aatcman

So, I finally got around to building my test rig... Unfortunately, I have not yet been able to reliably duplicate the units behaviour in the Jeep.. However, I have been able to make it produce its failed state once... When it did produce its failed state, it continued to fire the plug even with the trigger signal disconnected.. it did not respond to trigger at all (which means it did not change firing speed with RPM) and I could only stop it by removing power. Once I hooked power back up, it would not fire at all -- for a few minutes.. Then it went back to working like it should and I have been unable to break it again since.. Another thing I noticed was that right before the unit stopped responding, it started killing every other trigger pulse (pulling it to ground). --I had my oscope connected to the trigger line so I could make sure my input was similar to that of the ECM... In fact, it was because of the weird effects to the trigger signal that I disconnected the trigger line in the first place -- I wanted to see if something was wrong with my 555 or if this was a symptom from the ICM... --555 Circuit was fine.. For those that want to try this out, google 555 astable calculator. Pretty much any will do as they all show the same basic circuit.. I used a 1 uf cap for C, 12k for R1 and 18k for R2 and just set it up on a perf board... I ALSO used a diode (ANY small switching diode should work. I think mine are 1n4004 or 1n4007, but again, it doesn't really matter) This diode goes in parallel with R2, cathode at the C end. The point of the diode is to short around R2 while C is charging (while the output signal is high) and thereby allow us to generate a square pulse with less then 50% duty cycle.. With the values above, you will get a square pulse at appx 42Hz and appx 40% duty cycle. (40% duty cycle means the pulse is high 40% of the cycle). This equates to the appx equivalent of 840 RPM and 50 degrees dwell... For the purpose of the test, making an exact replication of the ECU output for a given RPM doesn't seem to matter -- you could use 10k and 20k to simplify it and still have a valid test.. At any rate, I powered the circuit with my 12v power supply and knocked it down to 5 with a simple regulator I got at radio shack. These look like a large transistor -- but require no extra components. Just power in, power out, and ground... This provided 5 volts to my 555 (supplied as marked VCC or +v in the drawings you will find) and produces an appx 4.5v square pulse on the output (my ECU varies between 4-5v on this signal, so this seems close enough) It should be noted that any source of 12v will work. Total draw should be 2-3 amps (note, this is appx -- I did NOT measure this...yet...) I then simply grounded a spark plug (mounted in a vice) to my 12v power supply - and also ran - and + from my 12v power supply to the - and + pins on the ICM. Next a spark plug wire from the coil to the plug finally a wire from the 555 (pin 3) to the trigger input pin on the ICM (the single wire on the 2 wire plug) NOTE: DO NOT attempt this test on the vehicle!!! This is a bench test!! No connection to the vehicle should exist during this test!! I am NOT responsible for damage to YOU, YOUR VEHICLE, OR ANYONE OR ANYTHING ELSE!! The coil produces HIGH VOLTAGE!! if you do this test -- BE SAFE, and understand YOU ARE DO DOING IT AT YOUR OWN RISK AND YOU ASSUME ALL LIABILITY!! This info is therefore provided for educational purposes only!! I ASSUME NO LIABILITY WHATSOEVER!!! Any use of this info signifies consent to this!! Sorry about all that, but we all know how people like to sue each other these days.... :o) I'll post another update if I find a better way to produce the fail I was seeing on the Jeep..

