Locking Open Loop Idle BLMs at 128?

Hi, I'm running a GM 350 TBI setup with a 774747 ECM on a '88 Jeep Grand Wagoneer with AMC 360 and a mild cam. Fuel control only, not using timing control. I've been running this setup for about a year now and everything pretty well dialed in (VE tables, Cold Start, WOT, DFCO, etc). I'm using Tuner pro and have an AEM wideband installed. Because of my cam, I need to run a richer open loop idle . The problem is that often times my BLM will change in my idle cells, even though the ECM says I'm in open loop idle. Sometimes it goes up, sometimes down. :mad1: A BLM change of just a few points is enough to skew my mixture way rich or lean. Once I enter closed loop in normal driving, it's seemingly random as to when it happens and weather it adds or subtracts fuel. It will never change BLMs while its idling in open loop, and INT is always at 128. If I enter closed loop (i.e normal driving) and then come to a stop and idle it may have dropped down to say 125 or up to 131 etc. Either way, it's really annoying and I've literally spent months trying to figure it out.

Some things I've tried so far are looking at factory 454 bins that have open loop idle and basically copying all settings that have to do with open loop idle to my bin. (I'm assuming that the 454 bins lock their idle BLMs at 128??? Otherwise idle AFRs would be all over the place like mine.) I've also experimented a bit with enabling open loop decel (so I'm in open loop long before I come to a stop to no avail. I have tried disabling DFCO as well no change. It still does it. I've also experimented with the proportional gain settings, O2 constants. I'm honestly at a loss to understand why it's doing this and how to fix it.
Is there any way to lock idle BLMs at 128 or even lock all open loop BLMs at 128? The only way I've found so far is to set all BLM MIN/MAX to 128 but that means I don't have any block learn function in closed loop which is not ideal.

Any help is much appreciated! Thanks!

Block learn works with the integrator as a way to store and apply long term corrections based on short term adjustments. Block learn values do not change from 128 unless the integrator is at or beyond a threshold value for a period of time. Integrator values are driven by O2 readings so the first thing to check is to see if the O2 is responding correctly at idle. If it is not, and you attempt to tune around it, you may introduce other problems.

$42 has a number of variables dedicated to allowing open loop idle. They are here:

Code:

---

; OPEN LOOP IDLE CALIBRATIONS
; L05
;
;----------------------------------------------
LD29E FCB 4 ; 4 MPH, OPN LP IDLE THRESH
LD29F FCB 5 ; 2%, OPN LP IDLE THRESH TPS
LD2A0 FCB 1 ; 25 RPM, Open loop RPM thresh IDLE OFF
LD2A1 FCB 2 ; 50 RPM, Open loop RPM thresh IDLE ON
LD2A2 FCB 160 ; 8 Sec's, Open loop IDLE AFR Ratio enable TIME DELAY
LD2A3 FCB 100 ; 5 Sec's, Open loop IDLE AFR Ratio TIME DELAY
LD2A4 FCB 137 ; 13.7 Open loop AFR MAX AT IDLE
;----------------------------------------------

---

By increasing both the idle off and idle on thresholds, and by dropping the time delay values to minimum, you should enable open loop idle.

Alternatively, you could try limiting the min and max INT value or you could try adjusting the value at which INT affects BLM.

Hey thanks for the reply. I'm not sure how to access the hard code that you have in the screenshot there but here are the values I have currently entered in tuner pro for those variables. Most of these values are copied directly from a stock 454 auto bin that has open loop idle. I've messed around with all of them a bit but this is what I have currently:

Open Loop- Idle Thresh MPH: 10
Open Loop- Idle Thresh TPS: 3.2
Open Loop- RPM Thresh Idle off: 1100
Open Loop- RPM Thresh Idle on: 1300
Open Loop- Idle AFR enable time delay: 0
Open Loop- Idle AFR time delay: 0
Open Loop- Idle Lean Limit: 13.5

To be clear, I have no issue with enabling open loop idle. I have never once seen closed loop at idle with these values in any of my logs. My INT at idle is always at 128, which is baffling because I am seeing an instantaneous BLM change upon entering my idle cells with no INT change. My issue is that sometimes a random BLM change occurs to my block learn cells skewing my idle AFRs randomly in whichever direction it sees fit. :mad1:

