A clue about light-load bucking

[Chucko]

Thank Ford (and Moates) for datalogs! I think I may have found a cause for bucking in heavy traffic.

My '89 Mustang has a stickshift, a cam, and a 65mm throttle body (stock was 55mm). My commute is short but the freeway traffic is frequently heavy. I try to use a delicate touch on the throttle to avoid braking and shifting as much as possible. I will occasionally get severe bucking at light loads around 1000-1200 RPM, usually in 2nd gear.

It happened a couple of times on the drive home tonight while I was datalogging. I found something very interesting in the log - the bucking occurs when TS is -1. It works out that my "delicate" touch on the throttle is TOO delicate - the throttle is barely cracked, not on the idle stop, but below the various "closed throttle" thresholds in the EEC. To put some real numbers on it, RATCH is at .885V and the problems are occurring at a TP value of .90V! So the EEC throws it into idle spark timing, and the trailer hitching begins.

My next tune for tomorrow's commute has smaller values of DELTA, HYSTS, DELRAT, DLHYST, etc., in hopes of reducing the window of opportunity. I'll let you know how it works out.

[decipha]

nice find... sounds like your on your way to knocking out that buck

i bet your smiling from ear to ear probably something like a fat kid in a candy store just for a visual reference.

[cgrey8]

Fat kid in a candy store with $100 bill probably accurately described how I looked when I 1st got my engine to turn over and was able to drive it out of the barn.

[Chucko]

Well, it ain't that easy, as it turns out.

The solution "should" work but I find there's not a lot of margin for error. This morning's commute had smaller increments for DELTA etc. but I still wound up triggering closed throttle spark when I didn't want it.

Bins and datalogs to follow, per request of another forum member.

[Chucko]

The plot thickens. The ISC plays a major role in throttle response at small throttle angles. I think the ISC parameters have a lot to do with my bucking problem too.

FN882 sets the maximum amount of ISC (DASPOT) air at any given RPM. In my early hacking on my tune, I had made FN882 steeper than either the stock A3M1 or X3Z bins. So the ISC had more authority to control airflow at low RPMs.

The other parameter that has an effect on throttle response is DASPTK. It effectively controls the "gain" of the ISC by scaling airflow proportionately with throttle angle. The units are lb/min/TP count. A "TP count" is about .005V. I had lowered DASPTK to .02 lb/min/TP count, per the X3Z value. Stock on an A3M1 tune is .06. That should have helped my cause, but the throttle response at low RPM and light load was still pretty touchy.

So here's where I'm going with this. With a throttle body that's bigger than stock, the last thing I need is the ISC making my throttle even more twitchy. By flattening out FN882 so that the dashpot doesn't have as much authority at low RPMs, and cutting DASPTK down to half the X3Z value (.01 lb/min/TP count), I hope to have more direct control at small throttle angles.

It should take me more throttle travel to get to the buck zone. That and the other changes described earlier in the thread should combine to keep the throttle out of the Closed Throttle region under these conditions.

I imagine I'll have to tweak the Accelerator Enrichment tables again.

[cgrey8]

Interesting. It sounds like you are digging into the ISC stuff like I had to a few years ago. It gets deep, but interesting how ISC is controlled.

However one thing I'm still not 100% positive of is if the ISC can cause bucking unless it is so aggressive of a control that you have the ISC opening and closing really fast. Generally, I associate bucking with a lean burn condition. A good 16.5 AFR will get my engine to hesitating and surging. However while cruising and typical traffic acceleration, the AFR can go up and flirt with 16.0 and often I never know there was a lean condition in progress unless I look over at the computer screen.

[vristang]

Have you played with Injector Timing?
It made a big difference with low speed bucking in my car... but don't ask me to explain how... lol
I just used EA to find better values.

[Chucko]

However one thing I'm still not 100% positive of is if the ISC can cause bucking unless it is so aggressive of a control that you have the ISC opening and closing really fast. Generally, I associate bucking with a lean burn condition. A good 16.5 AFR will get my engine to hesitating and surging. However while cruising and typical traffic acceleration, the AFR can go up and flirt with 16.0 and often I never know there was a lean condition in progress unless I look over at the computer screen.

The ISC is only indirectly involved. What's going on is a brief decel to a point where the required throttle is below the closed throttle threshold. Then the CT Spark vs RPM function, FN111, takes over, and things get seriously retarded. The datalogs show that nothing is happening with TP or ISC duty cycle until the bucking has been going on for several seconds. According to a book on tuning I've been reading, Engine Management Advanced Tuning, by Greg Banish, p. 74:

If spark advance is left too low in part throttle regions, the result can be a bucking or "trailer-hitching" sensation on the road as the engine misfires. Remember that larger camshafts increase the natural EGR of the engine and require more spark advance to compensate for the cooler combustion temperatures.

So the ISC gain change is really intended to limit ISC authority at light loads and low RPMs, so the throttle itself can be open wider under these conditions, leaving TP out of the CT zone. I hope.

