Ronny

iTrader: (1)

Search by my name and then keyword "proportional". Quite a few show up. I will check my current EBL.bin ( hopefully tomorrow) and report the changes I made from stock LO3.bin on that subject. Keep in mind I have 75 lbs at 21 lbs FP currently. Also do same search just use RBob for poster not thread starter for proportional and many will show.

I found this response from RBob:

And one of the proportional gains tables. I usually reduce the one that has larger values in it:

PRP - Gain vs O2 Error
PRP - Gain Muliplier vs Airflow

So possibly only those two to change?

Read this a few times!

https://www.thirdgen.org/forums/diy-...ng-points.html

Dominic Sorresso

Ronny,

I locked BLMs when I used a 7747 pre:EBL because the baud and BLM update rate was soooooo sloow. The INT changed faster.

dabomb6608

iTrader: (3)

What would cause a lean situation right after start up and when going under load first taking off (manual). It does it when motor is cold. Mainly in the mornings or after its sat for a few hours.

When it was warm outside it would only do it for the first block or so away from my house. Seemed like once I ran it above a certain rpm/speed it would then not do it at the next stop sign. Now with it being colder in the mornings it is doing it even longer. Noticed this morning it did it the whole way to work (not even .5miles but has 6 stops). Used to when it was warm outside temp I could make it to the 3rd stop (Market St.) and it would be fine when I went to continue.

At first I thought maybe my afr open vs cts was off. But it idles rich when started cold as it is supposed to. Its just going under load that causes it to go really really lean. It pegs out my WB Gauge and barely keeps chugging.

bphage

Here are some post I chopped up for reference: Might cut down on the searching some.

Smooth idle in open loop and surging in closed is usually due to too much proportional gain. Work with the gain table and value:

PRP - Prop Gain Offset for Idle
PRP - Gain vs O2 Error

Likely need to be reduced.

RBob.
It looks as though the prop gain is a tad high. Also, at the beginning the VE% is jumping a bit. This is in the 00:05:37 block. There may be too much of a jump in VE% at that point in the table.

As for the prop gain I would change this table:

PRP - Gain Multiplier vs Airflow

At the airflow shown in the data log: 38-40 gms/sec. Reduce the values at that area by 10-15%. See if that helps steady things down. The WB shows the actual O2 swing which is quite high.

RBob.
I don't mess with the prop duration too often. Mostly change the proportional gain. Then once the gain is low enough for a little rich/lean, if the INT starts to swing I'll slow that down.

RBob.
Originally Posted by gofast51
In one of the threads I was reading it was mentioned that RBob had written a paper on this subject. Where might we be able to find it? I did a search on here but didn't come up with anything. I think a reprint might be appropriate for this subject as it seems to come into play with the EBL. At any rate I would like to read it for educational purposes as I don't think I'm in a position to mess with it yet.
Check the tuning Book up in the stickies. Under the 8063/8746/7747 ECM, fueling I believe.

RBob.
I usually start with reducing the Gain multi vs Airflow. If the values become small (large injector flow rates), I'll then reduce the values in the gain vs O2 Error table. With this table I'll do an across the board percent reduction in values. This is to maintain the same shape to the table.

Then to fine tune use the gms/sec flow rate from the data log to define which are to change. This is with the Gain multi vs Airflow table.

Sometimes the INT starts to fight with the prop gains. Slowing down the INT update time helps in this regard. It can be helpful to slow the INT update down when doing prop gain cals. Just to keep it out of the way.

RBob.
A lot of the ideas that are being kicked around I covered in my first port (#3 of thread). It would be helpful to reread that post.

As far a tuning in the proportional gains tables, I go by exhaust sound and the INT. But first, I rough them in by changing them by the same amount of change in injector flow rate. If a larger displacement engine is in place, say a 383, that too needs to be taken into account.

It is all about ratios. Double the injector flow and need to halve one of the two proportional gains tables. The two tables get multiplied together, hence the need to only halve one of them.

