Will not start, codes P0122 and P0622

You need to connect a scanner to view the data the Engine Computer is seeing and responding to. Manually passing a bar past either of the sensors is pointless. The air gap is critical to proper performance. Even if you could generate a signal that way, you can only do one sensor at a time. The computer needs to see both signals to turn on the ASD relay. It also needs to see those signals long enough for it to determine when piston number one is coming up on top dead center. On some engines that can take as long as three crankshaft revolutions before it will start firing ignition coils and injectors.

Not really. We're in a transition period with a 2003 model. Up to at least the late '90s, you had to have solid signals from both sensors for the computer to turn on the ASD relay. If either signal was lost while driving, the computer turned the relay off and the engine would stall.

By somewhere around 1999 to 2004, both signals were still needed for the engine to start running, but once it was running, it would continue to do so if the signal was lost for one sensor. There was a back-up strategy that based injector and ignition timing on the one good signal. The missing signal adversely affected emissions, so the fault code that was set did turn on the Check Engine light. Once the engine was turned off, it would not restart with that failed sensor. On some engines, GM V-6s in particular, the engine could start and run with a failed camshaft position sensor, but you only had a 33 percent chance of that happening. The computer still knew when a piston was coming up on top dead center, but it had no way of knowing which one. It took a guess, then just calculated the next pair in the firing order to know which coil to fire next. If it guessed right, the engine would run. If it guessed wrong, it was firing the ignition coils at the wrong time. To take a new guess, you had to turn the ignition switch all the way back to "off", then back on and try again Each time it would guess at the stating point, and when it finally got it right, the engine would run.

On still later models, some engines will start and run on just one sensor signal, but it won't run at its best. Exact symptoms vary by brand and model. There's a series of notches, or holes, in a ring on the flex plate that are read by the crankshaft position sensor. I've seen groups of three notches for each pair of cylinders, and groups of four are common too. Chrysler used some that had three notches, three notches, then four notches. That extra notch produced one extra signal pulse that told the computer when piston number one was coming up on top dead center. With that design, the computer can still know which ignition coil to fire when the camshaft position sensor fails. As a side note, this is where a lot of people run into trouble when switching engines from a different model year. The Engine Computers will only work with the correct tone ring on the flex plate. If the number of notches is different, you'll have a crank / no-start condition. Always keep the original flex plate with the computer for that vehicle.

The point of this is with the older version that needed both sensor signals, it was a simple matter of measuring the voltage at the ignition coil or any injector. If 12 volts was there during cranking, the ASD relay had been turned on and we knew both signals were there.

For the next few model years, that still applied to crank / no-start conditions. If the 12 volts didn't come back during cranking, one of the sensor signals had to be missing. That took less than a minute to determine with the scanner, but remember, that only tells us which circuit needed further diagnosis. Most likely it was caused by a failed sensor, but it could just as easily be caused by a broken wire or corroded connector terminals.

This is where things can get a little confusing. The computer could be turning on the ASD relay, but with one missing signal, it won't know when to fire an ignition coil or injector. This is why we can't say for sure both sensors are working just because the ASD relay is on and we see the 12 volts at the coils and injectors.

To add to the misery, there's no easy way to measure those signals. That could be done with an oscilloscope like I used in tv repair, but those are a lot more cumbersome and time-consuming to hook up than a simple scanner. You still have to interpret the waveforms to see if they look right. With the scanner, you just let the computer tell you if it's unhappy with one of the signals.

The first thing I would pursue is to check at some local auto parts stores to see if they have a scanner you can borrow. Many of them rent or borrow tools. Next would be to find a mechanic with his own scanner who is willing to stop by and check this for you.

First let me tell you what I know about the cam and crank sensors. As with any electrical generator, you have to have three things to generate a voltage. That's a piece of wire, a magnetic field, and most importantly, movement between them. Many loops of wire is more efficient. Alternators, for example, use a coil of wire to create the magnetic field, and three groups of stationary coils that the voltage is "induced" into, then the movement comes from spinning the electromagnetic coil with the belt and pulley.

