Jump to content

Jeep YJ 2.5L TBI Rough Idle and Very Little Vacuum


Recommended Posts

a huge problem with my YJ brings me to this forum, so I'm a new member.

 

I own a 1989 Wrangler YJ 2.5L 5-speed. It's a '94 block, with the '89 accessories/head/etc.

 

I drove my Jeep to work one day a couple weeks ago. No issues. Went to leave to grab some lunch at a local restaurant and it started idling not well in about 20 seconds after starting it up. Tried to just drive it anyways, and the power would completely cut out as if I were idling, even though I had it almost floored. The RPMs would go from ~2500 down to ~1000 where it would stay until I got off the gas, when it would die. I started it up a couple times, but it would repeat the process. Luckily I was able to do a U-turn and coast into my work's parking lot.

 

tried to drive it home afterwards and it would do the same thing, over and over. It seemed to run well off of the gas I used to start the Jeep with, bringing itself up to ~3000rpms, and I would be able to quickly get it in gear and drive it. but after that gas was used up, it wouldnt want to run anymore and would die. Luckily there were no cars on the road and I was able to make it about a mile to my grandfather's house, where the Jeep sits now, unable to even idle well.

 

My Jeep will start fine after it's been sitting awhile, go up to 2k rpms as usual, but will quickly fall to around 600rpms and try to die. if I give it gas, it doesnt like it and ends up flooding itself out rather than gain rpms.

It won't start again after that unless the gas pedal is to the floor, because it's flooding itself out. While it's doing this, it produces white smoke and condensation that leads to black all on the cement under the tailpipe.

 

Now for what I've tried to do about it. I've done alot, so I'll try to be as thorough but organized as possible.

 

I dropped the gas tank, dried it COMPLETELY out, because I had water in the tank before this. New spark plugs, a full tank of new gas, a new fuel filter, and jumping the fuel pump relay to flush the throttle body. No difference.

 

I got a new injector a few days before this started, because the old one was seemingly clogged. When the throttle body bonnet is pulled off, I can hear a sort of hissing sound coming from the injector. I tried getting the old one cleaned out and trying it, but it's the same as the new one, so the problem probably doesnt lie there. However, along with the hissing, the spraying seems to mess up when it happens. However, it's flooding so I'm not sure if that is a cause or a symptom.

 

The fuel pressure was good, the return line I blew into with my lips and heard it bubble in the tank, so it's not blocked.

 

Tested for vacuum...5in HG if I'm lucky. The other day before I put the new injector in, it was bouncing from I think 10in HG back down to 0in HG rapidly. Now it just goes beween ~0-3.

 

Pulled the Throttle Body off, the gasket was good and it's clean. The fuel pressure regulator is seemingly fine. the IAC is acting normally and isnt very old. The CTS is within it's range at ~70F (~3400Ohms). The MAT sensor is a little off, but I have the '91-'95 sensor, so I have to look at the ranges on that. SHouldnt affect my Jeep that massivley anyways if it's only a little off.

 

The TPS is good and within range. The WOT sensor is working. New MAP sensor as well. Checked all my vacuum lines and they are all good and routed correctly. I plugged the brake booster, no change. Plugged the EGR/Charcoal cannister/Purge solenoid vac line @ the purge solenoid, no change.

 

I was worried about the head, and did a compression test, passed perfectly. All of them within 120-130. the timing test also passed great. I pulled the ECU out and there are no indications of heat or corrosion.

 

I'm not sure what else to do. I checked around the intake manifold for a leak in the gasket, but the gasket isnt very old and I didnt see anything. However, I can still do the WD-40 test, spraying around where it bolts up.

 

I havent checked the O2 sensor, but my manual says it doesnt take O2 into consideration until the vehicle is warmed up. I may have a clogged cat because the cat is old, but would that cause effects this major? Another thing is my power steering switch broke off the line and shorted on the exhaust manifold awhile ago that blew the computer, so I got a new one.

 

I cut the switch off, taped up the lines, and it ran great for a few days after that until this started, so I dont think it matters. But I figured I'd include it.

 

I know this was a long post, and I'm sorry, but I'm incredibly frustrated with it, after having all sorts of issues for the past two months. This one I just can't figure out. Any help would be greatly appreciated.

Link to comment
Share on other sites

  • Administrators

Just got back from the Wheelers for the Wounded Rubicon Super Event, Nevada drivers for the first time...I'm just seeing your post now...

