i didn't accurately explain the symtpoms. in hindsight, i think i made the mistake of using the term "tip in," when the issue was really related to throttle angle. i could hold this slight throttle angle and the engine would continue to misbehave until i either let off or fed it more.

the TPS base setting did indeed change this. i guess the '94-'95 ecm doesn't accurately baseline the TPS below a certain threshold. with the closed TPS voltage at 0.88V, it now instantly shows open throttle in datalogs as soon as i touch the throttle.

that said, i tried something today, mostly for fun. i forced the ecm to stay in open loop all the time. fueling appears to be based on load (MAP) and MAF flow. it drives unbelievably well, better than the stock engine did with a HOT cam. i'm tempted to stay like this.

also, with just the MAF determining fueling, the wideband shows 15.5:1 to 16:1 a/f at idle. it sometimes seems like it's on the verge of trying to surge a bit, but then it corrects itself.

i'll drive it some more like this and try some wot wideband runs, then report back.

-michael

 

i can see TPS V directly, yes. the TPS % seems to correspond directly to TPS V, no matter how the base TPS V is adjusted.

btw, this is the oem TB. i had a 52mm bored one i tried on the dyno and gained 0hp, so went back to a stock one. the wideband does actually lag a bit when i whack the throttle open, but i can't tell if it's really momentarily lean or if the sensor just isn't responding quickly enough. i can't tell anything is wrong in the tune from how it responds to big throttle inputs, and no detonation. all the problems were related only to very small throttle inputs.

but, today i had the opportunity to air out longer than usual, and the ecm tripped a TPS V too high code. so fixing the transition from idle to cruise caused the TPS V with TB fully open to exceed limits. i can't find where to change the upper limit, only to turn the code off.

is it acceptable to just turn the code off? or does the high TPS V affect engine management?

the combination of longtubes, slightly different O2 sensor locations, and the overlap definitely makes for some funky corrections in closed loop idle. i can really see how much worse this could be without the idle air distribution offered by the oem passages, yes.

right now my ecm will vary each bank (independently) quite a lot at idle, slowly rolling through 2.5ms to 4.5ms. sometimes the banks match, most of the time they're different. both sensors are new, maybe the exhaust pressure or temps are varying a lot, because they will show readings that look completely wrong. more load/rpm shows things that look more correct, and when in forced open loop the wideband shows a very steady and tight afr range.


i thought of the timing, but didn't realize it would be possible to add mph to the transition function. what table is that, and how does that work?

thanks.

-michael

 

Hmmm-

1. Re TPS - I've owned a bunch of GM cars over the years and haven't had a problem with a TPS
ever, so I can only speculate about things you probably already know:

On mine, IIRC the MAP and TPS share a regulated ~5v source and an ECM ground wire (check your
wire diagram to confirm your own box). Note the ECM ground does NOT have to be at the same
potential as the chassis ground. The TPS is just a voltage divider, returning a v signal to the PCM.
Noting your difficulty near the bottom of the scale, I'd start looking for a bad potentiometer or
dirty connector, or possibly a bad source v from the PCM.

If other devices share the same source and ground wires, it's also possible something else on those
lines is goofing up your TPS signal - in my case I'd check my MAP unit for shorts or something weird.

I'm inclined to find the source of a problem rather than just silence the Trouble Code, but that's just
me. The TPS hi alarm must be related to the low problem as well... at this point it doesn't sound
mechanical, but I'd check the throttle arm for free travel, anyway.


2. Lingenfelter mentions LT1 air distribution problems (briefly) in his engine book, and a lot of folks
report split BLMs at low engine speed. The only guy I know of to SOLVE the problem did it by building
an external air-tubing network that he then connected to each runner:
http://www.thirdgen.org/techboard/at...m-split-under-
Write-up here: http://www.thirdgen.org/techboard/di...ram-split.html
Dunno if it does any good, but it really is clever .


3. On my ECM the two permissives to transition from the idle-spark table to the main-spark table are
both on the Tuner Cat 'ECM Constants' screen - i.e. RPM and VSS enable minima. IIRC they formed an
either/or logic gate. If you don't see them on your mask, drop a note to TC and ask him if they are
defined for $EE.

BTW, the best thing I did for rear-end 'twitch' was to install those DRM camber brackets - world of
difference on uneven pavement.


