eword Posted June 26, 2018 Share Posted June 26, 2018 Just bought 96 legacy,173,000 miles, had a few things wrong, setting codes. Ran solid and smooth until it starts cutting out at random RPM/ load... Will still run and idle smooth, until it decides to start cutting out, then followed by uniform surging at idle. Doesn’t want to die, just Erving at idle.Replaced knock sensor and PCV, cleaned MAF, new plugs/wires/ fuel filter and pump. Seafoamed fuel and throttle body, to try to rule out all injectors being grimy. So when this condition is set, and irratic idle is happening, I can rapidly press the break pedal, and idle will smooth out. So I tested booster and booster check valve, and they seem to be in order as far as check valve operates in one way flow and booster checks with pressing pedal during engine running/off pressures on pedal and sinking when engine is turned on.... Was suspecting vacuum leak anyway, and thought booster diaphragm was ruptured because when pumping breaks smoothed our idle. But can’t confirm bad booster. Last codes after the replaced parts are 0100 and 0505, and catalyst, lean fuel, and O2.... anyone like the looks of this? Quote Link to comment Share on other sites More sharing options...
eword Posted June 26, 2018 Author Share Posted June 26, 2018 Hey, e, you getting sick of fixing these old subies? e: nope but more info...... when this condition is NOT happening, you can balence a marble on the stick shift. Smooth as silk. Plenty of power and hopes for another 80,000.......... Quote Link to comment Share on other sites More sharing options...
1 Lucky Texan Posted June 26, 2018 Share Posted June 26, 2018 maybe the big hose going from the idla air control valve to under the intake tubing is loose? maybe crank position sensor is getting intermittent? Quote Link to comment Share on other sites More sharing options...
eword Posted June 26, 2018 Author Share Posted June 26, 2018 Wouldn’t those be setting codes, though? I was going to clean the IAC, but am finding conflicting advice on it online. Some show how to do it and have gasket handy, some are adamantly NO WAY, dealer only.......... Quote Link to comment Share on other sites More sharing options...
eword Posted June 26, 2018 Author Share Posted June 26, 2018 It seems coil, injectors, sensors and stuff that throws a code, I’m back-burnering... evrything i replaced, except probably the fuel pump, needed to be replaced anyway. It was coding for MAF until I cleaned it. It’s not coming up anymore... Quote Link to comment Share on other sites More sharing options...
eword Posted June 26, 2018 Author Share Posted June 26, 2018 I’ll check that hose. I’ll see if I can test these other components. Might be bad but not storing code? Quote Link to comment Share on other sites More sharing options...
eword Posted June 26, 2018 Author Share Posted June 26, 2018 Also have cleaned upper throttle body and throttle plate.. Quote Link to comment Share on other sites More sharing options...
montana tom Posted June 26, 2018 Share Posted June 26, 2018 Pick up another MAF if you can, preferably a known good one(borrow a friends) or wrecking yard, do not buy a china knockoff. MAF are famous for slowly going bad in that year group. Seems a solder joint inside makes intermittent contact. They don't always throw a code . 1 Quote Link to comment Share on other sites More sharing options...
eword Posted June 26, 2018 Author Share Posted June 26, 2018 I disconnected the MAF and drove around. It runs rough, but still has the stutter thing going on. wouldnt a cracked valve seat or head gasket throw a code somewhere? Quote Link to comment Share on other sites More sharing options...
eword Posted June 26, 2018 Author Share Posted June 26, 2018 Show up as a misfire in the affected cylinder as a code? Quote Link to comment Share on other sites More sharing options...
1 Lucky Texan Posted June 27, 2018 Share Posted June 27, 2018 (edited) have you scanned for pending codes? it may be necessary to get live data - ELM327 BT adapters are cheap, and a smartphone with the Torque free app can go a long way towards getting sensor data that 'might' point to a problem. check your brake lights in case there's something weird with the bulbs or corrosion in the sockets - the lights put a drain on the alternator when you hit the brake pedal.....? there a lot of possibles sources for intermittent problems. You might examine ground connections, battery terminals, monitor voltage from the alt. and compare when it's running good vs bad.... youm ight try a battery disconnect reset - if the car has the same symptoms after that, I doubt a:f -related sensors are bad. The ECU is forced to start and run the car using the factory initial settings until data is gathered from the sensors. Could be something more physical; dropped valve, vacuum leak, ...?. Edited June 27, 2018 by 1 Lucky Texan Quote Link to comment Share on other sites More sharing options...
eword Posted June 27, 2018 Author Share Posted June 27, 2018 This torque app? It says it’s a payment calculator... Quote Link to comment Share on other sites More sharing options...
eword Posted June 27, 2018 Author Share Posted June 27, 2018 If I disconnect booster hose, or PCV hose, and partially plug fitting at throttle body with thumb, allowing unmetered air, it idles smooth as silk. Thinking the air filter was clogged, I removed it, no change. Disconnecting MAF will stall motor, but on restarting, surging idle remains, slightly less RPM. When plugging MAF back in, RPMs pick back up, surging idle remains. Am I wrong in thinking bad O2 will send a consistent bad/ default signal and not cause surging idle? Quote Link to comment Share on other sites More sharing options...