 

If you're talking about engine idle speed manifold vacuum this low, you either have a massively plugged exhaust system, retarded ignition timing or the valve time is off.  Compression being normal, I would begin with the ignition distributor.  See if the housing clamp is loose, whether the housing moved, and check for retarded spark.  Use a timing light for the base timing check.

 

If the distributor housing is secure and timing is retarded, suspect retarded valve timing.  This is usually caused by a timing chain that has "jumped".  Is the chain and tensioner new?  Properly installed?

 

Let's start here...

 

Moses

Link to comment
Share on other sites

Thank you for the replies! I wish I found this community a little sooner. but of course it's only after I make a topic that I see some progress finally! Something really odd happened when testing everything today.

 

I had it idling with everything vacuum related plugged up, just to see if it was the intake manifold gasket. It still had almost no vacuum, and was still idling very low. However, it started to run properly, building up plenty of vacuum and everything.

 

So I knew it was most likely not mechanical, but rather electronic. However, it returned to idling poorly and low vacuum after about 2 minutes.

 

I then found that if I let it run until it started to flood, then took off the injector, the vacuum would shoot up, as would the RPMs, going up to about 3000rpms. Meaning the injector signal, because I have tried 2 injectors now, was pouring out gas as aggressively as it could, flooding as soon as it was started, then unplugged and being deprived, back to flooding when I put it back on.

 

so I now know it's an electrical issue with the injector. The engine runs good when plugging/unplugging the injector, and the timing is great. The vacuum system is also fine, I'm pretty sure.

 

Next, I tested the injector wires. 3.7V to one of them, and the other gets 0V, then 12V while the fuel pump is priming, then back down to 0V. And ~14V when it's running. The 3.7V goes directly to the ECU at pin 21. Now to me, that sounds really odd that it would read 3.7V. SHouldnt it be in the range of about 5V?

 

SO I think the issue is either a bad ECU or a short between my ECU pin 21 and the injector. Does this sound correct?

Link to comment
Share on other sites

  • Administrators

When checking these voltages, where was the TPS (throttle position sensor) setting?  Have you checked the calibration on the TPS?  Injector pulse width looks for a TPS signal.  Try testing the TPS voltages, using the factory settings for the various throttle positions.  Let's see if the TPS is functioning properly.  You said earlier that the TPS is "good" and tests within range.  Is that while considering the actual angle of the throttle valve?

 

If you're experiencing a "voltage drop" anywhere, that is as likely to be ground related as hot lead with a 12VDC system.  The ECU completes the injector ground.  If there is resistance or an open in this ground circuit, you would experience a voltage drop or low reading as you hint.

 

Moses

Link to comment
Share on other sites

I tested the TPS sensor attached to the throttle body, with the throttle body off the Jeep. I brought it home and my dad used a 5V source to test it, so I know that's good.

 

What I did today was cut the wire at the ECU pin 21, and the wire at the injector that it leads to, and soldered a wire between the two to rule out a short in the harness. It acted strangely.  It still read 3.7 @ both the ECU and the injector and it was still flooding itself out.

 

but then it started to run properly for a minute, as it had before. it tested 12v at the injector and the ecu.

 

it then started flooding out again while still reading 12v at the injector and computer.

 

I also checked the system ground pin of the ECU, at both ground points (pins 1 and 2) and both of them read next to nothing for voltage, so the grounds are good.

 

Maybe the question I should be asking here is how can I convert this YJ into a carb setup.

Link to comment
Share on other sites

Okay, I retested the injector voltage this morning, because I was tired and aggrevated yesterday when testing it.

 

The wire in question, is a light blue wire running all the way from pin 21 of the ECU to the injector, connecting to nothing in between. So to rule out a short in the wire harness, I cut the wires at each end and spliced in a wire of my own running outside of the Jeep, separate from the rest of the harness. No change.

 

This is what the wire looks like now.

 

[ECU PIN21]---|Splice1|----------------------------|Splice2|---[injector]

 

SO, when the Jeep is off, it reads 0v at all 4 locations.

when it's set to run, it reads 3.7v at all 4 locations

when its idling it reads 12v, but onlyup to splice #2... Somewhere between splice #2, and the injector the reading changes from 12v to ~5.5-6.5V.

