Automotive Forums  Where Car Enthusiasts Go to Discuss Cars Online

your radiator cap is on the coolant tank..  where you pour the coolant into...

as for which spark plugs...

the only real issue is the dielectric grease properly applied to the inside of the boots on both ends without getting much if any on the internal contact..

without the dielectric grease.. the high voltage will leak between the boot and the plug insulator...  or the boot and the cap tower.. sending high voltage to the ground... causing your misfire...


please DO NOT USE dielectric grease on the spark plug threads.. that needs a different product.. spark plug anti seize...  

as for compression testers... some stores have a loan a tool set...   if you live near a harbor freight... they have them for 13 and 20 bucks for nice ones...  

I performed a few tests this weekend. Still haven't found the root cause, but I learned a little.

I pulled the plugs and examined them. None of them looked particularly fowled or wet, and all had carbon deposit. Plugs 2 and 3 had a little white colored deposit on them along with the black carbon, while 1 and 4 only had black deposit. I don't know if that may point to a coil pack going bad or not, but it certainly didn't seem conclusive.

I used an HEI spark tester and all plugs sparked. (Both coil packs passed a primary and secondary resistance test that I performed a while back.)

I ran a compression test and all cylinders appeared fine (220 PSI +/- 2 PSI).

No signs of carbon tracks on any plugs, boots, or the coil housing (which were all recently replaced anyway)

I applied fresh dielectric grease to all of the plastic interfaces around the metal conductors (being careful not to get any on the metal). Then torqued the plugs down to spec. I did not use anti-seize as GM did not use it, and their torque specs are were not designed to take it into account (13 ft-lbs).

I removed all wiring harnesses from the fuel injectors and attached a 12V LED light (I did not have the proper NOID light). For each light when the key is turned to the ON position, the light lit up briefly while the fuel pump ran so I know the electrical connections are good. The lights do not flash or light up at all for cylinders 1 - 3 when I crank the engine. The cylinder 4 light does light up (does NOT flash) when I crank the engine though. This seemed a little odd as I thought they were all supposed to flash while the engine was cranking.

Plugged the fuel injectors back in and started then engine. I then pulled them one at a time to see if I could isolate a single dead cylinder. Every fuel injector I disable affected the engine's idle. There doesn't appear to be a single dead cylinder. Whatever is causing the misfire is an intermittent issue.

Based on the funny NOID light test, I may take the ICM out and take it to Autozone for testing. I'm not sure if it would be useful to try replacing the coil pack responsible for cylinders 1 and 4 based on the difference in plug appearance. I have heard of these particular coil packs showing no signs of being bad in tests (resistance and HEI), but still turn out to be the cause of a misfire. If all else fails I may put in a set of AC-Delco Rapidfire platinum plugs to see if they perform better, or try replacing the fuel injectors. I hate to just throw money at it, but I've exhausted almost every test I can think of.  

thats strange with the injectors... during cranking...      does your digital volt meter have a millisecond setting or a duty cycle setting?  do you have a import junk yard around.. where you might scrounge some  connectors...

fuel injector connectors with a center keyway are easy to find..   if you search saabs or volvos... either on the ignition distributer connector or some of the oxygen sensors .. is a center keyway connector matching the fuel injector connector...  so you could wire the two together using female bullet connectors...   2 wires into the crimp end.. the female barrel open for attaching your test leads clip...  this would allow you to access the fuel injectors wiring without tapping into the insulation..

1.3 to 1.4 ms is what most cars should idle at..

as for the ECM to az.. they don't test ecms....

you can.. if you have a strong lens... with extreme care.. and a static band for your wrist..  open the computer and examine it for cold solder joints..  cold solder joints are where the wire through the circuit board has expanded the solder..  the heat at the base of the solder joint is pulled away by the circuit board so it does not expand as much.. this eventually breaks the solder joint in a tubular fashion..  so a circle may be visible in the solder joint around the conductor through the middle..  soldering computer circuit boards is not childs play..  but people experienced with it do it every day...    placing a black dot from a sharpie marker next to the possible makes it easy to find..

were you able to look down the openings where the springs are in the cap... to see the embedded conductors between the coil and the spark plug contact terminals... i have had the plastic cap burn and the high voltage jump from one spark plug conductor to the other...      is the cap and coils under warrantee at AZ..

coils.. cap.. boots... are the usual cause of misfiring on quad 4 motors..  

