5-15% might be the result of some adjustments...I remember we adjusted lambda and when the mixture was near the 50s the idle was higher than we liked like 800rpm at idle..
Idle RPM is not a variable in CIS-E like we have.
It is fixed at the MB value or else it is AFU (see threads for High-Idle).
DC needs to be 50% at idle. It is important to make sure you have your DVOM probes correct so that pushing down the airflow plate momentarily enriches mixture and the DC should go Down (Lower %). Right or wrong, that is how I interprete the manual. Its a way to determine we are all using the same reference.
If you show 5-15% (call it 10% Average) you are very rich and the system is doing 90% of its total capacity to keep over-rich mixture in Control (14.7 AFR). When you add throttle the airflow plate causes over-enrichment due to exceeding the control range. Might cause flubbing and hesitation.
If you have your DVOM Probes backwards and in fact you are Super lean and when you tip in throttle the car will fall flat on its face. Aint got no gas.
The CIS-E is a Lambda (14.7) governor and has a limited control range. When I say "Rich" or "lean" but you are still reading some percent DC between 10% and 90% you are still 14.7 AFR but the system is nearly maxed out and any addition increaese or decrease in fuel takes you out of the control range of 14.7 and you will see DC infinitely 0 or 1 out of the range.
I spent a year learning this system by hard knocks and have my hands comfortably around it now. I have 50% average at idle and 35% or so at 2500RPM. THis puts me a shade on the too rich side at Wide Open Throttle by reference to my WB/O2 sensor and dyno runs. Just where I want it.
Hope I can help and return the favor to those that helped me. Its a tricky little system.