 

the first 3 location are easy to test, because I left the splices untaped, but out of reach of anything to short against. to test the injector itself, I had to let the Jeep flood out until it was bogging, unplug the injector, which makes it run perfectly using the gas it is flooding out on for a good 5 or 6 seconds. While it was off I would stick the prong in the injector get a reading of ~6v, then immediately to the second splice next to it, 12v, and keep going back and forth, it didnt change.

 

Im thinking I should try a nice injector connector, but im still curious why it would do that.

Link to comment
Share on other sites

  • Administrators

From what you describe, this is the injector pulse.  The 6V range is normal, as the feed is 12V.  The ECU wants to provide a ground pulse, and the switching of current is approximately 1/2 the 12VDC.  Think of an alternator and its wave form.

 

Do check the connector just to be sure the leads are separated.  Otherwise, the ECU and wiring sound okay now, and you may have "cured" that problem according to the new voltage readings. 

 

You may have a defective injector or regulator setting.  Run pressure tests at the TBI input port with a "T" fitting and port plugs for the various hookups.  If pressure is high, make sure pump pressure is normal and that the return line to the tank is unrestricted for the full volume of returning fuel (not just "I can blow through it").  Narrow the troubleshooting down systematically, making sure the regulator and injector work properly and at correct settings.

 

You can hook a timing light to #1 cylinder spark lead and "watch" the cone spray with the timing light focused on the throttle bore of the injector with the engine running (however briefly before the engine floods).  The pattern should be conical and smooth.  If not, and especially if it looks like a non-atomized, distorted stream from the nozzle, there's an issue here.  The timing light view of the pattern is a quick troubleshooting measure.

 

It sounds like you've made progress with this wire repair, there was a clear voltage drop before the latest readings. Splices should be woven wire strands with rosin core solder and double thicknesses of heat shrink insulation tubing...Good job, keep us posted!

 

Moses

Link to comment
Share on other sites

I ended up re-checking my Coolant sensor and my MAT sensors. They were both open. It turns out, when I checked them, I had rushed it. They read ~4M ohms, but I had thought they read 4k ohms. Retested this time and they were anywhere from ~5.5M ohms to  ~18M ohms. replaced both, leaving them outside the Jeep with the old ones still screwed into the intake. No improvements.

 

I am almost positive that the injector isn't bad because I am now using my old one, and it's working the same as my new one did, I have swapped them a few times.

 

I haven't been able to test the fuel pressure yet, but I will soon, It's been a busy week. Although I checked the fuel pressure regulator by taking it off the throttle body. It seemed fine, no rips and the orange gasket looked good still. Made sure I put it back in correctly. Is it operated by vacuum?

Link to comment
Share on other sites

I didnt test the fuel pressure, but im positive it isnt that. It's gotta be either the wiring harness or the ECU, because high fuel pressure wouldnt cause the jeep to run well for a brief period then return to poor idle. it also would not have gone from good to immediately bad like it did.

 

Im gonna tow it home, pull the wiring harness, pull it apart and start looking for some burned up parts. if I cant find anything, it's time for a new ECU, even though it doesnt look or smell burned at all. Maybe it still is.

Link to comment
Share on other sites

  • Administrators

You're on the right track, 2point5liter.  The fuel pressure regulator diaphragm is spring counterbalance pressure, and you need the correct spring and regulator adjustment.  At these forums, we've discussed aftermarket regulators with "unusual" springs and pressure feedback.  Members often return to the OEM spring to get the right pressure. 

 

In any case, this is confirmed with a pressure gauge at the regulator.  The regulated pressure should range within spec.  See our forum discussions about 2.5L pressure needs and tests.

 

All of this could point to a dying ECU.  I like your methodical approach, though, as the "parts replacement" strategy should only follow known defective parts.  You're systematically eliminating faults and gradually narrowing to the wiring or the ECU.  Get the fuel pressure on spec before moving to the wiring and ECU. If that doesn't do it, wiring next, then a rebuilt ECU.

 

Moses

Link to comment
Share on other sites

  • Administrators

2point5liter...Is your MIL lamp (Engine Check light) glowing?  Any codes thrown here?  Other than an ECU failure, OBD can offer some insight, especially an O2 or other sensor issues.  With the flooding you describe, this sounds bigger than an O2 problem, which at worst would throw the system into limp home mode.  That mode or WOT mode would run rich but not deliver the excess of fuel you're describing here.

