Jeep 4.2L with Mopar EFI Conversion—runs well but NOX is too high

[CJN8]

 

 Vehicle background:

 

1983 Jeep CJ7 Laredo 

4.2 with Hesco MPI kit 

 

Westside Performance Valve Cover (thick cast aluminum replica of the stock cover)

 

T-18 four speed

 

Dana 300

 

Dana 44 axles front and rear (4.27 to 1)

 

36" Swamper TSLs

 

I bought this Jeep in 2003 odometer had 145,000 miles, now has 170,000, it does not get driven all that often.  Plaque on the motor reads "MOPAR Performance", it is a mopar 4.2 crate motor with a remanufacture date of 1998.  My estimate is that the motor has ~80,000 miles on it.   

Issue:

Recently failed low speed NOX during my last CA smog test, measured 1,322 and max is 1,066 (fail).  High speed NOX readings were measured at 948 and max is 968 (pass).  For the last 3 or 4 tests high NOX have been an issue.  Before I went to test my jeep I changed the oil, it was a quart low.  I removed and cleaned the plugs.  The fourth plug from the front had quite a bit of carbon buildup.  

After the test i decided to do a compression test.  Got the Jeep up to operating temp and did my first test dry.  Readings from front of motor (water pump) to back were 121, 128, 123, 137, 124, 122.  The fourth cylinder had a compression reading about 10% greater than the others.  This was the same cylinder that the spark plug with the carbon buildup came from.  Next I performed the same test but this time adding two squirts of oil to each cylinder before the compression test.  My readings were within 1-2 psi of my dry test.  My diagnosis is that the cylinder head needs to be rebuilt.  The cylinder head is worn letting excessive blow by enter the combustion chamber through the CCV valve which vents to the number 4 cylinder where the oil is burned.  This is why i was a quart low when I changed the oil and why cylinder 4's spark plug was all coked up.  Does this sound reasonable?  Can i rule out rings in this case?  What do you think?

Thanks for the help!! 

 

 

[stearnsm]

I'm afraid I disagree.  The cylinder head has no impact on blow by... blow by is compression leaking past the piston rings on the compression stroke.  The crankcase ventilation valve vents this pressure back into the intake. 

A bad valve guide can cause you to use oil, as the oil goes down the valve stem into the combustion chamber.  So a bad valve guide on the number 4 cylinder could be dumping oil into the cylinder and causing the plug to foul. 

Check the CCV valve and hose.  If there is an excessive amount of oil present then my bet is its the piston rings.  You can also pull the hose off the CCV and plug it with your finger while it runs, there should be almost no pressure against your finger.  If you can feel pressure or if it spits oil out then it needs rings.  Ironically, that could be why compression on #4 is so much higher, do to the extra lubrication.

[CJN8]

Thanks for taking the time to read my post and offering up some good food for thought.  I guess my next question is what is the factory spec for compression on the 258?  I have read 120-150 does this sound correct?  

[Moses Ludel]

Thanks for jumping in, stearnsm...CJN8, I really like your equipment.  As for the #4 cylinder higher compression, I go with stearnsm's suggestion to verify whether there are signs of oil entering that cylinder via the CCV and hose.  You definitely have more compression here, indicating possible coking of the combustion chamber and piston crown at #4.  None of the cylinders are reading particularly high, though each cylinder is adequate.  Optimally, they could be slightly more uniform.  10% variance between the highest and lowest cylinder is acceptable, the more uniform the better.  I'd like to know the idle manifold vacuum and base spark timing degrees.

First-off, I am very skeptical of cranking compression tests, there are many variables that impact the results.  Without going into the laundry list of read errors possible and their causes, I would move from compression gauge testing to a cylinder leak down test.  A leak down tester is far less expensive than a speculative engine tear down.

A leak down test can pinpoint ring seal issues, valve seating problems (specifically which valve if so), head gasket or casting seepage, or even cylinder taper issues if you use the gauge to advantage.  At these forums and the magazine, I cover the use of a leak down tester—and even how to make one inexpensively.  At the forums search box, enter "cylinder leak down" for a variety of situations where this tool is useful for pinpoint diagnostics.

Here's a key link at our forums...Let us know your findings, we'll get down to the bottom of your 258/4.2L troubles:  

 

[CJN8]

Thanks for the reply.  Yes, there is evidence of oil entering the combustion chamber through the CCV (which is in line with cylinder 4).  This is why I built a crossover breather tube for the valve cover. The CCV was too close to the baffle in the valve cover and it would pull oil into the manifold when I drove down hill and rolled off the throttle.  I verified this by putting a air/oil separator in the rear CCV to manifold vacuum line. Lee over at hesco recommend building the crossover tube so I did.  How do I check the manifold vacuum and base timing degrees? 

Thanks! Nate

[Moses Ludel]

CJN8...A manifold vacuum source is very easy to find.  This would be vacuum sourced at the intake manifold between the throttle body and the cylinder head.  Simply put, there are vacuum ports and devices on the intake manifold that will provide a manifold vacuum source and reading.  

Note: Above the throttle valve in the throttle body (or a throttle valve in a carburetor) is "ported" vacuum.  This is not a manifold vacuum source.

I sent an incomplete message when suggesting that you check the "base timing".  Your MPI conversion kit uses a remote crankshaft position sensor that attaches to the oil pan bolt holes at the timing  cover.  Base timing on the MPI conversion kit follows the same guidelines as a 1994-95 4.0L MPI inline six with a factory "60-way" PCM.  All spark timing measures are controlled by the PCM, including the relative base timing.  Advancing and retarding the base timing is possible at the PCM, within a small and emissions legal range, via the use of a Mopar DRB-III scan tool or aftermarket scanners that apply Chrysler's 1994-95 era OBD-1 software.

