Engine sensors and PCM

I suspect this will not work, but first, Best Buy, if you can actually find a three-terminal regulator there, is charging way too much. Searching the web sites for the common electrical component suppliers is a huge lesson in frustration compared to paging through their paper catalogs years ago. I used to buy hundreds of transistors from Dalbani in Miami, but their web site is useless. Digikey and Mouser Electronics are two more popular stores, but I gave up trying to find any information there. If you have a Radio Shack store near you yet, they will have this part in a card hanging from a peg board hook for a couple of bucks. Finally, I did a Google search for "voltage regulator IC" and came up with this:

UA7805CKCS - Linear Voltage Regulator, 7805, Fixed, 7V To 25V In, 5V And 1.5A Out, TO-220-3

That's the heading on the page from Newark electronics. They actually list the prices, and this IC sells for 77 cents, and as low as 27 cents if you want to buy 10,000 of 'em.

The "UA" refers to the manufacturer, so that will be different for the same part from other suppliers. "7800" series is standardized for voltage regulators. The "5" in 7805 means it's a 5.0-volt regulator. "TO-220" is the package style, meaning what it looks like. That is what I drew with the mounting hole near the top. This one can handle up to 25 volts coming in, it will regulate the output at 5.0 volts, and it can pass up to 1.5 amps which is much more than all the sensors require.

The problem with this is it is not addressing the failure internal to the computer. That same 5.0-volt supply runs through a resistor, then out a different terminal to the coolant temperature sensor. When you unplug the sensor, there will be no current flow through that resistor, so no voltage dropped across it, and you end up at the sensor's plug and the terminal at the computer, with what you started with, which is over 12 volts. This is where, if I remember right, you found over 8 volts. There's other circuitry inside the computer well beyond the simple resistor I described. That's why the voltage you found is less than 12 volts.

You also have an intake air temperature sensor and possibly a battery temperature sensor that work the same way. They all start out from that 5.0-volt supply, then each sensor is fed from its own terminal in the computer's plug. I'd be willing to bet if you unplug either of those, you'll again find more than the 5.0 volts that's supposed to be there. There's no way for you to get inside the computer to fix that.

Once you have a different computer installed and you know it's going to solve the problem, I would entertain a notion to take the old one apart to see if it could be fixed. Well, it can be fixed. What I mean is if I could fix it myself. The circuit boards are almost always sealed in a translucent jell to seal out moisture and dust, but very often power transistors are mounted to an aluminum heat sink, and they stick up out of that jell.

The next problem is you will never find service information for a computer module from any manufacturer. That is closely-guarded proprietary information that they don't give out, partly because they don't want a competitor to be able to copy it, and partly to avoid liability issues if someone modifies it or repairs it incorrectly.

With no service literature, all we have to go by are the part numbers stamped on the parts, and those too are usually proprietary numbers. The UA 7805 could be the exact identical replacement part needed, but the one in the computer will have a totally different number on it that can't be cross referenced. The only way you could figure it out would be to have a good computer with the cover off, in the car and running the engine, so you could take voltage readings and deduce which one is a 5.0-volt regulator. That's an awful lot of work for such a rare problem. I go through this when diagnosing a car radio problem, but I mainly work only on Chrysler products, and I see the same failures multiple times. Spending hours on the first one allows me to know instantly how to fix the next one. Neither of us is likely to ever run into this problem again, so it doesn't make sense to try to figure it out. I would only attempt it on my own car for the personal satisfaction.

It would only work for the sensors on that circuit. That would be the throttle position sensor, MAP sensor, and possibly the EGR position sensor. For temperature sensors, the 5.0-volt feed wire is the signal wire, so they need their own separate terminals in the computer's plug. Those start with 5.0 volts, then the sensors draw that down, depending on temperature, to develop their signal voltage. Suppose coolant temperature is such that it draws that line down to 2.5 volts. Now you're starting with, lets say 8.0 volts instead. If the sensor draws that down half way again, you'd see 4.0 volts for its signal voltage. 4.0 volts is a whole lot different than 2.5 volts to the computer. That would be interpreted as a really cold engine instead of one that is almost fully warmed up.

