Power loss = me stumped
 

Okay, so before I go into explaining this problem I would like to ask everyone to read through this entire post before spewing out ideas, as most of them have already been tried.

First off I'll give a rundown on the car, it's a 2004 6MT GT with 84k miles, aftermarket catback, all emissions related systems removed, but as far as engine goes everything else is stock.

About 3 months ago I noticed I was having power loss just before 6,000rpm and it would continue revving but with no guts, then once I got above 7,200rpm or so it would kick back in until redline. This issue is VERY intermittent and sometimes it goes days without doing it, sometimes two back to back pulls will behave differently, sometimes it does it for a week straight.

I only had this to go on and immediately checked plugs, coils, and wires. One wire was looking sketchy so I replaced it. I also made sure the MAF was clean as well as the throttle body and UIM. The problem persisted.

Next I started to think it might be a fuel delivery issue since it could be explained by being starved for fuel(although it kicking back in at 7,200 wouldn't make sense with that) and ordered a Walbro 255 pump to replace the stock, known to fail, unit. I also ran a can of seafoam through the tank to clean injectors. The problem persisted.

At the same time I was working on the fuel pump I was fixing a front main seal leak(which I thought oil getting on the ESS could be causing me issue) I checked the coils and plugs again and noticed that two of the coils had visible signs of arcing and got them replaced. During this time I also pulled the SSV and cleaned it as well as it's housing(the just before 6,000rpm would line up with the SSV opening). The problem persisted.

My next step was to get a friend to set me up with a way to datalog during the issue so I could see exactly what was going on. Of course the issue wouldn't happen with him in the car, so he let me borrow a his datalogger so I could attempt to get data on it.

Once I finally got data, it would appear that I am going extremely rich right when the power loss occurs and the airflow and calculated load are suffering and it is regaining a more civil AFR as the power comes back at the top of the rpm range.

Here is a plot of the issue occurring during a 2nd gear pull:

http://i45.tinypic.com/rtp3j4.png

After seeing this data we decided that it looked like a vacuum leak. The fact that it is occurring at the same time that the SSV and the secondary fuel injectors are coming online directed us to start our search there, we have now tested the SSV actuator(it's fine), all vacuum lines in that system(all were fine), the SSV solenoid(functions properly and swapped with airpump solenoid which also works fine), and we pulled the primary fuel rail to make sure all injectors were properly seated in their O-rings(everything appears normal). The problem persisted.

We then decided forcing the SSV open at all times would be a good way to see if the opening of it had an effect on the issue or if we could eliminate it from the list of suspects. As you can see from the following plot, the MAF readings seem to indicate that the issue ir present even with the SSV stuck open although the AFR is rich through more of the rpm range this way(not really sure what that would mean).

Plot with SSV stuck open:

http://i46.tinypic.com/21ucdv.png

Earlier today I was driving with cruise control on, watching the datalogger and saw something else peculiar. If I am cruising at a speed that would put me about around 3,800rpm in any gear for any length of time, my AFR slowly creeps richer. It goes richer the higher my rpm and will be fine if I let off the gas for a second then continue to cruise. The commanded AFR stays where it should while the actual AFR seems to have a mind of its own. While this is going on the engine load stays where it should be and the LTFT sits at +6.2%(which is where it normally is for cruising). The confusing part of this is that even in closed loop, the car seems to not try to correct the richening AFR at all.

Not that it contains much useful information, but here is the plot of it being rich while cruising:

http://i45.tinypic.com/i3dr15.png

And just for good measure, here is a plot of a 2nd gear pull that shows this car operating correctly:

http://i45.tinypic.com/2uz8kdk.png

After all of this we ran out of things to try and I swapped my MAF with another MAF off of a known to work RX8 and neither car saw a difference, so the MAF can be eliminated as well. And since it occurs in both open and closed loop I can safely assume it isn't an O2 sensor issue either.

ANYWAYS....

tl;dr

Power loss between 5,800rpm and 7,200rpm at WOT and gradually rich condition during >3,800rpm cruise

-Not the Plugs/Coils/Wires
-Not the SSV actuator
-Not the SSV itself
-Not the SSV solenoid
-Not the fuel pump
-Not the ESS being dirty
-Not the MAF
-Not the fuel injector 0-rings
-Not a vacuum leak(tested every way I know how)

So what could be causing my car to do this?

If you want any more info, feel free to ask, I'm sure I've forgotten plenty in this post...