On your application, unlike the 4.0L Jeep MPI engines, the distributor housing base is not pinned to fit in a specific position.  However, like the factory 4.0L MPI distributor, the distributor base is not a place to "adjust" the base spark timing.  The aim with positioning the distributor is simply to align the distributor shaft and rotor properly with the #1 spark plug socket in the distributor cap.  Once you make this alignment (placing the crankshaft damper mark at TDC or #1 piston at TDC on its compression stroke to be accurate), the rest of the spark timing should be controlled electronically by the PCM. 

Where a problem exists with the conversion kit is that the bolt-on crankshaft position sensor (CPS) can be manipulated in such a way as to alter the base spark timing.  If the CPS does not send a pulse at the same time the #1 piston reaches precise TDC on its compression stroke, the base timing will be either retarded or advanced, depending upon the location and mounting point of the CPS pickup.  Note:  On a stock 4.0L engine, the CPS mounts in a fixed position at the bellhousing/converter housing and picks up a signal from the flywheel or converter flex-plate.

On MPI conversions with the damper signal and bolt-on CPS at the front of the engine, I confirm that the CPS position is sending an accurate signal to the PCM at the TDC point for #1 piston.  An ohmmeter can check for continuity at the TDC position on the crankshaft damper, which should be the actual TDC for #1 piston.  Disconnect the connector/leads from the CPS switch/sensor, insert ohmmeter probes into the leads at the connector, and jockey the bracket until full continuity exists at the precise TDC point.  The signal/pulse is magnetic.  For proper function in service, the gap between the sensor pickup and the Mopar pulley damper (included with the MPI conversion kit) must be set accurately.

Once you have the CPS set accurately to #1 TDC, the PCM should adjust spark timing as necessary.  You can check base timing with a timing light, engine idling with no engine load, just to get an approximate idea.  You'll find that timing varies widely under different load, temperature and atmospheric (MAP) conditions.

Footnote:  It's likely your timing base is accurate.  Now that you're modifying the CCV to reduce oil sucking, try to remove carbon buildup at #4 cylinder and bring compression balance to norms.  This will reduce the risk of excessive heat at #4 combustion chamber, which leads to higher NOX readings.  Combustion heat increases NOX.  Once you restore compression to a reasonable match and reduce local combustion overheat at the same time, the NOX readings should come into compliance.  You should be running a catalytic converter without the Pulse Air input.  If an OEM style cat, the inlet tube from Pulse Air should be pinched and sealed.  (I close and weld the tube end to ensure no leaks.)  You do not need or require Pulse Air with the Mopar MPI Conversion Kit on a 4.2L engine.  (Confirm with your California smog or referee station if concerned here.)  For the cat to work properly and correct tailpipe readings, it cannot have leaks to ambient air.

Moses

[CJN8]

Status update. I removed the head and as thought had a considerable buildup of carbon across all the cylinders. I had the head fully reconditioned, reworked the baffles on my valve over, installed new thermostat, fully cleaned the intake manifold that was coated in motor oil.  Surprisingly the cylinders looked to be in good shape, some cross hatch visable with no ridge.  The motor had .060 Pistons installed so for sure this motor was rebuilt.  One of the last items replaced was the water pump. The strange thing was it had a reverse rotation pump installed on a V belt motor?? I replaced with a standard rotation pump. I have not fired the motor, plan to do that this weekend. Any words of wisdom to provide prior to firing either the new head?

[Moses Ludel]

CJN8...Sounds like you did a thorough cylinder head rework.  Smart!

From your description and experience with MPI and the OEM 60-way PCM/sensor feedback EFI that comes with the MPI conversion, carbon buildup is generally not extreme.  If anything, the carbon should be less than with OEM carburetion and the stock/original intake.  I'm guessing that you're running a catalytic converter with the Pulse-Air inlet tube to the cat pinched off/welded?  Pulse Air is not required with the Mopar MPI conversion.

I read back through your posts and am puzzled about the "Mopar Performance" (not Mopar Reman) label on the engine.  This raises a concern about the camshaft grind and your issue with tailpipe emissions.  While you have the valve cover off, a quick check of valve lift at the pushrod side of the rocker arms would be warranted.  You could do a valve opening duration check, too, if you have a degree wheel and want to go that far.  (Duration is more difficult to check with the timing cover in place, and valve lift alone should indicate whether this is a stock profile camshaft or not.)  

What you want to confirm is whether the camshaft is a performance grind with extra valve overlap.  If so, there would be a risk to NOX output (especially at low engine speeds) and a possible explanation for the carbon buildup.  Normally, the PCM and oxygen sensor feedback would make the engine less susceptible to carbon buildup than with a carburetor like the Carter BBD and its emissions equipment...It's fairly easy to get a valve lift reading with a dial indicator, and any performance camshaft grind would have more valve lift (intake and exhaust) than stock.  Assume that valve overlap would be altered, too.

Good that you corrected the water pump situation.  Starting the engine should not be an issue if you did not remove the valve lifters.  Oil should pick up quickly, though I would definitely pour some fresh motor oil over the rocker arms, pivots and pushrods before starting the engine.  It will take a brief spin to bring oil to the rockers.  Priming would be advisable if you rebuilt the engine or changed the oil pump, but that would not be necessary with just a cylinder head removal and re-installation.

If you used a FelPro (blue) head gasket, a re-torque of the cylinder head will not be necessary if you have set the torque carefully.  Some other aftermarket head gaskets also do not require re-torque.  See the recommendation for the head gasket used.  "In-the-day" type gaskets (shim or composition with gasket sealer applied) do require a re-torque after bringing the engine to full temperature and cooling it down.

That's what comes to mind...Let us know what you discover.

