1990 Jeep YJ 2.5 TBI Stalling during warmup

[madsquirrel]

I recently bought a 1990 Jeep YJ with 4 cyl 2.5 TBI engine.  The jeep runs great except during warm-up.  It starts fine and runs and idles well initially.  As soon as the temperature gauge begins to climb off the 100 degree mark, the engine begins hunting and surging and usually stalls.  If I push down on the accelerator and run it to 1000+ rpm, I can get the engine to heat up to operating temperature.  At that point (~160-170F) the idle becomes stable again.  So far I haven't done a lot of troubleshooting but have checked the EGR valve, changed rotor/cap, air filter, and fuel filter with no success.  I'm wondering if it might be a bad O2 sensor that's slow to heat up which will be the next thing I check/replace.  Any ideas or help is much appreciated!

[Moses Ludel]

Hi, madsquirrel...I would check the O2 sensor as a matter of course.  Also check the ohms resistance at the coolant temperature sensor (CTS).  The sensor is the signal to the ECU that the engine has actually warmed enough.  If out of sync with the real coolant temperature, your engine will come off its cold/warm-up enrichment phase too soon.  You should have enrichment until approximately 140 degrees F.

You've ruled out the EGR valve although I would confirm the temperature that the EGR becomes operational.  It should not function with the engine cold, and the EGR will respond in the same way that your warm-up cycle does.  Use tip-in throttle to actuate the EGR.  Make sure the valve closes with the throttle at idle position.

Let us know what you find...We can go from there.

Moses

[madsquirrel]

Moses, Thanks a lot for your reply.  I'm starting to think this may be a much bigger problem than I first thought.  I went out this morning to do a quick resistance check on the CTS. All the pictures and diagrams I've seen show it located on the thermostat housing.  I attached a picture of what mine looks like.  As I started crawling around the engine compartment I found  several vacuum lines disconnected.  At first I was thinking I had found the source of the issue, but after looking further it appears they were intentionally removed....hoses completely missing vs. just broken and hanging.  I'm starting to wonder if a previous owner may have bypassed a bunch of components.  Something that's really strange to me is that I haven't even been able to locate the CTS connector or see if/where it was cut off the wiring harness.  I only had a few minutes this morning to look so in the next couple days I'll really dig in to see what I can find.  I'm thinking a scan tool would be very helpful right now and it appears the Snap on MT-2500 might be a reasonably priced option.  Any thoughts would be very welcome right now....

[Moses Ludel]

madsquirrel...You now have a reason for why the cold start/warm-up is out of whack. That's why I suggested you check out the CTS.  In this case, there is no CTS function at all.

To help restore the system, I have provided a wiring diagram (below) of the CTS to ECU on your '90 Jeep YJ Wrangler.  The wire color coding should help you find the CTS harness.  I also added details on the TBI sensors and their ECU pinout:

Jeep YJ 2.5L TBI CTS Wiring.pdf

Jeep YJ 2.5L TBI Sensors and Pinout.pdf

Moses

 

 

[madsquirrel]

So, after my last update I felt the problem could be a lot of different things.  What I really wanted was an ECU code reader that could help me at least rule out some of them.  Unfortunately, the ’90 was before OBD II and it required a special reader.  I did a little more digging and found some information about the ECU serial format.  I didn’t want to buy a used and rather expensive single-purpose ECU reader so decided put my electronics geek-ness into action and build my own, but I couldn’t get started on that until I finished another electronics project.  So right after Christmas I got working on a code reader and had some success (and learned a lot about how the ECU & controls works).

Here’s the good news.  It looks like the MAP, Coolant Temp, Manifold Temp, Throttle Position, and engine speed sensor readings are good.  Spark advance appears to be working as I’d expect.  The ECU seems to recognize the proper throttle positions: closed, partial, wide open.  The ECU control indication will switch between open loop, closed loop, and decel modes.

So I attached some data and here’s what I see.   At startup the ECU runs open loop as expected.  The O2 sensor always says its full scale (~1000mV for 2.5L) at startup.  The fuel trim is at 128 which is center scale (lower values means fuel is subtracted, and higher means fuel is added over base). Like clockwork, as soon as the coolant temp hits 104F the ECU goes into closed loop mode.  In test #1 in the data, the O2 starts high when going into closed loop.,  The fuel trim drives low to bring the O2 down.  It appears to give up after 7 seconds and go back to open loop. 

In test #2, O2 starts high but then goes low as soon as closed loop starts (strange).  The ECU brings up the fuel trim to richen it but gives up after a few seconds. 

