This will be a continuing thread documenting the test procedure and results of WOODY'S (02 Simulator) fondly referred to as WOODY'S GIZMO.



The GIZMO was built to test the possibility of manipulating the fuel trim and timing advance / retard via the voltage output of the primary and secondary 02 sensors. Being the genius that our fellow member is, Woody whipped this thing up in his top secret laboratory in England and decided that I was the only person stupid enough to test it out on my engine and risk blowing it up Actually because I have removed all my CATS, my engine is the perfect test bed for this process.

That being said, I finally have a few days off work to install the thing and do some extensive testing and document the outcome... For those of you who are not familiar with how the 02 sensors operate, I have copied this section from the service manual.

UPSTREAM OXYGEN SENSOR
The input from the upstream heated O2 Sensor tells the PCM the oxygen content of the exhaust gas. Based on this
input, the PCM fine tunes the air-fuel ratio by adjusting injector pulse width.

The sensor output varies from 0 to 1 volt, depending upon the oxygen content of the exhaust gas in the exhaust
manifold. When a large amount of oxygen is present (caused by a lean air-fuel mixture), the sensor produces volt-
age as low as 0.1 volt. When there is a lesser amount of oxygen present (rich air-fuel mixture) the sensor produces
a voltage as high as 1.0 volt. By monitoring the oxygen content in the exhaust gases and comparing that to the
oxygen content in ambient air and converting it to electrical voltage, the sensor acts as a rich-lean switch.
The heating element in the sensor provides heat to the sensor ceramic element. Heating the sensor allows the
system to enter into closed loop operation sooner. Also, it allows the system to remain in closed loop operation
during periods of extended idle.

In Closed Loop, the PCM adjusts injector pulse width based on the upstream heated oxygen sensor input along with
other signals. In Open Loop, the PCM adjusts injector pulse width based on preprogrammed (fixed) values and
inputs from other sensors.

DOWNSTREAM OXYGEN SENSOR
The downstream heated O2 Sensor signal is used to detect catalytic converter deterioration. As the converter dete-
riorates, the signal from the downstream sensor begins to match the upstream sensor signal except for a slight time
delay. By comparing the downstream heated oxygen sensor signal to the signal from the upstream sensor, the PCM
calculates catalytic converter efficiency. This calculation is also used to establish the upstream O2 goal voltage
(switching point).

I'm off to the electronics store to buy some 20 gauge wire and snap connectors, to start the process... Stay Tuned...

 

Phear Woody's black box... PHEAR IT!!!

 

But do any of the ***** go to 11? That's the real question, isn't it?

 

interesting but why go this route? woody you're a mad scientist!

 

Why does the term "ludicrous speed" come to mind?

I'm thanking your engine in advance for allowing you to try this out!

 

Part 1 complete... I removed all four 02 sensors yesterday and added wiring taps on the sensor side. The grey wire is the sensor ground and the black wire is the 02 sensor signal to the ECU/PCM.

I then wired a 12v power supply for the GIZMO for testing purposes and used a voltage meter to mark the approximate rotational point of each pot from 0 to 1 volt.

I am now ready to reinstall the 02 sensors and measure the voltage of each 02 sensor after the engine warms-up and achieves closed loop operation... Prior to attaching the GIZMO...

STAY TUNED !

 

I have now reinstalled the 02 sensors and ran a baseline data log without the GIZMO in line. Note the values...

Next I will alter the voltage of the secondary 02 sensors and see where they effect the switch point of the primary sensors... And see if they have an effect on LTFT STFT Spark Advance, etc.

 

PHASE 2 TEST:

All connections from the GIZMO were attached and powered, the engine was run in closed loop operation at idle. The 02 sensor signal wires were jumpered from the GIZMO, but maintained the original signal wire to the ECU to see if the GIZMO would override the voltage from the 02 sensor to the ECU before I cut them...

I was able to adjust the voltage on the secondary bank even though the GIZMO was just jumpered / tapped into the sensor signal wire without cutting the original connection. If you look at the data log you can see where I turned the pot completely off on the 1/2 & 2/2 sensor bank and it logged .005v.

I did the same with the primary bank and the voltage was altered +/- momentarily, but since it was only tapped into the circuit I could only hold the voltage for a few seconds before it went back to normal...

Now that I know the voltage can be regulated by the GIZMO, the next step is to cut the 02 sensor signal wires to the ECU/PCM and have the GIZMO take over the voltage output...

THANKS WOODY THIS IS FUN.

 

Glad to see it all arrived safe and sound Derek, I'm only just catching up with this having been busy for the last few days. It looks like you have been having fun with the 'spaghetti' already.
Well, I'm probably mad anyway. The whole thing really sprang out of a conversation with Derek about some thoughts he had been having about a device for playing with the 02 sensor inputs to alter fuel trim. In the end we decided it was just easier to build something and see what happens - nothing ventured, nothing gained as they say.

 

Woody, You are mad... So, now I'm going to go cut some wires and turn over control to your GIZMO.

I'll be back with an update in a few hours...

 

Woody, for some reason the lights don't illuminate today and channel 4 seems to be dead (was working fine last night) So, I'm jumping both 2/1 & 2/2 onto channel 3 pot which is working fine... Channel 1 & 2 are working, but I have not cut the primary sensor wires yet, until I have full control over the secondaries... Will update soon...

