Intermittent; it is a word every service writer hates to hear. How do you diagnose a problem when it isn’t happening? How do you tell the customer that there is no flat rate for this type of repair? What tools are available that can genuinely help solve this kind of problem? If finding and solving an intermittent problem can be time consuming and frustrating for you, don’t forget that it can also be expensive and frustrating for your customer. We all know that there just has to be a better way; the question is, what is it?
While there are no easy solutions, there are ways to address these kinds of problems. There are several important starting points. First of all, you must genuinely understand how the system is supposed to work. Your diagnosis will depend on understanding what the systems is supposed to do, what information is required for the ECM or BCM, and then what actions the system is supposed to take to accomplish the desired performance. The reality that this is a little different for every make, model and sub system is no help, but it is the truth.
Secondly, working with your service writer, you must gather the best information the customer can give you. In a normal repair, the solution is obvious or nearly so.
We all know that trouble codes don’t cover everything and can even lead you astray. Still, they do help a lot, and we’d all be stuck without them. When the problem is something that only happens part of the time, the circumstances of when and how it happens are critical. Some techs may find it useful to get the story straight from the vehicle’s owner (also known as the “horse’s mouth”).
A third part is your knowledge of what can cause intermittent symptoms. There are reasons behind these types of problems. You need to know what can cause an intermittent, what it looks like when you see it and what possible techniques can be used for reproducing it. You may find it very useful to use the Internet and automotive data banks such as iATN. Sometimes there really are vehicle make and model histories that can be useful in dealing with that “tough dog” problem in front of you.
There also are tools that can help. Some of these are already in your tool box, while others are the special tool box used by electronic technicians to service computers and industrial electronics.
There are also tools that might be helpful, but are so expensive and out of reach that they might as well not exist. In the end though, it will be your knowledge, your patience and your diagnostic skills that will solve the problem and send a happier customer down the road.
Types of intermittent problems
Intermittent issues can be divided by type and cause into multiple categories. There are two values in pointing this out to you. The first is that a logical, organized approach to trouble shooting is almost always the best policy. The second is to push the point that not all intermittents are electrical gremlins. An organized approach, that doesn’t ignore any possibility, is the best and even possibly the only way to get to a solution. Anything else is pretty much guesswork.
When you hear the term “intermittent” it would be easy to say, “Oh yeah, I know what that means.” The truth is that intermittent problems cover more ground than you think. It could be a battery that occasionally goes flat, but it could also be a lean surge.
To be an intermittent does not mean that the problem has to be electrical. An intermittent is anything that works part of the time but doesn’t work under one circumstance or another. One little nuance worth mentioning here is that you do have to know what normal performance is, in order to understand what represents “broken.”
One customer complained that every time he took off from the parking lot, the doors would all of a sudden “lock.” Another fellow complained that his brakes rattled and vibrated when coming to a stop in the wintertime. There are so many new features like Lane Departure Warnings (vibration in the steering wheel) that can seem like an intermittent defect simply because the driver isn’t expecting the vehicle to behave that way.
Why intermittent problems happen
If you wanted to throw big words at it, you could call the automobile an integrated electro-mechanical device. All of it is mechanically, electrically, pneumatically connected and inter-related. All of it subject to hot, cold, moisture and vibration cycles.
This is further complicated by computer algorithms that take pieces of data from throughout the vehicle, process them, and produce commands that affect the performance of the vehicle. If the incoming data is wrong or intermittent, the result will be incorrect or wrong output commands. Your task is to understand the system well enough to work back from the wrong ECM command to identify what the garbage was going in that produced the garbage coming out.
A bit more about computer algorithms; don’t let big words confuse you. An algorithm is not very different than a recipe for baking cookies. The basic output function of the ECM is to control spark timing, fuel-air ratios, EGR flow and other engine parameters. To do this, the ECM needs certain information (throttle angle, engine temp, engine load, etc.). The computer gathers the information, processes it through the algorithm recipe and then outputs commands that set the operational details of the engine. The operational problems that we’re referring to as intermittents are often the result of intermittent data coming in to the algorithm. What you are seeing and what your customer is experiencing is intermittently wrong outputs from the ECM or other processing device.