PS: I have attached full-res copies of all the plots posted for easier reading.

Oh, two more things I forgot to mention, gas mileage and oil consumption are the same as always, about 17mpg and half a quart of oil every 500 miles and I don't premix(run Castrol 5w-20 still and always 93 octane)

APV? It should be opening at 6,250rpm, might not be.

I had considered that, but after seeing the strange behavior during cruise it made me reconsider. I should be able to just apply power to the APV motor and watch to see if it correctly actuates, correct?

Correct.

Another option is that it is actually an O2 sensor starting to fail. Could be screwing with all sorts of stuff, even open loop.

I had considered pulling the connection on both O2 sensors to force it open loop and see if the upper rpms still suffer... I will try both of those as soon as possible and post results.

Any other ideas to test are definitely welcome, as long as I don't have to pull the UIM again for the billionth time I'm down to try it.

Actually, upon further thought...(work is slow today) wouldn't the APV not opening cause the issue to persist through to redline, and power not to come back about 7,200rpm?

Not quite that easy, since there is are difference impacts between the valves. The APV changes the length of the intake runner, where as the 7,200rpm opening is the variable intake valve, which produces a mild supercharging effect where it uses the momentum of one rotor's intake phase to help increase the airflow into the other rotor.

However, agreed that it is unlikely that this change would produce this sudden of a jump for you.


Another thought just occurred to me. Someone that knows more about the RX-8's ECU will probably have to chime in on this, but for my 2005 Mazdaspeed Miata, the ECU has a trigger point at 5,000rpm where a variety stuff happens: it no longer pays attention to the knock sensor, it goes into open loop rich tables even if the load would suggest it doesn't need it, and there is a timing change. When you increase the airflow with intake and exhaust upgrades, it introduces an issue where under 5,000rpm it pulls fuel and timing under 5,000rpm when rolling into boost, but hitting 5,000rpm is like a sudden surge of power.

Obviously much of the causes and effects don't apply here, but perhaps there is something going on under 7,200 that the ECU is pulling fuel and/or timing because it sees something it doesn't like, but then crossing 7,200 makes it ignore what it's seeing and go to a default WOT table.

Just another guess...

From the looks of the plot with the bad pull I see it going open loop(the commanded AFR flatlines) but then as the stumble happen it goes back closed loop for a second(the commanded AFR jumps up to mid 14's and the actual AFR regains its composure) then the MAF/AFR readings look correct the rest of the way to redline.

From that I gather that you are correct that something is causing the ECU to trip up a bit, or at least realize that something is wrong. But the answer to what that could be is definitely something someone with more intimate knowledge of the ECU needs to supply.

I have been working with Kane recently on this issue, and he seems to think it's SSV or injector issues, but from the plots(and he knows a bit more about all of that than me) he seemed to believe that the issue is an engine issue, not a control issue.

Okay, so I'm about to test if it runs alright at WOT without either O2 sensor plugged in, I know it will definitely mess with the low end, but the top end should do fine if my O2 is causing the issue, I'll post the log once I have it.

I was wondering if just making sure that the motor in the APV actuator is running is enough to make sure that's not broken? Like put a hand on it and have someone cycle the ignition and see if it runs, or if I needed to really pull it off to check?

Alright, so I just pulled the connection to both O2 sensors and reset the fuel trims and drove around town a bit, the car has a complete lack of power(kinda expected that) but did not seem to have any hesitation at any particular point but did have more power in the very top end(closer to what it usually has).

Anyways... I'm not really sure what this means and would love someone with some more experience to chime in, I've attached a plot of a second gear pull with the car in this condition and what I find interesting are the MAF and calc load readings, they seem to hold pretty steady curves and don't have any major changes except for upon intake/fuel transition points.

Does this mean that my O2 is probably the issue?

http://i47.tinypic.com/2u44vt0.png

While not conclusive, I'd say that it's pretty suggestive that the O2 is the problem.

Front O2 that is, the rear only deals with emissions and helping fine-tuning the closed loop AFRs after long-ish steady state cruising.

Makes sense to me, yay for throwing more money at it!

On a serious note, could the strange rich condition after steady cruising by caused by the rear O2 failing as well?

I doubt it. the rear O2 is narrow band vs the wideband of the front O2, and from my understanding (which may be inaccurate) it can only refine what the front is telling the ECU within a limited range.

Looks like I'll order a new O2 on Monday and see if that fixes anything.