If at a steady state cruising speed the exhaust sounds like there is a bad u-joint, too much proportional gain. Look at a data log and look at the airflow at that time (gms/sec). Go into the proportional gain vs airflow table and reduce it at and around that air flow.

Flash and re-test.

As for using the INT, if the INT is rapidly increasing and decreasing there is too much gain. Would likely see this in the bad u-joint example.

A slowly increasing and decreasing INT means that there isn't enough gain. This forces the ECM to use the INT to create cross counts (O2 signal). This causes an overshoot of the AFR. Which can be heard in the exhaust an oscillation. Can also cause the engine RPM to follow the AFR change.
RBob wrote " If the idle speed has been increased you should also increase the RPM threshold for the BLM idle to decel switch over (BLM - Idle Cell RPM Threshold)"

ownor

iTrader: (5)

i need to save your post bphage, thanks for putting that into compact form!

on a side note, EBL is brilliant but imho it lacks a good manual. a wiki or the like would be really helpful for all of us, from rookies up to the veterans helping here. don't get me wrong, there's a lot of information out there and Bob is always helpful and approachable. i just think everybody would have a better time when not pointing out the same points over and over again. the information is just really spread out.

now my question
i recently noticed that my L03 when cold puts a lot of drag torque on the drivetrain in DFCO.. is there any way to relieve this condition? i know you can "adjust" DFCO drag torque by the IAC steps that are commanded during DFCO. but i'm afraid for this issue, a DFCO IAC steps vs CTS would be needed and i would probably be the only guy using this, but still just an idea.. probably most people just have DFCO disabled up to 50°C CTS or whatever the stock setting was. i have set it to like 20 or so for so long i already forgot.. anyone else?

morgsie

iTrader: (1)

I think rbob needs to host a wiki page on his site. This would not only be an easier to use reference (vs the 75+ pages in this thread) but it would cut down on the amount of typing rbob does! Could easily reference people with information by posting a wiki link. Easily updated by anyone, etc. The dynamicefi site is becoming quite cluttered with the 2 new ecu products he has released. Food for thought...

ownor

iTrader: (5)

that's exactly what i think, all the information is out there and people, first of all RBob of course, know their stuff but it's just poorly organized i always found..
i never understood how Bob is keeping up the good work with so many people asking stuff that has already been asked a 100 times before, not only in this thread but i can only guess at the amount of time he spends typing emails in a day
besides, a wiki is a really nice way of organizing a community's knowledge, which is definitely something that applies to not only TGO but also quite some other forums and plattforms the EBL has been becoming more frequently used. so, not just RBob spreading some of his brilliant wisdom but more of a knowledge base uniting all the different inputs from multiple brains

mike327

Which of these AE tables needs to be changed to ad fuel at wot "pump shot"? or do I add subtract to all of them and in what order. I am running a tbi on a carb manifold
AE-MAP filter
AE-MAP PW
AE-TPS Filter
AE-CTS multiplier
AE-RPM multiplier

Thanks

RBob

iTrader: (1)

The MAP and TPS PW tables are used to control the volume of AE. Higher value, more fuel.

The MAP and TPS filter tables are used to control the duration of the AE. A lower value is longer, but also slightly more volume.

The CTS multiplier adjusts the volume of AE based on engine temperature. Higher value is more AE.

The RPM multiplier adjusts the volume of AE based on engine RPM. Higher value is more AE.

RBob.

1991sleeper

iTrader: (-3)

I think we need a write up of what each parameter in tuner pro does what it controls and what a higher or lower value does for that specific parameter

tomclayborn

so im sure its here but I cant find it.. maf car uses a esc with knock sensor,, ebl flash with port mod switches to map,, map has no esc correct(in ecm). does it stay plugged in whats up with this.. after the pin moves of course.. leave it plugged in.. in place? some are 5.0 some 5.7 whats the low down on this?