Possible causes for garbage-in: Vacuum problems
Vacuum is used in a number of ways. As an actuator, vacuum powered devices using diaphragms are used in heater controls, power brake boosters and EGR valves. As information, vacuum levels are used to sense engine loading (MAP sensors).
Vacuum-related intermittent problems can be caused by what might be called the “bladder” effect. When there is a pin hole, or a split, the vacuum inside can either pull a leak closed or make a leak worse as the vacuum level increases. This can also happen as a function of temperature. When cold, the vacuum line is harder and stiffer and can leak more than when it is warm and soft. Mechanical vibration and position changes of the engine can also make the leak better or worse.
These hoses are also susceptible to heat damage. If they get up against a hot spot, they can melt just enough to create a pin-hole leak. How much this affects the operation of the vehicle depends on the function of the vacuum.
Vacuum lines and diaphragms are typically made of neoprene-like, synthetic materials. Viton, EPDM and silicone hose are variations. Good hose material must be resistant to oil, heat and anti-freeze. The hose needs to be flexible so that movement of the hose doesn’t wind up breaking tees and other interconnects. The hose also needs to be soft enough so that it can seal properly onto ridges or barbs on the plastic tees and sensor inputs. For the hose to function as it was intended, it must also be routed and held in place so that the hose is not damaged, pulled, stretched, pinched, etc.
The aftermarket has a place here, too. To make the engine look cool, hoses in a variety of colors are available. It is going too far to see a bunch of this on an engine and to declare victory in the intermittent wars. Still, every piece of it represents a factory connection that was taken apart and replaced. It is possible that the replacement hose might not grip the tees as well or may not have been fully or correctly installed. It is also possible that it is “generically” sized and really doesn’t fit metric fittings very well.
Possible causes for garbage in: Mechanical problems
Electronics and computers so dominate the conversation about intermittents that it is easy to forget that there are both mechanical inputs to, and mechanical outputs from, the engine control algorithm.
There is a wide variety of mechanical sensing that is part of the system.
Mechanical inputs that can lead to intermittent behavior included loose sensors that can shift position during operation causing a change in the information they were supposed to be sensing.
Contact type throttle position sensors can develop wear points that can put dead spots into the sensor’s output.
You can visualize how a computer could be confused by a TPS output that blips to 5 volts when it hits a dead spot.
Mechanical outputs that can cause trouble include actuators that bind up and cannot move in response to orders from the ECM. A carboned-up idle speed control motor or EGR would fit into this category.
Mechanical problems like this are very often heat sensitive.
It can fail to work either way. Some will work better cold than hot, while others work better hot than cold.
Possible causes for garbage in: Electro-mechanical
While it sounds like it would make a good title for a country song, it really is true that “trouble often starts when two worlds collide.”
Where this happens in automobiles is in electrical connectors and connections throughout the vehicle. These connectors (yes, there are lots and lots of them) are the interface between the wire harness, the ECM, and the sensors and actuators. The connections are not optional; you couldn’t assemble a car without them.
These connectors supply power (B+), ground (B-) and information to and from the ECM. How the information is relayed varies from one make to the next and when the car was built. The information can be analog or digital. The digital information can be routed on a bus system like CAN or it can be directly wired or sent to the ECM.
The key to this part of the intermittent problem is that all those connections and connectors are mechanical devices that are designed to carry electrical power and information. If the connections fail, the result will be either power loss or corrupted information.
The basic design of an electrical connection involves two pieces of metal, one of which is a spring. The key point here is that the two terminal halves are not locked or soldered together. A good way to visualize this is to take the pointing finger of your right hand and press it in to the palm of your left hand. Now wiggle and jiggle both hands. It is the spring pressure that serves the same function as the pressure on your finger. You can also feel the relative motion between your finger and your palm.
In terminals, this motion at the connecting point is called “fretting.” When the two terminals are plugged together, the spring should apply what is called a normal force that holds the connection points of the two mating terminals together. Standard construction is to make the mating terminals of brass, overlaid with layers of copper and tin.
