I just wanted to share my excitement of getting my WBO2 installation finished today. Woo hoo!

To mount all the boxes, I removed some useless white plastic thing from where the ECU sits. That cleared enough room to Velcro the Datalogit in place behind the ECU. I connected the Vout and Ground wires from the WBO2, connected the Datalogit to the PowerFC, the Commander to the Datalogit, and then tucked it all away in there. Next, I added some Velcro to keep the PowerFC in place. The Velcro works great all along the bottom edge of the PowerFC and there is a good spot in the upper corner near the door for a little patch. That holds the PowerFC in place pretty well. I then used some more Velcro to attach the WBO2 control box to the PowerFC. I did it without considering how the cover would fit at first, and I ended up having to tilt the WBO2 down and move it toward the front of the car so that the display connector would clear the trim cover. It is all secure are fits fine behind the cover now.

I used one of the HVAC vents to run the WBO2 power and display wires from the passenger footwell over to the driver's side. That works great, and has a lot of capacity in case I want to run some more wires in the future. The wires go under the passenger carpet over to my WBO2/PFC/DL stack. I have the display Velcro'd to the dash to the right of the steering column, where it is very easy to read while driving. I've got some pics of my new gauge setup (including the WBO2) posted on this page:
http://www.maxcooper.com/rx7/how-to/...ges/index.html

I drove around and did some logging tonight, and found that I was running really, really rich under boost (richer than 10:1). That explains the misfires I have been experiencing. I am running more boost than I was when the engine was tuned by XS, so I need to tune the higher boost cells to get it running safe and strong. I'll have an opportunity to tune the part-throttle stuff, too, for better drivability. It costs too much to have someone spend the time it takes to do that on the dyno.

A while back, I gathered a bunch of information that had been posted to the Datalogit email list and put it in the Files section if you are planning on installing a WBO2. I spent a little time verifying that the WBO2 display and the Datalogit were displaying the same A/F ratios, and found that they were within 0.1 of each other, which is close enough for me. Thanks to everyone who shared this setup info in the past. The info I collected is all available here (you must be a member of this group to access the files):
http://groups.yahoo.com/group/Datalo...Sensor%20Mods/

I was originally going to wait and install the sensor in a fresh bung near the end of the downpipe, but after I got my dashboard back together last night and had the WBO2 display reading 14.7:1 the whole time, I decided to remove the stock O2 sensor and install the WBO2 sensor in its place. I should probably move it down later, but it is kind of nice just to have one sensor. The TechEdge WBO2 v1.5 has a narrow-band output that would be great for that -- you could still run closed loop with just the WBO2 sensor installed.

-Max

 

love the pics max. you did an awesome job on the gauge install too. seems like you have too much GOOD time on your hands. something i might consider, i just hate doing stuff that others have done for personal reasons. all in all, you the man
kris

 

Max

Congrats on the install. Mine has been in for a little over 2 months and I cannot remember how I got along without it. I am still waiting to have a second bung installed although I can't say I miss the NB control. Even without 02 feedback I am getting 21 mpg on long trips so closed loop really isn't much of an interest point anymore (now I just have to figure out what new toy I want where the NB gauge went).

Now that you have it in, the obsession towards perfections starts. I have gone so far as to have installed a couple of spst switches to break the circuit between the vout+ and grnd and the datalogit. That way the questionable readings some people report on the 5301 when the datalogit is attached, can be ruled out. When the switches on in the on position the datalogit is powered and the O2 can be logged. When they are off, no data is sent to the datalogit, hence no possible voltage drain affecting the 5301.

Good Luck

Jeff

 

Max,

I just visited your site last week. One of the things I looked at was your wideband O2 sensor install.

Great site! Lots of good info. Thanks for sharing.

Tom

 

Jeff,

You aren't kidding -- I don't know how I got along without this thing before. I've had the Datalogit for months, and have suspected that I was running rich the whole time, but I didn't want to make any changes and pop the motor.

I did some logging yesterday evening and did some analysis to make a new base map last night. My logs showed I was richer than 10:1 under boost most of the time, so I leaned out the maps for my maximum boost (15-17 psi or so, P19 -- my MAP sensor cannot read high enough to hit P20 with the standard scales). I should probably invest in a new MAP sensor to read the higher pressures so that I can tune effectively up there.

I put the new map in and ran the car tonight for some more logging. I think it is still right around 10:1, but the misfiring is mostly gone, so it runs much better. It feels a lot faster than it did before, and I still have some more fuel to take out, which should make it even stronger. I'm just trying to be conservative, though, so I don't grenade the motor. I think I'll be shooting for 11.0:1 or so under boost.