In position sensors, the metal core is a permanent magnet. The coil of wire is wrapped around it, but the movement is created differently. A toothed "tone ring" or a ring with holes in it passes by the tip of the sensor. The varying metal and spaces disrupts the magnetic field that the coil of wire is sitting in. That movement of magnetic lines of force are what causes varying voltages to be induced in the coil. In its basic form, that signal can be used to tell the computer when an event has occurred. That type of sensor has just two wires and the signals they develop look somewhat like the sine wave that you'd see in house wiring. The voltage will vary gradually between a maximum positive voltage, then transition over time to a maximum negative voltage. The faster the engine speed, the higher the voltages will be developed. These are really good for speed sensors because all we're interested in is the number of events. We don't care exactly when the events occur. These worked okay for position sensors years ago because they were consistent, and we had adjustable distributors to let us set ignition timing to specs.

Much more precision is needed today with engines that have no provision for adjusting ignition timing. The heart of the position sensors is still the same with the magnet and coil of wire, but electronic switching circuitry has been added. The induced voltage from the coil of wire reaches a point at which the circuitry switches from one state to the other very quickly, so instead of the irregular sine wave, we see a square wave with nice straight "rise times" and "fall times". Those switching points are very precise and work perfectly for timing spark and injector pulses. The internal circuitry needs a power source and a ground, and then the third signal wire going to the computer. Most Chrysler computers use 5.0 volts to power the sensors, but other models could use 8.0 volts or 12.0 volts. In a few rare cases GM used 10.0 volts.

The reason I didn't comment on your testing the signal voltage right at the crank sensor is there's two different ways they can act. With one type, the movement comes from the tone ring moving past the magnet. The magic word is "moving" past the magnet. It doesn't matter if a tooth or a gap is under the tip of the magnet. If it isn't moving, no voltage gets developed in the coil. Depending on the design of the internal circuitry, the signal voltage will remain at 0.0 volts or at 5.0 volts even though you're turning the crankshaft by hand. If it helps, imagine you're holding a lit candle. The flame is going straight up from it. It doesn't matter if you're standing over here or over there; the flame is still going straight up because there's no movement on your part. This is the type of sensor that people might mistakenly call defective because they don't see a change in state as they rotate the crankshaft. They can't turn it fast enough by hand.

The second style has circuitry that generates its own high-frequency alternating voltage that is sent to the coil of wire. By changing polarity and / or voltage, that is the movement that interacts with the permanent magnet. The tooth or notch on the tone ring gets in the way of that interaction and varies how much of that voltage returns back to the circuitry. With this type of sensor, there is a very definite point at which switching occurs between 0.0 and 5.0 volts. This is where, if you measure the voltage on the signal wire, you'll see it switch state as you turn the crankshaft very slowly in either direction.

With the first design, there is a minimum speed the tone ring must be moving before a signal is generated by the sensor. Cranking speed is fast enough to accomplish that, but that might not happen if the engine is cranking too slowly due to a run-down battery. Or one with a dead cell.

As for the feed and ground wires, you'll notice for all the sensors, none of the ground wires actually go to ground directly. They usually share a ground wire in common with multiple sensors, then they go through the Engine Computer on their way to ground. That way the computer can monitor that circuit. As a result, you'll typically find 0.2 volts on that wire. The clue that there is a break in that ground wire is you'll have a whole bunch of fault codes related to all the sensors that use it. When you have a fault code for just one sensor, it's just that one and its signal wire that are the suspects.

The 5.0-volt feed circuit originates inside the computer and is very carefully regulated. It has to be stable because any variation will transfer to a corresponding variation in signals from sensors that develop a signal voltage related to what they're measuring. For example, at half-throttle, the throttle position sensor will develop a signal voltage of 2.5 volts. If its feed circuit went up to 5.2 volts, the signal voltage would show up as 2.6 volts. That one tenth of a volt isn't real serious for that sensor, but the same thing can happen with the MAP sensor. That one provides the main fuel metering calculation, and a few hundredths of a volt difference means a real lot to the computer and to how well the engine runs.

It is almost impossible for a sensor to develop a dead short between the 5.0-volt feed and the ground circuits, but it is entirely possible for the 5.0-volt wire to become grounded to the engine or body sheet metal. If that happens, all Chrysler computers will detect that and shut the supply down to protect it. That will result in a crank / no-start due to none of the sensors developing their signals. Once the short is located and repaired, you have to turn the ignition switch off, then back on, to reset that power supply. It won't come back up to 5.0 volts by itself until you do that.

