scarceller

The no fuel till 1400RPM: these cars have another map for decel fuel cut, this map only comes into play if the idle switch is closed and the RPMs are above this cutoff. In my chip I have set this cutoff to 1400RPMs, if I'm driving down the road at 3000RPMs and I simply let go of the throttle it returns to the idle stop and the idle switch closes. When this happens we transition from PT maps to Idle Maps and here we start looking at this fuel cut decel map and it sees 1400RPM is the threshold but we are at 3000RPMs so it commands the Fuel Injectors off (fuel cut) then if we continue to stay at idle stop the car starts to reduce in speed till we hit 1400RPMs once we hit this threshold the injectors come back on.

The factory fuel cut threshold is set very low, 1100RPM or so. This is part of the reason these cars dip below idle RPM when you return to idle. I think the factory set this very low for MPG but in my opinion you can easily set the threshold to 1600-1800RPM if you wish since most of us never drive are cars below 2000RPMs.

These DMEs have many more maps that are in play other than these 6 ign/fuel maps and not till you know how the entire system works can you properly tune these cars. For example there are other IAT and CHT compensation maps as well. I discovered all these maps and how they work, here's another interesting one: when in WOT the ignition can be compensated based on IAT comp map that many tuners don't know exists. So when you see the WOT ign map it's not the full story because at IAT temps above 40c these DMEs start to retard ignition, very very smart idea! and that's just one example of temp compensation. They do similar temp comps for fuel as well.

As for the cam timing I bought the car with these cams in place but I did sanity check them by taking out all the valve lash in cyl #1 and putting a degree wheel template on the crank pully and seeing that the exhaust and intake open when they should. They seemed to be properly timed at least within 1-2*.

This motor was built by Rob Parker in Atlanta and it has 2% or less leak down in every cyl. It burns no oil and screams, it turns about 270HP at the crank at 6000RPM (peaks at 6000) then drops off big time because of exhaust setup.

Here's the engine details:
Euro 3.2L
Intake port/polished and flowed
SSIs with M&K 2in2out
Still has stock AFM
Street and load dyno tuned for this setup.

The SSIs cause me to lose power after 6000RPMs, but this is a street car so no biggie.

Throttle response is incredible even with the AFM but this is because of the properly tuned PT Ign, if I use the stock ignition maps the response is noticeably different. I simply trusted the load dyno while I tuned the PT Ign and I have PT ign as hi as 52* advance at mid load in the 4000RPM range, the stock chips don't push much past 40*. On the load dyno the engine just wanted more ign advance and the torque just kept going up! The real trick is in properly tuning the PT Ignition in any engine.

KeithC2Turto

It is a typical stratigy to retard idle timing. It reduces a given part of the exhaust gasses and it heats the cat.

As to reversion, at the point in question I suspect you are having a reversion issue that keeps the cylinders pressurized as the exhaust valve closes. If you map out the exhaust events on a bar graph you can see that as one exhaust valve is closing another is near the middle of its cycle with the piston moving at it's highest rate.

At TQ peak the cam timing and the mass and speed of the gasses should be at there ideal and the cylinders are experancing the best scavagening and the motor will be operating at it's highest VE possable.

Your cams do not have that much overlap such that it wold probably be pushing exhaust up into the intake I am guessing or your idle would probably be rough.

That brings up in interesting question. Could what you are experanceing be a bit like what an agressive cam dose at idle?

whalebird

Sorry guys, I don't have an answer on this one. I'm a regular guy and this is getting thick for me.
Good luck

KeithC2Turto

Sal,

I do not know if I am seeing somthing because I want to.

First, I assunm that if the cell's run to a half way point of each cell. Thus, there would be an ignition change half way between 1760 and 2400rpm. This would put a timing change at 2080rpm.

On the ingnition side the delta is most significant at that 2080rpm and near where your issue seems to be. That is, the timing change is more significant at that point than any other. The width of the cells in that area my not be narrow enought to fine tune things really well there. Maybe Porsche picked that point along the rpm range for a reasion.


On the fuel side, if you look at the part throtle cell entries as being corlated to changes in VE, it is lowest in the area you seem to be haveing issue with. They start out, reduce throught the low 2000's and then increase with rpm after that up to peak RPM. Thus, the ,motors VE seems to be lowest in that area.