I swapped the R8 resistor in the WBO2 box out for a lower resistance one to reduce the error. It was 1000 ohm (where 14.7 would read about 14.7*33/34 = 14.27)and it should read about 14.7*33/33.1 = 14.66 now. That's close enough for me, and I've already got so much wiring and gadgets in the car that I don't want to wire up any switches. It was pretty easy to open the WBO2 box and swap the resistor. If you are on the Datalogit group, I included pics and instructions in the Files section there (use the link in my original message).

There's a good spreadsheet for making a new base map in the Files section of the Datalogit group site. I used that to do some analysis last night. I'd like to improve the pivot table for the A/F ratio stuff to convert to A/F or at least calc the difference between AN1 and AN2 before it does the pivot table for accuracy. Right now it just averages AN1 and then subtracts an average for AN2, which seems less than ideal to me. It would be nice to have a smoothing function for the curves, and perhaps a helpful interface for eliminating the less desireable logging data, too (like ignore the lines where the throttle is closed, etc.). Perhaps I will make some software to help with tuning via logged WBO2 data that has these features. It is just an idea at this point, and I'll be sure to post some messages if I write something.

The PowerFC + Datalogit + WBO2 makes a great combinaion for tuning. It seems even better than dyno tuning.

Tom, thanks for the compliment. The SPA gauges and WBO2 display complete my instrumentation mods at this point. I am really happy with how it all turned out.

-Max

 

Max,

Welcome to the wideband club. I simply cannot communicate how much better the car runs at 11:1 vs 10:1 AFR. Kepp us informed of your progress.

Kyle

 

Max

Oh yeah, you've got it bad already. Glad to hear you are having as much fun with it as I am. The combination of datalogit and WB makes life a whole lot easier (safer) when doing the put and take with fuel and timing.

The R8 swap is a great mod to start with but, like I say, I can get fairly obsessive with these details. I used a couple of submini spst toggles and they take up very little room. To be honest though, I did the R8 swap right after the original suggestion was approved by Peter and since then I seldom see a difference of .1 or .2 rich on the 5301 with the datalogit switched into the circuit.

I have ordered and am waiting for the new 16F84A PIC also. As you know the 5301 is light and cold sensitive and reads anywhere from .2 to .5 rich depending on the temp and/or intensity of the light. Peter suggests that the swap to the new firmware reduces this problem and for $6.60 delivered and about 15 minutes work (open the case, R&R the 16F84A, close the case and recalibrate) I would just as soon have the all condition accuracy.

I will be at the dyno this weekend to reconfirm the accuracy of the WB. I won't be tuning based on the dyno though. I am mildly interested in the HP/Torque curves but won't be tuning for the dyno. When I revise all my maps, it will be done on the track.

So much to do---so little time---I think I could use a new job that has two consecutively running six month paid vacations a year.

Once again, glad to hear you are enjoying the new toy. Keep us up to speed on your tuning adventures, pivot table software creations and don't forget to consider adjusting your ECU AIT tables to reflect the ambient intake air temp as the base (0.998) on the day you tune.


Jeff

 

Max:

I was curious to know if XS dyno tuned your car with these maps so rich? Or was a map just plugged in and ended up being so rich?

I need to get my own car tuned soon and if XS isn't doing a good job of it I will take that $$$ and use it elsewhere. I can see how it would be in their best interest to tune the car pig rich to prevent a blown motor.

Matt

 

I am running more boost now than I was when XS tuned the car on the dyno. Actually, considering how long I have been running in 'untuned territory' on my maps, I am glad they made them rich up there. You can see in my base map where they stopped tuning (since my boost wasn't high enough to hit the higher P rows).

Now, the maps are still pretty rich where they did tune, but I asked for reliability over power. And the car has never detonated audibly, despite much abuse from me. The car made about 370 RWHP at 12-13 psi or so at 7000 RPM, so I don't think it was tuned too awefully rich. Also, I have bumped the fuel pressure up a tad, so it is running a little richer than it was when XS worked on it.

I am definitely becoming obsessed. It is really great to be able to do the datalogging with the wideband data for analysis. I didn't know about the 5301 issues. Are you saying that it reads wrong when the sun is hitting the display? That seems odd. I've been doing all my work at night so far, but it seems like it might be a good idea for me to update my display.

