Your running a wideband in Open Loop though, no? You should be ignoring your BLM, not your O2mv's. Your O2's should be in the 450-500 range in idle, with an air/fuel of around 14.2 - 14.7, or even leaner if you want, not at 13.4. Your WUD doesn't report wideband? What is your wideband reading at idle, not the WUD, but the gauge itself? I flattened the same area under the IAC to a zero value as well, and the EBL VE learn brought it right back after a five minute drive. Running a much larger head/cam setup and idle is flawless so far. I can't go back in pages, what exactly are you experiencing, stalling at idle...?

 

Okay so your post made me think that I may have the processes wrong so first back to basics and here is is what I have been doing.

1. Did some NB VE tuning prior to WB got the car running
2. WB in OL VE learns. The idea as I understand it is that the commanded AFR will be achieved through sequential VE learns. The table is set for a commanded afr of 13.5 (at idle - maybe RBob can explain the OL settings as I don't understand them). So I am still rich at 12:1 ish values. Once I change the VE table to match the commanded afr of 13.5 I will have my VE table correct. (this is where I am at; today the dyno was used to ensure I wasn't going lean and had a reasonable spark table so for this smog tune I don't need to sweat the AE and PE stuff as I am safe.)
3. CL final NB tuning long drive to adjust VE to values CL wants
4. Re-enable the smog equipment then tune for active smog equipment not using the VE table but using other tables - I need to read up more on this.
5. Smog and pray I got it right.

Okay idle: I have had multiple issues getting the car to idle. The biggest issues have been lean blocks causing the idle to surge then eventually stall. This was compounded by overshoot (so nicely explained to me) from the SA compensation and the IAC steps. By raising the VE and flattening it out I was able to stop the lean surging and by decreasing the SA by about 60% for correcting the target RPM (SA idle compensation high and low tables) I was able to correct the rest of it; well I did need to decrease the stall saver enter rpm as that was another cause of surging rpm. After doing all of that I posted my first stable idle with the factory idle settings.

The problem I am facing now is with VE learns it goes back into surging then dying again. I think it is lean surge so my plan was to try and and slowly correct the VE table by hand as the WB consistently creates an unstable idle that I can't do VE learns in. But your post would suggest that I may need to change the target AFR to 14.7 for OL mode? Then continue with what I have been doing?

Thanks

-----

I went and did a series of VE learns and after about 6 learns the result is the dying idle in this data log.

I did try changing the OL cAFR table, changed all values to 14.7 but the cAFR did not change from 13.5; as far as I could tell it had very little effect on anything during idle.

Here is my game plan: I will work on slowly tuning the idle VE table to 13.5 to match the 13.5 cAFR... my last tune was fairly close. Then I'll try an Closed Loop idle and see what the AFR is and tune to the 14.7 for CL idle. Seem reasonable?

Thanks

 

You kinda answered your own question regarding the surging, as the wideband is commanding 13.4 fully warmed up because your running a richer Open Loop AFR that is designed for warming up only, not a hot idle. VE learn is more set towards correcting towards stoich, so they are playing tug of war with each other. You need to adjust your target Open Loop AFR to a more leaner 14.7 ratio when the engine is fully warmed up based on your coolant temp readings...

I guess I can post this video here because it is an open EBL thread, but this is with a heavily modded 305 running a very big cam. This is with a stock narrowband, and running Closed Loop. You should have absolutely no problem idling in Closed Loop with your setup as it is very docile in comparison...

 

This parameter:

Open Loop - Idle Lean Limit

RBob.

 

the way i got it is this: at least when doing WBO2 VE learns, it will just try to learn VE for whatever offset is necessary to make the reported AFR mirror the cAFR, regardless of stoich. for NBO2, well it can only work around stoich in a linear motion so yeah you would have to run around stoich (at least on average, which is what CL does imho) to make any use of that i guess.

i'd guess you'd have to play with inj offsets and prop gains for quite a while to get it idling correctly in CL?

 

Reductions of 50% are typical of a large cam. I am using today that reduction on a 350 cid 224/230 @ .05.

