I bought an 04 RX 8 about two months ago. 130,000 miles on odo, but about 30 on S1 engine and trans (bought it from a friend). Cat looks like it was replaces around then, too. Overall, car clean and well maintained by all indications. Everything stock, naturally aspirated.

Compression check is good and car was running perfectly. No smoke, easy starts, smooth idle -- for a rotary. Solid acceleration and power curve. No CEL or other lights.

A week ago, I was driving the car -- normal warmup, etc. --- and I hit a bump in the road while acelerating into a curve where they had lapped the asphalt, causing the car to do a bit of a side skip/bump. Nothing major, did not bottom out or hit the stops, but did jar the car pretty hard.

After that, the throttle response became sluggish -- I still had power, 70 mph shifting out of third, but it felt almost like a car with a worn clutch. Then, when I came to a stop sign, the car idled low for a few seconds and then stalled. The restart was a bit off -- like light flooding. Drove the car the rest of the way, about 15 miles, no CEL or other issues, but still felt sluggish and stalled again -- idle rpms low. Engine not vibrating, running or idling rough, no unusual noises -- just not it's usual self.

I turned the car off and let it sit for about an hour and a half -- was actually interviewing a 15-year-old racecar driver at their garage (and salvage yard -- gearhead heaven). Went back and started the car -- had to do flooding start protocol. Car cranked and warmed up, then drove home. Problems gone. Until I hit another bump about five miles from home. This time got a CEL with the power loss.

Went home parked. The next day, restarted fine, drove fine, no CEL, parked and started three times, no problems, until coming home, hit a bump, same problem. Parked and turned off for 10 minutes. Car started like it was flooded again, but then ran fine. Drove it intentionally over a bump to see if I could recreate the problem, did, got a CEL -- kept the car running until I could get the computer plugged in. Pulled a "lower than expected idle" P0506 code. When I turn off the car for 10 minutes, it comes back on and runs, no CEL and no code. Whatever is wrong is getting reset when it is off.

My guess is I have a sensor wire, O2, MAF, Cat -- something that comes loose for a second when I hit a bump, tells the computer I am running lean so starts running very rich. I have gone in and reseated all fuses and relays. I am going to clean the MAF and go over the 02 sensors and their plugs tomorrow. Also check the wiring to fuel pump, etc. Does anyone have any suggestions aside from those usual suspects that might be causing the problem?

This is my first rotary that I have owned, but used to manage an import dealer and we ran a lot of 7s. I got to drive them a good bit and spent time in the bays with service dept. learning the basics -- always wanted one. Have pretty solid DIY mechanic skills -- I keep my old air-cooled 73 Westy running like a top -- so I am not completely out in the cold on troubleshooting engines, but this is my first time working on an RX and any help would be appreciated.

 

Well, considering this car, the first thing I would check is the status of the motor mounts--they leak/crack easily and the jolt may have caused enough shifting to pull on the wiring harness. They won't throw a CEL (obviously), it's simply a common mechanical failure point. You can easily see the motor mounts through the wheel wells.

I would suspect something is wrong with either the Eccentric Shaft Sensor, or you managed to damage the throttle body motor. Problem with the ESS is, I'm not sure anyone has actually diagnosed a failure--a dirty one, perhaps, but never a case where they needed a replacement part.

Start with the workshop manual on foxed.ca. Do the throttle body inspection under Engine/Intake-air system and the Eccentric shaft position sensor inspection under engine/control system.

Good luck.

 

Much thanks tpb, I will check the mounts. They were good a couple of months ago, but the city and county have been repaving every inch of road around here and they ones they were not tearing up, the utility companies were adding potholes to, so it is possible one of those has gone bad.

That had not occurred to me because of the lack of vibration or other symptoms during idle -- but it may be enough to have yanked the wiring harness around and pulled something loose.

I don't think I will actually have a failed sensor, but more likely a loose connection to one -- I will double check the ESS wiring, too.

kudos.

 

Still having intermittent problem with the slow response -- but it has gotten better.
That may be in part because I am going over the car an inch at a time -- cleaned the MAF sensor, the ESS sensor, throttle body, checked both O2 sensors (and replaced manual oil with MT-90 while under there - no shift problems, but did not know what was in there so better to be safe).

Checked all other sensors and appear to be working. Not sure about the CTS sensor.

Engine mounts are good -- inspected visually, no unusual vibrations and passes the third-gear shift test without any lag.

I have a new set of coils that are going on next weekend.

Reset the ES position and fuel trim memory.

Overall, the car is running better -- but it was not running to much off to begin with. But there are still days when I hit a bump in the road and I get the flat throttle response and stalling at lights. Definitely running rich and flooding when that happens. Flood start procedure needed to get running again.

Could the Cat be bad and not throw a CEL light and code for that? I was wondering if there was a bunch of crud in the Cat that gets bounced into the mesh causing back pressure. Turn off the car and in drops back down off the mesh. That's a long shot.

Also, any possibility something is going screwy with the AC that is causing it to drag the engine. I do notice that the AC has been on every time that has happened. I have not had the power loss with the AC off, but that could just be coincidence.

I still think it is an electrical problem because it is intermittent, but am open to any suggestions. Any thoughts on a specific wire plug that if knocked loose for a second would cause the car to run rich until turned off and then reset?

 

Well, inspecting the cat is fairly easy, just pull it off and look at the honeycomb. AC drag would typically happen after the car is warmed up and the temperature outside is rather high. The bump could have damaged the A/C compressor to a point where you build excess pressure. The simple solution would to turn off the A/C next time it happens--make sure you can hear the compressor clutch disengage.

I actually talked with a friend of mine that happened to encounter the same code (p0506). In his case, the problem originated from a fuse that had been knocked loose (partly his own fault, hit a deer). Along the same vein, I would trace the throttle body and ignition as far back as possible and visually inspect connections to the ECU (for sensors) and fusebox. My thinking is this the car is running with working ignition and air control *bump* then air gets choked off or a plug stops firing.

After that, I got nothin'.

 

Put the new coils on today -- easy. Inspected everything but the Cat. So far, everything looks good. I suspect the coils I replaced were fairly new. Everything I am running across with this car maintenancewise looks like it was well taken care of. No bolts overtightened or stripped, all the rubber and other stuff taken care of or replaced. Everything that I would replace has been.
Which is encouraging.

After replacing the coils, I went out and ran it a good bit with no problems. Then hit a bump with the rpm around 7000, and presto. Slow throttle response. Turned car off, let sit, started up fine, with everything reset, drove great. Hit the same bump, but with rpm closer to 5000 and no problem. Hit it another time with rpm over 7000, and got the problem -- and a Cat CEL. It went away after the restart, but my guess is the Cat is dying and has some loose buildup in it. The high flow at top rpm plus getting knocked around is enough to create back pressure that isn't there at lower rpms.

Either tomorrow or next weekend, I am going to pull the cat and inspect it. If it is clogged or coming apart inside, I will grab a metal dowel and turn it into a midpipe until I can find either another used cat in good condition or a decent resonator midpipe and catback.

Thanks for all the advice.

 

Pulled the cat today and can absolutely guarantee there is no back pressure from there -- it is a perfectly smooth midpipe. Car running fine today, but problem is intermittent in the past. Also no CEL. Will see what happens.

Next suspect is the front O2 sensor and then I will start testing and cleaning various intake valves.

 

The answer to the problem?
Survey says:
Loose SSV solenoid plug.