Datalogs are coming... when I find time.

[cgrey8]

Interesting...so the engine can actually misfire with too little spark advance which to a datalog would look like leanout since by the nature of a misfire is not all the fuel burns, thus not all the oxygen does either and the WB registers that as a lean condition when it was just a misfire.

[Chucko]

Yep. Fuel burn is fastest around lambda 0.82 or so (i.e. A/F of 12ish). It gets slower as lambda gets leaner.

So lean mixtures require more advance. But misfires read leaner than they really are. So if we observe bucking along with a lean exhaust reading, which is the cause, and which is the effect? Even WB O2 sensors can't distinguish because they only measure oxygen, not combustion products.

I haven't yet played with injector timing in this region. I think it would help by ensuring that the injected fuel actually goes into the cylinder and not out the tailpipe or up the intake tract, so you don't get as much of an inherent EGR leanout effect. But that's just an unproven theory.

EDIT: Banish says fastest burn at 0.9 lambda (i.e. A/F of 13ish), somewhere between best power at 0.82 and stoichiometric.

[decipha]

well that conincides with my theory, its been recommended around here to reduce timing to aid bucking, i've found personally that increasing time reduced bucking, i never publish stuff like that because when you try to go against the massis folks tend to 'tune you out'

[cgrey8]

In my early days I too was told that decreasing timing would help with bucking. In fact, I probably helped to spread that belief from time to time. But I found 99% of my bucking was not from timing, but from leanout either due to slope issues, MAF curve, or other. Although as was mentioned above, it is often difficult to tell when you are appearing to leanout due to bucking or due to misfires. They feel much the same. Although this is the 1st I've seen the claim so blatantly stated that increasing spark can reduce bucking.

Now, do I believe that increasing spark will always do this? No. I've seen way too many times when making fuel adjustments fixes bucking so I still feel like a majority of the time, the engine really is going lean and bucking because of that. But for light-throttle bucking, not tip-in bucking, I could believe this might have some merit. It'll be interesting to hear from people suffering from this whether this is a legitimate fix for light-throttle buck OR if they've fixed this issue in the past with increased spark.

[EDS50]

well that conincides with my theory, its been recommended around here to reduce timing to aid bucking, i've found personally that increasing time reduced bucking, i never publish stuff like that because when you try to go against the massis folks tend to 'tune you out'

In my personal experience with my own combinations, I have found this to be true also. On my n/a car I have reduced timing in low load/rpm situations by as much as 20 degrees just to find that it agrivated my bucking situation. Increasing timing helped to ease but not completely fix my bucking issues.

[decipha]

oh great that makes me feel better, i take comfort in knowing that im not alone!

well that conincides with my theory, its been recommended around here to reduce timing to aid bucking, i've found personally that increasing time reduced bucking, i never publish stuff like that because when you try to go against the massis folks tend to 'tune you out'

In my personal experience with my own combinations, I have found this to be true also. On my n/a car I have reduced timing in low load/rpm situations by as much as 20 degrees just to find that it agrivated my bucking situation. Increasing timing helped to ease but not completely fix my bucking issues.

[Cougar5.0]

Quick comment - you can usually feel a loss of power & see a trending lean AFR before a buck/misfire occurs due to leaning out (i.e. - there's plenty of 'warning' in the datalog.)

[cgrey8]

I never thought about it, but you are absolutely right.

Cougar5.0, would you agree that a true leanout condition can be characterised/identified as happening when multiple datalog rows show a lead-up to the lean condition as opposed to a misfire which would likely be quick blips of very lean AFRs amongst near-stoic/rich datalogged rows?

[Cougar5.0]

I think that's true - so long as the logging rate is high enough so there are enough points to see the difference between a more gradual leanout and lean spikes.

[EDS50]

Quick comment - you can usually feel a loss of power & see a trending lean AFR before a buck/misfire occurs due to leaning out (i.e. - there's plenty of 'warning' in the datalog.)

How do you determine the cause of light load bucking when you dont have a loss of power or trending lean afr or lambses to show up in the datalogs? In my datalogs my afr's and lambses are pretty much even @ 14.0 during these conditions and dont dip lean or rich before, during or after the bucking occurs. I target 14.08 since I run e-10 and the bucking is not attributable to timing in my case since I have experimented with timing at thoses load levels +/- 20 degrees come to find out the car like more timing and eases the bucking but doesnt get rid of it. I have to command 12.0 afr to get rid of the bucking forcing o/l but I know thats not right.

[cgrey8]

This thread has been split at this point. To follow the split, click here:
Setting up for alternate fuel types

[Jdubes]

I am very interested in this subject.

My truck does a low RPM/load buck and it goes lean. I haven't looked very far into it yet because of all the other issues I have going on but it' extremely important.

My truck is a stick, and it has low range. So when I'm in 4wd Low, the throttle/engine input is much more sensitive and exaggerated. Bucking in this senario is unacceptable, lol.

[Chucko]

I haven't eliminated the bucking yet.