Update: YAY!!!! It runs (and stays running)!!! Ok, so, it turns out my guess of either the coil or ICM was correct... --I replaced both and now it runs like a top again -- and keeps going too. I drove it 42 miles today.. It hasn't gone that far COMBINED since this craziness started almost a year ago.... Apparently it liked its ground improvements as I've NEVER had the crisp throttle response from it that I do now AND the chronic hesitation (accelerator response delay) is nearly gone as well... However, being an actual elex tech, I want to figure out which of the two (coil or ICM) is actually bad WITHOUT destroying the new ICM (thus why I bought a new coil....) Now I know what yur saying, I posted above that I hooked up another coil to the ICM and it still didn't fire -- thus the ICM HAS to be the culprit..... Well, in a normal vehicle, yes... Not necessarily in an AMC/renix... Remember that in an effort to produce vehicles that were slightly A. better than the other manufacturers and B. lasted longer, AMC tried to make certain things work smarter than then norm. Thus, some of their ideas (and vehicles) were Almost Made Correctly... :-) A good example of this is the access panel in the floor (under the back seat) that gives easy access to the top of the fuel tank without removal on the FSJ's which is constantly joked about because many feel AMC/Jeep "made it super easy to get to the impossible to remove giant plastic nut that holds the sender/siphon assembly into the top of the tank"!! These things will drive you crazy until you drink yourself to oblivion and crash into your workbench in a drunk stupor (at which point your oil filter end wrench -- that little plastic or metal cup you stick on the end of the oil filter to be able to remove it with a standard ratchet -- falls off its pin and hits you in the head -- and thats when you realize its the missing key to the puzzle.... Yes, the AMC engineers intended for you to use this simple (and super cheap) tool to easily remove and reinstall that otherwise impossible nut -- yet there is nothing about it in any FSM, Chiltons, Haynes, etc... because the AMC factory manuals were Almost Made Correctly...... Gotta give them some credit for the MJ and XJ -- ya don't have to remove those tanks either -- run the fuel down below 1/2 and then just take the pump/sender unit out -- its mounted on the top side of the tank -- you can just slide under the jeep and remove it...... Point of the examples above is I think they did something similar (electronically) with the ICM... Which is why I don't think there is anything wrong with it even though it wouldn't fire a different coil yesterday.. You see when a semiconductor goes bad (thats all the ICM is folks, a couple transistors and a few other components) it STAYS BAD. It DOES NOT turn back on after it cools -- albeit in some extreme conditions a very large transistor or FET may start working again, but its function will be erratic and short lived... My ICM DID CONSISTENTLY COME BACK ON after enough wait time (WITH THE KEY OFF)... Coils, however, can fluctuate when the insulation between their windings starts breaking down -- first sign is incorrect or erratic timing and/ or WEAK spark (sound familiar?) --The "Other" first sign (which most will miss since there is no good way to measure on the Renix type of coil setup) is increased current draw. The worse the coil gets, the more current it draws.. In a regular simple electronic trigger ignition (Dura II [FSJ], GM 4 pin HEI, etc...) you will keep killing ignition modules for no apparent reason (anyone ever had a car that did that? replace the coil... --Most people miss this because the coil will still test good with a multimeter (ohms) as the breakdown is usually on the secondary side and usually only at high voltage (you lose spark way before it gets bad enough to see it with a multimeter -- usually) In a Renix, the ICM is even simpler.. I --In the above systems, the ICM has to take the input from a distributor (usually hall effect) and convert it into a TTL pulse, shape that pulse, amplify it, and then send it to the coil.. --In the Renix, all the ECU makes and shapes the pulse -- all the ICM does is amplify it and provide it to the coil as drive.. So WHY can I buy 5 GM 4 pin ICM's or 3-4 Dura II ICM's for the price of ONE "Dumber" Renix ICM? --My guess, which I intend to prove, is that the engineers that made it tried to make it smarter -- but forgot to tell the people that wrote the manuals how to use that extra smartness.... AMC..... --I think these modules have extensive circuit protection in them to prevent them from being damaged when the coil breaks down over time. --To put it simply, I think it has circuits that literally turn off the ICM when the coil draws too much current, thereby preventing it from being damaged.. --I also think that the DRB II tool so fondly spoken of in the manuals has the ability to detect this condition, probably even before the ICM shuts it's self off.. If my guess is right, the Renix ICM disables itself to protect it from overcurrent destruction... Kinda like a mini, automatic reset, circuit breaker.... Now that I know what the input to the ICM is (+12v, GND, and a 5v square trigger from the ECU) building a test to actually test it (not just fire an LED like at the parts store) should be super simple... --The input will be generated by a simple 555 circuit which I will rig to produce an apx 50hz 5v square pulse similar to the one generated by the ECU. ---Note: 50hz = 1000 RPM.. Formula is ((rpm/2) x cyl)/60 ... ---Remember, you only fire a cylinder every second revolution (google "otto cycle" if this doesn't make sense -- suck, squeeze, bang, blow....) ---So, 1000RPM = 500 Sparks per min per cylinder = 3000 sparks per min (500 x 6 cylinders) = 50 sparks per second (3000 sparks per min / 60 seconds per minute) ---Every spark requires a pulse. Thus 50 sparks per second = 50 pulses per second. 