RBob

iTrader: (1)

Start->All Programs->EBL Flash->Calibration Help

Dependent upon which EBL system is being used the "EBL Flash" may be "EBL P4 Flash" or "EBL SFI-6 Flash."

RBob.

RBob

iTrader: (1)

When replacing a MAF TPI ECM the knock sensor and ESC module stay in place wired to the EBL Flash ECM. No changes to the ESC system is required.

The knock sensor and ESC module differ between a 5.0L and a 5.7L engine. Changing the sensor isn't as important as changing the ESC module (knock filter).

RBob.

Fast355

iTrader: (2)

Got my hands on an EBL Flash today that belongs to a local guy. The EBL easily handled a Comp 269 roller cammed flattop piston 350 Vortec in a GMT400 truck. Played with the tune for a few hours and got it running pretty well for him. Last time I adjusted the minimum air rate, fixed a couple of vacuum leaks and tried to fix a connectivity issue that was a bad cable. With a working cable, today I started over with a fresh, untouched LG4 VE table, Vortec spark map and corrected the BPWC and Initial timing constant values to match the setup, then performed a few VE learns, smoothed out the VE tables, adjusted the open loop air/fuel table and highway cruise. I tried enabling closed loop but never could get it to transition. I think 02 sensor location is the issue, so I left it a touch rich, about 14-14.5:1 in open loop and 16-16.2:1 in highway mode and left it in open loop. Seems to run very well.

Dominic Sorresso

Fast,

Is your O/L AFR Bias constant > than the target AFR at the Coolant C for Clsed Loop?

Fast355

iTrader: (2)

Honestly the calibration was done on his laptop and not sitting in front of me. I can get a copy and look. His truck has a vortec manifold and he tapped the EGR fitting for an 02 sensor. The sensor sits up inside the manifold out of the exhaust flow. I think it is 02 placement more than anything.

Dominic Sorresso

Oh wow. That sounds a bit odd.

Street Lethal

iTrader: (16)

Boost control with the EBL-P4... which area in the XDF can I utilize, was thinking of the charcoal canister stream mainly because the connector lines up right where I need it to. Is there a way to get it to work properly, or should I just go with an aftermarket EBC?

RBob

iTrader: (1)

We haven't added boost control to the EBL P4 yet. The EGR output (E9) is what we are planning to use for the WG control. Easy enough to move the CCP output terminal/wire to that location.

RBob.

Street Lethal

iTrader: (16)

Anything is better than the mechanical one that I am using at this point...

ownor

iTrader: (5)

Nice...!
I haven't looked into it, but is BPC currently supported with the SFI6?

RBob

iTrader: (1)

I used an RJC unit for a while with good results. Haven't had much feedback from the others on the WG control in the EBL SFI-6 ECM. Myself, with the large-by-huge FMIC and charge pipes, there is a lot of lag in the system when looking at intake manifold pressure for the control point.

The RJC unit gets it's boost reference from the turbo outlet. This produced less under and overshoot.

The good news is that the EBL P4 now also has alky injection control. Just need the mechanical parts of the system and a Mopar fan relay module as a pump driver.

RBob.

RBob

iTrader: (1)

The EBL P4 and SFI-6 ECMs have separate injector flow and engine displacement parameters.

RBob.

ownor

iTrader: (5)

BPC - Boost Pressure Control, sorry for using an ambiguous acronym here
Interesting point on what you say about the control input signal from MAP vs. compressor outlet nipple...

Cflick

A tad "out there" yes. Sounds like something I might do !
I'm curious as to whether it responds at all, and how fast, though I realize that Fast355 doesn't have the vehicle readily available.
If it's too far out of the flow, that could explain a failure to transition.