Moses

[CJN8]

The reason i had excessive carbon buildup was due to the baffle design of the Westside Performance valve cover.  The baffle was essentially a cup with small slits in the bottom.  Not good, the oil would collect in the cup and when it got substantial enough the CCV would take a big gulp and that was that.  All that oil would run through the motor = carbon buildup.  I verified this buy running it with an inline air/oil separator.  This is how I corrected it.

Removed the original baffle

Made a proper baffle

Made a valve cover crossover tube to move the CCV further from the original location

  

The Cat I am running has the air injector tube capped.  

I have fired the motor and it runs great, even took it for a test drive and no major issues.  The only issue is the radiator is now leaking around where the filler neck is soldered on.  I was thinking I could just have it re-soldered but maybe it would be best to replace this 33 year old radiator.  The fins are a mess and it looks like one of the tubes was fixed in the past.  Any recommendation regarding where to order a new radiator?  I am partial to the brass/copper HD OEM rads.  

[Moses Ludel]

CJN8...Innovative fix with the CCV tube and baffle plates, should be resistant to oil splash from the valvetrain.  Is there enough low-pressure static (i.e., vacuum) to "suck" oil around the baffles?  Do the baffles have screens?  Is there a need for perforated spacers or screens between the plates and cover?  The water/oil separator should signal any oil pickup, right?  

I can see that you used J-B Weld for securing the nuts, as these plates and fasteners must not come loose.  Does J-B-Weld work well with this heat and oil exposure?  I'd like to know your experience with J-B Weld.  My first impulse would be TIG tack welds.  What's the strategy here?  An alternative might be hex headed bolts of the right length, heads tack welded to the plates where you have the nuts now.  These bolts/studs would run upward through the valve cover and could be captured with steel toplock nuts.  Tack welds holding the hex heads to the plates could be done with MIG, TIG or even by oxy-acetylene brazing...

I prefer the OE brass copper core radiators, too.  It might pay to find a traditional radiator shop and have them install a brass replacement core in your original radiator framework.  The frames hold up fairly well and can be restored if damaged slightly.  Talk to a local, quality shop.  The aftermarket radiators are mostly aluminum cores with plastic tanks unless you step up for an expensive "high performance" radiator, usually made from aluminum.

Moses

[CJN8]

Radiator fixed!  I had it cleaned, pressure tested, and leaks soldered.  I went for a test yesterday and I continue to fail high speed NOX, roughly about 1400 PPM with a max passing value of 1066 PPM (~30% over passing value).  I am very happy with how the motor runs after freshening it up.  Other than the valve cover splash guards that I installed I have no other baffles, I am getting zero oil in my clear catch can.  After I put the splash guards in I put red loctite on the nuts, I decided to JB weld the nut for added safety.  In hindsight I should have pulled out my MIG welder and put a few small tacks on the nuts.    

I can rule out?:
- Carbon deposits, head is new and piston tops were cleaned
- Cooling system, radiator cleaned and serviced, new water pump, new 195 degree thermostat, new hoses, new coolant (runs in the middle to middle lower green range on the gauge)
- Fuel pressure, 34 psi at key on, 30 psi at idle, 40 psi when i remove the vacuum line

Possible lean condition?
- Injectors?
- Map Sensor?
- Higher fuel pressure?
- O2? (I replaced it in 2009 but did not use a mopar factory sensor, heard that some O2 sensors do not switch fast enough to control lean conditions)
- Catalytic converter? (has a carsound unit that I purchased in late 2008, this was before you had to have a CARB stamped converter, maybe not strong enough to handle NOX?, this converter has never really produced good NOX numbers)
- ??

I know that the jeep is able to produce low NOX readings, in 2004 it blew 46 PPM at 15 MPH and 28 PPM at 25 MPH with 35" tires on.  

[Moses Ludel]

CJN8...I like your deductions and narrowing the prospects!  Unless there is a symptom of poor combustion or misfire, you would likely not have a fuel flow issue.  (Read the spark plugs here.)  

Have you run a cylinder leakdown test?  I'm suspect of cranking compression tests, as the moving pistons can "stuff" volume into the combustion chamber, whether the rings are sealing or not.  Is there cylinder taper?  

Footnote:  I worked on a Toyota 4-cylinder engine years ago with near optimal cranking compression and 75%-plus cylinder leakage due to taper and ring wear; the valves were sealing fine.  Lesson learned...Click here to see my forum post on use of a cylinder leakdown test:

A leak down test is revealing.  It takes into account static parts that cannot force a false reading, especially ring seal.  You can also narrow down the exact point of compression loss.  Keep in mind that cylinders fire under pressure and build even more pressure in the power stroke.  I'd run a leakdown test, if nothing else than to rule out any pressure losses and to comprehend the engine's sealing ability.  Pass the test?  Great!

The O2 sensor and cat could easily be faulting here.  I ran into the OEM versus aftermarket differences with both of these items on our 4.0L (stock) MPI engine.  I would at least use the OEM (outsourced from NTK/Denso) replacement sensor.  Performance cats are always on the margins for passing emissions.  I'm happy with my second Random Technology cat (see details at the magazine site: http://www.4wdmechanix.com/How-to-Installing-a-Random-Technology-High-Performance-Catalytic-Converter?r=1; the first one lasted 20K miles, this one has hung in there three times that length/mileage.  Either of these devices can create your current emissions issue.

Another issue can be late valve timing.  Simple test:  Check manifold vacuum readings at idle and as engine speed increases.  Pull the distributor cap and turn the engine in normal rotation to TDC point on the damper.  Then back up the crankshaft just to the point that the distributor rotor begins to turn.  The number of degrees approximates timing chain play and sprocket wear.  (Maximum 5/8" of damper movement is about the limit.)  Also confirm ignition spark timing with a timing light and the engine idling.  Despite EFI and electronic spark control, the base timing could be off if the aftermarket EFI crankshaft position sensor is not positioned properly.  This could affect emissions output.