The final authority as to whether the cam and crank sensor signals are showing up during cranking and have correct and usable waveforms is the Engine Computer. The only way we have of getting that information from it is to talk with it through a scanner. This can also be why it is possible to get enough of a signal during cranking to avoid setting a fault code, but the signal is not correct to cause the computer to fire coils and injectors. That's why I often mention that these defects may only be detected when a stalled engine is coasting to a stop, but not just from cranking the engine.

The starter system can be broken down into three circuits; the low-current, the mid-current, and the high-current circuits. The ignition switch is in the low-current part. Those tiny switching contacts only have to pass a small fraction of an amp to turn on the starter relay. That's the path shown in yellow in the second diagram.

The second half of the starter relay is the beginning of the mid-current circuit. Those contacts have to pass as much as 20 amps, but most commonly around ten to 15 amps. That current goes through the solenoid on the starter, in this case through the light green wire you asked about.

You won't be tested on this later, but this is where it gets interesting, and confusing. The third diagram shows the mid-current path. Once the starter relay has been energized, the contacts switch on, (red arrow), then that 20 amps flows through the light green wire to the smaller of the two terminals on the solenoid. Once inside the solenoid, current flows through two circuits. Each circuit has a very strong electromagnetic coil. Current flows through the green circuit, to include the "hold-in" winding, coil number "2". That one is energized all the time the relay is turned on.

The other half of the current flows through the "pull-in" winding, coil number "1", as shown with the blue line. From there it flows through the starter motor itself, then to ground. The starter motor has extremely little resistance to current flow, so for this part of the story, it's like it isn't even there. It acts exactly like a piece of wire. These two coils do two important tasks. The first is they create such a strong magnetic field that an iron plunger is pulled through them to push the starter's drive gear into mesh with the ring gear. Second, once the plunger gets there, a thick copper disc connects a pair of copper contacts to pass the very high current through the starter motor. The reason for doing it this way is it takes a lot of effort to push the drive gear into mesh. So much, in fact, that it takes the combined strength of both of the electromagnetic coils.

Once the plunger has moved all the way and the high-current contact is turned on, the fourth drawing shows that current path. The fat red line shows the path from the battery, through that internal contact, through the starter motor, then to ground. Here's the part I promised to not test you on. If you look at the pull-in winding, coil number "1", you'll see it still has full battery voltage on one side, point "A", coming from the starter relay. Now it also has full battery voltage on the other side, point "B" coming from the larger battery terminal. It has the same voltage on both sides. The net difference is 0 volts, so no current will flow through it. In effect, that coil is switched off once cranking starts. It was needed to help pull the drive gear out, but the hold-in coil can keep it there by itself. That turns off up to as much as ten amps and makes that current available to the starter motor. That few amps might be just what is needed to get an engine cranking fast enough to start on a cold day.

For this last comment, we understand there's a lot of other relays clicking on and off under the hood, so this observation isn't as valuable as it was in the past. Basically, if you hear or feel the starter relay click when you turn the ignition switch to "crank", we know the low-current circuit is working. That is the circuit that will include the neutral safety switch, or in this case, that function is done by the Engine Computer.

If you hear the fairly loud clunk of the starter solenoid's plunger engaging, we know the mid-current part of the system is working too. If the starter doesn't spin the engine at that time, the defect is in the high-current circuit. On Chrysler and Toyota engines that use the little silver Nippendenso starters, that is very commonly caused by the internal switching contacts being burned away. That is a repair that can be handled by a competent do-it-yourselfer. For all other vehicles, that is most commonly caused by a corroded or loose battery cable.

So, that light green wire is a switched power wire. Everything is grounded inside the starter, and that is bolted to the transmission, then current continues back to the battery on the negative battery cable, also bolted to the engine or transmission.