On some car models there is an identical 8.0-volt regulator circuit to run the crankshaft position sensor and the camshaft position sensor. Regardless if they run on 8.0 volts or 5.0 volts, for them that supply voltage is not critical. They develop timing pulses that switch from near 0.0 volts to some maximum, which is usually 5.0 or 8.0 volts, but the computer doesn't look at that maximum voltage. It only looks at when it switches from one state to the other.

There's two things to consider with fault codes. The first is there is always a long list of conditions that must be met for a code to set. One of those is certain other codes can't already be set. Take the coolant temperature sensor and the intake air temperature sensor, for example. The computer knows that if the engine has been off for at least six hours, both of them had better be reading the same temperature. If you unplug the CTS, it will set a fault code for "ECT voltage high", and at that point, the computer knows it can't trust those readings so anything it compares them to will stop being tested. You could still set a fault code for the IAT if you unplug it and send its signal voltage up to 5.0 volts, but if the sensor is just reporting the wrong temperature, as long as its signal voltage remains within 0.5 to 4.5 volts, it won't trigger a fault code. The computer won't know the IAT sensor is out of range because it stopped testing for that.

There's a similar relationship between the MAP sensor and the throttle position sensor. If the computer sees a sudden drop in MAP voltage for more than a couple of seconds, it knows that can only occur during periods of high load or hard acceleration. Only a high TPS reading would agree with that. If the computer sees the TPS voltage at idle, and the MAP at high load, those two don't agree, so a code would be set after the computer figures out which one is wrong. Some of the tests on those circuits will be suspended because there is nothing reliable for comparison.

The second thing is when there's a fault code and the computer knows it can't use a sensor's signal voltage, it can disregard it and "inject" an approximate value to run on. In the case of a failing MAP sensor, which we used to see real often in the early '90s, that is the main sensor for fuel metering calculations, but with no reliable signal, the computer can run the engine based on throttle position, engine speed, and other factors. The engine won't run well, but it will run. MAP sensor voltages are critical and a few hundredths of a volt means a lot to the computer. When the signal voltage is off enough to cause severe running problems, but not off enough to trigger a fault code, you will often find engine performance improves if you unplug the MAP sensor. At that point the computer has no choice but to detect the defective condition, and use other information to run the engine.

As long as I'm at it, I should mention there was a scanner glitch on some GM car models in the '90s and with their "Tech2" scanner. I don't know if this applies to aftermarket scanners. When the computer set a code for a sensor, and injected its own approximate value to run on, that injected value is what was displayed on the scanner. That led to lots of confusion. You could have a fault code indicating the coolant temperature sensor is out of range, but on the scanner the numbers look perfect and change appropriately as the engine warms up. Erase the fault code and it comes right back when the signal voltage still looks okay. Once you knew about that, the way to verify the fault code was to measure the signal voltage yourself with a digital voltmeter.

Yup. When you unplug the MAP sensor, the computer is going to guess at fuel needs based on the other sensors' readings, and engine speed. With some fault codes already set, many routine tests will be suspended, so some defective sensor conditions will not be logged as a defect. If a reading is wrong, but still within acceptable limits, the computer will try to use those in the fuel metering calculations. It can also try to add or subtract a little fuel to see how engine performance responds. That might help it calculate fuel needs better, and it could explain why the engine surges or hesitates.

I have over a dozen digital meters from my tv / vcr repair days, and now I have some of those cheapies from Harbor Freight Tools. Those cheap ones work surprisingly well, but I do have a similar one, (not from Harbor Freight), that also reads incorrectly on the "DC Volts" ranges. I thought I had the only meter in the world that did that.

The bigger problem I have is when I use a friend's auto-ranging meter, I usually overlook the range the meter chose. More than once I thought I had 24 volts in a vcr when I really had 24 millivolts from stray pickup, and I really had 0 volts. Lots of wasted time following the wrong readings, as you're finding out.

