Topic: 2004 STI Possible TPS issue (Read 545 times)

klarowe · « **on:** January 27, 2012, 09:58:46 PM »

The STI has been running a little odd lately and I'm trying to get some opinions. For a little while now its had a very slight rough / hunting idle... almost surging so to speak. Hasn't really bothered me so I haven't been working about it. But within the past couple weeks I've been having a hesitation issue. At cruising speed its a very consistent but VERY light miss but its barely noticeable. The big issue is when I try to get on it even slightly the miss is amplified to where in1st gear it will be as if I just completely let off the throttle and it just falls on its face for a second and then kicks back in. At first I though plugs so I put new plugs in and it didn't change anything. But then I noticed that it seems to be linked more to pedal / throttle position than RPM related, so I started looking into the TPS. I had to pull the TMIC off anyways to also look at my methanol flow sensor which hasn't been reading right and when I had just the ignition on (obviously, the car wasn't running), there was a higher pitched almost buzzing noise coming from the throttle controller... which as far as I know is also the TPS and idle control. I'm assuming this buzzing is not normal and that would explain both the hesitation and the idle issue... but before I go dropping $160+ on a new TPS I'd like to be as sure as possible.

So anyone have any opinions on the matter? Or even better if someone by chance has an extra throttle module laying around that they wouldn't mind letting me borrow I would greatly appreciate it.

Thanks in advance.

byroll01 · « **Reply #1 on:** January 28, 2012, 03:17:40 AM »

Mine buzzes when the key is on/car is not running.

Have you checked for vacuum/boost leaks?

klarowe · « **Reply #2 on:** January 28, 2012, 01:31:41 PM »

I haven't dove in to look for leaks, but I pull right around 20 at idle.

klarowe · « **Reply #3 on:** January 28, 2012, 01:32:52 PM »

Here is a quick vid of the noise... but if its normal then thats great because from what I can find that module is only sold with the entire throttle body... at close to $600...

KA_Killuh · « **Reply #4 on:** January 28, 2012, 08:45:28 PM »

You should be able to check the throttle to make sure it has ground, 5 volt reference, and then do a couple of sweep tests to see if there is a "dead" or flat spot in the sweep. That buzzing may be the throttle performing a self alignment or attempting to and failing. Not sure if Subaru throttles do that with Key on, engine off.

Also, how did the spark plugs look? Did any of them seem different from the rest? Do you notice smoke behind you when the car runs poorly and what color?

klarowe · « **Reply #5 on:** January 28, 2012, 08:52:45 PM »

All the plugs were good. They weren't too wet or too dry... and they all looked the same. I was hoping someone knew a way to test the throttle module. So it should have ground, 5v at no throttle, then slowly climb to 12v with the throttle? I know my way around a multimeter so I know how to test.. just not sure what to test for.

klarowe · « **Reply #6 on:** January 28, 2012, 08:55:32 PM »

And as far as running poorly, its so quick that I don't notice anything. Its basically just a quick hiccup then its fine again. Then when cruising at low throttle it has a consistent very light hiccup. I just wish it would throw a code so I can get an idea where to start.

klarowe · « **Reply #7 on:** January 28, 2012, 09:46:47 PM »

Quick update. Did a little more work / diagnosis and decided to hook up the code reader in hopes of finding something... even though the MIL was not on. And I lucked out a bit. I had 3 codes; TPS high input, IAT high input, and MAF high input. The MAF and possibly the IAT could be from when I was working on it. To test the flow sensor for the meth I was disconnecting / reconnecting the MAF with the ignition on.. which I'm sure was throwing shit for a loop. Not sure why the IAT was going off, but I was messing with just about everything near the TMIC. However I never once messed with the TPS so there is no reason that should have been tripped. I cleared everything and I'll recheck the codes after a little while and see if it comes back on.

hydrochloric · « **Reply #8 on:** January 29, 2012, 01:08:50 AM »

When I had a stumbling issue, it was my wires. But, of course, that's not an issue for you...

byroll01 · « **Reply #9 on:** January 29, 2012, 02:04:33 AM »

buzz in video sounds exactly like mine. IDK if that's the issue.

Have you changed any of the performance parts, like intake, exhaust or otherwise?

KA_Killuh · « **Reply #10 on:** January 29, 2012, 09:02:39 AM »

You should have ground, 5 volt reference, and a separate wire that is at 0 (or close to it) when the throttle is closed and ramps up to 12 when fully open.

klarowe · « **Reply #11 on:** January 29, 2012, 11:07:17 AM »

Quote from: byroll01 on January 29, 2012, 02:04:33 AM

buzz in video sounds exactly like mine. IDK if that's the issue.

Have you changed any of the performance parts, like intake, exhaust or otherwise?

Its Gus's old car and yes to all of the above.. lol. Here is the original modification list... I haven't changed anything engine wise since getting it.
FP Green; Ported TGV Deletes; Nismo 740cc injectors; APS 70mm intake / samco silicone inlet; SMC/Labonte methanol injectsion (SMC system with Labonte MAF based controller); Aquamist DDS3V8 gauge; Equal length header with 38mm external gate; Helix downpipe with cat delete; Borla hush cat back.

Quote from: KA_Killuh on January 29, 2012, 09:02:39 AM

You should have ground, 5 volt reference, and a separate wire that is at 0 (or close to it) when the throttle is closed and ramps up to 12 when fully open.