Completely unrelated but it decided the fuel line to the secondary fuel rail shouldn't be connected last night in my friends front yard -_-

Sounds like the O2.....at 8 years and 84K....lucky it lasted that long.

I use O2 sensors in a more critical situation.....abit they are different sensors...and they last more in the realm of 10's of hours before they start getting non-linear and give slightly wonky readings...

I think in general....most automobiles are running on less than optimum sensors after about 4 years....

Well I got the fuel line fixed and now it seems to be running fine, although that means nothing with how intermittent this is, so I guess ordering the O2 is the next step no matter what.

Any other cheap/easy things I could try before monday?

I have this issue as well. I noticed it after my car had sat on a rural property for four or so months. In the beginning I would often put a few gallons of fuel in it and run it on the weekends, but for about two months I didn't.

As for the loss of power, I too thought it was due to a vacuum hose... mostly because the field mice that made my engine compartment their home had chewed through one that seemed to have been in their way. You should have seen the number they did on the AC system. I should also note that they found a few of the wires tasty as well. I believe I got all of those, but I can't exactly reach all the places a field mouse might without removing the engine.

I also have another anomaly not likely related to this thread, but to be thorough I about the state of my car I will mention them. My TPMS is a bit iffy in the rear due to unrequested replacement of a seal by Discount Tire (a case of you get what you pay for I guess). My car also has loss of traction in turns that should not cause such. I think that is just a dire need of alignment. Another thing my car did recently was to light up the idiot lights like a Christmas tree. As I remember it showed ABS, Brake, TPMS, loss of traction light (swerving car) and the traction control disabled light. As I mentioned it will likely be fixed with a simple alignment, but I may need to replace the rear TPMS sensors as well.

It may take a lot of effort and money to maintain it, but I must say that I do love the car. Oh, and I guess this is important, it is a 2006 and I am the second owner. 70.5k miles. I know it needs a tune up too. I wonder if I can do it myself or if it is one of those things it is better to have someone with the correct gear do for me.

Take care,

Nexion

While the symptoms my car has might be the same you face, they are really common and there are plenty of other things that should be checked before you come to the same conclusion that I have with my car.

I suggest looking into many of the basic tune-up things first before trying any of the stuff I have(such as the fuel pump or the other things that require a lot of work). This thread is a good place to start when deciding what basic maintenance to start with(it's the same one linked to in RIWWP's signature).

With that said, if you have any questions or more information about your problem and why you think it's the same as mine, feel free to post up.

People of the internet, I pulled the connection to the rear O2 sensor only this morning and reset fuel trims. My LTFT seems to stay close to the same but my STFT's are negative and more what I would expect under different load situations, and the car seems to not be hesitating as bad, which I'll have to wait and see if it's something that is going to come back or keep getting better over the next few days.

Now, I'm happy that it seems to have helped the issue, but any clue as to why this would make a difference?

So after driving for a few days with the rear O2 unplugged the problem has come back. Which leads to me thing it's an issue with the fuel trims or e-shaft profile since it seems to go away for a while after those get reset.

Now the fuel trims look good, LTFT +2.7-3.9% at idle, between +1.6-3.9% while cruising with STFTs that sit between -.8% and 0. On throttle the STFT goes more negative to lean out the mixture which seems to make sense. At WOT the STFT goes to 0(as it should for open loop).

So since my fuel trims seem correct for what is going on, and the problem has come back despite them, what else does that leave to try?

Another possibly interesting fact, I just figured out I can also pull data on spark advance, but I honestly have no idea what I'm looking at... Does anyone know the standard values I should be looking for under different circumstances that would be considered 'normal' value? Also, am I looking at leading spark advance, trailing spark advance, some sort of split, or an overall advance of the two?

Not trying to endlessly bump, but I have MORE DATA!!!!

So I pulled with O2 sensors to try and get more consistent and clear logs, I drove the car until it seemed to have consistent pulls after resetting the fuel trims. The problem is definitely still there(which makes me think the O2 sensors are not the issue since the problem persists with them out of the equation).

Now my next question is do my spark advance profiles look correct? I logged them on these two plots as well and I honestly have no idea what I'm looking at, but isn't +30 a bit drastic?

http://i49.tinypic.com/2a9o9ck.png

http://i45.tinypic.com/1zqg1o5.png

a compression check should always be first, if it checks out then diagnose from there


just my experience, fwiw ...


.

It was tested in July and was above spec, so I figured it wouldn't have been shot since then. But it has been on the list of things to try.