Code P0122 TPS low circuit input

First you have to be aware that for any voltages to be accurate, they must be taken with the connector plugged into the sensor. Back-probe through the rubber seals on the back of the connector. If you unplug it and read on the terminals, the 5.0 volt feed will be correct, and the ground wire should have 0.2 volts, but the signal wire is supposed to go to 0.0 or 5.0 volts to force a fault code to be set. By leaving it plugged in, you should find between approximately 0.5 and 4.5 volts, from idle to wide-open-throttle. If you do, but you're still getting a fault code, you'll need a scanner to see what the computer is seeing for throttle position sensor voltage. If it's different than what you find at the sensor's signal wire, suspect that wire is broken or cut.

If you find 5.0 volts on the signal wire, since we can rule out the sensor, suspect that terminal in the connector is spread and not making good contact.

You can't have 12 volts on the ground wire. The only way for that to happen is if it is shorted to a 12 volt wire in a harness. That ground wire is shared with other sensors, so you would find the same reading on those other sensors. Rather, I suspect you have something wrong in the way you're taking the measurements. Check that the meter's ground probe isn't on the battery's positive terminal.

If you get a changing voltage when you tap on a sensor, it is much more likely one of the terminals is stretched or has a light film of corrosion on it. The only sensors that have a history of acting up from vibration are older GM mass air flow sensors and early to mid '90s Ford coolant temperature sensors. What is more likely to happen with any throttle position sensor is a small glitch in the voltage when you slowly move the throttle position. Those almost always occur way to quickly to see with a digital voltmeter because they respond too slowly. Even the best scanners don't update their live reading displays fast enough. The only thing that CAN detect those brief changes is the Engine Computer. It won't care about little voltage fluctuations. That's what it's used to seeing when your foot vibrates on the accelerator pedal on bumpy roads. What it DOES catch is when the movable contact has a momentary bad connection due to a piece of carbon that chipped off and lifted the contact away from the carbon strip. On almost all TPS circuits, there's a "pull-up" resistor that takes over and places 5.0 volts on the signal wire to force a fault code to set. It's very unlikely that will occur long enough to see with a voltmeter or scanner, but the computer will see it.

For training purposes, the standard acceptable range of signal voltage is 0.5 to 4.5 volts, corresponding to idle to wide-open-throttle. In actual practice you can expect to see roughly 0.4 to 0.7 volts at idle, and typically 3.9 to 4.4 volts at wide-open-throttle. Those numbers will vary between any two identical sensors and any two identical vehicles. 0.2 volts on your signal wire is rather low, and that would be on the threshold of setting a fault code, but that might be an acceptable value on your truck.

As I mentioned previously, one of the possibilities was that you used the wrong procedure which gave you the wrong value, and I was right. That's because you followed flawed instructions. In fact, I take issue with a lot of what I saw in that article.

First of all, to test the ground wire, they had you put the meter's lead on the battery positive terminal. Of course you'll have 12 volts that way, which is what you listed. That IS a valid test if you're testing a ground wire with a test light which needs voltage to work. It is not the way to take a normal voltage reading. Think of starting with 100 pounds of water pressure in a pipe. You use a gauge connected to the source and to a point a mile away and find the difference is 20 psi. Therefore, you can calculate you actually have 80 pounds at that point. Who would do that? You would just put a gauge at the point and measure 80 pounds directly.

I understand the thinking, but voltage is electrical pressure, and it is always measured in relation to another point. On cars we measure in relation to ground which is the engine block and the battery's negative post. That is how we take electrical pressure readings directly. Now, to be fair, testing the way your author said does have a legitimate benefit, but it wasn't stated, and that's why most people are going to be confused if they have some understanding of basic electrical theory. A ground wire is supposed to have 0.0 volts on it, and it will if it isn't "open", as in cut. Even if it's cut, you'll still have 0.0 volts on the "good" side of the wire, but there will be some voltage on the other part of the wire on the other side of the break. To add to the confusion, he gave you an awful lot of leeway on what's acceptable for the 5.0 volt supply voltage, and that implies you can have the same leeway on the ground wire. That means if you found 11.5 volts on it, that's okay. In fact, that is absolutely not okay. First of all, the 5.0 volt supply is very carefully regulated on all cars, and I'd raise an eyebrow if I found 4.8 volts, much less 4.5 volts. The same precision is needed on the ground wire. As I mentioned earlier, you should find 0.2 volts on it. Normally that two tenths of a volt indicates a loose or corroded connection, and it's that very tiny voltage that identifies bad connections in high-current circuits like starter circuits, but in this case, that 0.2 volts is dropped across the monitoring circuitry inside the Engine Computer, so it is normal. 0.4 volts is not normal and indicates a problem. Usually that's not an issue because most Engine Computers have four ground wires. Each pair is for a different purpose, but two wires are for the sensors, and failure of both isn't common.