Street Lethal

iTrader: (16)

I think I am going to use the wiper fluid resevoir (has gravity feed as well), as I hesitate to set it up in the coolant resevoir near the turbo as the exhaust housing is literally right next to the resevoir. Was by Paul last night, and he too underlined it would work, but was going to be test and tune from a % standpoint in Tuner Pro using the canister purge solenoid, so I need to dedicate this weekend to figuring out what percentage equates to 15-psi with an 8 pound spring. 93 octane w/alky injection makes Rob a happy guy...

nossbc

iTrader: (1)

hi I have a bit of an issue...my 90 gta with ebl p4, I have a problem when I let the car sit for more than 2-4 days I have to crank it for a bit then let it sit for few min, then crank it again and it starts....when cold.
now if I use it all day....then go to start it next day(car cold again after being parked for over 14 h) it starts just fine.
the problem only occurs when car sits more than 2 days...whats up with that?
fuel pump is new...aeromotive 340l...car has mild cam, tfs heads, hsr intake 58mm tb, eagle 383 rot assembly 10-1 comp

Cflick

For me, the stock TBI pump can't develop the pressure. ( in fact, stock has a built in relief at 15 PSI. ) Therefor, I have a Ford in-line booster pump.
My habit is to turn the key on, and listen to the booster pump. After a second, or two, or maybe four, it begins to sound loaded, and I know that fuel pressure is up. THEN I turn to crank. The engine might turn 4 compressions before firing. If it's been parked for a month, maybe it turns 8 compressions, or two full rotations, before running.
It never fires but not start. Never.
Normally, on a warm engine it doesn't get past the second compression before it's running.
I might suggest try key-on, wait 5, then crank and see if it's different. You may just need a second or four for the fuel to come up to pressure.

Ronny

iTrader: (1)

More cranking fuel? Sounds like it may be coolant temp low requiring more crank fuel.

nossbc

iTrader: (1)

i have a fuel press gauge on the car and usually I prime it 3 times before trying to start it...fuel set at 45psi and it builds pess on first prime usally.
once car starts...maintains 12.7 to 13.2 afr until warm when leans out to 14.5 at idle.
it use to have a burned chip before I installed eblp4 it would start fine even if I didn't start it for few weeks, same mods as well.
im just wondering if im missing something.
on the good note the chip I had before was done by local performance shop...made 418whp/512wtq...just dyno it yesterday with my tune on the ebl 434whp/544wtq...lovin the ebl.

1991sleeper

iTrader: (-3)

What could cause a dead spot in my rpm range at around 14-1600 rpm it just don't wanna go like its flooding out also my VE is really high here .. My wideband took a crap so I'm waiting for funds to fix that but it don't matter what I do in that range even if I mash to floor sounds like it's bogging out n just won't go till it gets to 1700 then takes off
Also if I stab throttle wants to sputter but if I ease into the throttle it goes pretty hard till I reach about 80% throttle then it just goes downhill
And how can I lean it out a little at cold cts temps right after initial firing until it gets up to say 130*

RBob

iTrader: (1)

Probably need to work on the spark advance (SA).

Which intake manifold are you using?

Sounds like the fuel delivery can't keep up. On the throttle stab does the engine hesitate and possibly backfire out the intake? Or does the engine chug and/or gurgle?

Hesitate is lean, chug/gurgle is rich. Either way need to adjust the AE tables. A dump of the data log is helpful as there is AE information in it.

This table:

Open Loop - AFR Multiplier vs CTS

As it is a multiplier, a more negative value is richer. reducing the values toward 0 will make it leaner. Can also make it even leaner by going with positive numbers.

RBob.

1991sleeper

iTrader: (-3)

I'm running a summit stage 2 dual plane intake with center divider notched 1.75" down and 2" across .. With a 1/4" steel adapter and very efficient CAI with carb hat ... IAT is in carb had directly in air stream about 3" from injectors ... And no popping or backfiring more of a gurgle / sputter

Also intake has no air gap and heat crossover is blocked off ... Also has been fully ported and polished on all short side radii

RBob

iTrader: (1)

With a gurgle/sputter it is rich. Reduce the AE.