Where people's thought train often gets derailed had to do with ignition system bypass circuits used years ago. A resistor was used in the ignition coil's circuit to lower current and voltage to the design values. During cranking, battery voltage gets drawn down to less than 12.6 volts; often to as low as 10 volts. That lowers the voltage feeding the ignition coil even more and would result in a very weak spark. That resistor is bypassed during cranking to get spark voltage back up to normal. Chrysler did that with an additional tap on the ignition switch. GM did that with a second smaller terminal on the starter solenoid. Ford also did that with a second smaller terminal on their fender-mounted solenoids, but they really confused the issue by using at least three different solenoid designs, and they are not interchangeable. Some use that second terminal for the ignition resistor bypass. Some use it for the neutral safety switch in the low-current circuit. Some don't even have that extra terminal. To my knowledge, Chrysler never had a third terminal on the starter solenoids.

That bypass circuit isn't used on newer cars. It is not exactly needed on any car for the engine to run. That circuit is used to make starting easier, especially on cold days or with tired batteries.

If an injector is stuck open, fuel pressure will bleed down within a minute or two when the engine is turned off. That causes a really long crank time later to restart the engine.
Fuel pressure has to build up high enough to spray from the injectors, and the pump is already running slower than normal from battery voltage being drawn down by the starter.
That makes fuel pressure build slower than normal. I've been involved with two cars with cut o-rings on the fuel pressure regulator causing pressure to bleed down within a few seconds. Both took a good 15 to 20 seconds of cranking to get the engines running.
Normally pressure should hold for weeks. If one injector is leaking, that won't cause the engine to stall because only the one cylinder will be getting too much gas. You'll see black smoke from the tail pipe.

Starting and running for just a couple of seconds is what happens when it's in theft mode. I do know that on my '93 Dynasty, if I reconnect the battery that was disconnected for storage, the system defaults immediately to theft mode with exterior lights flashing, but it doesn't blow the horn. I have to use the key in either front door lock to lock, then unlock the door, then the engine will run. The fuel pump is disabled in theft mode, but it still runs for that initial one second when you turn on the ignition switch. That gets fuel pressure up to normal, then the engine will run on that gas under pressure in the rubber lines. That pressure is used up in about two seconds.

There's one thing that's real common, but it never occurred to me yesterday. That is with all Chrysler products up to somewhere around perhaps 2010 to 2015 models, any time the battery or Engine Computer is disconnected, the computer loses its memory. Once the engine is restarted, most of the fuel trim data and sensor personalities are rebuilt right away without you even noticing. The one exception to that is "minimum throttle". That requires a specific set of conditions to be met for that to be relearned. Until that is done, the engine may crank but not start or run unless you hold the accelerator pedal down 1/4". You won't get the normal "idle flare-up" to 1500 rpm at start-up, and it will tend to stall at stop signs. For most do-it-yourselfers, the problem mysteriously goes away in a day or two, but the real fix is to meet the conditions needed for the relearn to take place, drive at highway speed with the engine warmed up, then coast for at least seven seconds without touching the pedals.

Under the right conditions, the injectors will be commanded to squirt a priming pulse when you turn on the ignition switch. That can introduce enough fuel for the engine to run for a few seconds, but then it can stall due to low idle speed. The real easy way to identify that is to simply hold the accelerator pedal down 1/4" and observe the engine stays running.

Once minimum throttle is relearned, the computer has enough control of idle speed to maintain it with more than half of the cylinders not firing.

Another way to know if minimum throttle needs to be relearned yet is to view the "idle steps" on a scanner. GM and Chrysler used the same automatic idle speed, (AIS) motor, but they call them different names. That is not a motor with brushes like we normally think of them. This is a "stepper" motor. The computer pulses the motor's four electromagnetic coils with varying voltages and polarities to place the armature in a specific position. As the step number changes, the armature rotates in small increments. As it does, a pintle valve on the end of the shaft retracts or extends to expose more or less of an air bypass passage around the throttle blade. The computer adjusts how long it pulses the injectors open to match fuel to the amount of air. For a properly-running engine, it's typical to find the AIS motor on step 32. The highest step number is 256, but you'll never see it get that high. I watched a Chrysler instructor do this on a V-8 Jeep engine. He disabled more and more injectors until only one cylinder was still firing. The engine obviously ran very poorly, but it still maintained the desired idle speed and wasn't even close to step 256.

Typical for a V-8 engine with one cylinder misfiring is to find the AIS at around step 50. The clue that minimum throttle hasn't been relearned is it will be stuck on step 0. The air bypass valve is fully-closed. That's why you have to hold the accelerator pedal down about 1/4".