I will check through my INT settings and report back

Just to explain this a different way, let's say I start up cold and begin driving in open loop coming to a few stops. Every stop I am at 128 INT and 128 BLM in open loop, and at the AFRs I want. I pull out on the highway and drive for several minutes. When I reach the proper thresholds, the ECM switches to normal closed loop operation. As I slow to a stop, before I'm fully stopped, the ECM switches to open loop exactly how I want it to. But now my BLM is now at 126 with INT locked at 128. A little richer than I want. I accelerate from the stop back into closed loop until the next stop which BLM may be down to 123 during open loop idle causing me to run pig rich. Later on the same drive I might see it at 133??? There doesn't seem to be any rhyme or reason to it. INT is always at 128.

Just to reiterate, I tried enabling open loop decel, that is to say any time I'm decelerating, I'm in open loop. So there is no closed loop operation even close to arriving at idle. It still does it.

I do not remember the 7747 using BLM corrections in open loop. I wonder if the changing BLM #'s you see are actually different cells as the engine's operating conditions change. Is it possible to track the BLM cell number?

I also do not see, during a brief review, a way in the code to reset or prevent BLM from being applied in open loop. So I have to wonder why this same problem does not affect factory open loop idle calibrations. a BLM change of only a few points, say from 128 to 123, should not cause the engine to run rich. In fact, this change indicates the ecm was attempting to correct for a rich condition.

Is your O2 signal supplied by a NBO2 or is it gnerated by the WBO2 controller? If you force open loop in all conditions, does the engine run well in all conditions? How closely does the O2 voltage track desired AFR when the ecm is in open loop?

Finally, have you done a bit comparison between your calibration and a factory open loop idle calibration to check for differences? Can you post the results here? Or I can do it if you upload a copy of your calibration and the factory open loop idle cal that you've been using for comparison.

I'm at work now but I plan on doing some more work on this tonight. To answer your questions: My idle in Park is pretty steady within the 40kpa/800RPM cell. In drive I'm either the 50 kpa/800 cell or the 40 kpa/800 cell depending on what altitude I'm at but I don't see it fluctuating.

I have basically copied everything from a 454 3 speed auto bin that I thought had to do with open loop idle. I agree it's odd that this is happening to me since stock open loop idle calibrations don't seem to have this problem. It is odd but just a few points of BLM change will drastically affect my idle AFR. I target 13.2:1 lean limit at idle. If I'm at 126 BLM I may see 12.8:1, if I'm at 123 BLM I may see 12.1:1. At 133 I may see 14.0:1. With BLMs at 128 I see the idle AFR very close to where I want it at around 13.2:1

I am running a Howell system with my own custom tune. The only thing I've changed is adding a park/drive signal to the ECM for smoother shifts in and out of gear (no idea why the Howell system doesn't include this). I am running a normal stock narrow band O2. Application would be a 1990 GM 1500 truck with a 350. I have an AEM wideband just for diagnosis and to watch/data log while I drive but no connection to ECM. I have not tried forcing open loop at all times, but I suspect this problem will not occur because this only happens once the engine has warmed up and entered closed loop (at non-idle conditions). Before reaching closed loop thresholds, no BLMs (including idle) will deviate from 128.

What I may have to resort to is starting from scratch using a 454 bin and use my current BPW and VE tables and go from there. Just seems like a lot of extra work because my tune is otherwise quite dialed in.

So I think I may have found the key. I raised my BLM- INT Delta to move BLM term from 4 to 6. That seems to have done the trick on my test drive. I'll drive the truck to work tomorrow and data log and report back. Thanks!

That might just crutch when Int affects changes to blm. There seems to be a lean and rich open loop parameter that ignores blm if it falls out of the range set. Suppose you could adjust those limits to “narrow” in to your target AFR.

I would agree with the plan to use a 454 base calibration. I had much better luck using a 454 cal for a 250 with fuel only conversion. The engine ran even better when we enabled electronic timing control.

I'd also agree with starting from a bin that runs open loop at idle if you want open loop at idle. There is likely some flag in the code that isn't mapped in the XDF you're using or some line of code that is different which also locks the BLM to 128 when idling.

Thanks for the advice everyone. For now raising the BLM- INT Delta seems to have solved my problem without a noticeable effect on my closed loop operation.