RBob.

1991sleeper

iTrader: (-3)

It does the same thing at that 1500 rpm area but in first n second gear it will pull through it only does it under load .. Say cruising at 40mph 1500rpm and if I try to accelerate and load goes over 50-55 kpa it just don't do nothin setts at 1500 an gurgles ... Also my PE sA vs gear are all 000

Also VE is really high in this area .. But learns show only -5

1991sleeper

iTrader: (-3)

here is my current bins and datalog . if anyone could help me out or has time to modify and resend i would b very thankful! if anyone can get this thing running like it should ill paypal u $ for ur time .

thanks all fellow thirdgeners!
Nate

nathanjagness@gmail.com

if u need info on the build just ask or go to my build thread @ "LO3Rebuild(HP Estimates)"

RBob

iTrader: (1)

Please post the .dat file of the data log, thanks.

RBob.

RBob

iTrader: (1)

The VE table is not correct, which may be caused by some other affect.

In the CSV file look at records 10,711 through 10,750. The BLM is bottomed at 118, the INT is headed downward, and the NB O2 value is in the high 900's. It is rich.

RBob.

drive it

iTrader: (1)

I know this is jumping back a ways-It's been running good-but it's Kalifornia smog check time again.
Here it's at 15 and 25mph on a dyno under load. CO was at the limit of .56, going for recheck because the gas tank didn't hold pressure-hose fixed now. Was afraid it would nudge over the limit on recheck so dug back into the O2 mean/rich/lean and at idle tables.
With this narrow band sensor I did get it to change to where I wanted it. ( I have an old 4 gas bench from pre dyno smog days to check tail pipe numbers, so can check tailpipe numbers without NOX).
But I had to adjust the O2 mean all the way down to 250mv with plus/minus 50 to obtain 14.5 to 14.7 by CO numbers at tailpipe. (0.2 to 0.13 CO)

I saw one post where someone had to adjust the MV mean all the way down to 100 I believe it was.

Question is how can a narrowband be off by that much calibration? On stock cars you just throw in a new sensor and it's good..........

Those tables are there so we can adjust, but do we really have to adjust for every narrowband to pass smog-just doesn't make sense compared to stock setups????? Can't wrap my brain around this one.

Cflick

Short version, it can't !

Slightly longer version, perhaps unburned fuel, which would leave a surplus of unburned O2, which would make a narrowband think it's lean and richen it up.
CO is, of course, unburned fuel, but it's vaporized at least. Your 4 gas *should* be able to detect that, but a narrowband won't.

Dominic Sorresso

This brings up a quick question for RBob and others. If you have flipped over from Bosch to AC Delco O2 sensor, or Denso for that matter, does that require a change in the O2 windows or does it not matter? I know that when I made the swap to ACs, my BLMs appeared to be leaner on the next scan where I had thought I reached close to a 128 avg.

drive it

iTrader: (1)

OK, the unburned gas theory-then HC should be high, which it wasn't.
Numbers for smog with mean at 450MVs.
15mph hc-71 (max-90) co-0.56 (max-0.56) nox-147 (max-774)
25mph hc-41 (max-72) co-0.42 (max-0.42) nox-127 (max-655)

and with mean adjusted to 200mvs.
15mph hc-65 co-0.36 nox-216
25mph hc-36 co-0.19 nox-178

So doesn't seem like excessively high HC to explain it either.

drive it

iTrader: (1)

RBob-any thoughts on this?

Are there other tables that could throw this off?

Cflick

Advance the spark about 3 degrees, and check it again.

RBob

iTrader: (1)

I'm not sure that I understand the issue. You have a 383, roller cammed, and SC'd engine and expect it to act like a stock engine? It won't.

Change the oil & filter and install a new set of spark plugs. That will help clean up the emissions.