1 pulse = 1 cycle. ---Thus 50 pulses per second = 50 cycles per second = 50 Hz (google "Hertz -rental" if this doesn't make sense) Now, I have no doubt the this input I'm going to make will be mostly identical to what the parts store tester makes... Its the OUTPUT that is important!!!! Its the OUTPUT that the parts store testers fail to accurately gauge -- ignition modules (ALL of them EXCEPT CD units like MSD) are CURRENT AMPLIFIERS!! You CAN NOT test a CURRENT AMPLIFIER accurately without a proper LOAD!! Two ways to do this proper. I recommend the second as the first, while valid, provides no indication of pass or fail.... --1. You could rig a 1 ohm ballast resistor between the two terminals mean to hook to a coil. ---BUT, you have no way to tell if a signal is being generated.. Unless you have fancy equipment like an oscilloscope (oscpe for short) ---If you had an oscope, you could clip onto the - terminal and watch the pulses. ---You might also be able to get away with a test light or ANALOG volt meter (one with a needle, NOT an LED display) and look for a fast flash or fast needle movement. ---BUT, the resistor will get VERY HOT, VERY FAST!!! ----ALSO, the resistor will NOT present reactance like a coil will (in a coil, this would be sudden surges of current -- especially when the spark plug fires...) --2. You could hook up a coil with a spark plug attached. This is the simple way and thus my recommendation.. ---BUT you MUST INCREASE the gap significantly!! (Note, this test will prove a coil good or bad too).. ---Ideally you would open the gap to 8x what it normally is (0.035 x 8 = 0.28 or appx 1/4"). This is because your cylinders run appx 8:1 compression... ----When you compress air, it becomes more dense -- essentially, when your piston is at the top of its stroke, the air in the cylinder is now in 1/8 the space it used to be -- which means it is 8x denser. ----More simply said, when the spark plugs fire in a cylinder, the spark has to travel through 8x more air to jump the gap then it does if you make it spark just sitting on top of the engine.. ---Again, thats ideally. You can't really gap a plug that far as it's only about 1/8" from the side of the plug to the electrode anyway.. ---So, for our test, I recommend just less then 1/8" (just a little less than the distance from the center electrode to the metal case) ---Mount the plug in something sturdy -- and something that doesn't burn! :-) ---Ground the metal base of the plug to the - of your power source (battery or power supply). 20 gauge or larger is fine -- plugs draw very little current... ----Note: Use a plug you don't care about -- you will likely damage it gapping it wide like that.... Plug should be the same type your vehicle uses... ----Note: Using a plug gapped to the regular gap is just as useless as the parts store tester..... The plug is the load. You need enough load to load test the coil and the ICM circuitry... ------Bigger gap = more resistance = higher voltage required to jump the gap = MUCH bigger current surge when it does jump (spark) So, we will have a simple 555 circuit making a 50hz pulse, a battery or power supply powering the ICM and the 555 circuit, the ground from said power supply going to the ICM, the 555 circuit, and the metal base of the spark plug, and a regular spark plug wire going from the coil to the plug. I plan to mount my plug in a drill press vice for convenience purposes but anything nonflamable will work... (even laying it on a non-wood work bench will work) When I power up the ICM and the 555 circuit, the plug should make crazy spark -- and continue to do so until power is removed from the 555 circuit... My intent is to test the old ICM first with a known good coil. -If it continues to make spark for more than the 5-10 min the jeep would run for, then its safe to assume the coil was bad... -If this becomes the case, I'll then repeat the test with the old coil to verify it turns its self off after such time... -If you had a known good ICM, you could test a suspect coil this way -- if the coil is bad, the sparks should become less blue (and more orange) as time goes by AND the module should eventually turn its self off... If someone can help me figure out how to post pictures, I'll post a picture of the test setup and a schematic. --Prolly $5-6 of electronic components (radio shack prices) needed to build it.. Again, thanks all for the help and I'll post one final update with the results of ICM test...