 

Okay so I have the idle the best I have ever gotten it. I am going to post the stuff I did/tried and found in hopes that if anyone else sees the same thing it might help... if others what to chime in on what they see I am sure it will help many noobs such as myself. If no one wants to read all this the last paragraph will have my current issue.

So I was successful at getting the car to idle at like 500-600rpm with a AFR in the 12s. Every time I attempted to increase the AFR via VE learns with the wideband i would end up getting a surging idle that seemed to be from a lean condition. No mater what I would do the idle curve always turned into a step curve that made no sense for such a mild cam (see "VE Learns WB").

It seemed to me that this was the result of the VE trying to learn with the surging idle were the SA and IAC would lead to rich/lean conditions. Then would exacerbate the condition as it would steepen the VE curve forcing it to go rich lean at a greater rate. I also thought proportional gains might be playing a role but that had no effect at this point on VE learns. I went through the data by hand to see what areas were rich lean and it was difficult to discern from the overshoot.

I figured I should be able to make the car idle at the stock idle settings, but I then realized that part of the problem is that the car was trying to idle much lower than the numerical value i had set. I tried changing all the values to 600 rpm etc and sure enough it wouldn't idle at that value it was of by x amount each time.

I was also fighting the stall saver which kicks in at 475rpm, I lowered that to 425rpm get rid of much of the surging, but still could not make it work with a 14.7 afr. I noticed launch mode coming on at times but at this rpm would not make a change in the SA so it was more of a symptom than a cause of the problem. At this point I concluded that motor just didn't like a lean afr at that rpm and that 500 rpm was well below what most stock iroc's idled at...

So I then increased my idle speed to try and get an actual idle speed of 600-700 rpm with a target goal of 650rpm (for a factory like ilde). I started with a stock VE curve from 0-55 map and 1000rpm. then did some VE learns with the WB and was able to achieve idle in the 600-700rpm range with an afr from 14.3 - 15.7; average afr of about 15. So slightly lean which from earlier posts is better than being slightly rich for passing smog. (see "current VE" - note I still need to work on the transition points from the idle area to the other areas I worked out)

I was nolonger fighting stall saver, but was still getting the surging idle. It looked like I had take out to much SA correction as my idle would decrease, SA would increase but not enough then the IAC would open up and my idle would shoot way up. So i changed my SA idle compensation settings and through trial and error along with data logging I found the best idle being close to that target afr of 14.7 I have had yet. I did find that I needed to go back up on my proportional gains... I had a value of 10, but 20 seems to be about the sweet spot.

--------------------------- Current Issue -----------------------------

So this is what i have now, (see "current surg" - this is in closed loop running of NB, WB is just logging) It is very mild but I think it could be enough for smog guys to want to fail me... they are jerks around here and I have had issues with them before over painting things being a modification and thus not passing visual! A note on the graph, each numerical increase on the x axis is equal to 1/4 sec.

I have several areas of surge on this graph this first being at x = 120; you see my idle start to drop which looks to be the result of the SA decreasing from 20 to 19 to try and decrease the idle (overshoot). At that point the idle plummets, SA increases back to 20 and the IAC increase from 40 to 41 at x = 130; RPM is still below target so SA increases again to 21 which hold at 21 until x = 14 at which point SA decreases to 20 and the IAC opens up 1 step from 40-41 (overshoot).

It looks to me that the IAC increase the idle much faster than the SA (i.e. 1 deg SA results in a smaller change in idle rpm than 1 step of IAC).

So at x = 140 if I were in control of the idle I would not have increase the IAC by one step, just decreased the SA by 1 deg. But I don't see how I have control over what the IAC does at this point. I am not sure how to correct this surge as everything I have tried thus far makes things worse.

Maybe I can increase the dead band around the IAC and try to force the idle to be controlled preferably via SA?


Okay the next point I think is worth looking at is at x = 380; the idle is slowly creeping up IAC decreases a step while the SA cant seem to make up its mind decreasing the SA 1 deg, then immediately increasing it a degree then decreasing it again... but again the IAC seems to have such a strong effect on the idle rpm it seems to be the cause of overshooting low, while the SA is increased to try to compensate. Finally at x = 394 the IAC increases again and the RPM overshoots again.

How in the world am I going to correct the IAC overshoot?