Moses

[CJN8]

Tested the MAP sensor today.  Does this explain my lean condition??

My Readings:

0" vacuum = 4.77 v (1.5% greater than below)
5" vacuum = 3.23 v (17.2% less than below)
10" vacuum = 2.38 v (20.6% less than below)
20" vacuum = .75 v (31.8% less than below)

Diagnostic info from an article in Four Wheeler Magazine:

0" vacuum output should be 4.7 volts.
5" vacuum output should be 3.9 volts
10" vacuum output should be 3.0 volts
20" vacuum output should be 1.1 volts

More voltage equals a richer condition, less is leaner. Even a small variance (0.2 volts) can make a big difference in air/fuel ratio and drivability.

 

http://www.fourwheeler.com/how-to/engine/1401-driveway-diagnostics-map-sensor/

[Moses Ludel]

CJN8...Certainly worth replacing the MAP if this data is correct for your engine...The MAP that comes with the HESCO EFI kit should be either 1995 YJ 4.0L or 1997 TJ 4.0L, depending upon dual-rail (early) versus single rail (later) EFI type.  The earlier MPI/EFI kits were two-rail; later is 1997-up pattern single rail 4.0L TJ.  All EFI Mopar kits use the 60-way/pin PCM.

Nice sleuthing...Confirm the data for your EFI type.  Confirm accuracy of your vacuum gauge and voltmeter.  What diagnostic tools are you using?

Moses

[CJN8]

Thanks for all the help Moses.  I ran a manifold vacuum test and am getting 16" with a steady needle.  Seems this is pointing to a timing issue?  I have not run a leak down test yet but it is in my plans.  

The MAP test was performed using a hand vacuum pump and digital multimeter.  Could a lean running MAP effect manifold vacuum?  I have read that lean conditions and low normal manifold vacuum go hand in hand, true?

Thanks again,

Nate

[Moses Ludel]

CJN8...Spark timing degrees of advance can affect manifold vacuum.  Advancing the spark timing will increase manifold vacuum up to the point that timing becomes excessive.  Retarding the spark timing will drop manifold vacuum.  In each case, test spark timing and vacuum with the engine at an idle and unloaded.  This will maximize the manifold vacuum.

There is a relationship between MAP, spark timing and manifold vacuum.  Retarded valve timing will also lower manifold vacuum.  This impacts the available vacuum reading for the MAP.  There is also a unique phenomenon:  When a (valve/camshaft) timing chain is too loose, the low-speed engine power suffers dramatically.  This is from poor combustion and compression loss due to the late valve timing.  This same engine may have a stronger pull at higher rpm, despite the retarded valve timing and reflective of this late valve timing.  Valve timing and spark timing each can impact the manifold vacuum.

Late valve timing can be the result of a loose or worn timing chain and sprockets.  Ignition timing is referenced to the crankshaft, so it can be adjusted with regard to the #1 piston's position in the cylinder.  

With the Mopar MPI/EFI kit, your crankshaft position sensor (CPS) must index properly with the crankshaft pulley.  This sets the base timing correctly.  All spark timing is controlled by the PCM—not the distributor, which is simply indexed in position at the base timing point.  Once you index the CPS properly, the base spark timing and timing under all operating conditions reflect the PCM signals.  Timing adjusts in milliseconds based upon the engine's sensor input to the PCM.

I'm not suggesting that your engine's valve timing is retarded, but chain wear should be considered and ruled out.  I'm not clear whether the base spark timing is set properly.  This depends upon the indexing of the CPS at the crankshaft damper.  Either of these factors have a bearing on manifold vacuum.

Moses

[CJN8]

Base timing is checked with a timing light, correct?

[Moses Ludel]

Yes...If the timing jumps around, this is not uncommon for an EFI/MPI ignition.  If the engine is not under load and idling stably, you'll get a reasonable sense for the base timing.  This can help clarify whether the crank position sensor (CPS) is indexed properly.

With electronic fuel and spark management (like your PCM), you usually have no means for setting the base timing.  Once the distributor is properly installed and indexed, the PCM takes over the timing functions.  The only provision for setting the timing would be a software re-flash.  Dealerships use a DRB scan tool or equivalent to do this.  (The software change in base timing can provide only 5-degrees adjustment range.)

Your Mopar/HESCO retrofit EFI's wild card is the CPS, which can be moved by repositioning its bracket.  This changes the TDC reference and as a result, the base timing.  The OEM Jeep 4.0L CPS cannot be moved, it fits rigidly at the bellhousing or converter housing.

Moses

[CJN8]

14 degrees base timing, it hunts a little but that is where the timing mark holds the majority of the time.  The slop in the timing chain is about 3 degrees when I rotate to TDC then rotate back until I feel the slack in the chain take up.

[Moses Ludel]

CJN8...I'd give HESCO a call and ask if this is within norms.  It's high at face value, but with the engine "unloaded" at an idle, this could be normal with the PCM controlling the timing.

Please let us know what they say and whether you need to adjust the crankshaft pickup or not.

Moses

[CJN8]

Called hesco, 14 degrees is good. Told me to check the VSS and for a plugged cat as well.

[Moses Ludel]

CJN8...Pleased that you shared this information with us, thanks!  A plugged cat could be the NOX issue.  VSS would likely not change the emissions results since you're not testing emissions while running the vehicle on a dynamometer.  

The CCV issue, with oil burning and exiting the exhaust ports, could create a catalytic converter issue.  A 3-way cat will clean up HC, CO and NOx.  