Secondary Ignition? That would mean plugs, wires, cap, rotor.... Hadn't thought about those because I just replaced them all a few months ago right after I starting having these problems.... Plug wires still look new... I'll pull a couple plugs and see what they have to say.... Dizzy cap too...

Ok, Another update.... So, I found the pinout.. Very simple actually... Three wire connector: A = +bat voltage (from latch relay) B = Ground.... C = Tach Two wire connector: only wire hooked up = spark pulse from ECM (5v sqaure wave) So, as I figured above, the ICM is just another version of the duraspark II ICM (or more accuaratly, Ignition amplifier, since all it does is take a TTL pulse and amplify it into ignition coil drive). This one just gets the pulse from the PCM instead of directly from a hall effect sensor in the dizzy.... So, I rechecked the ground wire to the ICM (unplugged the 3 wire connector, measured from the B pin in the connector to the - battery pole. 0.1 ohms.. I assume this is good.... :-) Then I rechecked the +voltage. Battery voltage present when key is on.. So, I hooked up my oscope to the 5 volt pulse line (back probed, both plugs connected to ICM) and grounded its ground clip The PCM provides a nice pretty pulse both when the engine will run (it was nice enough to run for 30 seconds while I did this test) and when it has no spark. When the engine died, the pulses continued until the engine stopped spinning (as one would expect with the key in the on position). At this point, it is my opinion that if I can send a pulse into the amplifier (ICM) and not get a spark out, then the amplifier (ICM) must be bad.... After appx 214k, I can see that being possible... To the poster who suggested the CAT, that was one of my first thoughts... I eliminated it as a possibility because I've had several go bad on other vehicles and they usually allow a restart (at idle with very little power beyond idle) when they plug up and kill the engine.. I have yet to have one kill my spark either...... So, now I get to see if I can find a store that actually has one of these on the shelf... I was just gonna modify the connections on one of my spare d-spark boxes I have for my other jeep and just use a can coil, but, unfortunately, those have no tach pulse output and therefore won't work as it would seem that the Renix ECU expects a tach pulse to be sent back to it from the ICM.... So much for avoiding the overpriced renix ICM....... Anyway, I'll post another update once i acquire another ICM... Agaon, Thanks!!

SORTA.... Parts stores can test to see if they will light up an LED on a tester... Problem is, an LED is 100,000x less current than what it takes to fire an ignition coil.. I've had lots of dead modules that won't fire a coil pass the parts store test no problem... Anyway, I unplugged the CAMps this morning. At it started like usual. Power and response down a little, but no significant difference... It would seem my running limitation is not distance, but time... I tooled around in my yard this morning for a few min until it quit.... A few min later i was able to restart it -- it wanted to idle, but as soon as I gave it gas it bucked and died. To be fair, it should be noted that I hooked up my timing light before I tried to restart it so I could see if I lost spark when it died... When I gave it gas, it felt really powerless and then lost spark like someone turned it off with a light switch... And three hours later, I still do not have spark (as usual, it will probably not restart until sometime this evening) I rigged a spare can style coil I had laying around in place of the renix coil... Still no spark... Trying to find a pinout diagram for the connections to the ICM so I can look at the input pulses with my oscope and verify I have pulses coming in while I have no spark... I figure its worth the effort to verify I really need a $115 part before I go and buy one.... That said, I can't imagine that it could blame a lack of input signal to the ICM -- since when it has been restarted after it quits the first time it idles and then immediately looses spark and dies when the accelerator is pressed because throttle would not change the input pulse much -- and not in any way power wise as that is the purpose of the ICM -- the ICM is just a current amplifier to drive the coil..... I would think that if it was wiring that either A, I would loose DC voltage to the + terminal of the coil (even when I have no spark, I have full battery voltage there) or B. slowly start missing, running horribly, start destroying exhaust components from backfires, and then die... However in my case, it simply misses and has very low power for a few seconds and then dies like someone turned off the lights..... Like a transistor overheating internally..... Again, thanks for your continued help and I'll post another update when I figure out how to check for the input signal....