-Max

 

Max

The 5301 display is both light and cold sensitive in all but the latest models. (shipped after February 21, 2003). See Peter's message (#2373) about the low temperature operation fix on the OZ-DIY-WB (excellent description of the problem, the cause and the fix). If you are not on that Yahoo group I will forward the message to you via PM.

Some 5301s also display light sensitivity and according to Peter's message the reprogramming of the PIC to cure the cold weather problem also resolves the light problem. Apparently daylight hitting the photocell increases the voltage applied to the LEDs (making them brighter) but also causing a inaccurate reading (slightly rich condition) to be displayed.

You can test for this condition by going to the car in full daylight and checking what the warmup period reading is showing (mine shows 14.5) and comparing it to the reading at night (mine shows 14.7). You can test everything during the day by reading the display in full sunlight and then covering up the photocell with black tape. If your unit has a problem, you will see a difference between the readings (the covered condition will show slightly richer).

Hope this helps. If you need anything more on this issue just let me know

Jeff

 

I just got my setup working... Wow what a great tool! Nocab72 and I went out and did some runs today and it revealed all sorts of interesting information! I can't wait to hit the road tomorrow with a few tweaks to my maps to see if I can get this bad boy running even better!

 

I just got my PFC/DIY/Datalogit setup working well this weekend and logged a couple of hours of data. A few questions:

I don't have any problem doing tuning for the boost areas of the map, but I would like to make sure that part throttle response is smooth and that I am able to maximize fuel economy. I understand this is a bit of an art, but I'm looking for generous individuals to provide a few tips...

My feeling is that the PFC doesn't transition between closed- and open-loop operation as well as the stock ECU. I get just a hint of hesitation at transition. Is this something I can fix with tuning or something I can change in the settings fields?

I also believe that the stock PFC air temp enrichment settings seem to be doing a pretty good job of maintaining stable AF ratios during ambient temp. changes. Anyone else have an opinion?

 

The OEM tables should do fine as long as you don't change the maps drastically. If you remap based on testing at a different IAT you should consider changing the basis to the IAT on the day you modify the maps.

In other words OEM maps are designed for 40C and 40C is entered into the correction table as 1.00 (or "reference") (ok actually it is .998).....fuel is added or subtracted when the temps increase or decrease (+ for decreased and - for increased temps).

If you create either significantly changed maps or basically whole new maps by adjusting the fuel figures based on AFR runs made with IATs at 50C the correction table numbers will not be correct.

 

All,

I've just a quick question: I am installing a PowerFC now and I am ordering the Datalogit package as well. When you talk about logging your A/F ratios, are you doing this with the OZ-WB02 package? Or with Datalogit? I am a bit confused. Also, has anyone used the NB outputs of the WB02 to drive the factory closed-loop control? Which pins on the CPU are the O2 sensor? Please advise,

-ch

 

Jeff48,

IF the correction factors are right, then it shouldn't matter what temp you tune at right? Say you tuned at 80 C intake temps and got it just right. If you go out the next day and the intake temps are -30 C, you will be fine as long as the correction factors used are good. That's the whole point of a correction factor based on temps.

So... are the Mazda correction factors good?

Brian

 

I have the WBO2 hooked up to the Datalogit so I can log the A/F ratios with the other information to correlate them. That way, I can get a better idea of which cells in the fuel map need to be changed.

I have the 1.0 version of the WBO2 (no NB output), but using the NB outputs on the 1.5 version sounds like a great setup to me. The 1.5 version also has a wideband output that is easier to convert to A/F ratios when logging.

-Max

 

Max,

Curious if your tuning with the Mapwatch function is working well. After the Datalogit guys updated the software we have found that the AFRs reported in the Mapwatch N/P cells are pretty accurate for fuel adjustments in the maps but a bit less so for timing.

 

Besides adjusting the trailing timing for split, I haven't messed with the ignition maps yet. I haven't even looked at the timing in the logs.

I have been doing logging with the laptop closed and just analyzing the data later, so I haven't been watching the map watch screen.