Running open-loop sounds promising for your car. I just had my mask open last week, to get this car
past smog inspection again - but I managed it in closed loop. Re the WB readings at tip-in, note that
the commercial WB units I've seen only displayed ~10 or so samples per second. That's the reason
I use a separate DATAQ a/d converter for my rig. It's picking up 120 samples/sec, which makes it
able to log short-duration bumps in the AFR - using it I can also see ignition misfires, weak injectors,
and whatnot.... it's not a digital O-scope, but for $25 I can leave it hooked up in the car all the time.
It (and the Honda WB sensor) are still running after 6 years of ~daily use.

Good luck with the project,
DrJ

 

The $EE has sequential fuel injection, if youre getting split BLMs you can adjust the multiplier for each individual cylinder. That will fix the split BLM issue.

Just at a guess, is it richer on the left bank and leaner on the right at idle??

Here is the 2 tables, one for idle, one for off idle. There is a setting in Constants that will dictate at what TPS% it stops using the off idle table and goes to even delivery for all 8 cylinders, factory setting is 12.9%.

 

but you're just guessing cylinders within a bank are running similarly. what if one bank has 3 cylinders at 11:1, and one cylinder at 17:1. the O2 sensor for that bank is basically averaging, and you'd be tempted to lean out that whole bank.

i've thought about plumbing 8 EGT probes into the headers in order to get the cylinder balance really close, but that would be difficult and expensive considering how much power it makes already.

-michael

 

Those are not my settings, they are stock settings. If you look at the coloured lines ive put on there, youll see opposing cylinders are equal at idle except 5 and 6 as well as 7 and 8. That is why i am taking a guess that your split BLM has the higher number on the right bank.

Right now youve got a split BLM at idle, and i can apreciate the fact that youre right about the guessing what to change, but if youre running from the stock settings on those 2 tables, they simply wont be right. Might as well start experimenting with it abit, cause i smoothed my idle out alot by doing that.

 

just put a cam in put car on a matco scan tool getting codes for qdm3,oxygen sensor and tps switch if any body could help me i would be greatfull,thanks

 

What voltages would be good to go down to. And my scales are different.
GM/sec lower/upper Mv
0 579/ 597
32 522/ 557
64 548/548
96 513/513
128 495/495
I am researching split BLMs at idle and this thread is good help.
I run the AS&M 58mm T Body, drilled 11/64 drill bit through idle hole and epoxied in the back to seal airway passage for idle.

 

What voltages would be good to go down to. And my scales are different.
GM/sec lower/upper Mv
0 579/ 597
32 522/ 557
64 548/548
96 513/513
128 495/495
I am researching split BLMs at idle and this thread is good help.
I run the AS&M 58mm T Body, drilled 11/64 drill bit through idle hole and epoxied in the back to seal airway passage for idle.

 

If it has a somewhat radical cam, the problem is that the excess overlap has an effect on O2 sensor readings because the O2 sensor was designed to read the results of combustion and now it is reading the results of combustion along with fresh oxygen that is being pulled through due to the overlap.
This means that even if you are at 14.7:1 at idle, the O2 sensor is going to see something leaner (even with a wideband). The INT. will raise and fuel will be added until the O2 sensor registers 14.7:1, the real AFR will be less than 14.7:1. AFR is related to the BPW, BPW=GMS of Air/(AFR, Injector Flow).

Changing the stoichiometric AFR will initially change the BPW. Decreasing the AFR will result in an initially higher BPW. This BPW then will soon change back to the normal BPW you previously had. The stoichiometric AFR does not control the BLMs., it controls the BPW. The O2 voltage points control the INT. It's when the INT swings a particular amount that the BLM then changes. It all comes down to is the O2 voltage points. Decrease the AFR and the initial BPW will increase. The O2 voltage swings will increase too. The INT is based off of predefined O2 mV swing points (constants), then the INT will decrease in order to stay in line with those O2 constants, the BLM will decrease and the BPW will return to normal even though a lower AFR.

In short, decrease the AFR and the initial BPW will increase. The O2 voltage swings will increase too. Since the INT is based off of predefined O2 mV swing points (constants), then the INT will decrease in order to stay in line with those O2 constants. The BLM will decrease and the BPW will return to normal even though you specified a lower AFR

 

Try dropping the O2 sensor voltage at 0gms/sec to around 400mv. That would be for the rich/lean/mean tables. Personally I think all of your sensor voltages are a bit high but start by playing with the airflow at idle first.

 

I think these voltages are factory. Not sure, but I believe they are. But thanks for the change idea, I will give it a shot for hoots.

 

That's the idea. You don't have a factory cam and the factory settings don't work. With a bigger cam you lose vacuum and combustion becomes less efficient. Once you get "on cam" the factory settings may be ok. That's why you start with the idle.

 

Thanks, thats why I am still learning.