bhammer

That might be doable for me to collect the data and help; so long as you are in C# or VB. I'll have to look back at my oscope and see. I know it has a save feature but cannot remmeber how much and what data can be saved.

paimon.soror

iTrader: (15)

I am far more proficient in C# than VB, but either will work

bhammer

And me, the other way. Just something about being a little lazy with VB that you can't with C#. And, I am by no means and expert. I am all self taught with both for work projects. The most extensive I got was a nice SQL backend database and GUI for users. Had to have one person responsible for updates as the concurrency issues got in the way.

I need to pick up a new BNC to banana adapter for the scope. It got smushed today under a car tire.

How much and what kind of data do you think we would need?

paimon.soror

iTrader: (15)

You know, now that I think about it I am not sure we would even need a scope, I believe there is a PID for the downstream o2 voltage reading across the CANBUS. I will have to check when I get home, but we could surely use that to see what the heck that O2 sensor is doing during normal daily operation. The next question would be exactly what the ECU does to calculate 'expected' value before it triggers a malfunction.

bhammer

Voltage I think is easy to get to but probably only half of the picture. My assumption is that the O2 sensors for these cars are more a frequency generator with an oscillating freq. than just a simple voltage device. My guess is that the ecu is looking for both voltage and frequency and compares that to a table based on current outside temps and the egt.

With a couple capacitors, resistors and a 555, it'll be no problem!!

jrx13

Yeah you don't need a scope for the rear 02 sensor. You can just graph the voltage. Any graphing scan tool can do this. The rear 02 is a regular 02 sensor and not a wideband.

The PCM runs what is a called a "Catalyst monitor" whenever certain conditions exist. It then commands the Fuel map to do certain things like driving the fuel trim slightly rich or lean and then monitoring that the if the rear o2 sensor responds correctly and also the time it takes.

For example, it will drive the Fuel trim rich just for a second or so. The cat converter is usually not able to convert all that extra "puff" of fuel that quickly so the rear 02 sensor should read rich in an expected number of milliseconds or whatever time they use.

In the case of no cat converter, the 02 sensor is constantly switching between rich and lean as the car is in closed loop. This automatically sets a "red flag" and the MIL will illuminate. On a good cat converter, the rear 02 is very slow and almost a flat line since the good cat is doing it's job and using most of the oxygen.

The problem is you don't know when the PCM is going to run the monitor, so how do you know when to make the rear 02 do what you want and fool the PCM?

I have heard of rear 02 spacers being able to fool the PCM because it takes the rear 02 sensor out of the main exhaust stream diluting what it senses. Has anyone tried this yet? I never advise a street car to run with no cat, so I haven't ever bothered searching to see if someone has already tried the spacer trick on an RX-8.

paimon.soror

iTrader: (15)

So the question is, when the PCM is doing the test, what exactly does the upstream O2 see? I would assume it sees that rich/lean mixture. Could we have a circuit that would monitor the upstream O2, notice when the PCM is performing its test, and respond accordingly?

As far as the spacers go, people have used them, and no one has been successful. It may take days, weeks, months, but usually people come back and state that they finally got the MIL

I too dont advocate catless street driving ... but with lacking support for the S2, I am willing to put my knowledge to good use to give people some options

Like i said though, for me, first things first, completing the rxduino so people can get it in their hands.

jrx13

The front 02 is a wideband sensor so doesn't work like a normal 02 sensor. A normal 02 sensor outputs a voltage of about 0-1 volt. Ok this is kinda iffy explanation but here goes, the wideband, on the other hand, is a type of "feedback" sensor. I consider it kind of like a Hot Wire Mass Air Flow sensor. The PCM uses milliamps to keep an element at a certain voltage. How many milliamps needed determines the air fuel mixture.

If you ever pull up the Front 02 sensor PID, you see the voltage hardly ever changes. But in fact, the PCM is regulating the voltage to keep that sensor at that voltage.