You would have found this a lot sooner if you were fortunate enough to have a scanner. You would have seen the correct sensor voltages right away, and the one that was wrong. If you're going to be working on these types of problems in the future, consider looking on eBay for a scanner. I have a Chrysler DRB3 for all of my vehicles, but there is a limited range of model years it will work on. Out of the box they work on models back to '96 or '98, depending on the year it was made. 2003 was the last year they worked on Dakotas and Durangos. A few 2008 Jeep models was the last they worked on.

With an extra plug-in card, the DRB3 will work on Chrysler models back to 1994, and it will do emissions-related work on all brands of cars sold in the U.S. Starting with '96 models. For that reason, a lot of independent shops bought them, and now you may find them for sale so they can invest in something newer.

With a different plug-in card, the DRB3 will work on every Chrysler product back to 1983 models. I have all the other specialty cards that can be used with this scanner, but I never used them. There is also an accessory kit that includes pressure sensors for transmission diagnostics, and a pile of other neat stuff, but I haven't used that yet either.

Also take a look at the Snapon Solus Edge. I got one on eBay less than a year ago for my 2014 Ram. The problem with these is the very high cost of the annual updates. I've heard different amounts from different people, as high as $1000.00 per year, and you can't skip any years. For example, if a shop was to buy one updated through 2014 and they wanted to get it updated to work on the latest 2019 models, they would have to buy the 2015 update before they could buy the 2016 update, then the 2017 update, and the 2018 update before they could buy the 2019 update when it comes out. The combined cost of those updates is higher than the cost of buying a brand new scanner. When you do buy a new one, you have to pay extra for European coverage.

This high cost of updates works in your favor if you don't need the newer year coverage. There's a bunch of these on eBay for less than $800.00 because they're only updated through around 2012 to 2016. Professionals don't want them because of that cost of updating them. That makes them attractive to do-it-yourselfers and inexperienced mechanics just starting out. One drawback to Snapon equipment is they are real proud of their stuff, and they charge accordingly for repairs and accessories. It can be a better value to a shop to sell a broken scanner and buy a new one.

Keep me updated on your progress.

You're going to get more than you bargained for now! I became a fan at first because so many parts interchanged between models and years. Later I became a fan of Richard Petty's, and he was all Chrysler. Next, while working at a Sears Auto Center in the '80s, I found all the common failures the manufacturers refused to address, and the big one was steering and suspension parts failures due to really poor designs, on Ford products. GM's engineers couldn't seem to standardize anything, and very few parts interchanged. Chrysler at that time had a choice of two PCV valves that covered every engine they offered. GM had over 50, but we only stocked the few most common ones and never had the right one. We used to joke that you needed to know the tire size, number of doors, and paint color to get the right PCV valve. One regular customer had to go to the dealer to buy an oil filter for her car because it was so rare, no one else made one. It was a real common model with a real common engine, but for that one model year, it took a totally different filter than for that same engine any other year or in any other model that year. Chrysler and Ford used the same oil filter for decades. Even an oil filter for a 350 c.I, in a Camaro was different than that for a 350 c.I. In a Firebird. If one was better, why not use it in both?

Once I went into teaching, I got to attend a real lot of high-level classes put on by car manufacturers and parts suppliers. This is where I learned there's a lot of love for Chrysler the company. They have been the industry leader in innovations that benefit the car owner. Those include the world's first "AC generator" for 1960 models, and they copyrighted the term, "alternator". GM and Ford had their copies in 1964 and 1966. Chrysler had the first electronic voltage regulator in 1970, first electronic ignition on Dodges in 1972 and Chryslers and Plymouths in 1973, first anti-lock brakes, (1969), first Engine Computer, (1976), first lock-up torque converter, (1977), first computer-controlled automatic transmission, (1989), first domestic front-wheel-drive car, (1978), first minivan, (1984), (that was supposed to be a Ford product, but the guy in charge laughed at the design), first air bag. And you can't overlook the 426 hemi that was built as a race engine, then detuned to put in passenger cars. GM's and Ford's versions were passenger car engines beefed up for racing. Even the little 2.5L could be modified to develop 400 horsepower, then adding nitrous got it to 500 horsepower, and the bottom end held up just fine.

