So I came across a new foreign company selling a O2 simulator, which Im looking for to get rid of my midpipe CEL light problem. Ive read all the past threads and the basic story seems to be that none of them work for the rx8... or at least theres limited success...here the email I got back from the company, and heres the ebay link http://cgi.ebay.com/O2-OXYGEN-SENSOR...sid=p1638.m122

Thanks for your inquiry!
Yes it will work on your RX-8 model.

RX-8's ECU has really has got a slightly overmapped ECU in contrast to the early RX series, but it considers all signals from exterior sensor as another unit.
The oxygen sensor systems work with the same principle in each make, only some engines are installed with special sensor (less than 1 %), but the main tasks remained the same, so you can be sure, O2 Sensor simulator will eliminate the oxygen sensor, accordingly it fits to your model.
We look forward to assist You again.
Feel free to contact me with further questions.



What do you guys think about the info he is giving me, and has anyone recently had success using a simulator in any way?

 

RX-8 in Hungary have a different ECU program than USA ones. I doubt he flew to the USA to test his $50.00 item. He also has this incorrect statement in the listing. " While check engine light is illuminating or flashing, ECU restricts the engine output and factory control program is running at medium / security level only, actually it limits fuel injection and injection timing." The no CAT cel has no effect on performance. Why would you buy this item when the seller has not tested it on an RX-8 and doesn't know how the removal of the CAT effects it? Looking at the other items in his STORE I bet every buyer gets the same answer to the question you asked.

 

good point

theres got to be a company out there that really knows what theyre talking about, and have tried it on an rx-8

 

do a search, this is not a voltage issue but rather a signal cycle issue, IMO it won't work

 

all it takes is a simple 555 timer chip and some resistors... just check with a VOM what resistances are being toggled when there is a good functioning catalyic and we can go from there

 

you might as well said that I need a flux capacitor
,
because I dont know what any of that means... if you can figure any of that out let me know, I know a lot of us would love to get rid of a CEL and would pay to do so

 

Tripod owns you

 

ive made o2 sensor spoofs before, i'll check and see what the ecu prefers to see coming from the downstream o2 and see if i can solder up something to fool it . much better than performing a bulbectomy in the dash!

 

Was searching today and found this thread...looks like it died about a year ago. Has anybody found a solution to this problem yet? I had made my decision to get a midpipe when I order my exhaust,but nothing bothers me more than a bright CEL in the middle of my dash all the time...

 

Take a black Sharpie and shade in the CEL light

 

Get an AccessPORT.

 

 

From my reading in the forums and elsewhere, the short answer is "NO".

In (very) layman's terms, the RX-8's ECU (apparently except for JDM '03's and early US 2004's which have not been re-flashed by the dealer) is not looking for a constant voltage from the rear O2 sensor - the ECU is constantly altering the mixture and looking for the results of this change via feedback (differing voltages) from the rear sensor. It is the lack of variable response from the simulators that throws the code.

If you have some time, then you may want to try a variety of "spark plug fooler" devices that move the rear sensor out of the full exhaust flow (or introduce some fresh air into the stream). This would be trial and error until you find something that approximates the (lack of) catalyst. Even then, you may find that your set-up doesn't work at all on other cars.

The above may oversimplify this topic quite a bit, but is a quick explanation. I (of course) stand to be corrected.

By the way, even the Accessport (and Hymee's device, I think?) does not solve the problem; it only suppresses display of the light / code. So far, this is the best solution to your problem.

Please post any results.

 

Being an EE, this problem intrigues me. The 555 timer solution, while interesting is probably too little. A microcontroller with A/D for the real O2, and D/A for the fake O2 seem like the minimum for a simulator. If the ECU is really tricky, other inputs will be necessary, such as temp, and I don't know what else.
Given that the AP can mask this issue, and I don't have a midpipe right now, I don't have a lot of motivation. Has anyone collected any data on the behavior of the 2 O2 sensors?