My idle high compensation settings are:

RPM Deg
0 - 0
50 - 1.05
100 - 1.05
150 - 0.70
200 - 0.70

My idle low compensation settings are:

RPM Deg
0 - 0.35
50 - 0.7
100 - 1.05
150 - 0.70
200 - 0.35



Any thoughts on what I should change to fix this? I am so close, but I just haven't gotten it yet.

 

So I tried increasing the Dead Band of the IAC from 12.5 to 25 and that made things worse with a clear IAC overshoot in both directions even with a 50% increase in the SA comp tables. Then I decided; what the heck I have my VE table better aligned and the cam is so mild the stock settings should have worked better than they did so I should re try the stock settings.

I first added back the original SA comp tables and DB back to 12.5, which idled... you guessed it... much better. So I decided to increase the Prop Gains back from 20 to stock of 32. I needed to make some minor changes to my VE table as a result of that change as I was getting some lean surge so I took the car for a spin (in the driveway) using the NB for tuning in CL as it was really close and used the WB to monitor actual afr.

Here is what I now have; I am still getting a little surg (x = 420) but I am so hyper focused on this problem now that it might be completely normal and I am just over reading. The afr is on the lean side I figured a target of 14.2 - 15.2 would be good... the mean afr = 15.1, max afr = 16, min = 14.7. I figure I could just add a little fuel across the board, but the VE learn doesn't want to add any so what do you guys think leave it alone, good for smog, or add some fuel as its a little too lean?

Would someone please take a look at the new idle table and let me know what they think... good enough... needs work etc; big thing for me is obviously passing smog.

I posted the New VE table, still need to work on some missed spots, I have been using a smoothing factor of 6; I was thinking of going down on the smoothing factor any input for the smoothing factor with the final NB tuning stuff?

Thanks everyone.

 

OK, ebl p4 is ordered with 730 ecm!
Question on cable-I believe the cable I've been using for cats rt tuner will also work on ebl p4, yes?????

 

For idle a slightly richer then 14.7 AFR is better, say 14.2. To do this in closed loop (required for cat-con to operate), there is a table of three parameters that the ECM targets:

O2 - Idle O2 Window Terms

450 mV is stoich, higher is richer.

The idle shouldn't be that uneven. The IAC is also doing too much work to maintain the idle. Double check the mechanical aspects: firing order, wires, plugs, cap & rotor, base timing, vacuum leaks, valve adjustment and so on.

RBob.

 

If you want to scan via the ALDL port that cable will work.

The What's Up Display (WUD) has it's own data port. It runs at 115K baud serial, to a USB/serial convertor that is embedded in the supplied cable.

See here:

http://www.dynamicefi.com/EBL_P4_Ins...p#comm_install

RBob.

 

Got it! So if I'm dyno tuning and want to lenghten the cable I can just use a standard usb to extend it then?

 

It is better to extend the serial end of the cable. Serial data is more robust then USB in a noisy automotive environment. An 1/8" plug to 1/8" jack cable, for stereo, such as this:

http://www.radioshack.com/product/in...ductId=2103868

RBob.

 

Just about everthing is new and is correct: Firing order, wires, plugs, cap & rotor, were just changed with MSD stuff and are correct; base timing = 10 deg I didn't change it as the motor seemed to like it, no vacuum leaks (almost everything is plugged up & all the gaskets are well sealed), valves should be okay... there is the exhaust leak on the passenger side header that I can't seem to permanently fix... but this is far from the WB and NB O2 sensors so shouldnt effect that reading as there is no leak in the driver side.

I just broke the stem off the fpr trying to pull the vacuum line off to see if it was leaking.... stem pulled off the fpr with the vacuum line attached. (so there wasn't a leak before me ) The MAP shares the same port as the FPR. There was some debate as to this being an issue. Should I tap a new port while I have everything taken apart?

Maybe i should plug the 9th injector while i am at it?

I ran a few more ve learns, it took more fuel out and i get nothing but zeros on the ve learns; just to be clear, should i be adjusting the o2 window terms to bring down the afr to th 14.2 - 14.7 range?

I was looking for how to tune the egr on the tpi... is the egr - PW multiplier vs vac table the one to use?