I am a little amazed that I am still getting A/F ratios below 10:1 for most of the high boost cells. I leaned the maps out a bit and it runs so much better now (lots more power, no more misfires) that I am pretty sure I was running pig rich. But I am hesitant to lean them out much more since I am only hitting 85% injector duty cycle (550/1300 with 40 psi base pressure), maximum and more like 81% or so average on the high boost cells. Maybe the stock regulator is adding some fuel pressure or something. My fuel pressure gauge sensor is connected to the aftermarket FPR in the return line, so it isn't really measuring the pressure in the rail. I thought that it would read correctly as long as the aftermarket regulator wants a higher pressure, but perhaps there is some pressure drop between the rail and the FPR, or the stock regulator is adding a few pounds. Or I'm just not making as much power as I should be. 81% duty @ 40 psi & 0.62 BSFC & 15% DL loss is about 375 RWHP, and I would expect to be making more at 15.5 psi. Perhaps that isn't so far off if I reduce the DL loss to 12% (light wheels, flywheel, double underdrive pulleys) and up the fuel pressure to 45 psi, which gives 412 RWHP. That seems more like it, extrapolating from my last dyno chart. I was expecting 400RWHP at 15 psi or so. Or it might just be wishful thinking. I thought I was running more boost than that, too, but I just checked the logs and it was more like 110 (*0.1422 = 15.6 psi) across the high boost cells.

I have the WB sensor in the downpipe where the stock sensor normally goes, so I think it might be getting too hot and screwing up the readings or something. Does anyone know which direction (lean or rich) the readings drift when the sensor gets too hot? I should probably mount it at the end of the downpipe before I take more fuel out.

I've got a 3 bar MAP sensor on the way, too, since I'd like to run more boost than my stock MAP sensor will read. I'll do some more tuning after I install the MAP sensor and move the WBO2 sensor.

I'm not getting any detonation showing up on my J&S, but I probably need to increase the sensitivity and then back it off once I start getting some activity. I turned it down after I was getting lots of false readings from a loose knock sensor a while ago.

-Max

 

This is what I was trying to get at earlier. My runs this weekend were in the morning with air temps of 20-25C and then later in the afternoon I was seeing temps closer to 50C. I assumed that the stock PFC air temp correction worked well for me because the AF ratios were very stable between these runs.

Perhaps I didn't have a wide enough variation in temp to notice a difference, but I would be interested in knowing what others have seen as well.

Brian, did you pick up any good part-throttle tuning tips last weekend you'd like to share?

 

Just wanted to compliment you guys on the quality of this thread! I've printed it out for future reference!

It seems to me that wideband + datalogit is the only way to go. More and more I'm feeling like dyno tuning is a waste of time. Besides, a couple days of dyno tuning nearly pays for the wideband equipment, anymore. Plus it would sure be great for my piece of mind to be able to constantly monitor the A/F ratios. I know what I'm saving up for next.......

 

Max,

I was reading the OZ-DIY-WBO2 site and noticed this:

"The sensor reads the partial pressure of gases in the exhaust and infers the AFR, rather than by measuring some magical AFR directly. This may be an issue on forced induction, and in particular, on turbo-charged engines.

AFRs will indicate richer than they are, causing you to run leaner than you think.
Lean AFR's will be richer (or less lean) than indicted.
A solution is to ensure you locate your sensor away from the turbo, and certainly on the exhaust (low pressure) side of the turbo rather than the engine side."

I wonder if the pressure at the stock O2 sensor position is artificially high under boost. My guess is that it would be higher than a NA car until the exhaust gas gets up to speed. Perhaps more argument for moving the sensor to the end of the DP?

 

Thanks hyperion, that may be what is happening to me. I appreciate the clear info on what the effects would be. It could be pressure or simply the high temps right after the turbo that are messing up the readings for me (if they are indeed messed up at all).

I know I should install it further away from the turbo, but I couldn't resist trying it out once I got the rest of the system installed, so I just put the sensor in the stock (well, stock on the RX6 downpipe) location. I think I'll wait until I have it in a better position before I lean the maps out anymore.

-Max

 

Max

Like you, I have my sensor mounted in the stock RX6 DP O2 sensor location. I just had my car at the dyno last weekend and my datalogged WB numbers (Aux - AN1-AN2) consistently read between 0 and .2 AFR leaner than the dyno's WB at the tailpipe. Using my setup, I am not concerned about tuning too lean using datalogged numbers as a reference. Maybe you could take a quick trip to the dyno for a comparison run.

 

Max,

my Uego WB reads consistent from O2 sensor bung to the tailpipe.

 

jeff48, that is good to know that they were close since your's is mounted in the same position is mine. Hmm... maybe I will keep tuning with it as-is.

twokrx7, is your bung (I can't believe I just typed that) close to the turbo(s) like the stock sensor position?

I have an early (RS*R) RB dual tip catback with a straight-thru muffler in the midpipe. Do you have a more open exhaust than that? Perhaps I am getting a bunch of backpressure screwing things up. Or I am just still running really rich.

-Max