A couple other thoughts...

I have heard the MAP sensors go bad about the same time their vacuum hoses go bad. That vacuum hose was rotted out (replaced now) so I may take a look at the MAP voltage output at idle and under load. Any idea where that voltage should be.

I have also heard the crankshaft position sensor can start to become weak. It would cause timing issues at idle, and perform better at higher RPMs where the electro-mechanical sensor produces a stronger signal. Can't think of a way to test this though without just trying a replacement - I don't think my scan tool has access to that sensor's output.  

I believe my multimeter has a duty cycle setting, but that would report a percentage reading not a pulse width. I believe it also has a frequency setting though, so between the two I should be able to get a good idea of what kind of signal it outputs.

I was under the impression that AZ could hook up the ICM and test it. I though the same box they use to test alternators and starters was capable of testing it as well. I may have to call to confirm, or barge in and demand answers.

I didn't thoroughly inspect the inside of the boots... just shined a light in to make sure the springs inside didn't fall out. If there was a voltage leak I thought it would have been apparent on the plugs, outside of boot, or on the coil housing.  

somebody sent me some scan tool number .. their engine is surging at idle..
this is to help you understand what the scan tool is displaying..


                                   cold                                       warm
coolant temp-------------65 degrees                              186
throttle sensor-----------.58                                           .58
         set at .56 not running
injector pulse ----------- 2.5                                         2.2   this is too high at idle.1.3, 1.4ms
desired idle--------------712rpm                                   600rpm
actual idle  --------------780/800 open loop                  680/750 closed  this is too high
map sensor--------------1.68                                         1.98
throttle angle--------------0 %                                        0 %
block learn --------------110                                          110   128 is design perfect
spark advance ----------14.7                                         14.7
block learn cell------------5                                              5
integrator-----------------126                                         124   128 is design perfect
idle air mtr steps----------0    ???????????????????????       0         20 to 50 counts is perfect
fuel pressure-------------48 psi                                       48psi

this car is idling too fast.. but the computer cannot close the idle air motor below 0 to slow the engine down.. so it keeps the fuel injector pulse wider than idle programming should be..   to fix this car.. readjusting the throttle plate..  unplugging the idle air control motor when the idle air counts are at 0..  backing out of the throttle stop screw... till the engine is just barely idling..  shutting the engine off if the TPS voltage is below 0.35 volts.. and bringing it back to between .35 and .60.. some cars have specific voltages for his.. starting the engine..   bringing the idle up by hand on the throttle lever to about 2000RPMs.. then plugging the idle air control motor back in..  slowly backing off on the throttle until you are back to idle..  the idle air counts should be within 20 to 50 counts at this point..  rpm 450 to 600.. depending on the motor..    this should also reduce the fuel injector pulse width to 1.3 to 1.4 MS..   the block learn numbers should come back close to 128..   there are 255 possible cells for the program to select from..  yep.. computer programming.. learning where the engine is running.. adaptive..  to a point..

 all these specs brought back by just a throttle stop and then resetting the TPS ..

again.. this was just to help you understand what the numbers mean...

on obd2 systems.. looking at the fuel trim percentages... short term and long term.. instead of block learn.. or integrator..  

I'll see what data I can pull up in regards to this... Is the injector pulse number an average from all injectors with vehicle running? Or is it just a reading from one injector?  

with a single oxygen sensor after the motor.. and then one after cat oxygen sensor..

you will only have one bank of data stream to calculate fuel trim..

V6 and V8 motors have separate exhaust systems to at least the cat.. so they can modify the fuel trim per side..

so on your .. even though its sequential fuel injection.. the injector pulse width should not vary between injectors..

longer ON time means more fuel..   shorter on time. less fuel..  but you understand that..

did you ever run a compression test???  

Yes... all cylinders appeared just fine. 220 PSI  +/-2 PSI