Will do.. Thanks.. Do you have an easy way to test the ICM? I cleaned the coil to ICM contacts as directed in your tips -- no change...

gogmorgo, I have the 5 speed manual if that helps... Cruiser, I read somewhere that while the Renix WILL run without the CAM position sensor hooked up or functioning, the ecu will calculate appx timing based on the last input it got from the CAMps and it will fire all 6 injectors as one bank instead of individually like normal. The same writeup also said that once stopped, some years of the Jeep Renix will not restart with the cam sensor not functioning... At any rate, this mode of running is supposed to provide much reduced throttle response and kill your mileage. Supposedly it will keep the MIL on in the later HO (chry obdII) 4.0's as well - - and the code will be one that says both CPS and CAMps error...... I found a procedure to test the CAMps -- but do you have one to test the ignition module and coil? Note -- neither are warm at all when it quits running... Thanks!!

OK, so update.... I went through the grounds. Several were VERY high resistance... Some were 40 ohms or more to the - battery post! Unfortunately, I don't have a big enough soldering iron to coat them and the wire around them in a sheet of solder, so for the mean time I have smashed/recrimped them with a vice grip. Yes, I did the one behind the driver side tail lamp too... All sensor grounds are now less than 1 ohm to the - pole of the battery... I also checked, cleaned, and tightened the grounds at the dipstick stud (my - battery cable is hooked up to this stud as well, rather than a separate stud as suggested in the writeup) I also checked (but had no reason to suspect) the ground between the firewall and the block. I replaced the braid with a #4 cable shortly after I bought the jeep about 10 years ago.. Still good.. So, I then took the jeep for a test ride... I got EXACTLY 2 miles down the road and it died... Apparently the grounds were not the problem... :-( -The balast resistor was cold -- so its not the culprit.. -The fuel pump was still running, so not that either... --just to be sure, I pressed the pin in the fuel pressure test fitting -- healthy blast of fuel -- tells me fuel supply is good... --The CPS measured .2v and 230 ohms.... After a 1/2 hour wait, it started up again. So I stopped it, unplugged the CPS and hooked up the meter again to compare. The CPS was between .4 and .6v I plugged the CPS back in and drove home. 1/2 way back it started missing badly and it died 1/2 way down the driveway (1/4 mile long driveway) and I managed to coast back to the house.. -Fuel pump still running (of course, it stopped after a few seconds) -CPS measured .2v So, I figured, based on other posts I've read, MUST be the CPS.... AMAZING since I JUST replaced it not 50 miles ago or so... (See my OP) So, I got another CPS -- this time the red box standard motor products one... Today I installed the new CPS. I even used the suggestion in cruiser54's writeup to drill the upper hole in the CPS bracket to 3/8 so it can ride slightly closer to the flywheel and provide a higher voltage... Before I even plugged it in to the harness, I plugged it into my meter.. .7v So, I hooked it up and went for a test drive. When it first started, it ran smoother (and slower idle) than it EVER has since I've owned it -- actually ran like I would like it to. Within a few minutes of driving away it started having a nasty miss -- like as if one cylinder was firing really badly out of time.... Then at almost exactly 1 mile away, it died... After a few minutes I was able to restart it, but it only made it 1/2 way back before it died hard. No restart -- just like before, it will likely be HOURS before it will restart now... I dragged it back to the house with my other jeep... Its sitting there cooling now... BTW, just in case my NEW new CPS was bad, I checked it again. .8v - 1.1v... Still no start.. So, it would seem that CPS is NOT the problem... I then tried to test the other CPS :-) (Yes, there are two... The first is the one we all know, the Crank Position Sensor, in the bell housing..... The other is the Cam Position Sensor, in the distributor) So, does anyone have a how to for testing the CAM position sensor? It has 4 pins, and even on the highest resistance setting, my meter says they are all open (resistance so high no possible way for current to flow). The meter doesn't even flinch, it just stays at infinity no matter which two pins I touch with the probes.... That said, I do have some good news to report... It would seem that when the engine does decide to run, the throttle response is MUCH better with the fixed grounds (probably because the TPS ground was one of the super high resistance ones...) Anyone have any ideas where to go next? Thanks!