Fretting and fretting corrosion is what results from micro-motion at the point of connection in the terminal. Fretting corrosion is a progressive damage that results in data drop outs, lost information, and quite probably intermittent loss of vehicle performance.
Quite a number of protections are built into the system to prevent fretting corrosion. Long, stiff, unsupported wires leading away from the terminals make it worse because they act like levers transmitting their motion to the contact point. Lubricants, including the tin plated onto the terminals, help to minimize the fretting corrosion.
The plastic shells that hold the terminals minimize the motion and hold the connectors together. Weather seals like the silicone rubber multi-ribbed types are designed to keep water and abrasive dust out of the joint.
Despite all the precautions, electrical terminals in the vehicle are subject to stress and damage. This is particularly true where time and mileage have worn down or deteriorated the protections the car companies put in place.
It is common to see pieces of wire harness that were originally taped, strapped, tied off, fitted with push on (Christmas tree) fasteners, etc., dangling loose. This puts terrific stress on the terminal surfaces and is one known, root cause for intermittent problems.
Possible causes for garbage in: Thermal
Heat related problems can result from too much heat or from cold operation. All materials have a coefficient of thermal expansion or CTE. This means that the materials will grow with heat and shrink with cold.
When the size is important, like in the fit of electrical connection or the fit of a vacuum hose, changes in temperature can result in an improved or worsened fit.
This is especially true where the materials are different and the shape of each will change differently as the result of a temperature change.
A special case for thermally related problems involves electronic assemblies. The construction of sensors, modules and ECMs involves possibly thousands of connections between tiny, intricate components made of metals, ceramics, plastics, silicon and so on. These devices are generally self-heating in that they dissipate power in their normal operation. They are also susceptible from nearby heat sources such as motors, solenoids and exhaust systems.
One of the things you are doing when you take a vehicle for a test drive is to heat up the electronic assemblies and their connections to see if you can reproduce a heat-related intermittent.
Mechanical problems will often make themselves known related to heat. Heat can also make “sticky” assemblies move differently. Part of the intermittent troubleshooting procedure is to listen to what the customer says is happening and then try to relate that to changing heat conditions. “It only happens when the engine starts up” should suggest “cold, tight and possibly sticky” to you.
Possible causes for garbage-in: Vibration/shock/non-OE installation/alterations
Before any vehicle model can go into production, it must first go to the test track or proving grounds. When you see what is done to those cars you can better understand how the car companies prove that their vehicles are good to go.
Much of the testing involves corrugated roads with killer pot holes. It varies with the automaker, but much of this testing could be called “destructive.” They shake and rattle things so severely that eventually things start to fall off, fail or become intermittent.
All of this is done with production-intent wiring, vacuum hose routing, module mounting, sensor location, intact heating and cooling and so forth.
A good first place to start your troubleshooting is to inspect the vehicle. You want to look for things that have come loose, been moved, been disassembled, or had aftermarket parts and assemblies added.
Possible causes for garbage-in: Moisture, salt water and corrosion
Pure water does not conduct electricity, but water with ions does. Adding just a dash of salt or other similar materials can make the water conductive. In terms of intermittents, the water that does conduct electricity can corrupt digital and analog signals in two ways. The conductive water can short, or partially short, signals off to ground. The water can change position as the vehicle moves and turns one way or the other. What was supposed to be digital data in the form of 5 volt square waves can become reduced in amplitude and clarity due to the leakage.
Water also leads to galvanic corrosion. Two dissimilar metals in the presence of a conductive electrolyte like ionized water can create a small, unintended battery whose electrical flows wind up causing corrosion of the metals.
It is not a long path between corrosion and missing or degraded signals.
Places not to look
While it may sound odd, there really are places not to look for trouble. This is based on probabilities and service histories.
The wire harness is a good example. The terminals on the ends do give trouble and are a reasonable place to look for problems. The wires themselves are usually well insulated and then covered with tape and or flexible plastic conduit.
It is not to say that wire is never the problem... it is just not a very likely place to look.
There are two exceptions to this rule. One exception might be where there is evidence of substantial collision damage.