One problem if you have the OEM converter would be inadequate capping off of the Pulse-Air inlet tube at the catalytic converter.  Do you have an OEM type '83 cat with the tube, and if so, is the tube pinched off?  The Mopar/HESCO EFI conversion, like '91-up Mopar stock MPI 4.0L induction, eliminates the need for Pulse-Air.

I pinch the cat's Pulse-Air inlet tube end snugly and weld a bead across the pinch closure.  That makes the cat air-tight and prevents dilution.  A leak here would also create toxic fumes.

Moses

[buzz54]

Hey CJN8, I also have my jeep with the Mopar MPI and was curious on how you did your setup on the valve cover with the hoses...very clean look.  Mine is messy and was wanting to do something similar with the hoses that connect to the valve cover/intake etc...attached is what mine looks like with stock valve cover.  very ugly and messy

[CJN8]

It has been a while since i updated the status of my NOX issue.  I finally broke down and took the Jeep to a reputable emission repair facility.  They smoke tested the exhaust system and found that the header collector before the catalytic converter was leaking so they replaced the flange before the catalytic converter as well as the one directly after the catalytic converter.  They were able to pass the low speed test but it still failed the high speed test.  They continued working through the issue by testing the fuel trims, smoke testing the intake, looking at timing, etc. etc.  In the end after ruling out all other potential issues they performed a compression test and noted that all six cylinders were in the 170 to 172 psi range.  They said, which I verified, that factory compression numbers should be between 120-150 psi.  In the end the diagnosis was that the higher compression is creating more NOX which is no being sufficiently scrubbed by the aftermarket magnaflow catalytic converter.  They recommended either the factory 1995 4.0 catalytic converter or a larger catalytic converter with the appropriate EO number.   

 

The saga continues...can you even get a factory converter anymore      

[Moses Ludel]

CJN8...Yep, that's high compression for the inline sixes, especially a 4.2L in the '80s!  Could be a previous rebuild with considerable cylinder head milling, a thin head gasket or higher compression ratio pistons.  Retarding the spark timing slightly might help with the NOx issue.  The shop likely knows that there is some latitude of adjustment in the Mopar PCM (5 degrees or so).  You can also jockey the crankshaft position sensor a bit if it's attached near the front damper in typical Mopar EFI Kit fashion...You must be running higher octane fuel at this compression with a 4.2L cylinder head?  Otherwise, I'd expect knock/ping unless the timing is retarded.

Here are the OEM part numbers for a 1995 Jeep YJ Wrangler catalytic converter if its available.  These should be good numbers for crossover to current offerings:

CONVERTER, Catalytic
4.0L Eng.
52017717 1994-95, Federal
52017718 1994-95, Calif.

Moses

 

[CJN8]

How do you space the pickup on the front of the motor down? How much spacing? Do the spacers go between the aluminum bracket and the oil pan?

Regarding the head gasket I used a run of the mill felpro. Is their a better choice that is thicker? 

[Moses Ludel]

CJN8...You can explore and compare head gaskets for thickness although FelPro is usually the same as OEM or allows slightly for milling work, which would make it thicker than stock.  How long has this head gasket and engine build been in place?  Is there any possibility that the higher compression is from carbon buildup?  This is rare with EFI, the whole point to an O2 sensor feedback system is controlled air/fuel ratios, which reduces carbon buildup drastically.

The goal with the pickup is spacing between the pickup tip and the crankshaft damper.  This should be at least 0.020" gap;  William H. shows a 0.050" gap on his Mopar EFI Conversion like yours (see first photo at topic):

Make sure the damper runs true and that the gap is adequate for any damper runout.  Spacers would be used according to where you want the pickup to position (at TDC, BTDC/advanced or ATDC/retarded) while maintaining the right tip gap.

Moses 

[CJN8]

I just recently had the head reconditioned.  However, the Jeep failed NOX, roughly with the same numbers before the head was reconditioned too.  

The pistons were cleaned prior to the install of the reconditioned head.

I have spent thousands trying to get my Jeep back on the road...my options have run out.

My base timing is 14 BTDC, when I put spacers in will I see the timing drop?  Will this be visible with a timing light?

 

[Moses Ludel]

CJN8...Lots of work and cost here!  Thanks for the photo, you did a very nice clean-up job during the head installation.  The pistons look stock by design and should be correct for lower compression ratio.  The compression figures you offered earlier, were they before you did cylinder head work or after?  The cranking compression was actually within norms from your test figures, certainly not abnormal or high.  The smog shop test figures were considerably higher.  Between the 14 degree base timing advance and the engine's 175 PSI compression readings, you should hear spark knock/pinging on anything but high octane fuel.

Quick questions:  1) Is your cylinder head a 4.2L type or a 4.0L conversion head? (It looks 4.2L in terms of the valve cover bolt holes.)  2) Did you hold the throttle open when performing the compression test?  3) Did you crank enough for "maximum" reading per cylinder?  I'm puzzling over the discrepancy between your findings (121 to 137 PSI) and the smog shop's readings of 175 PSI. Which reading is right?

If the high cylinder pressure is related to compression ratio, the cause could be smaller combustion chambers than stock 4.2L or the thickness of the head gasket. Combustion chamber volume is different between a 4.2L and 4.0L cylinder head.  Milling or surfacing a stock head "too much" would also increase compression ratio by reducing the combustion chamber sizes.  "Milling [surfacing] the head" is a time-honored way to boost the compression ratio.

The shop that diagnosed the NOX issue seems knowledgeable about the relationship of compression ratio and NOX.  The actual factor is heat.  Any cause of upper cylinder heat to temperatures above 2500 degrees F will create NOX.  An Mopar EFI conversion should, if anything, reduce NOX by more precisely controlling the air/fuel ratio and spark timing.