Another problem with using the battery's positive post as the reference is there can be some variation in battery voltage. A fully-charged battery will read 12.6 volts, so if the ground wire has the normal 0.2 volts, you're going to measure 12.4 volts. If the battery was just charged with a portable charger, the surface charge can make it read as high as 13.0 volts. If it is partially run down, it might read 12.2 volts. All of those are legitimate values and do not indicate a problem, but when used as the reference for other voltage readings, you can see how they could vary by six or eight tenths of a volt. When only 0.2 volts, (related to ground) is acceptable, you can't interpret a reading as good or bad when it can legitimately vary that much.

Measuring the voltage on the ground wire my way will give you 0.2 volts. 0.0 volts, and 0.4 volts are not okay. Doing it as described in the article, you said the author said "11 to 12 volts is okay". No it isn't. The only acceptable reading would be 12.4 volts, but you still have the battery voltage variable to make the reading useless.

So, instead of following the directions to move the meter's ground probe to the battery's positive terminal, leave it on ground or the negative post, and take all three voltage readings directly. If you want to experiment, after you read the signal voltage with the plug connected to the sensor, unplug it and read it on the terminal in the connector. The first way you'll find close to 0.5 volts. The 0.9 volts the author mentioned is too high. With the plug removed, you should find the full 5.0 volts, but on some vehicles it will go to 0.0 volts. There will be a pull-up or pull-down resistor to force the signal voltage to go to one of those maximum values and set a fault code. That way it can't "float" to some random value that the computer would try to use.

The other red flag in that article was when the author said to pierce the wire insulation. That made the hair stand up on the back of my neck! There are test probes that help you do that, but in my Electrical classes I watched for students doing that very closely on my prepared cars. If they did, they were allowed to replace that entire piece of wire from one terminal to the other. Most importantly, that hole is where moisture is going to sneak in and corrode the wire. Before it breaks completely it will develop some resistance and a circuit can still work, but not correctly. In a high-current circuit that could mean a starter motor that cranks too slowly, but it still cranks, or a heater fan or wiper motor that doesn't run as fast as it's supposed to, but it still run's. In sensor circuits, an error of as little as 0.01 volt can affect what the computer thinks is happening and how it reacts.

People who use "Scotch-Lok" connectors for trailer wiring run into the same problems because they don't seal out moisture. To add to the misery, GM used aluminum wires on some of their vehicles, and those corrode very quickly. If you pierce that insulation, you can expect to find a broken wire within a few weeks.

The other issue on my "bugged" cars was a wire with a hole in it would be a clue that someone else was looking at that wire when they were trying to diagnose the problem previously. Absolutely do not poke holes in any wires. If you have already, use some RTV gasket sealer from a small tube to seal the hole when you're done. Electrical tape should never be used on a car or truck. It will unravel into a gooey mess on a hot day.

The last concern is your last sentence, "I was able to test the OHMS and the readings for all 3 wires is.7mV." Ohms and volts are two different things. If you simply mixed up the terminology, I'll get over it, but voltages are read in a circuit that is powered up and working. Resistance, (ohms and kilohms), must always be read in a dead circuit. The meter has its own internal battery for that purpose.

What I suspect you were doing was measuring the continuity of the three wires between the sensor and the computer, and you found.7 ohms. You'll actually have more resistance than that just in the meter leads, so 3 or 4 ohms would not be unexpected. Regardless, you don't have to do those tests unless the preliminary voltage readings show to suspect a broken wire. If any one of those three wires did not pass its continuity test, you would have found a seriously incorrect voltage on that terminal at the sensor. The voltages can only be correct if the wires are okay.

Also watch the "volts" and "millivolts" designations. There's a big difference between the two. If you have an auto-ranging meter, it can be real easy to overlook the scale the meter picked for you. I have over a dozen digital meters I used in 35 years of tv repair, and I never used an auto-ranging meter. It is way too easy to make a mistake when you're in a hurry and taking lots of readings very quickly.

Dandy. The ground and 5.0 volt feed wires are perfect. Look at the signal wire and its 0.2 volts. That's the one that must be between approximately 0.5 to 4.5 volts. Less than 0.5 volts is what will trigger the fault code you have.

For your resistance readings, 7 ohms is acceptable. That would be excessive for most high-current circuits, but sensor circuits have extremely little current flow, so that resistance isn't a factor in its operation.

What we DO have to do is figure out why there's 0 volts on the signal wire. That wire could be shorted to ground, or there could be a break in that wire AND your model uses a pull-down resistor to take the voltage to 0 volts to insure it sets a fault code. Since you checked the continuity of the wires already, we can rule out a break in the signal wire. There still could be a stretched terminal in the connector. In fact, that is the most common cause of this fault code.

