Jim85IROC

iTrader: (2)

A friend has a 20k mile 89 IROC with a completely stock LB9. For quite a while he's been having a problem that's been getting more and more frequent. He'll start the car, and it'll run fine for a while, then all of a sudden the idle will drop 300 rpm, and it runs real rough. If you give it any throttle, it'll almost stall. So far, he's changed the ECM (twice), plugs, wires, cap, rotor, coil, ignition module in the distributer, fuel pump, and he's had a fuel pressure gauge on it. He's checked the spark and it's nice and bright.

I went over last night with my scan tool to see if we could learn anything. After startup, everything looked normal. Coolant temp sensor is working, MAF airflow is where I'd expect it to be (10g/s at 1k rpm), oxygen sensor seems ok, crosscounts are going up, blah blah blah. Suddenly, the idle drops by 300 rpm and it's running poorly. Oxygen sensor voltage seems to be hovering around 400mv although crosscounts are still going up. Block learn jumps to 160, and injector pulse width jumps from 2.2ms to almost 4ms. Every indication is that this thing is suddenly going WAY lean. We throw a fuel pressure gauge on it, and with the key on, it's right at 43lbs. Start it up and let it run, and it's in the 35-37lb range, just like I'd expect. When it starts running poorly, it goes up to around 40, which makes sense because there's not as much vacuum when it's running poorly.

My guess is that his fuel pump is dieing. He's got pressure, but I think there's not enough flow. What is confusing me is that this issue isn't constant... it runs fine for 5-10 minutes, then suddenly gets this way. We've also tossed around the idea that maybe it's the chip/memcal, but I'm reluctant to think that's the problem. The computer seemed to be doing all the right things in an attempt to combat the lean condition. We've also discussed the possibility that maybe the screen in the tank is plugging up with crap, or the hose between the fuel pump and the steel line is collapsing, but I'm under the impression that those issues would cause a loss of fuel pressure that we'd see on the gauge.

Anybody have any suggestions? Before he goes through the effort of replacing the fuel pump, he wants to be reasonably sure that it's the problem.

Thanks

305sbc

iTrader: (3)

I would watch the datalogging a bit more to see if this problem coincides with any ECM trigger points, like EGR or CCP, etc...

I would also do some other tests like watching it with a timing light to verify it matches what you're seeing on the datalog.
I would also verify the actual AFR.

I would then start eliminating variables, like vacuum lines, AIR activation, IAC control, etc.
If you eliminate the MAF and the ESC, but the problem still happens, it starts pointing towards an intermittent mechanical, or main electrical system problem.

After that I'd do some mechanical checks: Spark plug visual, compression check, and pull the valve-covers to watch for a problem like a sticking valve.

HTH

LAFireboyd

iTrader: (3)

As I read this, those details jumped-out at me. This is exactly what happens when you manually-test the EGR system when the engine is at idle. By sticking your fingers underneath the diaphram of the EGR valve and pushing-up on it to open it, the engine should bog-down, and it will idle roughly. So I'd suggest checking the EGR system before you do anymore parts-changing.

Jim85IROC

iTrader: (2)

Would the EGR cause it to run so bad that it's undrivable though? When it's like this, it runs so poorly that I can't even give it any gas with it in park.

I suppose it's easy enough to unplug the vacuum line from the EGR valve and see if it helps.

naf

iTrader: (7)

EGR could cause this but I'd expect it to be a little more intermittent. Unplugging the vacuum line won't rule it out though as the valve could be sticking open even without vacuum. Do what LAFireboyd said: when it acts up manually open the valve and it should worsen OR manually open the valve at normal idle and see if it displays the same symptoms.

Fuel pressure increasing slightly when it runs rough doesn't indicate a bad pump. If the pump wasn't supplying enough fuel to meet flow requirements the pressure wouldn't be good. It could be that the injectors aren't opening enough. Pump is providing fuel into the rails but it's not getting out.

Jim85IROC

iTrader: (2)

With it idling ok, injector pulse width is at 2.2ms. When it's running poorly, pulse width goes up to about 4ms... I'm assuming the computer is trying to dump more fuel in to compensate for the lean situation. I suppose there could be an issue with the injector(s), but this isn't a problem with a single cylinder as far as I can tell... it's something that's effecting all 8.