Your base timing sounds too far advanced.  If you're trying to relocate the pickup (alter the base timing), you need to shim or space the bracket in a manner that will 1) keep the pickup centered on the damper and 2) move the pickup toward the retard or trailing position.  The pickup (CPS) sends a signal to the PCM that the #1 piston is at TDC.  To retard the base timing, you want the pickup to index in the retard direction.  Basically, you're indicating that TDC is at the retard location.  

The PCM has no idea whether the pickup lines up with the crankshaft TDC mark for #1 piston or not, it's simply assuming that the pickup sends a TDC signal.  (A stock 4.0L CPS is non-adjustable and factory aligned for an exact TDC for #1 cylinder reading.)  Confirm the base/idle adjusted timing with your timing light.  I would try a new setting at 6 to 8 degrees BTDC.  Check the NOX, and if dropping but still too high, retard a bit more.  AMC ran the carbureted 4.2L conventional distributor engines between "0" and 4-degrees BTDC at idle with the vacuum advance hose disconnected.

You can check the manifold vacuum before and after this timing adjustment.  Vacuum should drop as you retard the base timing.  When you retard the base timing, you will likely be retarding the entire timing curve, as the 4.0L Mopar PCM relies on MAP and other sensors while not using a knock sensor signal.  You've already checked the MAP.  You might test the ohms reading at the coolant temp sensor if you have not done that already.  Also, the throttle position sensor (TPS) is a fixed setting (non-adjustable) but could be off-voltage.  This could cause a rich or lean condition that would affect upper cylinder temperatures.

If your CPS fits like William H.'s, lowering the bracket should retard the timing.  Try putting flat washers or suitable spacers between the oil pan and backside of the bracket.  This will place the bracket/pickup lower and drop the pickup clockwise on the crankshaft damper (when looking at the damper from the front of the engine).  Take a timing light reading to confirm that this has retarded the base timing.  You may need to elongate the bracket's mounting holes to create "slots" for re-adjusting the gap between the pickup tip and crankshaft damper...

Lastly, as a point of interest, the 4.2L engine used an EGR system.  The sole purpose of EGR was to recirculate exhaust gases and dilute the incoming air/fuel mixture; EGR cools the upper cylinders to reduce NOX.  The 4.0L engine with Mopar MPI (1991-up) does not use EGR, and these engines met even more stringent NOX reduction levels than the 4.2L engines.  The Renix EFI 4.0L system (1987-90) did use an EGR to reduce NOX.  Your Mopar EFI Conversion kit patterns after Mopar MPI.

Moses

[CJN8]

The compression numbers offered earlier were before the rebuild.  When I tested after the head rebuild I got about 135psi even across all cylinders.  The shop doing the testing say it is 171, 174, 175, 174, 174, 170.  I plan to test it again this weekend to verify the numbers.  I am using a cheap harbor freight compression tester, maybe mine is not accurate?  

The real interesting thing is that I am really not getting any spark knock.  I drove it home from the smog repair shop yesterday and pulled a long grade in 4th going 65mph and had no real issues with pinging.  I do run 91 octane.  

I have the 4.2 cylinder head still.

Yes I removed all spark plugs and held down the throttle during the compression test.  I plan to re do the test this weekend to check my old numbers (maybe my HF tester is not as good as the shop tester).  

Could a small casting flashing bump cause the high NOX?  I have been going through my head and I seem to remember that there was a small bump on one of the cylinders near one of the valves. 

[Moses Ludel]

See my red highlighted comments below, CJN8...Curious about your pending compression check and also what develops from re-adjusting the spark base timing at idle:

6 hours ago, CJN8 said:

The compression numbers offered earlier were before the rebuild.  When I tested after the head rebuild I got about 135psi even across all cylinders.  The shop doing the testing say it is 171, 174, 175, 174, 174, 170.  I plan to test it again this weekend to verify the numbers.  I am using a cheap harbor freight compression tester, maybe mine is not accurate?

Compression is two-fold, it can be boosted by smaller volume in the combustion chamber or even get a bump from above normal cylinder seal (i.e., zero gap rings).  Valves need to seal all the time, piston rings have proscribed gaps, and that leads to the acceptable and desirable 8-10% cylinder leak down in a newly broken-in engine.  I prefer a cylinder leakdown test for diagnostics and have promoted that approach through my years of work.  A compression gauge is good for quick assessment but often overlooks cylinder taper, leakage under actual combustion pressures and other factors.  Your compression gauge should be suitable for what you're trying to determine, which is simply the relative compression ratio.  Here's the difference: 

 

The real interesting thing is that I am really not getting any spark knock.  I drove it home from the smog repair shop yesterday and pulled a long grade in 4th going 65mph and had no real issues with pinging.  I do run 91 octane.

I'm not convinced yet that your compression is too high.  The NOX issue could simply be spark timing or even a manifold vacuum leak (intake leak from the base of the throttle body to the engine), which could lean out a cylinder/cylinders and raise combustion temps.  I would try a light mist of low-volatility spray like WD-40 along the intake manifold flange and around the throttle body base.  See if you can find cause for a lean cylinder(s) condition.  That could cause local heat in a cylinder and raise NOX.  

If you have an infrared surface temp tester, also take cylinder external block, head and individual exhaust header tube temperatures to see whether you have an individual cylinder(s) running hot from poor coolant flow or a localized lean burn condition.  

I have the 4.2 cylinder head still.

Understood...Not optimal in terms of detonation/ping or flow, but this is a durable and functional head.  If you're not running this engine above 4500 rpm much (if ever), the 4.2L head is okay.  You've spent plenty on it, I would leave the head alone at this point.

Yes I removed all spark plugs and held down the throttle during the compression test.  I plan to re do the test this weekend to check my old numbers (maybe my HF tester is not as good as the shop tester).

