@ Lambda CIS Cracks –> Duty Cycle, AFR measuring, CO etc.

[AndrewCologne]

Hello everybody,

I started this thread here, where I ended up in developing my own digital ECU for my US 1982 930/16 engine based car, providing the needed duty cycle to the frequency valve for achieving the proper 14,7:1 mixture.
Thoughts on replacing 911 SC ECU with Raspberry solution

It uses also a removable LCD unit where all actual values are shown while driving:
Sensor voltage
AFR
Duty Cycle
Fuel Control pressure
Engine (Oil)Temp
RPM
CO



Unit runs flawlessly – works with the original Narrowband sensor but even far better with an external wideband lambda controller and Bosch LSU 49.9 sensor. In my case an Innovate LC-2 controller but would work with all other AFR controllers out there as well.

When setting the CO to 0.5 % factory specs (with sensor disconnected and CO measuring before Cat) than this results in an approx. 45% duty cycle at 950 RpM idle – everything normal so far.
But I recognized (wenn sensor reconnected and engine is warm) when driving the engine at about 2000 RpM and higher the duty cycle decreases significantly down to 15% to achieve the 14,7:1 stoichiometric value based mixture.
I checked my pressures where the system pressure is 4.7 bar and the control pressure is 3.5 bar, so everything perfectly matching the factory specs.

2000 rpm —> ca. 30% duty cycle
3000 rpm and above —> ca. 15% duty cycle!
These are values when crusing very cautious, so no 30% throttle state (which would end up in 65% duty cycle anyway). Spark plug readings also showed an optimal color/state/face, means perfect combustion/AFR.

To me that means, that without an ECU control, the car would run very rich at higher rpm’s, even if not needed.
No further enrichments occur like in case of Euro WLRs do here occur.

Means if 15% duty cycle at higher rpm’s provides the needed 14,7:1, what would happen if you hit the gas for WOT, … here at WOT a „static“ 65% duty cycle mode in my case ends up in 11,0:1 AFR!, which is far too rich.


So my question is, … is there anybody running an AFR gauge/display with a proper idle CO setup of 0.5% who can confirm such low AFR results with a very rich resuting mixture at high rpm’s or at least tell me what AFR value is resulting at WOT or passing the 30% throttle position?

It would be interesting cause after this test, now in my ECU the optimal needed AFR of 12.5:1 when accelerating will be calculated based on the actual running duty cycle when not accelerating.
Means if 15% is the duty cycle when i.E. running the engine at 3000 rpm and 120 km/h then – when accelerating – no static 65% dc will be forced but a calculated increase of the dc from 15% to for example 25% dc will result in the perfect 12.5:1 stoichiometric value based mixture. And will be kept at 12,5:1 by reading the Wideband sensor Bosch LSU 4.9.

[2jmotorsports]

Man oh man...subscribed!

[9dreizig]

not sure what ECU you are using but here's a thread that is a wealth of info https://forums.pelicanparts.com/911-930-turbo-super-charging-forum/820323-digital-wur-plus-frankencis.html

[AndrewCologne]

Hi Todd,

thanks but here I do not refer to WUR or fuel control pressure related issues.

So here I just want to know from people who have installed an AFR gauge what AFR-Values do come up, when driving continuously with middle and higher revs on the streets/highway.

The ECU mentioned above is a digital build of my own and works as it should, means like the original one, .... but this one shows the values and ... the Wideband sensor proofs its proper functioning.

It comes even in an original Bosch ECU case so the connection is kept --> plug and play

The Acceleration unit still exists but its bypassed, means not needed.

[AndrewCologne]

Noone with a 911 SC and installed Wideband-Sensor and AFR Display here? :-)

[dhanl82]

Hello,

"So my question is, … is there anybody running an AFR gauge/display with a proper idle CO setup of 0.5% who can confirm such low AFR results with a very rich resuting mixture at high rpm’s or at least tell me what AFR value is resulting at WOT or passing the 30% throttle position?"

In response to your above question – I have a 1982 US SC with a fully functioning Lambda control system and an additional bung in my cat bypass running a AEM wideband A/F ratio gauge in my dash.
I do not have a CO meter to set idle mixture so I disconnect the O2 sensor ( as per factory direction) and set idle A/F mixture utilizing AEM gauge. I have the idle mix set at about 12.8 which when Past 1/3 throttle( with O2 sensor reconnected) results in an A/F ratio of right at 12.5. I realize that the – O2 sensor disconnected - idle mixture is a bit rich but this does not concern me due to the fact that with the O2 sensor connected the Lambda system corrects this to dithering mixture between about 14.3 to 14.9 with the same at cruse below 1/3 throttle, the only time not around Stoich is with a cold engine and past 1/3 throttle when it goes to 12.5. The car runs extremely well at all speeds.
Anyway, I hope this is some help.

Dave

[mysocal911]

Originally Posted by AndrewCologne

Noone with a 911 SC and installed Wideband-Sensor and AFR Display here? :-)

I doubt you will find detailed data, e.g. a graphical chart of AFRs or tabular data as a function of RPM & engine load,
on the Bosch CIS system you are looking for on this website. You may have to find another 911SC to test.
Have you considered that your sensor plate movement may not be properly calibrated?