I know that the higher fuel pressure doesn't indicate a bad pump. The pressure is higher because there's less vacuum. I'm questioning whether or not we can measure adequate pressure and not have sufficient flow. It's not something that I've experienced, but I've read over and over here and other places that a fuel pressure gauge won't tell the full story because in some cases you can have adequate pressure but inadequate volume when a pump is failing. I'm wondering if that's what we're seeing here, or if I'm barking up the wrong tree.

305sbc

iTrader: (3)

Watch the battery voltage closely while you're datalogging.
I doubt that it's running out of fuel at an idle.

naf

iTrader: (7)

You could, in certain conditions, measure adequate pressure but still not have adequate flow. It would be a case of your pump not being able to overcome static and dynamic losses in the fuel conduit system and flow suffering. A clogged filter, crimped fuel lines or some other restriction in the system would be the most likely causes-for a near stock system. This is unlikely to be your issue, being intermittent (the restriction part) and highly unlikely to be a pump problem for you. I can't imagine how a pump could fail and still provide adequate pressure WHILE exhibiting failure symptoms.

Jim85IROC

iTrader: (2)

Battery voltage seemed to stay rock steady at 14.x volts while running.

I'm really starting to hope that it's an EGR issue, because beyond that, I'm really at a loss. I can't see how it's mechanical in nature, because it's too severe of an issue to be something that's impacting only one cylinder. I'm confident that it's something causing a severe lean condition, yet the ECM & associated sensors show no signs of being to blame... and if we're ruling out fuel, then EGR really seems like it could be the only other variable.

naf

iTrader: (7)

Keep in mind that it only has to impact one cylinder on the O2 side to severely whoop the ECM's ability to maintain a consistent mixture. 'Course the O2 reads 'O2' and not 'rich/lean'. If the ECM is responding by trying to richen the mixture, it's sensing too much O2, on the left bank.

Check your plugs-right after it does it-see if any look different. Ohm the injectors?

Jim85IROC

iTrader: (2)

Perhaps. I've had a lot of situations where I had a single cylinder go ****-up and I've never had it effect the idle/performance so severely that the car barely runs.

naf

iTrader: (7)

Don't think it would in this case either, didn't mean to imply it would. Response in most failure cases would be a rich condition, which, as far as the engine is concerned, is more pleasant than a lean response (and less likely to do damage). Was just thinking that if I were to check plugs, I'd check the left bank first-cause the ECM is definitely seeing a problem there.

ASE doc

Hi Jim, Youre correct in thinking that a fuel supply issue which would affect idle would cause a drop in fuel pressure. The inegrator numbers point away from a fule control issue as the system is adding fuel to compensate for a lean condition. I have seen several cases of what youre describing. A few were broken EGR valve springs. A few have been bad O2 sensors. As soon as the vehicle went to closed loop, a few minutes after start up, the engine would drop idle and run poorly. This is hard to see on data at idle as the single wire O2 doesnt read properly at idle to begin with. It needs to be heated 650f to read its full range. But if you see high integrator numbers and dark spark plugs. Suspect a bad O2. The EGR Solenoid is a vent type. It vents vacuum to atmosphere til the ECM closes it allowing EGR. Then the valve is controlled by proted vacuum vs spring tension. A broken spring will cause the valve to open with minimal vacuum. Test it by disconntecting the hose from the solenoid to the valve when the problem is present. Apply vacuum to this hose. The engine should stall. If no significant change in idle occurs, you have a faulty EGR valve confirm this by blowing on the same hose. See if idle improves. Hope this helps. ASE Doc, L1 Master tech

ASE doc

Hi Jim, Sorry. Im getting old and I sometimes forget things. A quick test for O2 is to disconnect the sensor when the problem is present. This will put the ecm in fault/calmap mode. The O2 data pid will go to neutral, about 450mv. See if the idle improves and the inegrator numbers return to normal. Ive had O2 sensors be bad out of the box. Hope this helps. ASE Doc, L1 Master tech

sergio

Jim,
I a currently experiencing a similar problem. Have you solved this problem? Was it the fuel pump?