 

A repetitive code could be insightful, and that's the reason for storing codes.  The only code that never seems to appear is a failing ECU.  The microprocessor has trouble troubleshooting itself.  Stored codes can be picked up with anything from a simple code reader to a DRB or equivalent scan tool.  This system is not OBD-II, however, so there's a limit to its diagnostic and troubleshooting capabilities.

 

Moses   

Link to comment
Share on other sites

I dont think I have a check engine light on my 89.

 

I do have the diagnostics terminals though, and have been using them to help troubleshoot.

 

I checked the fuel pressure and it was good, so the next thing I did was pulled the wiring harness out. I can't see much that's bad, but I can see that someone else already had this thing out, and used duct tape instead of electrical tape in all the spots where multiple wires connect. Unless it came like that from the factory? That would be awful.

 

Anyways, im currently trying to organize this mess, using nails and rubber bands with it laid out on a large piece of plywood. So many connectors are running to the diagnostic terminals I can barely even see what's going on.

Link to comment
Share on other sites

  • Administrators

Your motivation and thoroughness is highly commendable, 2point5liter!  What an undertaking, you obviously like electrical work.  Some would consider this a deal breaker, you're taking it all in stride.

 

So, the wiring restoration and then the ECU itself?  There's an outfit at Idaho (Module Masters)  that does module rebuilding and restoration, perhaps they can tackle a fuel-and-spark management computer.  There may be other sources for rebuilt ECUs.  I'd like to know what you do with the ECU if that turns out to be the trouble spot. 

 

Keep the grounds in mind, this is D.C.!  On project vehicles, I upgrade all engine-to-chassis-to-body grounds with 1/0 cable and industrial terminal connections.  Grounds: the other half of a D.C. circuit.  Good connections, no corrosion "wicking" and no paint barriers!

 

Moses

Link to comment
Share on other sites

Thank you! I love this Jeep, It was my first vehicle, bought it barely running for $850. It makes me mad at times, but I wanna see it up and running again.

 

I'll keep that in mind with the grounds, because something is going wrong here. I forget if I mentioned it or not, but this will be my 4th computer actually. I get them from Advance, remanufactured from a company called "Borg Warner" I think.

 

the issue the first time was my Power Steering Pressure Switch broke off and melted against the exhaust manifold, causing a short. it blew my ECU, the wire leading to my fusible link, and the fuel pump relay. replaced them all, before I knew that was the issue, and they blew again. Found the issue, fixed it, replaced everything, and it ran for 3 days, until the ECU itself went bad. That was a cardone reman, and when I opened it up, I was appalled at what they said had "passed". blobs of solder connecting things that werent supposed to be touching, and white powder corrosion under the board. It was an entirely new issue too, so I knew it was just the ECU that went bad.

 

Now I had this ECU in, from Borg Warner, and it ran for about a day, started cutting out, and got worse and worse, and now it just refuses to run without flooding itself out. 

 

That's why I was so skeptical of the wire harness, maybe blowing the ECUs, but it seems to be perfectly fine. This ECU looks perfect, doesnt look or smell burned at any points. But neither does the wire harness, and everything seems to be routed correctly, even if it is super messy.

 

Ill keep that in mind with the grounds and give that a shot. if it runs the same after I get this harness all cleaned up and reinstalled, and the grounds redone, Ill go for a new ECU.

Link to comment
Share on other sites

  • Administrators

The aftermarket rebuild industry sounds like it's let you down...Mopar has its own Reman system, and perhaps the product would be better than Borg-Warner or Cardone.  Borg-Warner was a quality supplier in the day.

 

The OEM Mopar part number is 53005023.  The Mopar Reman (if still available) is JR005023.  My Mopar Reman catalog is 2010, and this is the listing for 1986-90 2.5L TBI with manual transmission.  These units, as you know, are Fenix types.  Check price and availability on this Mopar unit.  Also check the part number on your existing ECU.

 

Moses

Link to comment
Share on other sites

Well, I checked everything in the harness. It all checked out in my continuity tests, meaning, I got sold ANOTHER bad ECU.

 

I wish I had a friend with the TBI 2.5L wrangler so I could check and make 100% sure, but hopefully Advance Auto fulfills their 1 year warranty after I told them I had to do the same with AutoZone when I picked this one up. If not, I'll use that part number you gave me and find a more reliable source. Thanks!