Numbers given in the service manual are for explaining theory of operation, and are used as a guide. Some people replace fuel injectors because they're off by a few ohms, but you're just as likely to find injectors that are working perfectly fine and also don't measure within specs. For most things with coils of wire, we're interested in whether they measure shorted, open, or something in between, and that something in between is almost always okay even though it isn't exactly as listed in the service manual.

I drew you some coils of wire to help explain what can happen and what can't happen. The first one shows a typical coil in something small like the items I listed. There will usually be well over a few thousands loops of very fine wire coated in varnish for the insulation. That prevents the loops from being shorted to the adjacent loops.

The second drawing shows the coil is shorted. It's almost impossible for that to happen internally to the coil itself. You're more likely to find corrosion built up between adjacent terminals in the connector.

The third one shows the wire has broken off one of the terminals. This is the only failure that's pretty common. It is also common for the break to be so small that the wire touches the terminal and the item works intermittently. Distributor pick-up coils commonly fail in this way when they get warm, then work again after cooling down for an hour or two. To be valid, any testing has to be done while the problem is occurring.

The fourth coil shows the least common failure where some of the loops of wire are shorted together. The varnish had to be compromised, either from overheating or moisture got inside and corrosion started to eat the wire away at a scratch or nick in the varnish. While building "bugs" into a car for my students to diagnose, I left a Ford starter solenoid hot-wired for about three minutes. Those are meant to be energized just long enough to get the engine started. It got my attention when it started smoking. It still never measured lower in resistance, but it did warp the coil to the point the plunger couldn't slide freely to engage. I turned it into junk but it's resistance was still normal.

Drawings 1 and 3 are what you're going to run into on a regular basis. It's good, regardless of its actual resistance, or it's open. One quick measurement is all that's needed. The defects in the second and fourth drawings result in resistance lower than normal. The fourth one won't develop its full magnetic field so it could operate sluggishly and / or intermittently. It might only operate when the engine is running due to charging system voltage being a volt or two higher than battery voltage. This can also be why a starter solenoid engages when the vehicle is getting a jump-start, but not when trying to crank just on its own battery. The important point is resistance didn't go higher than normal in the three defects except for the open circuit.

There's three things that affect the coil's resistance. One is the type of material, but that is always going to be copper, so that variable is eliminated. Every wire has some resistance per foot, so the longer the wire is, the more resistance it will have. As a wire gets fatter, it's easier to get more current through it, so that tells us resistance had to go down. The wire's resistance is affected by its length and diameter, and neither of those things are going to change. The only way for the coil to measure higher than specs is it has to get longer or thinner. Since those things can't change, it's unrealistic to call a part "defective" because it measures higher resistance than what's stated in the service manual. The part worked when it was new, so if it's not working now, it has to be due to some other cause. That's why measuring injectors and other parts is a waste of time. I know some instructors who still teach measuring parts because it is listed in the service manual, but if you're doing warranty work at a dealership, you won't get paid for wasting time with these kinds of tests. What they are good for is doing an autopsy on a failed part to learn why it failed or why it caused the problem it did, but you would do that on your lunch break.

The same things apply to your idle speed motor, but having peeked inside a few, those coils use rather fat wire, so resistance is going to be a lot lower. I didn't think it would be as low as ten ohms, but I wouldn't question that or be concerned with that reading. There's four identical coils inside that unit. Unless they're doing something inside it that I'm not aware of, you should get very nearly the same reading between any two terminals.

Since I'm in a drawing mood, I drew you an AIS motor in the second drawing. To save time, I used colored rectangles to represent the four coils of wire. We'll use your ten ohms for this description. If you know how to calculate values with Ohm's Law, this will make sense. If you don't know what that is, you'll have to take my word for it.

If you measure between terminals 1 and 2, the current supplied by your meter will flow through the blue coil, but a little will also flow through the orange coil, then the purple coil, and then then green coil. Those three coils are in "series" during this test. Their total resistance adds up to 30 ohms. Unless you're a circuit designer, doing the math is of little value. With two paths for current to flow, overall resistance is lower than just that of the blue coil. Okay, I did the math anyway and came up with 7.5 ohms. What is important to know is you should find very close to the same reading when you measure across any one of the coils. That's between terminals 2 and 3, terminals 3 and 4, and between terminals 4 and 1.

