Tuning the Z...

[Superdave]

So i figure i'll start 1 thread to keep all my problems in.. lol

I'm working on a street tune, got the BLM's all near 128 for normal street driving and overall the driveability is much better now..

The first problem that i have is very annoying.. in open loop on a cold start the car fires up fine, runs for a bit then just dies. Then i gotta do a clear flood to get it started again and hold the throttle so it stays running.. even holding it at 2K RPM it will eventually just flood out and die. After several restarts it will eventually run fine and idle on it's own.

I've attached a datalog of this happening.. check out line 171..


Any input is greately appreciated

Thanks,
Dave

[merlot566jka]

me too. same damn thing. cept i can hold it at 1300 and it will continue to idle.

you may need warmer plugs to keep it from fouling.

EDIT:

bpw spikes up really freakin high right before it dies. voltage changes too. dont know what to make of it tho

[Joseph Upson]

I should be at it again starting this weekend. Perhaps delaying closed loop onset will help and maybe some changes to the crank fuel table which varies with engine temp also possibly causing problems with start fuel as the temperature rises. I recall a thread where closed loop was disabled until a good base tune was acquired then reactivated for further fine tuning.

The base pulse width spike could be in response to the rpm drop causing the MAP value to increase as if the throttle was being opened. I know first hand how crucial the MAP value can get.

I suggested to another member having turning troubles that maybe we all should look into trying to get the idle TPS and MAP sensor readings as close as we can to what they should be stock to try and help the situation because immediately on startup there is a MAP signal that does not corraspond with what TPS and RPM values should be at idle because of the camshafts we are running the one modification we all have in common which speed density engines don't do very well with if they are radical enough to cause a considerable drop in idle vacuum.

I've gone from around 18 inches of vacuum with the old engine to about 12 with the new and that's about the limit for low idle vacuum. I've discussed the idea of offsetting the MAP value with a variable resistor to lower the idle value a little to give the impression of a normal reading at idle, sort of like a fine tuning adjustment.

Perhaps a look at all of the individual tables involved with maintaining idle and when they are activated relative to open and closed loop will be a good place to start also since obviously the ecm is making a adjustments involving an input that is not active during open loop.

We should also be taking a look at the integrator, I read where someone set their minimum BLM to 128 and watched the INT, if it dropped low the fuel delivery is still rich and visa-versa. I show a BLM of nearly 128 steady but the INT often bottomed out at the 76 limit showing the engine is still rich at idle. Since the INT changes rapidly it gives a better representation of fueling status because my BLM didn't change much and I understand at some point it just holds steady regardless. You may have to cut ecm power and let it reset and compare the results to what you have now.

Another issue I have to consider is whether or not the hole I had to drill in the Northstar throttlebody for an IAC air passage is big enough since the adaptor plate was not designed to allow for IAC air flow into the engine.

I forgot to mention the trouble larger injectors and higher fuel pressure can add also. Here is a good read from Third Gen on 730 tuning;

http://www.thirdgen.org/techboard/tpi/43496-fuel-pressure-pulse-width.html

[Superdave]

you drilled a hole for the passage? I just marked and cut a bigass groove with the die grinder..

the new style IAC is very nice though, it's so quiet


I changed the closed loop timers back to the stock 3.1 values, i'll see in about 10 minutes if that helped. it's like 180 sec on cold iirc

[merlot566jka]

the 4.3 has the same hole in the throttle plate, yet its very very small. i did help some, but it needs more.

i opened up the back of the tb up so i dont need to make that port into the adaptor plate. its done like the iac on the 3.4 dohc tb.

maybe the iac's are different from what our ecms control?

[Superdave]

welp, i dumped the stock values in the open loop AFR tables then bumped up the idle RPM to 1400 around the startup coolant temps i've had.. she fires right up and idles like a dream. Althought the target AFR's are back in the 8's and 9's again.. guess i'll have to play with those tables a bit more.

i still had to pull some fuel out of the idle VE table, once it went into closed loop the idle BLM dropped to 124.. no biggie.