If you look at the WOT trim map, the value is highest where VE is the greatest. Porsche quotes peak TQ at 4800rpm which is where the cell value is the highest.

On a side note, I am suspect in any fuel cell value after about 5200rpm where the AFM seems to stall and looses its fidelity as it seems accepted that Porsche may have dialed in more fuel in thouse areas as a protection against going lean.

Thus, there may be somthing in the 911 basic design' be it a significant area of exhaust reversion or be it the cam's timing effect that could be converging at that point. I do not know if the width of the cells are such to be able to tune that area perfictly. It may be a comprimize has to be reached.

scarceller

Keith,

My idle is good but I did bump base idle to 920RPMs and I run a tad on the rich side at idle AFR=14.2 or so with 7* ign advance, that's the sweet spot for idle in this motor. However, I suspect I have some unburned fuel at idle because even at 14.2afr I smell it slightly rich. I'm guessing I have some raw fuel crossing over to the exhaust.

I have no complaints about idle or any other area of drive-ability other than this surging at lo-load lo-rpm. I'm not done testing this surging. I'm backing off ign in this area to match the stock 89 ign map. Then I'm going to richen fuel to afr=13.5 and see if it helps. I'll also try leaning fuel to afr=15.0 just as another test. I already know that backing off ignition helps.

scarceller

Keith,

The DME interpolates the values from one cell to the next so ignition is smoothly changing as RPM increases. The DME has code in it to do this, it does not stay at one fixed ign value till it sees the next cell. All the maps interpolate and calculate proper values on the fly.

At WOT my AFM signal taps out at 5200RPMs, I measure it in real time. But at PT I think it really never taps out unless you are at very hi-load near 6000RPM. I think the AFM has plenty of range for PT loads.

KeithC2Turto

Thanks on how the cells work. That makes more sense.

Also good to hear valadation on where the AFM tops out.

I think of this more in terms of Hp more than rpm. Thus, if one were to cruse at 6000rpm and say 100hp, the AFM should be effective.

On the flip side, if one built a suck through turbo and it could make about 185 hp at 3500rpm, that should be where the AFM would stall.

The best.

scarceller

Correct, the AFM was built to handle most PT loads across the board. If you tried forced induction of any kind with the stock AFM you would be toast! It has very little head room.

At WOT it runs out of head room at 5200RPM after this the fuel is calculated from RPM, IAT and CHT alone. But below 5200RPM the AFM signal is used even at WOT. Some misconceptions exist out on the internet and forums about this. Some folks seem to think AFM signal is totaly ignored at WOT, they are 100% wrong! I've seen the code and I know AFM signal is always used to calculate load and certainly used to calculate fueling but once the AFM tops out then only RPM, IAT and CHT is used.

scarceller

Here's the 95 993 ign maps, notice they also de-tuned ignition on these cars in lo-load below 4000RPM. I'm seeing a trend.

scarceller

I wanted to update what I learned about surging at idle or PT. Through lots of testing I found that surging occurs for 2 reasons and they are both related.

1 - Mixture is to rich, this increases flame front speed and changes when max cyl pressures occur. If the flame front speed increases then the burn occurs to quickly and max cyl pressures will occur soon in the cycle, meaning you may be creating max pressures before 10-15 degrees ATDC. This is bad.

2 - To much ignition advance will also cause the same condition. If you fire the burn to soon you have same issue in that you achieve max cyl pressures to soon in the cycle.

So fixing surging can require you to adjust both fuel and ignition. But in a stock setup ignition usually does not change but AFR can easily change. I found that usually surging is a result of to rich of a mixture.

I have done lots of testing on the above by intentionally altering ign and/or fuel and can easily cause surging on purpose by richening fuel or adding to much ign advance. I have a chip emulator in my 3.2L Motronic DME and I can change fuel and/or ignition on the fly with car running.

What I discovered is that surging is the result of altering Flame Front speed by increasing flame front speeds with richer mixtures. Or advancing timing to much and thus starting the burn to soon.

Here's the important relationships to remember:
1 - Rich mixtures have faster flame front speeds and thus require less ign advance.
2 - Lean mixtures have slower flame front speeds and thus require more ignition advance.

