Another Rich Decel Thread..Need Help

[WeathermanShawn]

2002 Camaro Z-28 H/C as noted in signature. Stock 2002 injectors. Car is mechanically sound.

I have had difficulty accomplishing OLSD as my primary tuning option. Many hills and elevation changes in my area. I end up sampling multiple and widely varying g*K/kPa values at the same RPM. I have noticed the BARO value doesn't change accordingly while driving, and without holding pressure constant I am sampling data at RPM/MAP cells that becomes unreliable.

But my question is on the widely familiar Rich Decel problem many of us have. Whether I run CLSD, CLMAF, or 'regular' CLSDMAF (stock closed-loop), if I let off the throttle at any RPM, my AFR reads 12.9-13.2 for at least 9-10 seconds. Why does it do this?

I have lowered VE in the 15-20 MAP to almost zero g/kPa values, and it has no effect. I have even logged FTC cell 21 and it always indicates it is pulling 7-9% fuel, but my wideband reads rich. Is this just a reality with injectors, even stock? Is it just a reality of having a bigger cam?

The wideband is calibrated, I have no vacuum or exhaust leaks. Ultimately I realize I can just utilize DFCO and my problem is solved. But, it makes lower MAP tuning difficult, and I am just very curious why.

Anybody have any constructive opinions or comments to enlighten me?

[alwaysinboost]

bump for you, I'm having the same issue...

[jetblast]

go to the minimum pulse tables and decrease them by 5 to 10%.

[WeathermanShawn]

I'll give that a shot and let you know how it works out.

[hquick]

I've noticed the injector voltage offset table makes a big difference, B3701.

[WeathermanShawn]

Thanks hquick:

Yea, it is definitely odd to be so rich on decel with stock injectors. Perhaps I never noticed, because I previously used DFCO 100% of the time. It is only in doing the more advanced tuning (DFCO disabled) that I noticed it. My narrowband O2's are less than a year old, but maybe they are are switching too slow and contributing to the issue.

Just recently began logging GM.Volts. While my voltage looks good, I have seen a number of threads addressing B3701, especially with larger injectors.

Any new insight into adjusting this table, or should I read through some of the search posts?

[5.7ute]

I dont really see the point in changing the B3701 until this is known to be the cause. With stock injectors & the stock values it should be accurate.
There is the possibility that the transient fuelling tables may be adding fuel in this scenario. Or that the small pulse adders, min pulsewidth etc are active. I have a thread on here somewhere on calculating actual IBPW that may help pinpoint the problem you are having. Once you have the calc.pid made for your tune, and change the cal link txt to reference this pid things should become more evident.

[WeathermanShawn]

For those interested and especially running closed-loop...

Appeared to be a function of B4309 Throttle Cracker in Gear Table. In order to address some 'bucking' I had previously 'zeroed' out B4309. So what was happening upon throttle let-off was the MAP was following rapidly..much faster than the RPM's were falling. So it always took about 8-15+ seconds before wideband AFR's came back to 14.63.

On a hunch, put B4309 back to stock. It worked in that there was no further rich decel. Can't say I like the snail pace RPM's fall, but perhaps the previous setup was too fast for the STFT's to adjust to. Had no idea 'zeroing' out TC would do that, but guess it makes sense. Hope to find a happy medium.

[redhardsupra]

Shawn do you have logs for the 'before and after'? i'd love to see it on a graph

[WeathermanShawn]

Yes, I do. Got a before and after run this morning.

I'll attach and send them out later this evening, when I get off work.

[WeathermanShawn]

Here is a couple of quick snapshots of a log run I did this morning.

One is with the 2002 Camaro M6 B4309 'zeroed' out. The second log I simply re-loaded the stock 2002 Camaro B4309 Table.

The first decel shows External Wideband readings well below 14.63 AFR. The second shows stoich was hit on decel very rapidly, compared to the first. A few disclaimers. The Pids I selected were primarily for another tune/log goal. Willing and able to log identical IAC DMA pids and do one run down a hill with no TC, another with.