 

Ill check back in after I get my vacuum lines back on, my harness cleaned up and installed, and my new ECU in and installed. Looks like I still have a lot of work ahead of me.  

Link to comment
Share on other sites

  • Administrators

2point5liter...I'm an advocate of lamp load testing, especially ground circuits.  Continuity tests can be deceiving.  A friend recently wrote me with an issue related to his Dodge Ram Cummins truck with 370K miles on the clock.  I'll share my comments because your hunt for "continuity" may not be enough on these wiring circuits, especially ground related.  Here are my suggestions and comments about ECU troubles:

 

"…A great test for wiring circuits is the “lamp load test”.  This is a simple test involving a spare headlamp, wire leads, an inline fuse and clip or terminal ends.  The test indicates not just continuity but also the voltage drop in a wiring circuit.  I learned this trick from Eric B., the Carson City Dodge dealership specialist on Cummins engines.  We had an issue at 800 miles with our new ’05 Ram 3500, the engine dropped cylinders and threw an ECU code for a #1 cylinder misfire.  Eric B., dutifully following the orders of Chrysler engineering and his service manager, swapped and moved injectors around, did all of the “#1 cylinder misfire” troubleshooting and repair measures—to no avail.  Other cylinders began throwing “misfire” codes, and Eric suspected the ECU itself. Chrysler said, “Impossible!  If defective, the engine management computer [ECU] would self-interrogate and throw a defect code!”  There were 25 new Dodge Ram trucks in the region (West Coast to Denver) with the same issue, and Eric persisted.  Finally, bending to Eric’s stature and track record, Chrysler approved an ECU replacement on our truck, the first such repair authorized in the zone.  As suspected, the problem ended with a Cummins “Recon” ECU in place.  At 130K miles, there have been no subsequent issues, and last year, Chrysler sent a VIN driven notice to owners of ‘05/’06 Cummins models, extending ECU replacement to “lifetime” warranty coverage...In the troubleshooting process, Eric analyzed the injector ground circuit with his lamp load test, including injector firing.  Injectors fire from an ECU ground signal.  (The injector is “hot” at all times and waiting for the ground signal to open the pintle.)  Eric shared that the lamp load test was the fastest, most accurate way to find an open or high resistance point in a wiring circuit.  Makes perfect sense, this is D.C., and grounds are as critical as the hot leads.  Eric ran lamp load tests on both positive and negative circuits to quickly isolate any high resistance points (caused by a short, open or loose connection).  He found none…If you find yourself wanting or needing to make this electrical repair, the dimming of the test lamp indicates a short in the circuit...Keep in mind that grounds on a D.C. system are important, they carry as much load as the hot leads."

 

Does this note to my Cummins Dodge friend resonate, 2point5liter?  A short or open could even be related to that power steering switch short to ground that you experienced some time back...Continuity is only part of the story.  A voltage drop could be the rest.

 

Moses

Link to comment
Share on other sites

Hmm...really interesting. I'd hate to have to pull the harness out again! but if I need to at any point in the future that sounds like it'd be really helpful, thank you!

 

Advance honored their 1yr warranty, so I got a new ECU in, and it's running great, idling as it should and adjusting as needed to keep a nice idle.

 

I haven't gotten a real ride out of it yet though, waiting until I have some backup to call if I end up broken down on the side of the road, until AAA can get to me. However, when putting the harness back in to test the new ECU, I noticed something I hadn't noticed before. The relay connectors.

 

I had been so focused on the actual wires that I didnt even look at the connectors. the black boxes the relays get plugged into were all melted, and the terminals had some kind of goop on them. I swapped new ones in so HOPEFULLY if I did have an issue, that was it.

 

Thank you for your help!

Link to comment
Share on other sites

  • Administrators

Good news, 2point5liter!  Sounds like defective relay plug sockets and/or the ECU.  Waiting for your test drive update and a successful ending...

 

As for voltage drops and lamp load testing, you should be able to perform this test without removing the harness.  You'd leave the harness attached at the power source.  Disconnect the harness connector at the powered device and "jumper" the lamp between the connector and the device.  (A ground check would be similar.)  Check for a voltage drop or dim lamp when powered up.

 

If your current work solves the problem, lamp load testing can be a tool for down the road!

 

Moses

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.



×
×
  • Create New...