eword Posted June 27, 2018 Author Share Posted June 27, 2018 My bad. I am wrong about the O2.... also dumb, as it was setting codes and telling me....... going to replace now. ill get my dunce cap and let you know........ Quote Link to comment Share on other sites More sharing options...
eword Posted June 27, 2018 Author Share Posted June 27, 2018 Changed 02. Nothing changed. Rescanned codes, got p0100, p0133pd mil on, monitrs catlyst 02 snsr O2 htr went to test MAF,, voltage too low to start test. test and replaced battery, test MAF, tested ok. drove great! For 20 minutes... Now back to same surging idle, seems like brakes are partly engaged? Doesn’t roll, clutch in, freely..... Quote Link to comment Share on other sites More sharing options...
eword Posted June 27, 2018 Author Share Posted June 27, 2018 Going back to test codes...... Quote Link to comment Share on other sites More sharing options...
montana tom Posted June 27, 2018 Share Posted June 27, 2018 That year subi has a hill holder on it. If you were trying to roll backwards it will mess with you. If you were rolling forwards then you should check the caliper slide pins and grease them with Sil-glyde. Quote Link to comment Share on other sites More sharing options...
eword Posted June 27, 2018 Author Share Posted June 27, 2018 Thanks, MontanaTom! I’ll definitely check this out! Quote Link to comment Share on other sites More sharing options...
heartless Posted June 28, 2018 Share Posted June 28, 2018 (edited) interesting tidbit for you.. one code being set can set off a whole sequence of other codes... start by addressing the FIRST code listed when reading the codes.. fix that. reset codes, see what comes back then fix the next FIRST code, reset, see what comes back... repeat as needed. Edited June 28, 2018 by heartless spelling 1 Quote Link to comment Share on other sites More sharing options...
eword Posted June 28, 2018 Author Share Posted June 28, 2018 That’s what I’m shooting for. P0100 MAF is likely a vacuum leak, having visually inspected connector and observing proper voltage at idle and uniform increase with opening of throttle plate. Oddly enough, P0133, as I’ve just replaced O2, is also likely a vacuum leak or bad MAF which has been tested, and likely also a vacuum leak. so when I disconnect ANY vacuum from throttle body the idle smooths right out. Disconnecting the fuel pressure regulator smooths out idle until you plug hose going to TB, and closing that “vacuum leak” , and it goes right back to surging idle. Leaving fuel pressure regulator branch of vacuum connected, but disconnecting down stream switch also smooths out idle.. unless you plug the vacuum line, again surging idle. disconnecting the vent tube running from driver’s side head to air intake, after the MAF, will even out idle. Disconnecting brake vacuum line smooths out idle but idles high. I can partially plug fitting at intake manifold and drop RPM to a smooth idle. so, I’m trying to fix a vacuum leak when creating a vacuum leak seems to work. Is it just a rich fuel mixture that the unmetered air is helping to burn up? Why so rich? MAF and back to a vacuum leak that I kind-a want? everything I read and from past subie experience tells me p0100 isn’t about a bad MAF... which I tested... id love to hear ideas for tests, I don’t have the Bluetooth reader.. just voltmeter..... Quote Link to comment Share on other sites More sharing options...
el_freddo Posted June 28, 2018 Share Posted June 28, 2018 If unmetered air is smoothing out the idle have you thought about the possibility of one or more leaking injectors? Cheers Bennie Quote Link to comment Share on other sites More sharing options...
eword Posted June 28, 2018 Author Share Posted June 28, 2018 I have absolutely considered leaky injectors. There is no code being generated for individual cylinder so I would have to assume they all leak uniformly, though it doesn’t rough idle, it is a surging. Creating vacuum leak on either side of intake manifold smooths out idle. This also suggests to me it is a uniform issue with all injectors, if this is the case. I’m not wanting to risk breakage by removing them. Anyway to test this theory with injectors in place? Quote Link to comment Share on other sites More sharing options...
eword Posted June 28, 2018 Author Share Posted June 28, 2018 I’m at the shop, they have confirmed all my tests for sensors, including MAF being within specs, also cannot find vacuum leak, are pretty confident an exhaust leak wouldn’t cause this (I’m probably going to check this later on cool exhaust with soap water), and are also baffled that creating a vacuum leak evens out idle, the shop owner walked away at that point... I’m trying to find out if the computer is shot, they said they have no test for ECM(?) Any ideas? in the meantime, I’m trying to reset computer by disconnecting battery, and try relearn.... thanks all. Hope to close this thread... Quote Link to comment Share on other sites More sharing options...
1 Lucky Texan Posted June 28, 2018 Share Posted June 28, 2018 pull the vacuum line on the fuel pressure regulator - see if it's wet Quote Link to comment Share on other sites More sharing options...
eword Posted June 28, 2018 Author Share Posted June 28, 2018 I think you recommended that, I’ve done it, no gas in line, no gas dripping out unplugged, BUT when I disconnect that, or ANY vacuum line, the idle evens out. the mechanic didn’t believe when this happened at shop this morning. But when I plug fuel pressure back in, or any other line, motor goes to extremely low idle then directly to surge... Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.