There's two tests that should help find the cause of the code. First, unplug the connector, then measure the resistance between the center signal terminal, and ground. You should find an open circuit. Your meter will give the same indication as when the meter probes aren't connected to anything. Now switch to one or two higher scales. You still should get open circuit readings, although on the highest resistance scales you may get some reading due to all of the other circuitry inside the computer. That is normal and acceptable. If you find a low reading on the lowest scales, you'll have to inspect the harness to see where that wire is grounded.

The second test involves checking for continuity between the connector terminals. To do that, reconnect the plug, then back-probe through the rubber seals around the wires to take the readings. The actual value between the ground and 5.0 volt feed terminals is irrelevant but you can expect to find around 5,000 to 10,000 ohms. That's 5k to 10k. You only need to do one test on the signal wire, (because if it tests good on one, it will on the second test too), but if you do the two following tests, the total will add up to the resistance you just measured between the ground and 5.0 volt feed wires. Measure from the signal wire to the ground wire, then from the signal wire to the 5.0 volt feed wire.

As an example, if you find 8,500 ohms between the ground and 5.0 volt feed wire, you might find 500 ohms between the ground and signal wire and 8,000 ohms between the 5.0 volt feed wire and the signal wire. Those two numbers will change a lot of you place the throttle in any position other than idle, but they'll still add up to 8,500 ohms, in this example.

There's going to be quite a bit of tolerance, or leeway in the numbers. Don't be concerned if they don't add up exactly. To be off a couple of hundred ohms is not significant. We're looking for an open circuit.

Voltages are always taken with the ignition switch on. Resistances are always taken with the ignition switch off because the meter supplies its own power. In fact, if the circuit is powered up, resistance readings will mean nothing.

Your readings have me scratching my head. Here's what I'd like yo to do. Unplug the connector, then measure right on the sensor's terminals. You will find some resistance value between the 5.0 volt feed and ground terminals. 5,000 ohms is typical. You may find as much as 10,000 ohms. The exact value is not important, but you should not find an open circuit, (OL), or a real high resistance, like in megohms.

Next, measure from the ground terminal to the signal terminal, then from the 5.0 volt feed to the signal terminal. Those two readings should add up to what you found in the last paragraph.

If it will help, write all these values down. Now reconnect the plug, then measure the same three values by back-probing through the rubber seals where the wires go through. You must find the same values. If you find an open circuit on any measurement, you either aren't making good contact with the meter probe, or that terminal is stretched and not making contact with its mating terminal on the sensor. If a terminal is stretched, you can usually pull the plastic locking wedge out of the connector. That will allow you to reach inside with a pick to lift a locking tab which will allow you to pull that wire and terminal out. Then you can squeeze the part that touches the sensor's terminal. If it's beyond that, you can get a whole connector or just the terminal with some wire from a salvage yard. The third option is to buy a new plug from an auto parts store or a new terminal from the GM dealer's parts department. The auto parts stores have huge catalogs now listing thousands of replacement connectors because with all the insane computers, terminal problems have become very common, and most mechanics won't waste the extra time, or don't know how to repair individual terminals. Even when they do, they don't like to risk having a comeback if their repair doesn't last. GM dealers have huge kits with dozens of different kinds of terminals.

If all of your resistance readings are okay, unplug the connector, turn on the ignition switch, then measure the voltages on the three terminals in that connector. Be careful to just touch the terminals with the meter's probe. If you push it into the terminal, that will spread it and create an open circuit. You must find 5.0 volts, and 0.2 volts on the ground wire. You might find 0.0 volts on the signal wire, but typically you'll find 5.0 volts on it. Nothing else is acceptable at this point. If you do find something in between, you'll need a scanner to see exactly what the computer is seeing. If you don't have access to one, I just found one on eBay for less than 50 bucks that will get the job done.

If all of those voltages are okay, reconnect the plug, then back-probe the wires and take the readings again. This time the only one that should be different is the signal wire. If it is still the same as what you found with the connector unplugged, that terminal is stretched.

Write all those values down, then holler back.

In your first two sets of readings I think you have two terminals mixed up. The signal to ground resistance and the signal to 5.0 volt resistance should equal the ground to 5.0 volt resistance, which they would if you turned around the last two readings in both groups of three. In the third set, with the connector unplugged, and reading right on the terminals, resistance readings have no value except when we're looking for a short. There's no way to tell what is normal or correct because you're just reading the circuitry inside the computer.

In your next reply where you found.74 volts on the signal wire, that is just about perfect, but be sure that's really three quarters of a volt, and not 75 millivolts. Auto-ranging meters always trick me by making the scale hard to determine and notice. .74 volts is a typical value at idle. If you're on the right terminal, you'll see that voltage go up when you work the throttle. It should be around 4.2 to 4.5 volts at wide-open-throttle. If you have 75 millivolts, that will set the fault code for signal voltage too low.