Two common quotes from a national-level trainer for mechanics from independent repair shops are: "Chrysler is the only manufacturer that has been able to make an engine run right without using a mass air flow sensor", and, "the list of manufacturers with the most 'customer-friendly' business practices starts with Hyundai, Toyota, and Chrysler". GM, BMW, VW, and Audi are at the bottom of that list.

I first noticed GM liked to build everything in large assemblies so you had to buy more than you needed. Their second version AC generator that showed up around 1972 was a very nice unit and easy to repair, but it was a complete system that included the voltage regulator built in. They said that was for their mechanics who didn't understand how the systems worked so all they had to do was bolt on a new assembly without any diagnosis. Same for the "High Energy Ignition", (HEI) system. Drop in a new distributor and connect one wire to solve any problem. That too was not a bad system, but at the time no one knew how to repair it.

We used to have two small round head lights or one large one. In the mid '70s, GM lobbied NHTSA to allow for two sizes of rectangular head lights. Their assertion that won the argument was those were not so tall, so they could design a lower hood line for better visibility in front of the car. First model they used them on, they stacked them one on top of the other, so the hood line ended up being even higher than before.

I worked at a tv repair shop in the '70s that was authorized for GM warranty work on their radios. We were one of well over a hundred such shops in the nation that served all the GM dealerships. In 1976, a fellow from GM came around and said they were going from 100 shops to six, then to two, and he was pretty proud of that. In 1999 I was invited to visit one of the warranty shops, in Denver. They had been shipping out two UPS trailers each day filled with GM warranty radios, but that wasn't a large enough volume for GM. That shop lost their authorization, then there were two places left in the nation, just like that guy from GM said would happen two decades earlier.

GM was also bugged by the fact little hobbyists like me were fixing their radios, so they stopped allowing us to buy radio service manuals and parts after 1994. Next, they had a 100 percent failure rate of their CD player laser assemblies throughout the '90s, and since customers were now required to go to the dealer to have them remove the radio and send it in, a lot of those car owners just bought aftermarket replacement radios that held up better. To combat that, starting with some 2002 trucks, they built the Body Computer into the radio so it couldn't be removed. The aftermarket industry came up with "radio relocation kits" to let you mount the original radio under the front seat, then you'd cut the speaker wires and run those to the new radio in the dash. Without the original radio you'd have no turn signal click, no cruise control, and other functions. The dirty trick they pulled was that Body Computer is the master computer that all the others respond to. When you turn on the ignition switch, you're turning on the radio, and that tells all the other computers to turn on. What they did was they added a "Lock" selection on one of the drop-down menus on their scanner, that once pressed, it electronically locks every computer on the truck to that Body Computer. That programming can not be undone. The customer will never know a disgruntled mechanic did that until the radio fails and has to be replaced. All the other computers will only answer to the original radio, so now those are worthless to a salvage yard as they won't work in any other truck. If the radio fails, you have to replace every other computer on the vehicle. "Got'cha". There is no valid reason to pull a stunt like that other than to siphon more money out of their customers' wallets. In fact, GM has a whole pile of these "customer-unfriendly" business practices designed to cost their customers dearly after the sale.

In the mid '70s, someone at Ford figured out that grease fittings on ball joints and tie rod ends cost a nickel a piece, and if you leave four of them off each car, build a million cars, you save four million nickels. They weren't too concerned about ball joints that separate leading to loss of control and a crash. That got a lot worse with the '80's Ford-built "killer car" Escorts. Outer tie rods wore out in 15,000 miles, so fast, we couldn't keep up our inventory. Every Wednesday we got a shipment in from Sears headquarters that included a dozen ball joints and tie rod ends for GM products, a half dozen each for Chrysler products and all the imports, a dozen for the other Ford products, and 44 outer tie rod ends for Escorts. By Saturday we were buying more Escort tie rod ends locally. I did so many of these, I could replace both outers in less than five minutes without even taking off the wheels.