[SappySE107]

Why is your TPS at 2.34% at idle? If its not 0%, then its a problem so I can't tell you anything until I see a datalog with cold start up at 0% idle.

[Superdave]

that's my foot on the gas a little to keep the RPM's up...


Looks like i might have spoke too soon, i went back out to work on the idle VE and now it's acting up again.

I worked on getting the IAC counts to average around 30, i usually keep them higher but wanted to give this a try. It still does the same thing. All of a sudden it just freaks out and floods.

[IsaacHayes]

Crazy MAP spikes confusing it? LOL

[Superdave]

meh, i think i'm gonna put a regular heated o2 in the rear bank and let the ECM run off that.. just to take the wideband controller out of the picture.

[IsaacHayes]

Ahh yeah do that didn't know you were running a narrow sim from a wideband.

[SappySE107]

Send me your .bin. and I will try to sort it out for you.

[Joseph Upson]

I should be resuming my tuning effort on the 3900 today. I found in one of my GM manuals on fuel injection diagnosis, support for the suggestion that it is probably better to tune according to integrator changes instead of BLM values.

According to the manual, long term fuel change does not get involved until the system detects that short term is being overloaded. In other words your table has to be far enough off optimum for a long enough time before BLM changes get involved. On the other hand the integrator values representing short term correction changes rapidly so the closer the integrator remains to 128 the more precise the tune.

As a result I'm going to start by disabling closed loop and using my WBO2 for an initial tune since a lot of trouble tends to start when the ecm switches to closed loop, in an effort to have the fuel tables closer to a good tune before the ecm fully takes over fueling.

I'm also going to swap to an electric cruise control module and use it as a form of dyno tuning method by installing a switch that allows me to momentarily override the brake switch so that I can set cruise on the interstate for a speed that will maintain a corrasponding rpm on the fuel table like 2500 rpm for example and then slowly apply the brakes. The cruise control will try to hold that rpm and speed, at the same time the load will increase taking the TPS and MAP values into their upper ranges at that rpm for a more accurate status of fueling conditions in those areas that are often zipped past to quickly for an accurate reading.

That would provide for an excellent datalog because you will have a broad range of TPS and MAP values over roughly the same rpm permitting you to go across that particular row and nearly penpoint rich or lean areas.

Since 6th gear at 65 mph holds rpm at about 2000 in my car it should work very well requiring a drop to the next lower gear along with changes in cruise settings to get the upper rpm points.

The same method can also be used with timing advance for those with an emulator and ability to observe for knock retard while increasing timing in small increments

[pocket-rocket]

I'm also going to swap to an electric cruise control module and use it as a form of dyno tuning method by installing a switch that allows me to momentarily override the brake switch so that I can set cruise on the interstate for a speed that will maintain a corrasponding rpm on the fuel table like 2500 rpm for example and then slowly apply the brakes. The cruise control will try to hold that rpm and speed, at the same time the load will increase taking the TPS and MAP values into their upper ranges at that rpm for a more accurate status of fueling conditions in those areas that are often zipped past to quickly for an accurate reading.

That's a pretty good idea man Hope your brakes are cheap and you don't warp a rotor that way. Either way it would be worth it to get a good tune going.

[Joseph Upson]

That's a pretty good idea man Hope your brakes are cheap and you don't warp a rotor that way. Either way it would be worth it to get a good tune going.

It'll do little harm if any to the brakes since each rpm range tested for would only require about 5 second intervals worth of braking to allow for a a good duration over several points up until I reach 100 kpa considering cruise speed will be around 35-40 kpa. I may not even have to load it to 100 kpa by just getting close enough to extrapolate through that point.

I can also accomplish some of the same although not as effectively by accelerating in a high gear from a low speed. It's an idea that I believe will offer some benefit.