MR2 Owners Club Forum banner

1 - 20 of 40 Posts

·
Registered
JDM '94 gen3 turbo
Joined
·
39 Posts
Discussion Starter · #1 ·
Hey Guys,

Finally got the 'new' MR2 registered and I've been driving it around the past week just to get a feel for it before. Unfortunately, after 3-4 days of driving it I found an issue today.
'94 JDM Turbo (gen 3 3sgte) 75k miles

I was driving home, and the car starting acting very funny on me. I would describe it as "jumpy".

The check engine light comes on, and the moment that it does, my MR2 starts acting like it has no idea whether it wants to accelerate or decelerate. It'll give a ton of power, then immediately cut that power, then give a ton of power again, then cut it again, over and over. It occurs at a somewhat regular interval, but its not perfectly synced by any means. About every like 1 second or so it'll switch from low rev to high rev. It was very minimal in 4th gear, but I went to slow down and changed gears, and it seemed to get progressively worse the lower gear I went into. After I got it in second, I just threw it in neutral and coasted to the side of the highway. It died on me in neutral as I coasted.
While stuck there, I could start it and idle, but if I so much as TAPPED the accelerator it died immediately. If I wanted it to NOT die while idling, I had to actually hit the accelerator decently hard to get myself into like the 2.5k. But while doing this, it was not particularly happy, just sounds like a dulled noise compared to normally revving in neutral, and again it was slightly jumpy.

Was in a dangerous spot on the highway, so I decided to risk getting out of there and I was able to get into first, and move (while it still did the stuttering thing). The weird thing is that after that, when I hit 2nd gear, the check engine light shut off, and I got all my power back. I was less than a mile from home, so I manage to get it back to a large parking lot in my apartment complex before the behavior returned.

I was confused, and tried to drove it around my parking lot for a sec to see how it behaved. It HATES 1st. The stuttering is REALLY bad in first but lessens in 2nd/3rd. The other weird thing is that it seems to just go away completely randomly at times, acts happy for about 10-20 seconds, then acts weird again. In all cases though, when the car behaved weird the check engine light is on, and it turns on and off perfectly when the behavior starts and stops.

Other things I noticed afterwards in the parking lot I thought I heard a very strange noise occurring simultaneously with where I would normally hear just the whooshing of air intake.
When I finally got it in the garage, I let it just sit and idle, and it definitely sometimes idles in a 'jumpy' manner.


Having only recently familiarized myself in the last year or two with the mechanical knowledge behind vehicles, my immediate thoughts are:
1. Sensor (afm? tps?)
2. Vaccum leak
3. Fuel filter/pump is clogged or something

Anyone have any ideas? Just looking if anyone else thinks I'm way out in left field or not.
Thanks ahead of time for any thoughts.
 

·
Super Moderator
91t gen5
Joined
·
6,766 Posts
Gen3 doesnt have a AFM just a map sensor. You can unplug the tps and test the car it will still drive without it. I think you might be onto something with the vacuum leak.

Have you checked to see what codes the ecu has stored?
 

·
Registered
JDM '94 gen3 turbo
Joined
·
39 Posts
Discussion Starter · #3 · (Edited)
Well that explains why my "afm" didn't look like what I'd seen in videos haha. I havent checked the ECU codes.

Is this the correct method for doing so? Or is there a higher tech method?


I've actually done some software work involving CAN on OBD-2 at my job, but im completely in the dark about OBD-1, other than knowing that its not "simple" and very vendor specific. I'll read up on it some more.

Thanks
 

·
Super Moderator
91t gen5
Joined
·
6,766 Posts
Just like on that website you linked but TE1 and E1. Jump those two together and then turn the key to the ON position, the engine doesn't have to be running. The check engine light will blink for the code number and continuously if there aren't any stored codes

 

·
Registered
JDM '94 gen3 turbo
Joined
·
39 Posts
Discussion Starter · #5 ·
Just got a chance to check it this morning. It gave back only one code, code 31 which is apparently 31: Air Flow Meter signal (Vacuum Sensor signal). Reassuring to me that the car at least agrees with us.
 

·
Super Moderator
91t gen5
Joined
·
6,766 Posts
I assume you're good with a voltmeter from your job description. The MAP sensor has 3 wires, a 5v supply (VC) , a ground (e2), and the signal (PIM). I bet something is going on with the wiring conversion from the swap causing an intermittant connection. Gen3 MAP sensors aren't cheap so you need to be absolutely positive it's bad before you spend the money on one.
 

·
Registered
JDM '94 gen3 turbo
Joined
·
39 Posts
Discussion Starter · #7 ·
You are correct with that assumption. Cool, I'll give it a test.

Unfortunately a botched swap wiring conversion wouldn't be a problem because all the wiring is original haha. My car is a complete '94 RHD JDM import, so same engine/wiring it came out the factory with.
 

·
Registered
JDM '94 gen3 turbo
Joined
·
39 Posts
Discussion Starter · #9 ·
Yeah I figured getting the extra 40 horses and a mostly mechanically stock vehicle to either keep stock, or turn into something of my own was worth the RHD downside. I think I got a decent price on it, so I'm not complaining haha. (was also a sucker for the color).
 