Could be the gauge, but even an HF gauge should not be off by more than 5%-7% or so.  I'd emphasize that the 120 PSI described in manuals is actually quite low.  These figures show up in factory shop manuals because this gives the vehicle manufacturer leeway to avoid warranty claims.  True cranking compression for a 1983 4.2L inline Jeep six should be more like 140-160 PSI in a healthy or newly broken-in engine, maximum variation of 10% between highest and lowest cylinder.  

I do have a question not raised before:  Do you have a stock camshaft?  If not, your NOX issue at low speeds could be very much related to a modified camshaft grind.  If not a stock camshaft, what is the profile?

Could a small casting flashing bump cause the high NOX?  I have been going through my head and I seem to remember that there was a small bump on one of the cylinders near one of the valves.

The 4.2L stock head is not a ported, polished and cc'd head.  The cc'd part is what you're considering.  If there is a significant amount of added casting, that could reduce combustion chamber volume and create locally higher compression on that one cylinder.  This is somewhat ruled out by your uniform cranking compression (according to the shop's test).  More important is dynamic compression, the actual load each cylinder is pulling, which can be affected by valve lift, valve opening duration, compression and manifold vacuum at individual cylinders.  A dynamic compression test, traditionally done with an oscilloscope engine analyzer by shorting each cylinder and measuring the loss of power, can be very revealing.

This raises another question.  If you were performing a leakdown test and pinpointed valve leakage (intake and/or exhaust), I would question whether you have an issue with cylinder seal caused by the pushrods being too long.  This occurs on engines like your Jeep 4.2L inline six that do not have a provision for valve adjustment.  When the head or block deck gets surfaced beyond a point, the valve stem heights or pushrod lengths must be adjusted to compensate.

If the valve stems stand too tall, or if head and block surfacing lowers the rocker arm positions, shorter pushrods may be needed when the valves cannot close completely.  Most "normal" machine work will not create this situation.  If you want to know more about valve "adjustment" for a 4.2L Jeep inline six that has no adjusters, only fixed pedestals, check out my vlog:

http://www.4wdmechanix.com/jeep-232-258-and-4-0l-inline-six-and-2-5l-straight-four-valve-clearances-and-adjustment/

A leak down test would be helpful to see whether the valves are seating when static.  Again, your compression readings do not suggest that there are unseated valves, but you should at least be aware of this Jeep inline six head and block surfacing issue.

Moses

 

[CJN8]

From your write up on the FJ40...

This engine ran sluggishly, the result of inadequate cylinder seal at the rings.  The valves seated, the head casting and head gasket were fine, however, the combustion gases leaked past the rings and into the crankcase.  Engines like this have a distinct "blow-by" smell in the crankcase oil.  While the PCV system can recycle blow-by gases, the remnant of combustion gases remains in the oil.  I have often referred to this odor, notable with the oil filler cap removed and the engine running, as the "smell of death"

Bingo...this is how my oil smells, even after a change and I drive for 20 miles its there.  I ordered a used oil analysis test bottle from blackstone labs on Monday to test this smell.  I need to find a leakdown tester.  Great help Moses.  Thanks so much.  

 

[CJN8]

A little more diagnostics...I looked into the catch can which is between the valve cover and the intake manifold.

The crankcase gases appear to have both oil and water present, probably due to the short trips I have been taking.  I looked at the engine oil in the crankcase and there is no water or coolant there.  I also looked at the coolant and there is no oil.

I ordered a leak down tester today.  The unit is made by OTC and got good reviews on amazon.  I hope the test is telling.

I will report back.  Thanks for all the help Moses, it is much appreciated.  

[Moses Ludel]

CJN8...Smart to get the oil analyzed.  This is a a conundrum but also an opportunity to raise the bar on engine troubleshooting and analysis...  

Leak down tests never fail to pinpoint issues.  I cannot overstate how impressively these tests narrow down a casting crack, ring blowby, a valve(s) leaking, head gasket seepage, the list goes on.  All of this with the engine not running! 

Think of cylinders as a sealed unit.  You're looking for the percentage of air leaking out of the cylinder with the piston at TDC on the compression stroke, both valves closed.  The air being outside sourced from a compressor/tank is constant.  The percentage of leak allows for leakage "norms", more specifically the piston ring gaps allow a controlled amount of blowby, a percentage that is acceptable on all but ultra-high compression racing engines with zero-gap rings.  

What I like the most about leakdown testing is the ability to judge cylinder seal with the rings at the highest point of cylinder taper.  This is the least tensioned and widest ring gap point.  A fresh engine with straight cylinder walls would have approximately the same seal with the piston rings at any position on the wall (assuming that the intake and exhaust valves are closed).  This is not the case with an engine that has cylinder wall taper.  The highest degree of taper is at piston TDC.

Note: If you remove the oil filler cap with the engine idling, and pull the PCV valve out of the cover as well, you may see the rhythmic puff of blue smoke exiting the filler opening.  This is blowby that might be otherwise sucked through the PCV and manifold induction system, ultimately burned in the combustion process.

You can't go wrong investing in a leak down gauge.  I'd be glad to comment on its varied uses and ways to interpret the readings.  You bought the OTC 5606 tool?  That's the one I recommend...I paid more for my Snap-On unit new in the early '80s.

The condensation in the can raises some questions.  Pull the oil filler cap and see whether there are signs of coolant or condensation at the valvetrain.  If not, you're likely right about this simply being condensation.  You might reconsider the PVC hose routing and breather/baffle arrangement at the valve cover:  A prime attribute for a PCV system is reduction or elimination of condensation in the crankcase and throughout the engine.  The question would be "why is there condensation at the can?"  Once the engine starts and warms, there should be no condensation.  Maybe this can is a catch basin for condensation?