Ford is also famous for leaving off the most important alignment adjustment on their front-wheel-drive cars. The front wheels on most cars lean out on top just a little. On the Escorts and Tempos, they're tipped out so far, (and look so ridiculous), the tires wore down to the belts on the outer edges in as little as 15,000 miles. It seems their rationale was that made the little cars ride much better than those of their competitors. They just didn't want you to know you'd be buying tires every year. Most tire stores wouldn't even warranty their tires on those cars.

The Taurus didn't fare any better. They used the "rubber bonded socket" design for their outer tie rod ends. Basically, you drop the ball into the socket, then glue it together with molten rubber and hope it holds together. The alignment adjustment I just mentioned, "camber", causes a tire to want to roll in the direction it's leaning. Besides adjusting that to specs on both sides, it's critical they both be the same to insure the car goes straight when you let go of the steering wheel. Well, this is another one of their, "what you got is what you get", and there's no changing it when it's wrong. For years, as an alignment specialist, I struggled with how to make those cars go straight, then I asked a friend who was the alignment guy at the local Ford dealership. His method was if the car always pulled to the right, for example, he would break the tapered studs free for the two outer tie rod ends, turn the steering wheel to the left, then retighten the tie rod studs that way. Now, when you straighten the steering wheel, those two tie rods are twisted and trying to straighten out. They're putting pressure on the steering system to turn to the left, and that will counteract the car's pull to the right. This goes against everything professionals do to prevent installing a part with a permanent twist to a rubber bushing, but it was the only thing he could do to stop the customer complaints. Of course that led to even faster part failure, but once the tie rod tore apart and separated, there was no way to prove it had been under stress by the way it was installed. This is even worse than tying a rubber bungee strap between the body and the steering linkage. It's a cobble job, but it keeps the customer happy.

The Mercury Tracer was advertised as having 17 "features" not found on their competitors' models. Those included "steering", "brakes", and "wipers". The ads looked pretty enticing to anyone who didn't know the least little thing about the machines they trust to get them back home.

Ford advertised a much lower cost of maintenance than for all the other competitors' cars, but they based that on the "normal" use maintenance schedule. That specified oil changes every 7,500 miles. Standard for the quality of oils back then was to change it every 3,000 miles to replenish the additives that naturally wear out. If you look closely at the requirements to meet that "normal" driving, it can't be done. If you drive at night, in cold weather, in hot weather, on dusty roads, prolonged high-speed driving on paved roads, extended slow-speeds, etc, you fall under the "severe use schedule that specifies the same 3,000-mile interval that every other manufacturer called for.

GM did that with their use of Dex-Cool antifreeze. They advertised it as "lifetime" coolant to make it look like you never needed to replace it, but on the sticker under the hood, they say to replace it every three years. Even the Dex-Cool company says to replace it every two years. This was just another marketing trick. It resulted in lots of corroded heater cores and radiators, and the term "Dex-Mud" which is what the product turns into when mixed with other brands of antifreeze.

Toyota used to brag in their ads that the contacts in their air bag crash sensors were "gold-plated, for your safety". What they didn't tell you was all crash sensors on every car in the world have gold-plated contacts in their sensors. This is no different than the Tracer with its "rack and pinion steering", (the same as found in most other cars), their "steel-belted radial tires", (the same as found in most other cars), and their "delay wiper system", (the same as found in most other cars). All marketing hype that has no value.

These are just the most noticeable tricks the manufacturers pull to insult our intelligence. Unfortunately, whatever one of them dreams up, the rest all follow later, good or bad. GM gave us computers that have to be programmed to the vehicle's VIN, but of course only the dealer can do that. Now they all have that nonsense. It hurts the car owner and it hurts the salvage yards. It only benefits the manufacturer. That's why my daily drivers will be '95 and older Caravans for as long as I can find them. My 2014 truck is nice with all its toys and gimmicks, but I'm a nervous wreck worrying about the first computer to give up and what it's going to take to fix it.

We can tackle the fog lights next. That circuit feeds off the head light switch, but I'll have to dig up a wiring diagram to have something to follow. Consider starting a new question for that when the time comes. That helps the site to organize solutions to aid others who are researching the same problem.