If you measured between terminals 1 and 3, there will be two identical paths for current to flow. One path is through the blue and green coils. Each is ten ohms, and being in series during this test, that adds up to 20 ohms. A second identical path exists through the orange and purple coils, also a total of 20 ohms. In effect, that puts two 20-ohm resistances in parallel. Two identical resistors in parallel equals half of one of them. Half of 20 is ten, so again, you'd find ten ohms during this test.

You'll also have a good two to five ohms of resistance just in your meter's leads, so you have to subtract that from your reading to find the actual resistance of what you're measuring. To say that a different way, if your meter leads have two ohms of resistance, and your reading is 12 ohms, subtract the two ohms to find the item under test has ten ohms.

Where you might find value in this test is lets suppose there's a break in the blue coil. When you're measuring any of the other coils, there won't be that second alternate path for current, so you'll be reading just the one coil under test; that's 10 ohms. The Engine Computer isn't going to differentiate between 7.5 or ten ohms, so no defect will be detected, and no fault code will be set. The symptom would be the motor just vibrates but it won't rotate, so there will be no control of idle speed. This is very unlikely to happen since the coils use such beefy wire.

You're much more likely to run into a break in one of the four wires going to the AIS motor, or a terminal corroded off inside the connector. With that type of defect, the computer will see the no-current flowing through that wire. That will set a fault code related to an "AIS circuit defect". I don't think the codes are specific enough to tell you which wire is broken. Fault codes just inform enough to get us into the right area that needs more diagnosis. Then, this would involve measuring the resistances of those four wires. For that we're interested in continuity, and not the actual resistance, but here again, there will be a few ohms of resistance in each wire. That is insignificant as far as the computer is concerned.

There's two more-effective ways to test an AIS motor on a GM or Chrysler product. The easiest is to just listen to it. Sometimes they can get out of sync with where the computer thinks it is, so at some point, it will run it down to step "0", and a little more to insure it's fully closed, then it will open it up to roughly step "50" in preparation for the next engine start, to provide that "idle flare-up" for a few seconds. I seem to notice that taking place on GM engines when they're stopped. I notice it at times on Chrysler engines when the ignition switch is turned on. It sounds like what a model railroad locomotive sounds like, but only for one or two seconds.

The second way to test them again requires a scanner. That allows you to select an idle speed up to 2,000 rpm in 200 rpm increments. The scanner will command the Engine Computer to raise idle speed to what you selected. If engine speed responds accordingly, the AIS system has to be working.

We don't see this much any more due to better additives in gas, but we used to run into the air bypass passage plugged with carbon. Even though the AIS motor was working and the valve was opening, no additional air could get in to raise idle speed. When this was common up to the early '90s, diagnostic fault codes weren't advanced enough to detect things like that. We had to go by the description of the problem, which was low idle speed, hard starting, and frequent stalling when coming to a stop. Cleaning out that carbon took about a half hour to an hour.

Some older ones, particularly the '90s Caravans needed a thick paper spacer stuck on the end to set the air gap, but those had a slotted mounting hole to allow for that adjustment. Newer ones use a metal bracket with two mounting holes, or the sensor has an ear with the single mounting hole. Those don't need the paper spacer and the gap is set automatically.

We were always warned that failure to use the spacer when required could allow the crank sensor to be pushed in far enough that the tip of the magnet would be hit by the ring on the flex plate, and that could break the sensor. I had one that I thought I was too smart to need that spacer. I pushed it all the way in, then pulled it out what I thought was the right amount. Two weeks later it was back with a complaint of intermittent stalling. Not knowing the recent history, someone else put in a new sensor and solved the problem. That problem may never have occurred if I had used the spacer like I was supposed to.

Both engines use the same crankshaft position sensor shown at the top left of this photo.
The air gap is set with this mounting ear. Your camshaft position sensor mounts the same way and can't be installed incorrectly.

I don't mean to confuse the issue, but the lower photos show the camshaft position sensor for the 3.5L engine. This one does need the spacer. It's shown attached to the bottom of the left photo, and the replacement is shown at the right side for when you want to reinstall a good sensor that was removed for some other service. In the middle, the arrow is pointing to the elongated hole for the mounting bolt.