RBob.

drive it

iTrader: (1)

The issue I am puzzled about is the apparent calibration of the narrow band O2 sensors.
It seems I'm not the only one that has had to adjust the mean/high/low O2 Mv way down to obtain stioch-14.7.
My understanding was that 450mv will be 14.7 with all narrowbands.
Going to the ac delco O2 sensor did get closer than the bosch but was still running rich.(all blms already at or close to 128s)
(does have new oil/filter/plugs)
Thanks for any insight.

Dominic Sorresso

Stoich is stoich depending on the fuel u r using. Unless it's pure gas, it isn't 14.7:1. So the 450mv is a crossing point between rich/lean but it's not 14.7 necessarily. It more or less defines "no change". Here in the Midwest, we have "10%" ethanol, so I tune using 14.2-.3 for stoich in my cal. For WOT TUNING, I am using Lambda, not AFR. The only area I play w the O2 window is down at idle due to cams. Other than that, I leave the other values alone and concentrate on correct injector bias, prop gains, and VE.
I know I have some inaccuracies in my explanation. No doubt RBob will add his clarifications to it.

drive it

iTrader: (1)

I understand stoich/lambda (after a lot of help here!). I left my wide band at 14.7 so look for 14.2 on the wide band since we also have the 10% here. And I also use tail pipe gases.
I went with injectors with published constants.
adjusted prop gains also. (with a lot of help here)
I know I've run into numerous parts new off the shelf that were bad/shorted out.
Maybe I've just gotten bad O2 sensors?

Cflick

Unlikely.
The internal chemistry of a "narrow" band O2 sensor is such that it can only indicate two things.
1. Too rich.
2. Too lean.
The ideal stoich ratio varies somewhat with additives, somewhere between 14:1 and 15:1. Usually, pretty close to 14.65:1 or so. We typically round it to 14.7:1.
The O2 sensor indicates whether the burned exhaust came from a mix either too rich, ( generates roughly 0.9 volts ) or too lean. ( stops generating anything, ergo zero volts )
The chemistry is such that the response of the sensor is near instant. Averaged over time, ( 0.9 + 0.0 ) / 2 = 0.450
A narrow band will always indicate one or the other. Rich, or lean. Always.

The mix between the piston and cylinder wall above the top ring, for instance, never burns, ( because it is cooled below the combustion point by the heat extracted into both the piston, and into the cylinder wall ) but is blown out with the exhaust, so there is always some part of the exhaust that contains unburned fuel, and unburned O2. If even one molecule of O2 hits the sensor, it swings too lean. If even one molecule of unburned fuel hits the sensor element, it indicates too rich. There will always be a few molecules in the exhaust stream.
If, on average, the exhaust indicates too rich as often as it indicates too lean, we can presume that on average, the mix is close to stoich.
Another fortuitous reason for swinging between too rich and too lean, is so that the catalytic converter always has both fuel and air for IT to burn ! If a perfectly burned exhaust hit the cat, it would see only CO2 and water, and do nothing.
Ideally, we hit the cat with just enough unburned mix to keep it hot, so that it will do the final clean-up for emissions purposes.
The cat is only capable of a limited amount of burning. It is intended only to clean up and finish burn that mix that was caught between the piston top and cylinder wall. If it is overwhelmed, then the exhaust emissions go up. Late timing can cause some of the mix, even if it's correct, to not have burned before it blows through the cat, which then becomes partially overwhelmed, and show up as increased CO in the exhaust stream.
( partly because it's a catalytic reaction, not an actual "flame" burn )
That's why I suggest you advance timing a couple degrees, and look for a change.
Of course, it's also possible your cat is going, or gone.

Dominic Sorresso

Actually, if you are looking for stoich while using 10%, your WB would still indicate 14.7 unless you have function for adjusting the calibration curve. If not, then an AFR OF 14.7 would need to be a Lambda of 1 indicating an ACTUAL AFR of 14.2 or so.