I found where you said this "For the EGR the 'EGR - BPC Multiplier vs VAC' table is the one to tune. Can do VE Learns with the EGR active and see how the corrections go. Then mod the EGR BPC table to bring the corrections in line. Just don't use the BIN from the learn.

Then the 'EGR - SA Added' table. Try a little more and a little less. See how the engine responds. The other values could probably remain as they are." - but i cant find that fist table... as i recall this is for a different ebl version.

Anyhow now that I screwed myself with the FPR I am going to tear down the TPI unit and replace the FPR, put in the 60lb injectors, and I'll run new vacuum lines in the few places I haven't done so already. I might pull off the valve covers and check the valves while I have this all taken apart. I need to fix the oil pressure sender too those things always break... so I will fix that and run oil lines to be used for the future remote turbo.

sigh

 

My fpr/map also share a vacuum line, far as I can tell it does ok. I don't think it would hurt to check for vacuum leaks, the setup on my 383 has all new gaskets, and it still had a leak, causing me to fail emissions (also has a rolling/surging idle like you).

 

Thanks, that helps. I have sprayed everywhere and cant find a vacuum leak. The only thing hooked up are the FPR, MAP, and PVC. Everything else was blocked off without a leak. When I take everything apart I will be using new gaskets etc.

 

Well the only source of a leak I've found is the GM 454 TBI/Spreadbore adapter, it has very little surface area around the outer edge of the adapter, it isn't flat all the way across like many carbs are.

 

That was one heck of a find, was it large enough for you to get an RPM change when you sprayed the area or could you just feel the vacuum? I have gone through at least half a bottle of windex looking for little leaks... I originally found a whole bunch of leaks but I can't find a darn thing left. With all new gaskets etc I'll be certain to get everything leak free.

 

That is not a good idea. Map should be a dedicated vac source

Tap the brake booster fitting and place a nipple and use it for VAFPR

 

I'll have to double check the ports on my throttle body, there are a couple unused ports on the front and an unused one on the rear, they're smaller... the brake booster is currently running off a cast/threaded hole in the manifold itself and the port on the throttle body is blocked off.

*edit*
Found this image, anyone verify? I used the ported manifold on accident when I first installed the EBL/383 combo, caused a lot of tuning issues.

 

I just installed the TT-1. It seams to me like my readings are a little off. At idle the NB reads about 460 while the WB shows a ratio of about 20-21? At WOT the NB reads 980 and the WB 13.4
I also noticed when doing a learn it says something like "BLM not 128. Open loop forced" or something like that. Is this all correct or is something goofy?

 

NB read is about right. Lean Rich Median values are about 450 median where it cycles rich lean over that V. 22/1 car would not run. That is max lean read for WB. NB WOT your read is rich at 900+ but hard to define rich with a NB. WB at 13.4 is lean for WOT. Typical is 12.5-12.8. Did you calibrate WB? Does your WB require a calc. Mine does.

 

Yes, the MAP sensor should have it's own port. Running it with the FPR will slow the response time of the MAP sensor.

At a minimum it should be unplugged so that it can't spray fuel.

They are what affects the closed loop AFR. Note that there are three tables for off-idle, and a table of three values for when in idle.

Yes, do a VE Learn and use the corrections results to modify the 'EGR - BPC Multiplier vs VAC' table. Be sure to convert the MAP Kpa to vacuum by subtracting it from the barometric pressure. Can also use a dump of the log for the VAC & Baro values.

This is a single value (not a table): EGR - SA Added

RBob.

 

Yes, those images are correct. The left & right ports are full time vacuum.

The ported vacuum port is for CCP.

RBob.

 

As delivered the AFR range is 10:1 to 20:1, so if the WUD is showing 21:1, something isn't set up correctly. Or the TT-1 set up has been changed. There is a built in TT-1 WB device in the WUD.

Is there an exhaust leak in the manifold/pipe that has the WB sensor in it?

Need to force open loop to do VE Learns via the WB. Set the Closed Loop - CTS Threshold high. However, until the WB reporting is correct I wouldn't use it for VE learns.

RBob.

 

Any idea where it would be best to tap the plenum to add a new port? I'll do that while I have everything apart.