Perhaps the IAC function is simply allowing the RPM/MAP to fall slow enough to allow for adequate STFT switching?

Anybody else had a similar experience?

[5.7ute]

WeathermanShawn, all you have done is add air via the IAC valve to bring your AFR into line. This has not addressed the issue just masked it.

[jetblast]

with a h/c mod, seems to me that your condition would be expected especially, if while in decel., you have altered the ve tables in that particular region. if you are not getting any poping noises while on decel perhaps all you need is a slight adjustment in the mininum pulse tables. remeber that injector pulse width is now greater than stock at any given egine running state, so the injector should be comanded to closed a bit quicker and/or to a shorter time for decel, hence minimum pulse width tables.

[WeathermanShawn]

WeathermanShawn, all you have done is add air via the IAC valve to bring your AFR into line. This has not addressed the issue just masked it.

Whats wrong with that? Isn't that what we are doing with RAFIG with idle?

I think I understand your point though, but why would my minimum IPW be to rich for a stock injector?

Not saying I have discovered anything new..but isn't the idea of adding air an acceptable tuning method to address problem fueling areas?

[5.7ute]

Whats wrong with that?

Isn't that what we are doing with RAFIG with idle?

I think I understand your point though, but why would my minimum IPW be to rich for a stock injector?

Not saying I have discovered anything new..but isn't the idea of adding air an acceptable tuning method to address problem fueling areas?

Appreciate the replies(s),..just wanting to learn to do it right.

Thanks.

..WeathermanShawn..

If you are running rich under decel it is obviously caused by too much fuel for the conditions. Adding air will make the AFR read right but as you have found it will prevent the engine from decellerating properly, may as well hold your foot slightly on the gas.
The problem you are seeing is the inability of the injector to inject a small enough fuelmass. Whether this is in the tune or a characteristic of the injector is what you will need to find out. You will need to back calculate the actual commanded IPW to see this.(Not IBPW as this includes small pulse, offset modifiers etc.)
FWIW camming an engine will also lower the VE in the lower RPM/MAP areas adding to these problems.

[redhardsupra]

(Not IBPW as this includes small pulse, offset modifiers etc.)

where did you get this info? i've been looking for it for ages!

[WeathermanShawn]

5.7, thanks for the explanation.

The more I think about it, what you said makes a lot of sense. You are correct in that I would not be happy with that type of deceleration.

I will try that out.

[5.7ute]

where did you get this info? i've been looking for it for ages!

I worked it out when trying to make a pid for transient fuelling. I had noticed a discrepency when back calculating IPW from airmass. After further investigation by using a lookup() pid for the voltage Offset tables I could get values to correspond as long as the calculated IPW was above {B4006} small pulse threshold.
I then made a calc pid using the IFF() function & lookup() using {B4005} values which then had the rest of the values come into line.
This also confirmed my theory that the {B4005} small pulse adjust should not be linked to GM.IBPW but my new CALC.IPW pid.
thoughts?

[joecar]

Mick, can you post that calc pid... thanks.

[5.7ute]

No problems Joe. It takes a succession of pids that are all tune related.(They need specific info from your tune file to work)
CALC.IPW
*CLC-00-015
ms 0.00 30 0.3 "({GM.DYNCYLAIR_DMA}/{GM.AFR}/{GM.INJFLOW})*1000"

This one works for you lucky people that can use GM.INJFLOW to give CALC.IPW
CALC.IPW
*CLC-00-018
ms 0.00 20 0.3 "({GM.DYNCYLAIR_DMA}/{GM.AFR}/lookup({GM.MANVAC}, 0,8.7266, 5,8.7813, 10,8.8359, 15,8.8906, 20,8.9453, 25,9.0, 30,9.0547, 35,9.1016, 40,9.1563, 45,9.2109, 50,9.2578, 55,9.3125, 60,9.3594, 65,9.4141, 70,9.4609, 75,9.5156, 80,9.5625))*1000"

This is for us poor buggers that need to input our IFR into the equation to get it working. Again this would give CALC.IPW. You can obviously use only one or the other. The colored numbers are the ones that need to reflect your tune file.