Awesome. Once I finish working on my buddy's dumpy VW i'll dive into the TPS and test the wires coming off of it. I'm assuming the ignition will have to be on to test?

klarowe · « **Reply #12 on:** January 29, 2012, 05:49:40 PM »

Here are the results of the TPS test. I had 6 wires at the switch, not just 3.
From the left side of the plug (connector on top:(
Red/black - tested ground regardless of key position or throttle position
Red - showed 1.05v at resting throttle and SMOOTHLY increased to 3.72v at WOT
Yellow - tested ground regardless of key or throttle position.
Lt. Green - 5v constant with key on
Dk Blue - showed 1.18v at resting throttle and SMOOTHLY increased to 3.84v at WOT
Lt Blue - 5v constant with key on

Both wires that changed with throttle position increased smoothly. It was tough to do through slowly, but I was able to adjust it in .02v - .04v increments.

So now I have no idea what is going on. And I rechecked for codes and brought up nothing. I'm heading out of town for a couple days tomorrow and have a rental car for the trip so I won't be needing it (or be able to look further into it). But when I get back I may just have to drive it until it throws a code and keep my code reader on me for when it does.

klarowe · « **Reply #13 on:** January 29, 2012, 06:05:28 PM »

I did some Google searching and found an article that described the Subaru DBW system in depth. Apparently it uses a redundant signal as a "back-up" which would explain why I had doubles of everything. So from what I can tell, other than a slight difference in voltage on the "position" wire (1.05v-3.72v vs 1.18v - 3.84v) everything seems ok. When I get back I will test the wires at the actual module on the TB and see what I'm getting there.

byroll01 · « **Reply #14 on:** January 30, 2012, 02:22:13 AM »

Isn't the reason you're checking the TPS because you thought the buzzing was abnormal? I'm telling you my '04 has buzzed like that for as long as I can remember. My opinion is to forget the TPS and start looking elsewhere, Have someone do some logs with you and send them to Dom. He tuned the car. Or pay Morgan@synapse to do some logs with you.

KA_Killuh · « **Reply #15 on:** January 30, 2012, 06:04:10 AM »

I feel like whenever a big turbo Subaru runs like crap, compression and leakdown tests are in order. It's a stock block, right?

Turbaru · « **Reply #16 on:** January 30, 2012, 06:57:02 AM »

Good idea. I think a compression test is in order, at the least you will know where you stand.
The 'hiccups' are what alerted me first that something was not well.
Hope its turns out to be an easy fix.

Clean the MAF sensor yet?

klarowe · « **Reply #17 on:** January 30, 2012, 04:04:29 PM »

Quote from: byroll01 on January 30, 2012, 02:22:13 AM

Isn't the reason you're checking the TPS because you thought the buzzing was abnormal? I'm telling you my '04 has buzzed like that for as long as I can remember. My opinion is to forget the TPS and start looking elsewhere, Have someone do some logs with you and send them to Dom. He tuned the car. Or pay Morgan@synapse to do some logs with you.

I have moved on from the TPS, I just wanted to make sure since I had the information to test it, and the test only took a few minutes. I'm going to try to do a few logs and see what is going on... just have to get some time to dig in further. I also want to find a way to test the cam sensor, crank shaft position sensor, and knock sensor so I can rule those out as well. I really don't think its the motor. it is a stock block, but a compression test was done when I bought it and everything was 100%. I will do another check when I get a chance just to be sure though.

KA_Killuh · « **Reply #18 on:** January 30, 2012, 06:38:01 PM »

Cam / crank / knock sensors will generally throw codes. I wouldn't waste time trying to test individual sensors on a hunch. Record some data and see what's going on. The longer you drive it, the more you chance causing more damage.

klarowe · « **Reply #19 on:** January 30, 2012, 07:08:05 PM »

I'll have to figure out how I can log. I have an OBDLink setup in the car for giving me engine parameters, but I'm not 100% sure if it has logging capabilities or not. I also have my innovate LC2 that does datalogging, but it only logs a/f which may not help me at all. Either way I may have to find someone to give me a hand logging... I pretty much have no idea what I'm doing. Performance is a hell of a lot different than offroad... haha.

byroll01 · « **Reply #20 on:** January 31, 2012, 12:04:58 PM »

I have a tatrix cable and could do some logs if I come down for an alarm install. Otherwise you could come up to Saratoga and we could the logs. I'm off tomorrow.

klarowe · « **Reply #21 on:** January 31, 2012, 12:29:54 PM »

I sent you a PM about the alarm. But of course I read this after the fact. I'll knock some off of the labor for your help / time. I work during the week and don't usually get home til 5ish so in order to do both the alarm and the logging, we would probably have to plan for a weekend if that works. Just let me know whatever works for you.

klarowe · « **Reply #22 on:** February 01, 2012, 06:54:03 AM »

Update: I started up the car this morning and the MIL immediately came on AND the cruise control light was flashing. I checked the DCCD to make sure it wasn't related and the DCCD was working fine. I plugged in my scanner and got a P0172 System too rich (Bank 1) and the same code but PENDING. Not quite sure how it knows the system is too rich because the upstream O2 sensor is my wideband and it is not connected to the ECU at all (as far as I know). I'm starting to wonder if there is a dead spot in the tune... but I could be way off.

Bioya · « **Reply #23 on:** February 01, 2012, 12:56:16 PM »

How is the upstream o2 your wideband? Did you change something? The car when you got it had 3 o2s, one in the header, one in the downpipe (wideband) and one "post cat" The upstream (b1s1) was new at +/-55k.

klarowe · « **Reply #24 on:** February 01, 2012, 02:24:36 PM »

Oh damn... I looked for the upstream a while ago for something and didn't see it so I assumed it was replaced by the wideband and that the tune took care of the signal for the ECU. Now I feel like a complete idiot... lol. Thanks once again... I'll look more closely at the O2 sensor and see what I can find.

Author Topic: 2004 STI Possible TPS issue (Read 545 times)