This means that messing with AFR alone can be a very dangerous proposition! If you alter AFR you must revisit ignition. I have learned about these relationships first hand on a load dyno while doing Part Throttle chip tuning for light loads at PT.

Since I started this thread about surging, I figured I'd update on what I've learned.

whalebird

Thanks for the update. One can see why twin ignition was used so early and introduced in the production 3.6s.

KeithC2Turto

I understand the relationship between AFR, ignition, and flame front speed.

Not sure I get how the surging happens. What is actually happening that makes the rpm surge? Is there some type of instibility causing the rpm's to increase and decrease?

I am sure you know this but:

I belive a richer mix runs cooler because combustion is speeded up. Lean makes it hotter because combustion is slowed down an more heat is retained in the cylinder/head.

Cam, cam timing, and compression (that is effective compression rato) also effect flame front speed.

The faster we can make the burn rate, the later we begin ignition, the less pumping losses, the more power.

This is also what twin plugs do. Insted of starting ignition at say 30 deg BTDC at WOT to achive peak pressure at the ideal time, timing can be set later, say maybe at -25 deg. Thus, less pressure is working against the piston on the way up before the power stroke.

scarceller

Keith,

I'm not sure why it surges but if max cyl pressures occur to soon the car surges. In the 3.2L at idle you can try this test for yourself: While at idle simply disconnect the CHT sensor this tricks the DME to think the CHT is dead cold like -20c and richens the heck out of the mixture to 12.0AFR range the minute this happens the engine starts hunting for idle.

I have played around with idle AFR and Ign by simply altering the idle maps. If I add more fuel at idle it starts hunting then if you back out timing while still having the extra fuel it stops hunting. Same goes for PT lo-load, I have also done same test in the lo-load PT rows of the map I add 6-10% extra fuel she surges at cruise back out on timing it stops. I was most surprised by these findings.

So if you have a car that's acting up check your base mixture you may simply be to rich. Timing on these cars is very rock solid and does not fluctuate based on age or wear of components. But fuel mixture certainly can.

Also, for those cars that idle hunt during cold start same thing. Usually it's because you have to much fuel during cold start.

KeithC2Turto

There is probably an advantage to runing a little rich at low load points and adjusting the timing.

Might help throttle response a bit. I know this works with CIS.

As we know the AFM Motronics does not have a TPS to instantly alert it to acceleration. It must depend on sensing a change in air flow or engagement of the WOT switch. With this is probably a delay between when the cylinder start pulling more air and when the ECU gets that there is a need for more fuel (lean surge) and can deliver.

At least with CIS this strategy seems to help where there is a more significant delay and a more significant lean surge.

Sounds like your car keeps getting better and better.

scarceller

Keith,

Yea the car runs great now. I also learned a lot from this book:
http://www.amazon.com/Engine-Managem...9083596&sr=8-1

I highly recommend it for learning howto tune on a load dyno.

Another little trick to help with throttle response is to retard ignition on tip-in acceleration. Just like you add accel fuel at the same time you retard ignition. The idea is to have the energy from the burn reach max cyl pressures a bit later (25deg ATDC) this helps chase the piston down on the stroke and gets the engine accelerating (picking up RPMs) a bit quicker. You can achieve this by dialing back the ignition in the last 2 hi-load rows of the PT Ign table, since you never really hit these rows except for tip in acceleration it's not really necessary to tune these last few row for Max Torque. Some modern EFI systems allow for this feature but the stock Motronic does not so all you can do is mess with ign in last 2 hi-load rows and dial it back some.

Then here's the full story on how fuel mixture effects ignition advance. As already discussed Rich mixtures need less advance while lean need more. But there is still more to this: the theory is that if fuel particles are tightly packed together they need less advance because the flame front sort of hops easily from one to the next. So if we think about this, how can fuel particles get tightly packed? 2 things can do this:

1 - More fuel in the mixture, richer AFRs
2 - Pressure, higher loads pack fuel/air more tightly

This is why you'll see less ignition advance at high engine loads (like WOT) than at lower engine loads. The higher the cyl pressures the less ignition advance required. Also why hi-comp motors or turbos need less ign advance. Always keep in mind that denser mixtures have fuel particles tightly packed and thus require less advance. Once I got this idea straight in my head tuning became much simpler.