When you get your new cylinder leak tester, I can share how to use the device to detect a casting crack or head gasket seepage.  I doubt you have either, your cylinder head should have been magnafluxed during the rebuild, and your workmanship looks right with regard to the cylinder head work...

Moses

[CJN8]

Preliminary test while motor was not warmed up.  I wanted to get my technique for finding TDC down.  I used a vacuum/fuel pressure gauge so I could see exactly when the piston was at TDC.  Worked like a charm

Front of motor

1. 15.6 PSI (intake/exhaust manifold no hissing noise, valve cover hissing noise) 

2. 12.2 PSI (intake/exhaust manifold no hissing noise, valve cover hissing noise)

3. 8.8 PSI (intake/exhaust manifold no hissing noise, valve cover hissing noise)

4. 7.7 PSI (intake/exhaust manifold no hissing noise, valve cover hissing noise)

5. 8.8 PSI (intake/exhaust manifold no hissing noise, valve cover hissing noise)

6 11.4 PSI (intake/exhaust manifold no hissing noise, valve cover hissing noise)

Back of motor

 

I am going to warm the motor to operating temp and repeat the test.

Nate

[Moses Ludel]

Nate...I'm guessing you mean percentage readings?  This is with your new leak down tester?

If so, these readings are plenty acceptable.  Up to 20% leak is a good running engine.  The hissing at the valve cover is the normal amount of air getting past the rings, moving into the crankcase and finding its way to the valve cover as normal...This normal leakage rate accounts for the ring gaps.  Even the 15.6% leak is okay and no cause whatsoever for alarm.

What setting of air pressure are you applying?  Sometimes raising the pressure to 100 PSI will improve the reading and drop the percentage of leakage.  This is due to more ring pressure against the cylinder wall.  My Snap-On gauge calls for 60 PSI of line pressure.  I usually do the test at 60 PSI then bump it to 90 PSI.  Start with the instruction recommendations.  Also, too much testing blows the oil seal off the rings and that has a subtle impact on the readings.

Will be interesting to compare warm testing, this can sometimes make a difference.  I wouldn't expect much difference.

Cool tool...Congrats!

Moses

[CJN8]

Yes, percentages. My mistake. I did the tests at 90psi. 

[Moses Ludel]

Excellent, Nate...The only test you might add would be watching the radiator (cap off during leak testing) for any bubbles while you're pressurizing the cylinders.  This would rule out a casting crack or head gasket seepage.

Moses

[CJN8]

So my detective work continues. Have you heard of setting fuel sync on the mpi kits and the jeep 4.0? The distributor does not adjust timing but I guess it does adjust fuel synchronization? Could this be an issue?

[Moses Ludel]

Nate...It's important to align the distributor housing and rotor properly on a 4.0L EFI/MPI engine.  (That's the distributor furnished with your 4.2L EFI kit.)  The distributor and timing are not adjustable, and the distributor housing should not be moved to "adjust" base timing.  Timing reference is from the Crankshaft Position Sensor, which tells the PCM when #1 piston is at TDC.  (We discussed ways to retard spark timing by moving the CPS.)  All other spark timing functions are within the PCM.

Note: The distributor is strictly for distributing spark to the plugs at the correct point.  It does not serve any function of timing advance or the base timing.  The distributor housing must be in the proper/stock designated location with its driveshaft and rotor aligned properly so that the rotor points to #1 spark lead with the engine at TDC for #1 cylinder's  firing cycle/stroke.

Beyond this, the distributor does function as the camshaft position sensor on these engines.  The PCM is looking for the difference between CPS at TDC #1 cylinder and the position of the camshaft as determined by the correct positioning of the distributor driveshaft, rotor and cap.  If this is too far apart, there is a signal that the valve timing is wrong.  This is typically due to excessive timing chain wear.

Setting fuel sync on a 2000-up coil-on-plug (distributor-less) 4.0L MPI system requires indexing the camshaft position sensor properly.  This also may impact the fuel sync reference signal for distributor type ignitions like 1991-99 model 4.0L engines use.  This would include your engine's 1994-95 type distributor.  

Otherwise, fuel flow is strictly a function within the PCM and its software.  Individual cylinder/injector timing cannot be changed or adjusted independently, although "trim" can be checked with a DRB-III scanner or an aftermarket equivalent tool.  Fuel trim can be observed during diagnostics with this type of scan tool as well.

I would check your distributor housing position and also the indexing of the distributor's driveshaft and gear to make sure the rotor aligns with the #1 cylinder position on the distributor cap when the #1 piston is at TDC on its firing stroke.  There is a clear procedure in the FSM for setting the distributor properly.  If you need that information for a 1994-95 distributor, let me know.

Moses

[CJN8]

Distributor checks out and is set to spec.  

Also, the leak down test percentages evened out after the engine was warm, averages were about 9% to 11% which is pretty good I guess.

I emailed Hesco and they said put 1/2" spacers under the CPS bracket and re-adjust my CPS to proper clearance.  I did that today and it adjusted my base timing from 14* BTDC to 7* BTDC.  I took it for a spin and no sign of any issues.  My manifold vacuum went from about 17.5 to 16.0.  I suspect this timing adjustment should make a difference in NOX production, only one way to find out.

I will report back.

Thanks Moses! 

[Moses Ludel]

Exciting, Nate...You're making progress!  Manifold vacuum is a bit low if this is with the throttle closed at idle.  If so, check for an intake manifold to cylinder head vacuum leak or a defective brake booster check valve.  Something has dropped or bled off the vacuum—yes, timing advance is a factor, but both readings, if at idle with the throttle valve closed, are a bit on the low side.

The key now is NOX testing...Good luck, and share the results!

Moses