I can't for the life of me find the "EGR - BPC Multiplier vs VAC" table: See image... what am I missing?

Thanks again.

 

With the EBL P4 it is a PW multiplier table. This stuff always gets the best of me as there are subtle differences between the EBL products.

RBob.

 

Rbob, I have a few questions regarding the sfi6 ebl. Do you want too keep our in this thread, start a new one, or maybe put it in the tb boards?

 

If it is more about features & such then tuning a new thread may be best.

RBob.

 

... hmm, like like adding an SA learn screen in the WUD for ION sense tuning.

 

Yes, as I was looking through the schematic there were some addition features that I would like to get clarified. I am getting closer to raising the money to buy it and had a few questions. I'll start and EBL SFI-6 thread.

 

I tapped the brass fitting serving the brake booster. I think it was a 5/16 thread with nipple(brass).

 

would you like a hypereutectic cylinder block to go with that sir? (.. this is what bmw does on some of their recent V8 engines)

i'm following your tuning issues bphage, i'm curious how it turns out.. i don't really have anything to input at this point tho.

on another account:
anyone tuning an EBL on a CFI in here? gonna give it a shot next week for a 84 C4 that has some heavy exhaust work and is not running all that well in its stock form now... i told him to upgrade/overhaul his TBs etc top parts first but he didn't listen (yet). so we'll see how that turns out.

 

interesting post. so basically idle speed is mainly controlled by SA (or it should be), because it provides a much faster/more responsive and both precise control?
about those 14.2 ... that also true for a sniffer test?

 

There was an exhaust leak but I believe I have it all fixed now. At idle it reads about a 16 afr. I will check again to make sure I didnt miss a leak. Just to double check is there anything you have to do for the NB if the smog has been removed?

 

The IAC does most of the idle speed control. The idle SA compensation is fine tuning the idle speed. It also prevents a rolling idle.

An idle AFR of 14.2 may help with the sniffer. Running a lower spark advance usually helps too.

RBob.

 

Nothing needs to be done in the calibration.

RBob.

 

A CF is no different than any other TBI. All i can think of is the IAC(x2) which reduces IAC counts by 50% and the 4+3 manual trans(see recent comment by RBob).

 

thanks for your pointer ronny. will look up on the tranny comment.
so i guess i'll probably be starting from the '89 L05 bin that is provided with the EBL.

 

Due to the heads used on the '84 x-fire engine, the base EBL_F_TB2 BIN has better SA tables. Using an L05 BIN will have low SA and a lazy engine.

This is the thread Ronny is referring to regarding the 4+3:

https://www.thirdgen.org/forums/diy-...-lock-4-a.html

RBob.

 

Hey RBob which bin would u reccomend staring with for an LO3 with TFS super 23 175 heads , flat tops , and 216/220 cam ... Check out my thread "LO3 rebuild (HP estimates)"

 

Start with the EBL_F_3001 BIN. The cam is a bit more then what your engine has, so if it stalls after 15 - 30 seconds of running will need to increase the lower end of the VE table.

To do that add 12% to the entire 400 RPM row (multiply by 1.12). Then decrease it by 1% for each row as you go higher in RPM. So the 500 RPM row gets 11%, 600 RPM gets 10%, and so on until you reach 0%.

Once the fueling is better see how the engine likes the higher RPM under load. If it is rough or pings try the Main & Extended SA tables from EBL_F_3006. Just copy them over into your current BIN.

All of the above is easy to do in Tuner Pro using the toolbox functions.

RBob.

 

Thanks ! Yea I have most of the EBL down pat .. As far as using it anyway .. I had te car running 9.800 1/8 with 98% stock Lo3 so I'm thinking about just using the bin I had and just swapping over the SA tables and changing BPC for new injectors

 

An automotive alternator is capable of about 0.7 volts minimum, to something above 200 at high RPM, and is proportional to rotational speed.
The current output is dependent on the alternator voltage vs. system voltage. As the alternator voltage rises, the current rises linearly with the voltage difference.
The regulator drops the alternator field to limit the current so that the alternator doesn't pull the system voltage too high. It can still produce 200 volts if the armature is disconnected from everything, at near zero current. With any current draw at all, ( voltage any miniscule amount above battery ) current is drawn, and drags the output voltage down to the limit set by the field.
The battery is not capable of anywhere even remotely close to the voltage differences the alternator is, so the battery will "take over" as soon as the alternator voltage drops ( due to insufficient RPM ) to equal to the battery voltage. Then, the voltage will slowly drop proportional to the battery discharge. *IF* the battery drops sufficiently low, the alternator will attempt to bring it back up for all it's worth at the present RPM, at whatever voltage level that might be.

