Cobb SFR Issues
 

I have been mostly quiet on the forums as of late and I hope to beat WantedTwo to the punch to be the first Cobb SFR kit. Obviously I lost ;)

History:
I ordered my upgrade to my current kit last November??? From there I knew i needed several steps. First I needed to swap my injectors back out, i was currently running the 550cc RC P2 and Secondaries. I had to replace these with upgraded stock. I needed a Cobb which I got from MM. I bought the kit used and there appeared to be a bad seal when i got it so that needed to be rebuilt. I was running the Int-X so that had to be removed. In general there was a large task ahead.

I started the rebuild when I and amassed all the parts I thought I required. At that time I took the turbo off the car and sent to get rebuilt. I started the removal of the Int-X, I changed the cold side piping out. I swapped out Low Impedance RCC injectors for the upgraded stock versions. I have to say you have experienced pain until you have to splice wires with a solder gun under the hood of a Rx.

From the start like any project I was met with delays. While the accessport arrived as promised. The turbo rebuild did not, I was quoted a 3 week turnaround. The more likely outcome was about 4 months. Tim from SFR was very good with helping me and I have to say is the best in the business and helped me the best he could. Since I was essentially a experimental customer the kit had a few flaws. The biggest being the relocation of the stock O2 sensor and a few missing pieces.

The first time I got the car together after waiting about 4 months for turbo, maps, injectors, kit parts etc. Immediately the car started and for about 30 seconds made music to my ears. The joy was short lived as the car changed modes and started to hunt. AFR's went from 11's to 18's at idle. The car would sputter then reset. The entire saga would repeat.

I called Tim and sent an email to MM(I dont have Jeffs number). I didn’t know what would cause this. Tim right away apologized for the issue and said that the issue was related to the stock location of the O2 sensor(As documented in WantedTwo's Build thread). He stated that the factory O2 sensor needed to be moved to just before the turbo or 18" from the exhaust ports. This meant tearing the turbo, wastegate and the uppipe out of the car.

After removing the entire setup. I bought a saddle 02 Bung and had it placed on the up-pipe of my setup. I had to decouple the O2 sensor from the lune that held the transmission sensors. While not technically hard did require some time. After all this I put the car together. To my dismay the result was exactly the same.

Today:
As of today I still do not have the car stable in idle. At this point I have yet to even come close to the resolution. I have taken extensive steps to try to troubleshoot the issue.
1. Relocated Stock O2 to virtually the same location as WantedTwo’s
2. Clean the MAF sensor. (Used the store bought Sensor cleaner)
3. Use the Mazdaspeed intake and place it on the end of the SFR kit (No filter but did include the 3 pieces and both mesh grates.
4. Use the mazdaspeed intake in the “Stock” location bypassing all the cold side piping including both grates.

Where I stand:
As of this evening, changing the location of the MAF sensor has made no improvement.
The AFR’s of the car cycle between 13 and 17 like stated in wantedtwo’s post.
The stock O2 sensor has been moved to the location of WantedTwo’s
The ST Trim of the car cycles very quickly from -8 to -25 and back


If anyone has any input onto this issue please let me know. I have been trying to get this resolved and have been waiting 6 months to get the car back on the road. I will attach the location of my stock O2 and a data dump of my issue

Its all in the tune. My setup is VERY similar to the SFR and I went through exactly the same issue. Relocating the O2 sensor helped, but the resolution was in the tune.

subscribrd

Do you know if your maf sensor is giving stable accurate readings ?
If not you need to find someone who has a sweet running top mount with the AP and find out what they did to get the maf accurate and consistent .

check the chart of your logs out below ..

It has a hunting cycle of approx. 6s between peaks

your maf readings seem about 20% higher than they should be for the speed of the engine .
Mine are typically around 5.5g/s at 800rpm and 6 g/s at 1000 rpm
yours are 6.6 and 7 respectively .
So - get your maf rescaled ...

I thought about the bad MAF reading as well. That is why I went ahead and put my mazdaspeed intake on at both the end of the turbo piping as well as at the stock location. I couldn't use the entire intake but I was able to get 3 sections including both screens. I received the same result.

I am running the first map MM sent to me. I am about 90% sure the MAF hosing is the same size as the Mazdaspeed.