CALC.COM_IPW

*CLC-00-019
ms 0.00 20 0.3 "iff({CALC.IPW}>0.699085,{CALC.IPW}+lookup({CALC.IPW}, 0.426,0.167173, 0.4860,0.136778, 0.547,0.121580, 0.608,0.106383, 0.669,0.091185, 0.729,0.075988, 0.790,0.060790, 0.851,0.045593, 0.912,0.030395, 0.973,0.030395, 1.033,0.045593, 1.337,0.045593, 1.398,0.030395, 1.581,0.030395, 1.641,0.045593, 1.945,0.045593, 2.006,0.030395, 2.553,0.030395, 2.614,0.015198, 4.012,0.015198),0.881456)"

In red is the value from {B4003} minimum injector pulse width. Blue is the values from {B4005} small pulse adjust. Yellow is the value in {B4004} default minimum pulse width
CALC.OFFSET
*CLC-00-016
ms 0.00 20 0.3 "{GM.IBPW1}-{CALC.COM_IPW}"

This pid will work out what is left of the GM.IPW that can be attributed to either transient fuel or the voltage offset table. I then made another pid using the "average" values of the voltage offset table as it is too large to fit in a custom pid. I only used between 13&15 volts as this is where you should be operating 99% of the time.
CALC.TRANSIENT_FUEL
*CLC-00-017
ms -10 10 0.3 "{CALC.OFFSET}-lookup({GM.VOLTS}, 13,0.3647, 13.5,0.3343, 14,0.3191, 14.5,0.3040, 15,0.2736)"


I need to clean all this up but you get the idea.

[joecar]

Mick, thanks... I'll use those to play around with...:cheers:

[5.7ute]

Mick, thanks... I'll use those to play around with...:cheers:

No worries. I believe the yellow in the last post(I cleaned it up a bit) was the default min pulse width for the injectors I have but I will need my other computer to verify. Unfortunately I dont keep good notes like Marcin. :doh2:
I will post back after lunch when I can get my hands on the laptop.

[5.7ute]

Once again I messed up. After redoing my notes on the weekend & looking into Shawn's logs I noticed a few mistakes in that previous post. Red should be your minimum pulse width, Yellow the default pulsewidth.
One other thing to note is that the small pulse adder table is not added if the default pulsewidth is used. This means that in a case like Shawn's, only a lowering of either the default pulse width or voltage offsets will allow a IBPW of under 1.7xx ms.
Changing the voltage offset will have an effect on the fuelling in all conditions, and should be avoided.(Not to mention this is calibrated in the factory & should be correct for his injectors)
Lowering the default minimum pulse width could be an option, but only if the injector can be kept under control at these smaller pulsewidths. Go too small & the injector will be basically shut off or too inconsistant.

[WeathermanShawn]

Mick was good enough to take a look at my tune and log. Perhaps his efforts will help others also..even running stock injectors at low VE's creates some rich AFR's.

Though not a real solution, I usually enable a 'soft' DFCO when I head up to the mountains. More for engine braking, but it cuts out the fuel also.

[5.7ute]

No problems Shawn.
One thing I should have mentioned in that last post is that sometimes the injector characteristics leave us with no room to move. So if we cant get the injector to have any control with a PW of under 1.7xx total then thats it. This isnt really a problem unless you need this low a PW under light cruise as DFCO can be utilised under decel conditions.
FWIW, in your log you would need an IBPW of @1.4 ms to keep AFR's in line under decel conditions. I do not believe this will be achievable or really that necessary. I will send you an email in a couple of hours on a few things we can try.
Cheers Mick