·
Super Moderator
91t gen5
Joined
·
6,766 Posts
Yeah I figured getting the extra 40 horses and a mostly mechanically stock vehicle to either keep stock, or turn into something of my own was worth the RHD downside. I think I got a decent price on it, so I'm not complaining haha. (was also a sucker for the color).
You're right gen3 is much better and worth it over gen2. You might also want to open up your ECU, the capacitors on the gen3 ecus leak and cause corrosion, this could be part of your problem.
 

·
Registered
JDM '94 gen3 turbo
Joined
·
39 Posts
Discussion Starter · #11 ·
Thanks. I came out to try and test the sensor this morning, but realized I don't have a good way to actually access the signals while its connected.

How do people typically measure it while its plugged in? Do they suck it up and splice a part of the wire so they can actually measure off of it?

I pulled it off though and 25 years worth of dust came out of the thing. I cleaned it up and plugged it back in, and it seemed like the car idled way smoother. Idk if thats normal for having the thing unplugged vs plugged in though. Went and drove it around to see if it exhibited the horrible behavior. It did not. But, to be fair it has only done this to me once so far out of 5-6 drives.

Either way though, my vacuum hoses look rough to my eyes. I'm thinking about replacing them for peace of mind.
74389
74390
 

·
Super Moderator
91t gen5
Joined
·
6,766 Posts
Where did the dust come from? Where do all of those vacuum lines go to? The MAP sensor should have its own dedicated vacuum line to the intake manifold right next to the fuel pressure regulator line. If you need a source for vacuum I would tee off the fuel pressure regulator or pick another port on the manifold.

You can use a t pin to "backprobe" the connector I wouldn't cut wires or insulation.
 

·
Registered
JDM '94 gen3 turbo
Joined
·
39 Posts
Discussion Starter · #13 ·
I just snapped some pics of some of the hoses in general for reference. The one going from the MAP sensor actually didn't look horrible.

Sorry, poor wording with the dust comment. I meant I pulled the CONNECTOR off, and dust fell out of the connector. The dust was kinda trapped between female connector's plastic guard, and the sensors plastic guard. I had a HELL of a time trying to get the connector off though. It was like he dust buildup had trapped the connector in place. Once I brushed off the dust buildup, I was able to much more easily properly connect and disconnect the connector.
 

·
Registered
JDM '94 gen3 turbo
Joined
·
39 Posts
Discussion Starter · #14 ·
74391
74392
74393


So. It looks like the vacuum tubes are all hooked together, in a 3way + 3way which go off to the 4 following locations.
1. MAP sensor connects to a hard-line bolted on the manifold, but that hard-line is connected into the large intersection. (You can see the hardline in the leftmost picture under that rectangle bracket on the manifold)
2. Intake manifold (also visible in leftmost picture kinda hiding under those two lines that disappear into the top left corner of the photo.)
3. The sensor in my third picture next to the map sensor is connected to the intersection next near the intake manifold as well.
4. There is a connection that heads back towards the radiator, and disappears behind some material in front of the firewall. Not sure where it ends.
 

Attachments

·
Registered
JDM '94 gen3 turbo
Joined
·
39 Posts
Discussion Starter · #15 ·
After looking up that other sensor.. It's also a MAP sensor? I am somewhat confused now.
 

·
Super Moderator
91t gen5
Joined
·
6,766 Posts
Your car has some crazy stuff going on. The MAP sensor with the label worn off is the factory one, the one next to it with one bolt holding it in is not factory. It seems like that MAP sensor was used for boost controllers or an Apexi power FC. The vacuum line that disappears is probably going to a boost gauge.

If it were me I'd tear it all out or at least figure out where all of it goes and replace the lines that look rotted.

Do you know where the second MAP sensor's wires go to?
 

·
Registered
JDM '94 gen3 turbo
Joined
·
39 Posts
Discussion Starter · #17 ·
Haha. Perhaps its not so stock after all! I'll hunt down where that 2nd MAP sensor is going and then figure out from there whats going on.
 

·
Registered
Joined
·
7,722 Posts
Often PO’s install a mechanical vacuum gauge in cab and run a hose off manifold to it. These lead to issues with pressure leakage which in turn affects MAP sensor. As mentioned above I’d eliminate these additional lines and ensure your factory MAP has good connections. Once engine is confirmed operational install the boost controller and gauges off seperate tap from plenum away from MAP.

Jimb
 

·
Registered
Joined
·
104 Posts
How did I miss this thread? This sounds almost exactly like what my car started doing right after I bought it. The check engine light would come on and the car’s acceleration would cut in and out like an on off switch. Totally undriveable. Engine would die at idle or just idle weird, smoke from the exhaust... it was all over the place. My mechanic replaced the MAP sensor and the car was fine for a day, but then the issues came right back. He then suggested the ECM to be the real culprit, since it had directed him to the MAP sensor in the first place. He put the original MAP sensor back, refunded me, and suggested I seek out a new ECM. I ended up buying a refurbished unit and installing it in my driveway, and it fixed everything immediately, just like that. None of the issues have come back in the three years since. I don’t know how much this could apply to your car, but I just thought I would share, because man do those main symptoms sound familiar.
 

·
Registered
Joined
·
311 Posts
My car was doing that and it was due to leaky capacitors in the ECM. Fortunately I had some spare computers to make the diagnosis quickly. The caps go bad on everything. Speedometer, tach, headlight retractor module, and ecm. I replaced the capacitors after testing with a ESR meter.
 
1 - 20 of 40 Posts
Top