Yes, balast resistor gets very hot (figured it was just supposed to as the one in my points ignition boat does too) -- but when the engine stops, the fuel pump is still running (for a few seconds anyway). Bypassing the balast resistor makes no difference.. If I try to restart right after it quits, it will usually idle fine (only for a few min) but will die immediatly if I step on it at all... I'll read through your list and then post an answer -- sorry, haven't read your list yet... Thanks!

Funny you should say that.. The guys on the GM forums say the same thing about the GM ALDL bus..... :dunno: My scanner is supposed to be able to read ALDL, EEC, whaterever Chrysler calls theirs, GM obd 1.5 and all OBD2 + CAN I'm hoping the Renix stream is similar enough to something else to make it happy....

Or check with Uhaul. They said they had one available for my '88 Comanche a few years back...

Hi all! I have been having a problem with my 88 (4.0, Renix) lately. When it runs, it runs perfectly smooth with no hickups, no oil loss or burn, 22 MPG, just like it should (even with 214k miles on it...) However, after 20-30 miles, it will seem to lose power and within a minute it will stall and refuse to restart... Sometimes I can restart it and it will run for a few min (IF you feather the throttle AND with lots of hesistation and misfires that come and go), backfire, and quit again... Temp gauge does not change when engine dies -- couldn't say what the fuel gauge does as it regularly dances anyway... No tach on this model.. Temp runs between 200-210 -- needle sways back and forth as the thermostat opens and closes... Sometimes there is an odd burning smell when this happens -- but I have not yet managed to track it down... If I let it sit for an hour or so, it will go for another 20-30 min.... So, my first thought was the common CPS. And I was sure that I had found the problem, upon discovering the last 1/4" of the cable at the sensor end to be completly devoid of any kind of insulation what-so-ever... Well, I replaced it (cable is routed as per the diagram provided by cruiser elsewhere on this site)... And the problem goes on... Next thought was the fuel pump -- except that it RUNS when the engine will not start (to be fair, I tested this by cycling the key AND listened for it when the engine quit -- no change in sound, pump ran for a few seconds after the engine quit) -- and there is pressure in the fuel rail when the engine quits (sorry, can't say how much pressure as I can't seem to find a parts store that wants to sell me a pressure gauge for fuel injection... It squirts out quite readily when you press the pin though). I replaced the regulator AND the vacuum line to the regulator in case it was causing the pressure to go low... Does anyone have any suggestions? Also, I know that this Renix thing has a data bus, despite the fact that it loses its mind when you turn it off... Does anyone have a schematic with which I could build an adapter to be able to read this thing with my scan tool -- especially so I can use the live sensor mode to see what is not sending something to the compuer when this happens? My tool supports OBD-1 AND OBD-2 formats, and has cables for Ford, Chrysler, GM, and generic OBD-II... Also has OBD-1.5 GM cable... Thanks for any help!