In test #3, the O2 is low (zero) when it goes into closed loop.  The fuel trim is increased as expected and the O2 appears to start to respond, but the ECU drops back after 6 seconds.  Test #4 & #5 are the same. 

So in this data, the O2 sensor seems to go low and stay there and the Jeep runs mostly in open loop.  It runs pretty good by the way.  I have some other data runs where it’s the opposite.  The ECU goes through warmup and then the O2 sensor reads max scale most of the time. When the ECU goes into closed loop, it drops the fuel trim as expected and the engine dies, which was the problem I was trying to solve in the first place.

So at this point it appears that there’s a problem somewhere causing the engine to not be able to run closed loop.  I’m guessing it’s the O2 sensor so I’ll try replacing it.....I'm just thrown by the fact that the sensor seems to read sometimes.  I'd expect it to fail and always be at max or min scale.  Any other suggestions??

At least with the ECU reader I have a diagnostic tool that I didn’t have when I started!

Log pm 1-23 r2.pdf

[Moses Ludel]

madsquirrel...This is really commendable!  You made a highly accurate and detailed reader that appears to work well.   I went over your data in the PDF download, very interesting and detailed...I'll throw some ideas your way...

I would question the ISA motor function and the O2 sensor.  I'm not a fan of "parts replacing" for testing purposes but would comment that O2 sensors are not all the same.  Check your O2 sensor, share the type and part number, I'll do some research here.  As for the ISA motor, try to explore this further, you may have a motor or drive that is simply non-responsive to the sensitive transition between cold to partial warm (your temp sensor setting is low here) and then warm.  On that note, consider a temp sensor that would reach a higher temp before triggering the ECU to go to closed loop and lean out.  Fuel pressure (regulated) could play a role here, too, if pressure is low, that could cause stumble and dying in general.

You're describing a weakness or stumble in the transition between enrichment mode and a closed loop, warmed engine.  Focus on that transition, it sounds like the engine could simply be too cool to handle closed loop that early.  Your data acquisition is remarkably good, so you can quickly see whether any modifications or parts changes have a useful effect.

By all means, keep going!  Could simply be a bad or out of sync O2 sensor.  This is a heated sensor, so I would take into consideration the wiring and voltage to the sensor heat element, too.  Also make sure that the EGR is not opening slightly due to throttle valve position, an incorrect vacuum source or a leaky (carbon clogged) EGR plunger.  Check your EGR hose routing and any one-way check valves.  Consider vacuum leaks like the transfer case fittings or disconnect actuator motor at the front axle...A lot of stuff, but you can handle it!

Moses

[madsquirrel]

Moses, Thanks for the input!  I hadn't considered the ISA motor since idle RPM seemed okay, but I'll do some checking.  There's also some ISA signal outputs on the D2 connector (where I'm hooking into the ECU).  I may try grabbing and logging the extend/retract signals...there are a few bytes in the data stream that are unknown which may contain this info.  On the O2 sensor, I can't tell much about the brand or how long it's been in there. The engine has 147k on it and it may well be the original.  Do you have any that you'd recommend or stay away from?   On the EGR, that's a good idea.  Could I simply plug (temporarily) the engine vacuum at the EGR/Purge solenoid to see if there's  a performance difference as a starting point?  I'll look into the other items as well.  Thanks!!!

John

[Moses Ludel]

madsquirrel...For O2 sensors, I've learned to use the OE manufacturer's brand.   Not necessarily Mopar, as they have virtually abandoned the 2.5L TBI Jeep parts.  I'm suggesting the O2 supplier to Mopar.  On our '99 XJ Cherokee 4.0L, it was NTK (i.e., NGK of another era).  I took the part number directly from the OE sensor and bought an NTK replacement via Amazon.  You can take a similar approach.   

Do yourself a favor and avoid the Brand-X sensors, I bought one from a retail parts house, historically a very good product line part whose name I won't mention, but these mass market part numbers have too many "similar" applications.  These are sensor that physically fit and have readings somewhat close to your application–but not right on.  I learned not to waste the time and money.  Determine the OE brand, trace down the part number, and get that specific O2 sensor. 

The EGR valve test you describe would be a quick and general test.  I like to take a gloved finger (wear a mechanic's glove to prevent burning your hand and fingers) and actually pull up on the EGR's diaphragm to unseat the plunger with the engine idling.  There should be a distinct change in rpm and a roughness to the idle.  If this does not occur, the valve is likely stuck open or the intake/exhaust ports are plugged.

Please update, I like your thoroughness and am curious how your "new" reader follows these troubleshooting steps and parts changes.  If you need help with an O2 sensor part number, let me know...

Moses