It takes real power to drive the alternator. ( or generator )
Simply put, the alternator output voltage multiplied by the alternator output current ( which includes everything needed to run the vehicle, not just charging current ) equals the alternator output in watts. There are 746 watts per horsepower.
The purpose of underdrive pulleys is to prevent horsepower from being taken by the alternator at the expense of the drive wheels. On the race track, we are perfectly happy to let the battery discharge to run the vehicle for a few seconds, diverting as much engine power as possible to the wheels. On the street, at night, in a rain storm, not so much.
Alternatively, underdrive pulleys are used where the engine RPM gets high enough that there is a real risk of tearing the alternator apart from overspeed. ( which is not pretty, trust me )

I've never run underdrive pulleys, but I have run cut out switches.
Underdrive reduces the power used to spin the alternator, making more available for the wheels. A cut out switch makes ALL of that power available to the wheels, turning the alternator into nothing more than a heavy idler. Don't forget to turn it back on, though.

If scrounging the 4 or 5 horsepower to run the alternator is important to you, then you might also consider the considerably more power used to drive the water pump, but do so knowing that idling in traffic can produce a "less than desirable" result. ( and that would include the staging lanes at the race track )

 

Now, that's something I'd be REAL interested in, as Rbob knows well !
It'll complicate things beyond the average tuner, though. SA requirements can be far more dynamic than one would think. I'm afraid such a feature would confuse more than it would help most of us.
( I've many, MANY hours worth of ion sense spark records )
What I've learned about that, is that the optimum spark for that particular combination of parts at that particular time on that particular day is not so easy to determine, nor is it that critical +/- 4 or 5 degrees.

 

Are these the tables you are talking about?

INT - Lean O2 (Lower)
INT - Mean R/L
INT - Rich O2 (Upper)
O2 - Idle O2 Window Terms
Closed Loop - Stoich AFR

Would you have a bin for car that you tuned a (for example) 14.2 closed loop AFR? Considering most of north america is running blended fuel (10% ethanol), this should almost be in all the tunes!

 

The first four are what defines the closed loop AFR. Stoich isn't important as long as it is in the ball park.

O2 sensors operate on Lambda, they don't know anything about what the stoich AFR is. Which is why you don't need to set the Stoich AFR to 14.2 for 10% ETOH blended fuels.

RBob.

 

i was under the impression the cfi had something similar to the swirl heads.. guess i was wrong then thanks for your pointer! any additional word on the IAC issue that was mentioned, or how to offset BPC with those split injector feeds, if that is necessary?
edit: i just remembered that this car was a A/T car.. so that would be a 700R4 then.

 

on those 14.2 for ethanol blend fuels or other types of blends... i was of the impression that was the stoich ratio and lambda is 1 at the point of stoich. so while a wideband would report 14.2 the nbo2 would read say 0.450mV/stoich for whatever fuel that is the target ratio for?

 

to take this a step further maybe, multiply electrical output power with the generator (alternator) efficiency and you get the mechanical power needed to drive it. that would likely be something in the 0.4 to 0.5 range for those generator designs?

I would imagine there's quite a bit left on the table with frictional losses in the bearings and windage though, plus you still have the rotor inertia. Then again UDPs just change these numbers a certain amount (10%, 20%?) either.. That and you don't remove the "electromagnetic" load imposed by the field excitation with UDPs, which might still be the biggest "power robber" when analysing alternator loads/losses

I have a question on this... With just the battery running the engine, is there anything lost concerning engine output power because of the lower system voltage? Which would be a reason to still run an alterator for 14.x output voltage during any racing..

Oh and what did you turn to make that poor alternator burst? Makes me wonder what the stock ones can take..