Lambda reading issue

Post Reply
HondaDrag
Posts: 60
Joined: Wed Mar 06, 2013 5:12 pm

Lambda reading issue

Post by HondaDrag » Wed Apr 19, 2017 7:41 pm

Hi guys, today i have finished the drag car and i noticed that on the dyno when i apply throttle the Lambda reading goes up to 2.46 even upto 4.6 as well.
In idling the lambda works fine.
I have fitted an external wideband and mixture is spot on when i apply throttle.
Now this car has worked perfectely for two years and this year i have took engine out for a refresh and also upgraded the battery to 16 volts.
I have checked wiring and looks good and even tried another sensor and works the same.
The sensor i use is the NTK LZA-09-E1
I have checked the voltage on the Blue wire which is the Fuel output pin and with sensor connected it measures 5.45v.....i am not sure if it should be 10.7v
Could it be that sensor is not working good with the 16volts in the heater?
Any help appreciated.

stevieturbo
Posts: 834
Joined: Sun Dec 07, 2008 2:04 pm

Re: Lambda reading issue

Post by stevieturbo » Thu Apr 20, 2017 12:37 am

Can you add a resistor or something to drop the supply voltage to the sensor ( yellow ) down to 12-14v ?

Although no idea how 16v would affect it.

The fuel output pin, ie blue wire will provide the ground to the sensor heater. So I guess depending what sort of PWM it is doing at the time that voltage may appear variable, but the ecu is only grounding the output, not supplying a voltage.

Although if it measured say 12v between there and chassis ground, then pretty much no current would be passing through the heater. If it read 0v with a meter between it and chassis ground then the heater would be fully active and drawing max current

Are you sure no other wiring changes or ecu changes were made ?

HondaDrag
Posts: 60
Joined: Wed Mar 06, 2013 5:12 pm

Re: Lambda reading issue

Post by HondaDrag » Thu Apr 20, 2017 8:07 pm

Thanks for help.
I have cut the yellow wire and tried with several resistors in series but without success......i used from 5ohm to 15,000ohms
voltage did not drop much even with the 15,000 ohm but what i noticed was the lambda reading came very very slow.

I rechecked all the wiring and looks perfect.

Engine runs perfect just Lambda reading goes up to 4.66 and ecu stops the CLL even in idling and especially under full throttle!!!!

I am sure thats because of the 16volt battery the issue came.

Now i am going to fit an external WB with a 4.9 sensor and set Lambda2 for Closed loop and see from their till i find a solution for this issue.

I have also sent an email to support hope they can shed some light on my issue.

TimH
JT Innovations
Posts: 605
Joined: Wed Sep 01, 2010 2:51 pm

Re: Lambda reading issue

Post by TimH » Fri Apr 21, 2017 8:52 am

Just a half-educated "muse" here...

The NTK heater is set to 10.7V (from memory) in the CAL. This is clearly not battery volts, and it's a PWM output that's used. So there is quite likely to be internal monitoring and control of the "effective" voltage on the heater. Measuring this with a voltmeter will not give meaningful results because the PWM frequency.

There is no table for battery volts compensation - this would implicitly be taken care of: Veff = (SQRT(DutyCycle))*Vbatt

So, in other words, I would be very surprised if the change in battery voltage caused any change in the effective heater volts.

But, as said, this is a muse, not a given.

However...there is a period of time that the ECU is set to delivery maximum volts. This defaults to 30s and could now be delivering 16V not 14V...that *might* be outside the spec of the NTK heater so *might* have damaged the sensor element perhaps? The LSU4.9 has a maximum heater supply voltage of 16.5V, but a maximum Veff of 13V. If the NTK is the same, then it might hinge on whether the heater control strategy simply delivers 100% duty during warm up, or controls it to a specific maximum Veff.

Post Reply