At some point soon, I may try starting from scratch with a 454 bin. It would take a little more work that I initially thought though because in addition to copying over my VE tables, AE, PE, etc. I'd need to go through and disable everything related to the TCC, EGR, Timing, ignoring the Knock Sensor, Disable flags so I'm not constantly setting the check engine light, etc.

You might find it interesting to compare the 2 bins and see what comes up different that is unknown. This would tell you how much different the bins are besides the stuff mapped in the xdf you are using.

I agree. We have plenty of documentation around for this ecm but there are still occassions where one can be surprised by a variable with a name that doesn't really tell you how it works.

Forgive me for being a bit of a newbie at this but how do I view the hard code of a bin to see/compare the individual lines of code? Is there a program that I can download to open them with?

Quote:

---

Originally Posted by 1project2many

Block learn works with the integrator as a way to store and apply long term corrections based on short term adjustments. Block learn values do not change from 128 unless the integrator is at or beyond a threshold value for a period of time. Integrator values are driven by O2 readings so the first thing to check is to see if the O2 is responding correctly at idle. If it is not, and you attempt to tune around it, you may introduce other problems.

$42 has a number of variables dedicated to allowing open loop idle. They are here:

Code:

---

; OPEN LOOP IDLE CALIBRATIONS
; L05
;
;----------------------------------------------
LD29E FCB 4 ; 4 MPH, OPN LP IDLE THRESH
LD29F FCB 5 ; 2%, OPN LP IDLE THRESH TPS
LD2A0 FCB 1 ; 25 RPM, Open loop RPM thresh IDLE OFF
LD2A1 FCB 2 ; 50 RPM, Open loop RPM thresh IDLE ON
LD2A2 FCB 160 ; 8 Sec's, Open loop IDLE AFR Ratio enable TIME DELAY
LD2A3 FCB 100 ; 5 Sec's, Open loop IDLE AFR Ratio TIME DELAY
LD2A4 FCB 137 ; 13.7 Open loop AFR MAX AT IDLE
;----------------------------------------------

---

By increasing both the idle off and idle on thresholds, and by dropping the time delay values to minimum, you should enable open loop idle.

Alternatively, you could try limiting the min and max INT value or you could try adjusting the value at which INT affects BLM.

---

I quoted^^^because that's what it looks like.

Not getting into too much detail, I think what you're looking for is what they call a "hack" and can be found in it's respective ECM number/mask i.e. 1227747/$42
http://www.gearhead-efi.com/Fuel-Inj...ams-Tuner-Info!

Ima thinkin like PE, some ECMs when entering PE, will lock BLMs at 128. Perhaps "open loop idle" bin follows this same methodology but since you started with a "closed loop idle" bin, maybe it uses the last known BLM upon returning to open loop idle.

I was talking about using Tools\Difference Tool in Tunerpro, not comparing line by line by hand. This will list the differences between the bins and show what is different that isn't mapped in the xdf.

To compare the straight code you would just open the bin with an hex editor. You'd likely find it completely pointless to do so though. What you're likely meaning is a decompiled version of the bin turned back into assembly language. The problem is that you'll not find that decompiled versions of every different bin file. Typically, there is one or 2 versions floating around that people have worked through and added comments to so they can understand what the code is doing.

Thanks for all the advice. For now my issues seems to be resolved by increasing the INT Delta required to change BLM. I went through the difference tool in tunerpro like you suggested and browsed through all the differences between my custom bin and the 454 auto bin I'm referencing. Honestly I don't see anything related to open loop idle that I haven't already copied directly to my bin and/or played around with to get where I want.

In the future, I'll probably start fresh with a 454 auto bin and go from there. For now this is working just fine for me, so I think I can put this thread to rest for now. Thanks again for the help!

You completely missed the point. You were looking for unknown differences, not mapped differences. If you find unknown differences, then the bin for an open loop idle application working right vs yours not working right could be due to the code in those unknown differences.

I see what you mean about there being some items titled "item not defined" by my XDF that are likely the cause of my problem. I will definitely be starting over with a new open loop idle bin when I have the time. I'm just not in a rush now to do it since I seem to have resolved my issue or at least made it significantly better by tweaking one parameter rather than starting from scratch with a new bin. I've got a few other projects that are taking priority right now including pulling/rebuilding an engine out of a 1960 International. :jfj: But I will get back to this one in time.

Thanks again for the info it was a big help.