 

Hmm... that's strange. On light means the unit has 12V, and the other means you have the regulated supply running. Is it possible that you are feed ing 12V back up on of the inputs? If so you might have burnt out one of the pot tracks. Its all fixable; you should be able to get spares from 'Mr Cucumber'.

 

Here's the bad news... My PC Laptop took a crap in the heat so I have not been able to run data logs so far today... But, before it crapped-out, I had set the secondary bank voltage at 2mv on both 2/1 & 2/2...

After I spent 30 minutes trying to get my freakin' #$%$%&%*%&%^& laptop working... I gave up on the data logging, but kept playing with the GIZMO even though I was unable to see the output voltage or timing / fueling data... But I had marked all the pots last night to the corresponding voltages...

Knowing that the secondaries were at appx. 2mv, I cut the signal wires on both primaries and let the GIZMO take over... I had the 1/1 &1/2 pots set at appx. 8mv when the wires were cut... The engine was at idle in closed loop operation at appx. 750 rpm... It took about 5 seconds before the engine started sputtering, not bad, but a notable change that almost seemed to be a misfire, and rough idle...

By the time I got back into the cockpit, it was idling very rough with the tachometer bouncing +/- 100 rpm... Then I noticed a CEL and was pissed that my laptop had crashed and could not check the DTC...

Since I was already knee deep into this process, I started adjusting channel 1 & 2 (primary sensors 1/1 & 1/2) to see what happened... I slowly reduced the voltages to about the 6mv mark and the engine started idling nice and smooth and the tach leveled off...

I then decided to give it some throttle and brought the RPM's up to 2k and held my foot very steady to maintain that RPM... The engine felt a bit out of sort, not a misfire, nor running rough, but something felt / sounded different... I held my RPM's perfectly at 2k and then reduced the voltage slowly on 1/1 & 1/2 and the RPM's slowly raised till I hit the 5mv mark on the pots and then flattened out at 2,300... When I reduced the voltage more, the RPM's dropped back down to 2k... And conversely when I increased the voltage the RPM's dropped as well... This told me that I could find a "sweet spot" in voltage output for the 2k RPM range...

I then took it up to 4k RPM and held it steady... then I adjusted the 1/1 & 1/2 channel downward till I found the "sweet spot", about the 4mv mark... As I reduced the voltage from 6mv, the RPM's increased to about 4,500 until the 4mv mark and then as I reduced the voltage more, the RPM's started to decrease back to 4k...

So MAD SCIENTIST WOODY... Even though I was unable to log the data to verify the results, I can tell you that your little GIZMO does something to alter the performance of the engine by over-riding the voltage output of the 02 sensors...


My PC Laptop has come back to life now that it has been sitting in the air conditioned house for an hour... So, I'm back to testing in a few minutes (and a few beers) Hopefully I can capture some data this time and see what this voltage adjustment is doing with the AFR / TIMING...

 

I think this is heat related... It's about 105 f here and was about 130 in the car... I took the box and my laptop into the house for an hour of cooling... and your GIZMO is back to work...

 

That would make sense, the regulator has a thermal cut-out at about 85 'C, and it runs quite warm so the high ambient may be enough to trip it out especially in the enclosed box. 105'F ambient is not something I usually have to worry about here LOL.

This all feels like some sort of strange remote control experiment sitting here.

 

Alright, everything is up and working... The PC laptop needed a cool down... And the GIZMO needed a cold beer... Wait, that was me...

So, here's what I got for ya...

Manipulating the 1/1 & 1/2 02 sensor voltage definitely does something...

I need to study the test 3 data log and compare it to the baseline log... It looks like .465 mv on the primary sensors gives me a balanced fuel ratio when the secondary 02 sensors are at .170mv... I ran the engine for about 30 minutes at various RPM's and at idle... Not under load... And adjusted the voltages based on the tachometer readout and the feel of the engine... I have to study the logs a bit more to see where the timing changed...

But I think we can get something from this process...

 

Now i understand what you are trying to accomplish and i tried to do a search for the following idea but didn't find anything. This is a MAF car and there is a lot of people on imports and LS1s that run either Apexi S-AFC or some type of MAF simulator to adjust the MAF signal to richen or lean the fuel on the car. Have you or anyone else ever explored this method??

 

Yep, its the next thing on the list, the O2 simulator box will let us test this too.

 

As Woody stated in the above post... Yes, unfortunately both the 02 sensors and MAF are tied in together in the grand scheme of fueling and timing on the Crossfire, so by just altering one, the other will nullify your efforts... So, to do this properly, you must alter the signals simultaneously...

We were just testing the effects of manipulating the 02 sensors alone, and it has been determined that the secondary sensors play a minor role in setting the switching point of the primary sensors which can control fuel trim to a certain extent. But, without simultaneous MAF adjustment, the timing dwell will choke or starve the engine.

I'm sure Woody is in his MAD SCIENTIST LAB cooking-up another box that has one more potentiometer for the MAF. I hope I get to test this one as well.

 

What computer program are you using? And can you do anything with it, beside read dtc's and read the data?