Make sure, SURE, your fuel injector harnesses are on the proper injectors and that none of the wires have been improperly located in their respective connectors. I learned my lesson a few years ago and your symptoms replicate mine.

/\ good call Charles . I'm guessing that would also mean very rich afrs at low loads ?

Well at idle the car would only be running on the primaries correct? They were not changed since the last time I had the car running.

Also with the maf sensor disconnected the car runs fine with afr's right around 14.

disconnecting the maf would stop any pcm corrections . So the fact that it is rich (14.0 vs 14.7) could still indicate either maf scaling or the injector issue Charles mentioned .

Do you think it's the sensor or the fact it goes into closed loop when it's connected? The sensor has been located in the charged piping for the past year. With the int-x I obviously was not using the sensor.

If the maf is disconnected how would scaling be an issue? Likewise with the injectors wouldn't the fact that it is running stable without the maf indicate the injectors are running correct?

Before I tear down the uim to check this I would like to make sure it's thr culprit.

Because it is running rich . When you reconnect the maf the pcm tries to correct the rich condition so if injectors were wrong it would make the wrong adjustment . Likewise if maf scaling were the issue . The pcm just keeps chasing its tail trying to correct the uncorrectable ....

from what I have seen people who put pressure to their maf sensors, like you stated before with the int-x setup, the sensors go bad, I have seen this in the mazdaspeed protege community...

I'm sorry I am trying to realize how the injectors could be wrong. Since the ECU only fires the primary injectors up until around 2500 rpm's and the car is running the only way they could be messed up would be if the P1 for Rotor 1 and P1 for Rotor 2 were switched. Since the P1's inject into the rotor housings and if they were switched I would think the car would not run at all.

Having had all that apart a couple of times I find it hard to work out how you can do it wrong .But I think what Charles means is that plugs from the sec/p2 bank end up on the p1 bank .

Yea I'm almost sure that is not happening. During this go around i didn't even remove the rail that the p1's sit in. The wires were spliced when I installed the kit the first time. I had incorrect information stating the rc injectors went in the p1 position. I swapped them back when I realized they didn't fit.

I think you still don't understand that they are referring to the fuel injector wiring harness connectors being plugged into the wrong fuel injectors. Mine just ran rough and rich when I did this, not like yours, but it could vary just depending which connector was on which injector. You really need to get an idle log to MM for review. I suspect it's more of a scaling/tuning issue too, but pulling the uim is not that bad and well worth the double check to make sure the fuel injector wiring harness color code matches the proper injector if you didn't verify this previously.

See the installation note in the lower half of this attachment:

https://www.rx8club.com/attachment.p...4&d=1230876815



.

Ahh i see what your saying, like i said before It may be possible I do know I followed the diagram MM posted and double checked when doing my rebuild to the letter. I double and triple checked the wires. There isnt much play with the P2 and secondaries to begin with. Like I said this go around I didn't even touch the P1's.

If MM doesn't get back to me with a new map to try perhaps I will remove the UIM to verify the P1's are correct but like I said the P2 and the Secondaries I looked at, then got lunch looked at again before I re assembled the UIM.

Ah . I see your issue now . Eating on the job :p:

based on all the knowledgable forum people who made the same mistake apparently the wires have more stretch that you realize .... :dunno:

The injector harness connectors are strangely, and counter-intuitively, laid out in the context of the entire injector harness and swapping them around is a very easy mistake to make.

as we both know .... :wallbash:

So I went ahead and removed the UIM to look at the injectors today. They all appeared to be wired correctally. I used the manual I had for the color codes and verified they were as follows.

P1 Front: LG/R W/L
P1 Rear: LG/B W/L
Secondary F: G/R W/L
Secondary R: L/B W/L
Front P2 : L/B W/L
Rear P2: W/G W/L

I did notice one of the tubes running to one of the sensors in the bad must of been pulled off when I was re-assembling it the first time. I corrected this issue when I put it all back togeather.

The car started and went back into its same issues. I tried several different configurations to attempt to narrow down the problem. I will include them in seperate posts below.

Configuration 1:

Used my mazdaspeed intake to circumvent all the charged piping like shown in the image attached. I also attached a datalog of how the car ran with this configuration.