500 manifold flanges from British Car Conversions and made gaskets from some exhaust gasket material i got off summit. I also ended up redoing the whole system to make it fit and am very happy with how it came out. No more leak, I really like the flanges made by bcc. I kept busting the rubber mounts I had made so while I was at it I also re did the motor mounts. I used 2 poly bushings from here:http://www.ballisticfabrication.com/...ve_p_1453.html. Made a mount on each side of the motor and bought some poly transmission mounts from Rodney Dickman. The engine is not going anywhere anymore, its great. The engine vibration is more noticeable now but it doesn't bother me.

Thanks to Roberts Wideband patch I am tuning with a wideband and no leaky exhaust. So far everything is going very well. I don't have my camera with me so sorry for no pics yet. Will get some and a vid of the exhaust soon.

1) After the car has been ran for a while, warmed up, shut off, and let partially cool back down (like if I go get some groceries), it has a hard time starting again. According to the wideband it is lean at start up (is the wideband even accurate during start?). What table controls start up fuel relative to coolant temp?

2) Also, it runs much leaner than "target AFR" at idle when cold, what table will adjust that?

Thanks

[robertisaar]

widebands are NOT accurate right after startup. reason being that when a lot of fuel is being dumped in, a lot of it doesn't burn in the combustion chamber and can still be burning/unburnt as it passes the sensor. will cause wonky(technical term) readings.

with a correctly tuned VE table, then you can mess with that kind of stuff, since VE is taken into account for everything except for cranking fuel, which is done in raw BPW... something you may need to adjust anyways since your injector size relative to cylinder size is most definitely different than the LH0 your tune was likely originally setup for.

[LZeppelin513]

I spent some good time (ga$) tuning this weekend. Mostly part throttle tuning and a couple of low rpm WOT just to see AFR. I was doing it all *open* loop with the wideband. What is more accurate for part throttle tuning, *open* loop method or using BLMs to tune?

Also, at WOT 2000-3500 rpm my AFR stays between 10.5-11.5 (haven't touched PE yet) but I usually will get 1 knock count and a tad bit of KR but not at the the same RPM/load. That seems rich enough to not be getting any knock? Using 93 gas.

Edit: i misspoke, meant open loop said closed

[robertisaar]

closed loop IS BLM tuning... if you were in open loop, BLM and INT wouldn't move, meaning you would be watching only your wideband to see what kind of changes to make.

widebands can be fooled and they can lie to you.... so if all else fails, do plug cuts. those don't lie.... they may confuse and possibly draw you to the wrong conclusion, but the spark plugs will tell you what is really going on.

i've had people mention a "rich knock", which goes against all conventional wisdom, and that it cleared up once they ran a more realistic PE AFR. 10.5-11.5 is REALLY rich. i'd look more along the lines of 12:1 at richest, but that's just me. the higher octane fuel will certainly help.

if you want to undeniably determine if that's false knock or real, you would need to run even higher octane fuel. since you're already playing with 93, that's gonna get expensive real quick. as an alternative, you could drop down to 91/89 or even 87 and see if the knock still occurs in roughly the same situation and with the same intensity.

and 1 knock count isn't bad at all.... a lot of factory tunes do far worse with factory engines.

[LZeppelin513]

Thanks for the fast reply, I edited my last post. I wrote closed loop but meant open.

I'll look at my plugs and see if I can find anything. Any links to good info on reading plugs? I have zero experience looking at plugs and would imagine it to be tricky. In your experience, does wideband or CLOSED loop tuning tend to be more accurate?

Good to know about rich knock. Once I get my VE tables dead on and start playing with WOT more I'll shoot for leaner AFR.

[robertisaar]

you can get to a more accurate VE table faster using a WB, IMO... since waiting for the ECM to move the INT to the correct value to cause the O2 to swing around the stoich point can take too long to reach in the conditions that you want to play with. decel is a good example of this. you can let off the throttle quickly and if your DE and VE isn't correct, then you'll see the INT shoot off in one direction, then eventually move back toward 128 or move to a position where the O2 voltage will start swinging around stoich. can take a while though, longer than the amount of time you're in certain conditions at least.

reading plugs: Bruce Plecan comes to mind.

http://www.gearhead-efi.com/Fuel-Inj....P.-dedication