Won't rev past 8300.
 

Hi,

I have a 2004 RX8 race car. In that car is the engine out of my previous car, which was new last year. It has four weekends on it. After the second, the compression tested very well.

The engine refuses to rev past about 8300. Sometimes it gets to 8600, according to OBDII. And when it gets there, it's not a soft cut. It hits a wall. The first time I noticed this was on the track, because the car is not registered or insured.

Here's a link to a video of the problem. https://drive.google.com/open?id=1vf...vknLO6F3U6q1lT

The car had a MazdaManiac tune on the ECU. I switched that for the 'stock style' tune.

I have changed the alternator, because the 15v+ the RaceCapture shows seemed high. It turns out that's just wrong. I get 14.4v at multiple points when the car is running.

I have changed out the ESS and tone wheel, in case those were the problem. I reset the NVRAM afterwards.

I also took out the BHR coils and put in the stock set that I had in the old car last fall when I wrecked it.

At this point, the only thing I can think to check:

1. Get an adapter for the fuel line and check pressure.
2. Undress the engine completely and start fixing every damned little thing I can find.

If anybody else has any other ideas, I'd love to hear them. I'm less than two weeks away from my next time trial.

Thanks,
Nick

Can you get past 8300 at partial throttle? If you can, I would think that narrows it down to fuel/air delivery, so fuel pump, intake valving. If you can't it's more likely some combination of ignition, sensor, tune or timing problem.

In the video linked above, I'm applying partial throttle - only enough to nudge it up to the point where it hits that wall.

The tune on the ECU currently is the Cobb AP 'stock style' tune. It did the same thing (cutting out abruptly at 8k-ish) with the MazdaManiac tune that was on it.

Have you tried just uninstalling the Cobb to put it on the stock tune? I doubt it is that but some data logs would help as well and definitely check the fuel pressure.

Could be a air solenoid issue. If the solenoid has failed then the actuators for the SSV and VDI will not function

I saw something very similar on a race car last week . It wouldn't rev past 7500 in 4th gear but was fine everywhere else . I logged it and found the throttle was opening and closing by itself ! The owner put a new throttle body on and problem went away .

The Cobb is its own problem. I can get it to write maps that are stored on it to the ECU, but I can't get it to connect to a PC to write new maps to it.

The AP has a 'stock style' map on it that I wrote to the ECU. Is it an actual 'stock' map? I have no idea. It doesn't seem to have one stored on it that was the map as the car came from the factory.

I'm waiting on a USB OBDII interface so I can use a more detailed diagnostic tool like Forscan to get a full log of what's happening. That should come tomorrow, and hopefully it helps.

Fuel pressure seems...unlikely. It seems probably caused by something electric or software. The way that OBDII data seems to drop out completely is very suspicious. I do have a Deatsch 200 pump to install if I need to.

I have wondered that myself, but I think an air solenoid issue would probably result in a lack of power, not an actual inability to rev past a certain limit.

Oh that is a weird symptom. I'll look at the TPS and throttle command when I can log it tomorrow.

How was that car stopping at 7500? Was it just a loss of power, or a sudden, hard cutoff like mine?

You just have to uninstall the Cobb and it will revert it to the OEM map.

What do you mean by 'uninstall?' The Cobb AP writes a map to the ECU. Then I unplug it from the car. The only way the stock map gets saved for later use is if the user actually takes that step first. The previous owner of my car doesn't appear to have done that, because the only maps on the AP are the Cobb ones.

It would accelerate fine in all gears except in 4th when it got to 7200ish rpm and then it just wouldn't go any faster . If he changed into 5th , it would be ok again . I think it was one of those combination of issues things ...ie weak throttle solenoid, low voltage , pressure in the intake due to a front scoop ...etc etc . I actually still have the log if you want to see it ? You can clearly see throttle at full in previous gear then at full as he changes , then it goes all over the place as he reaches 7200ish.

There is an uninstall feature on the Cobb, it's something you have to do to revert the car back to stock and something you would have to do if you ever decided to sell the Cobb because if it is not uninstalled then it is useless on any other car.

Oh right. Well, that's worth a shot. I'll try it over lunch. Thanks.

That's freaking weird. I'll keep an eye out for that. Thanks!

I'm trying to get a USB OBDII interface working with the car. In the meantime, I'm tempted to unplug every damned thing I can that won't prevent the car from starting. Maybe there's electrical interference?

Aside from ABS, what else can I unplug that won't disable the car? Where can I find those things?

Thanks.

It could be anything, but it's not tune related if you uninstalled the AP.

Do that same test you did in the video at full throttle and log throttle position . If it's similar issue you will see it on the log . Doing it at part throttle might not be so obvious . :dunno:

I will try that, but I'm pretty sure my problem is not the throttle body - unless it's electrical interference of some kind.

The way OBDII seems to stop communicating and warning lights flicker tells me something really weird is happening electrically.

Probably just your battery cables or some dumb shit.

Hmmmm ground loop? How is the logger grounded, if at all?

It actually turned out to be settings for the OBDII PIDs it was watching. They needed to be set to a 'passive mode.'

Right? It never occurred to me that it would just be shoving data into that network, messing with the frames, by default.

Bus networks suck. One bad device attached and you get tons of collisions and errors.

There's a reason why 10BASE-5 and 10BASE-2 Ethernet were phased out decades ago.