A/F ratio at high rpm

#1

This past weekend was my first weekend with an A/F gauge on the dash. Was interesting. We may have a problem that we didn’t know about.

In 3rd gear when I accelerate to redline the A/F doesn’t seem to do anything odd. My DME is the one that we selected at the dyno that did the best both in terms of keeping a fairly rich mixture and also for not screwing things up to much between 4200 and 5200 when the DME apparently pulls back timing.

I say again, in 3rd gear accelerating to RL the A/F ratio seems to be fine. But in 4th gear it’s not. It takes longer to accelerate to R/L in 4th gear. And between 5500 and 6200, as the rpms slowly climbed, my engine would start going lean. Scary lean. The gauge would show 17 and then it would go off scale. That’s bad.

I unplugged the DME to “reset” it and the behavior was not changed. Then I swapped to a spare DME and the behavior was improved a little, but not a helova lot.

On a dyno the engine accelerates quickly so I don’t think that we see this leaning out. But in 4th gear, it’s bad. Ok, it’s equally bad for everyone, and our pistons haven’t melted from high temps due to the lean mixture, but it’s still a bummer.

#2

Dyno pulls are usually done in 4th gear as on these cars that is a 1:1 gear ratio. When your car was on the dyno it was run in 4th.

It is normal the mixture lean out to stoic above 5k, even when the AFM has been tweaked for about 13:1 below that point. But it should not go much past 14:1. The probable cause or running too lean above 5k will be a fuel delivery issue or dirty injectors. The AFM is maxed out above this point and the DME is running off rpm & temperature dependent maps. So if it is running lean the first suspect would be dropping fuel pressure.

#3

I have a FP gauge on my dash now for just these kinds of mysteries. FP was fine. My injectors were flowed and cleaned prob a dozen events ago.

I think that the key piece here is the acceleration rate. A dyno pull and a 3rd gear accel to RL seem to me to be pretty similar in how fast the engine gains rpm. But accelerating in 4th gear the car takes a while to get from 5400 to 6200. And it was during that slow climb that the engine would go lean.

My guess is that this behavior is happening to all of us and we’re just now discovering it because someone is for the first time running an A/F meter in his car. I’d bet $$ that if I went on a dyno there’d be no indication of this problem. Heck, I was on a dyno 2 events ago at VIR and this problem didn’t exist. The problem only seems to be visible when the high rpm acceleration rate is slow.

#4

“even when the AFM has been tweaked for about 13:1 below that point.”

What legal tweak would that be?

#5

Not again!?!?!?!? :S

::sticks head back in sand and enjoys care-free race weekends::

#6

See Jim’s response in this thread:

http://www.spece30.com/forum/42-engine/55499-king-tuts-dyno-experience

#7

See Jim’s response in this thread:

http://www.spece30.com/forum/42-engine/55499-king-tuts-dyno-experience[/quote]
9.3.1.1.4 The stock air flow meter plastic lid may be opened and a slight adjustment may be made to the internal components to allow proper electrical contact, away from a worn groove.
The rule only allows adjustment away from a worn groove in the contact medium. It is pushing the grey area to say that a clockspring adjustment is necessary to accomplish that.

Personally, I think Ranger’s car is leaning out due to his ram air setup.:wink:

Alternatively, is this a sensor issue? Is the sensor located too far upstream so it is getting too hot under long WOT runs? Has the sensor been calibrated? Good clean air source for reference while installed?

#8

I saw that, and read this:

That is pretty good for a 200k engine. You could improve the A/F ratio below 5k by an adjustment to the clock spring tension in the AFM, Right now it is correct for a street engine, but about 13:1 is better on race engine.

Not part of 9.3.1.1.4, and its not pushing anything.

The clockspring or cogwheel have no contact points that ‘wear’, and adjusting them creates no electrical ‘new track’.

It’s clear what the rule is meant to allow.

#9

I have always understood that if you adjust the contact point of the AFM to a different contact point on the carbon strip that you are affecting the resistance that is read and therefore need to adjust the clock spring to adjust for that. Just changing the contact path will mess with the A/F in bad ways if you don’t counter that with adjusting the spring.

Michael

#10

Are you moving that contact path up or down the resistance measurement, or in & out from the same point?

It would be challenging, to move that contact up or down the sweep, as its broadly and firmly secured at two points to meet at -a- specific point up and down the sweep. Bending that arm, only moves it in one of these directions…

#11

[quote=“Tow Mater” post=55618]Are you moving that contact path up or down the resistance measurement, or in & out from the same point?

It would be challenging, to move that contact up or down the sweep, as its broadly and firmly secured at two points to meet at -a- specific point up and down the sweep. Bending that arm, only moves it in one of these directions…[/quote]
I believe the goal is to modify the length of the arm by bending it. In other words, if the radius of the swept quarter-circle is 1.5", make it 1.4".

[attachment=1403]AFMboard.jpg[/attachment]

#12

Yup…but that’s not changing the contact spot up or down the sweep…which would change the resistance value returned by the unit, which may require clockspring work.

The contact remains in the same plane…resistance wise across the sweep within those laser cut resistors…altering no values to the ECU.

#13

[quote=“Tow Mater” post=55618]Are you moving that contact path up or down the resistance measurement, or in & out from the same point?

It would be challenging, to move that contact up or down the sweep, as its broadly and firmly secured at two points to meet at -a- specific point up and down the sweep. Bending that arm, only moves it in one of these directions…[/quote]

The contact is moved as you describe. Just in and out. I still think with the contact riding on a different carbon path that the resistance is different.

Michael

#14

[quote=“ilateapex” post=55623]The contact is moved as you describe. Just in and out. I still think with the contact riding on a different carbon path that the resistance is different.

Michael[/quote]
Looking at the circuits printed on the board, I don’t see how change in radius would affect the reading, but I ain’t no EE.

#15

If so, should stand out on an dyno A/F chart then…right?

Ive owned many a number of cars with flapper AFMs, turbo and not, and the clockspring trick applied to every one of em…not once have I heard that “Oh ya, if you fix that problem (carbon track), you gotta re-tune the device for it”.

Not trying to be a pain, I just dont see where the clockspring is remotely in the range of the existing adjustment allowance on a separate part of the device.

SE30 rocks…because of the simplicity of the rules…and they’re cool cars too…not…poodles…

#16

I’m not sure if the values can be read with a good Fluke or not but someone could try. I had trouble when I built my SE30 getting the A/F right and after messing with old AFM’s I finally broke down and purchased a new one. Things are better now.

It makes no difference to me if you adjust the spring or not although I do believe it is a gray area. If you don’t think moving the contact hurts anything then great. I am sure there have been springs out there moved and no way to police this. I actually liked it better when we were allowed adjustable FPR as then you would not have the lean problem.

Michael

#17

Mike:

Just curious…im not seeing a fleet of people with vented pistons from lean…is it a problem, or just an observed tuning issue that BMW left as a “thats odd…” gift?

Ricky Bobby would run hot & lean…you know it…

:woohoo:

#18

This has been a common mod for tracked E30’s over the decades and buying a random E30 off the street or from a BMW Club member means that in all likelihood this was already done at some point - completely randomly and with no research either. Dyno? What’s a frikin Dyno!?!?!?! LOL