Setting the closed throttle thresholds lower has definitely helped keep the buck zone in the part-throttle regime, where I have more tools to deal with it. I suspect I have a vacuum leak, which means anything in the very light load range will be leaner than commanded. I need to solve this first.

[Ford347]

I have just started my tuning with BE. I am thinking this may be a reason why my car is bucking during low load part throttle. I noticed today while cruising at 40mph when I let off the throttle the spark advance drops from 30-40 to 19. I believe it is thinking it is at idle. My car bucks at low loads in any gear.

Now, I have not fully tuned my injector slopes or MAF (just put in stock values, but drives good). Has anything more been found on this subject?

[decipha]

its suppose to do that at closed throttle it should reference the spark idle strategy and throw timing around to target desired idle rpm

I usually force open loop below 13% or so on most tunes I do to help avoid this situation all together, I also spend some time adjusting the lower voltages of the maf to maybe help out as well. *knock on wood* no bucking problems to date, I think I have been very lucky thus far, also I mess with the cbaza strat which doesn't seem to be plagued by the bucking issues of the older ecu's

[Chucko]

Well I guess I should post a followup. I did fix the vacuum leaks and the tune in general is a lot better integrated.

Bucking is not eliminated, but it's limited to a much smaller range of conditions now. Aside from changing the scalars mentioned above, I changed the idle spark vs RPM function FN111, which is usually flat, to ramp from 24 degrees at 400 RPM to 40 @ 1200. So when the throttle is closed anywhere above 1200 RPM - or the EEC thinks it is closed - spark is at a healthy advance. Loads are super light under CT so this poses no real risk for my premium-only tune and stock compression. This also helps eliminate backfires from decel fuel shutoff.

I also set injector timing at idle and the region just above it to 1200 RPM to a fixed 328 degrees - where the intake valve opens for my cam. I haven't noticed any real change in behavior. I'd like to think it helps with idle during warmup.

No idea when I'll get the last of the bucking out, but it's now a small enough issue I can frequently drive out of it, even in bumper-to-bumper traffic.

[Chucko]

I usually force open loop below 13% or so on most tunes I do to help avoid this situation all together, I also spend some time adjusting the lower voltages of the maf to maybe help out as well.

I've just recently bumped the open loop threshold on my tune to 13% to try and get more control of transient fueling while shifting under closed loop.

I haven't tweaked on the MAF curve. I have a Pro-M MAF I bought new from them and the 9 point sheet that came with it. I'm not seeing any indication that it needs tweaking.

[Ford347]

its suppose to do that at closed throttle it should reference the spark idle strategy and throw timing around to target desired idle rpm

I usually force open loop below 13% or so on most tunes I do to help avoid this situation all together, I also spend some time adjusting the lower voltages of the maf to maybe help out as well. *knock on wood* no bucking problems to date, I think I have been very lucky thus far, also I mess with the cbaza strat which doesn't seem to be plagued by the bucking issues of the older ecu's

I'm doing a lot of searching and reading hopefully I'll be able to reduce my issue. How do I force open loop below a certain percentage?

[2Shaker]

IMO spark can contribute to bucking, but playing with spark control will never totally solve it. I'm in the ISC and lean mix camps when it comes to the total solution. In fact I think if fuel is correct and no bucking is occuring you would have a hard time inducing bucking by changing spark control values. I gave up on spark as the cause of bucking long ago and sicovered if I set the minimum PW scaler a little rich (CBAZA) it cured the bucking.

[Chucko]

I'm doing a lot of searching and reading hopefully I'll be able to reduce my issue. How do I force open loop below a certain percentage?

Scalars LOLOD and LOLODH control the lower threshold for closed loop. The EEC switches to open loop when load < LOLOD - LOLODH and back to CL when load > LOLOD .

[decipha]

you force open loop at low loads by the scalar " LOLOD " anytime the load goes below the value it will force open loop

this brings up a new thread i'll be posting shortly

its suppose to do that at closed throttle it should reference the spark idle strategy and throw timing around to target desired idle rpm

I usually force open loop below 13% or so on most tunes I do to help avoid this situation all together, I also spend some time adjusting the lower voltages of the maf to maybe help out as well. *knock on wood* no bucking problems to date, I think I have been very lucky thus far, also I mess with the cbaza strat which doesn't seem to be plagued by the bucking issues of the older ecu's

I'm doing a lot of searching and reading hopefully I'll be able to reduce my issue. How do I force open loop below a certain percentage?

[Ford347]

IMO spark can contribute to bucking, but playing with spark control will never totally solve it. I'm in the ISC and lean mix camps when it comes to the total solution. In fact I think if fuel is correct and no bucking is occuring you would have a hard time inducing bucking by changing spark control values. I gave up on spark as the cause of bucking long ago and sicovered if I set the minimum PW scaler a little rich (CBAZA) it cured the bucking.

I wasn't totally suggesting that spark was the problem. More that the EEC thought it was at idle instead of actually cruising at 40mph.

Thank you for those scalars.