Now comes the problem of trying to determine when the PCM is alternating the fuel map to run the Catalyst Monitor or just making minute corrections. It will be very hard to spot. There may be a hidden PID in the PCM stream that tells you it is performing the Catalyst Monitor but other than that it is beyond me.

OBDII is a pretty good system and was meant to try to foil basic attempts at skimping smog regulations. So you really need some major software hackery like the COBB to make the MIL not turn on for Catalyst Monitor but still show the test passed when hooked up to a scantool/smog station. So until someone wants to R&D the steps on a Series II, I don't think there is an easy fix with resistors and *****.

For now, just have the RxDuino monitor PIDS. You can hack the PCM another time on then next project!

paimon.soror

iTrader: (15)

No worries silly pants, thats the goal

xexok

Lots of people have tried that on an s1 and for some it works, some it does not. Someone will use a certain length spacer and it will work fine for them, then another person tries that exact spacer length and they get a cel. S1 was a little weird though I remember reading some of the 04s and maybe 05s were going catless and never getting a cel without even doing anything.

What I plan on doing for now anyway is just putting the cat back on for smog, not that hard. We dont have those mobile smog vans around here that catch you and send a letter saying you need to get smogged again so that should be fine for me until something gets developed.

paimon.soror

iTrader: (15)

damn fab shop needs to hurry up lol

paimon.soror

iTrader: (15)

They must have heard me :D ... they have shipped out the new boards, should have them in the next few days.

bhammer

Great news..

paimon.soror

iTrader: (15)

New board has arrived, left is the new, right is the old. As you can see the new one is significantly larger.

bhammer

That is a bit bigger, but the soldering should be easier.

What are the dimensions of the new board?

And, more importantly, when are you going to ship mine out?

bhammer

How's the assembly going?

paimon.soror

iTrader: (15)

Was away this past week, but the order for all of the components has been placed, so I should be getting them in the next few days.

paimon.soror

iTrader: (15)

Dont mind the leftover alcohol and flux on the above.

bhammer

Looking real good!! How long do you think it will take per board?

paimon.soror

iTrader: (15)

That is a good question. This one is pretty much the proto so it will take the longest as I get acquainted with the placement. I am thinking that on average a board will take a couple of days to assembly and test. Once we get this proto out to you guys to do some testing, I will probably order the blank PCB's in bulk so that I have them handy and as an order comes in I will work on placement.

I actually sent the design to a few PCB assembly shops and they are charging an obscene amount of money for a few boards fully built, so much that I would need to charge $550 to break even (LOL yea right...). Assuming this prototype works, I will see if they have some bulk discount that would drive the price down.

Worst comes to worst I will just have to do each board by hand, but the good thing is that each board will get some TLC and tested with my car before I send them out.

xexok

So you do not have any real idea on a price yet? I know it will be hard to pin down an accurate price, but maybe you could PM me what you are thinking of charging if you don't wanna post it. I know it would not be the real price, but just a ballpark figure would be nice.

paimon.soror

iTrader: (15)

The price I am shooting for is around the 199.99 range. Again it could be less or just a tad more, I dont expect to go really any higher than that unless we start further customizing the boards (like if some of you want bluetooth, or wifi, or ethernet etc).

I hope that is a reasonable price for you guys...again, it isn't set in stone, once I have a "production ready" board we can all talk.



All done, just need 4 more pieces (resistor arrays that i forgot to order bah ).

paimon.soror

iTrader: (15)

FYI Just ordered a surplus of the resistor arrays. Looks like they are parts that are a bit hard to find and are only made from Panasonic. Eventually I will revise the board to not use the arrays but I ordered enough to make a bunch of boards . Ordered via USPS so they should be here in a few days

seanp

So, are you accepting "bribes" for beta testers? Need any in the UK? :D

bhammer

Very cool on the tentative price and is much doable for me. Dang, ain't got one made and you are already having supplier issues.

I can think of a few "mods" to grow/enhance the RxDuino but I'll keep my mouth shut unitl I can get a working / testing version.

Let me know when you are ready for PayPal!