A second place to look might be where the wire harness bundle enters the door with power locks/power windows, etc. The wire harness bundle needs to flex as the door opens and closes and there are models out there that have had trouble with wires at these locations.
Isolating the fault: Replicating the problem
Your first requirement as a service technician is to isolate the intermittent to the sub system that is responsible. You can’t work on everything... you need to narrow the search. The obvious first step is to use your code reader to narrow down to the system.
We spoke earlier of the importance of getting the story of the intermittent behavior in as much detail as possible through your service writer or from the customers themselves. You can add to what they are telling you by asking questions and getting even more detail than was originally offered.
You need to “pump” the person who experienced the problem for every detail you can get. Be careful not to put words in their mouths, but do get all of the details that might help you.
A word is in order here about respecting your non-mechanical customer. For better or worse, no one knows better than the vehicle owner what is “normal” for their car. They live in that car as they drive back and forth to work or whatever. They may not know a muffler from a tailpipe but they do know how their car usually behaves. It is amazing how often their descriptions are spot-on.
If they tell you “And then the car goes gwhumpa-da-di-diddy-dot doo,” you can probably assume that it does make a noise pretty close to that.
The good news is that there probably is real information in their description. That information can help you only if you have the good sense to genuinely listen and then try to tie their description of the fault to something that really would make such a noise like that. Genuinely listening is not easy to do, especially when your customer really doesn’t know much about cars or how they work. Listening is also good for customer relations. They know that they and their car have trouble. They want to help, so do yourself a favor, and listen.
Your first goal needs to be make a list of all the systems that are affected and under what circumstances the problem occurs. Even though it might seem silly, writing the list down can be very important and useful. The list is a way to clarify in your mind where the trouble exactly is and isn’t. When you write it down, you may see relationships or linkages between parts of the problem that you just didn’t notice when it was all floating around in your head.
Your second step is to try to repeat the circumstances that the customer says produced the fault. It may be useful for you to let the customer do the test driving while you ride along. They may be doing things they are not aware of that might have a bearing on the problem.
The third step is a thorough examination of the car and the systems that seem the likely location of the fault. At one time the U.S. military did a study on how trouble shooting really works. They found that over half of all repairs really started with the technician looking to see if they could “see” what might be causing the problem. Dangling wires, loose or missing vacuum hoses, broken springs... the list of things that can be “seen” if you simply take the time to look carefully. This visual inspection is an important part of the diagnosis.
The fourth step is to take what you have learned and then sit in front of the computer. Google, iATN, identifix.com, OEM service sites, TSBs and so on are possible sources of information that can help you. There is something like 150 million passenger vehicles out there on the road in the U.S. The chance that your customer’s problem is the only instance that this has ever happened is very small. The important point is to do your investigative homework first. You need to make sure you have the problem well defined and clear in your own mind. In this way, your searches of the various fix data bases will be as focused as possible. A five-minute search may really help. An hour-long search may just be an exercise in “surfing the Net.”
Hopefully by this time, your search for the problem has narrowed considerably. Your efforts to replicate the problem should be focused on the areas we have talked about in terms of what could be wrong. Search the vacuum lines associated with your problem. Wiggle the electrical connectors. Exercise mechanical functions and make sure they move freely through their normal range of motion.
Specific things you can do
There are a number of tests that can be performed to help further define the problem. For the sake of this discussion, we will assume that you have checked the trouble codes and the sensor data to make sure that they are in normal range.
The two most important test instruments are the digital multimeter and the oscilloscopes. There are standard tests for items such as TPS sensors that you are familiar with, so there is no need to review them here. Besides, if the fault was in a specific sub-assembly, it should have set a code. What you are looking for is a data input or an actuator output that is generally normal, but occasionally wrong or corrupted.
What you can do is to monitor the sensor or actuator to see what the output looks like. Then make the vehicle perform as best described by the customer. Some of this data is available via the ALDL and can be monitored as a helper drives the vehicle.
If the problem is truly an electrical connection, there are a couple of things you can do. The first is to set up a meter or scope to monitor the data on the suspect line. Then squeeze the connector like you are trying to push it further together. This will shift the contact point where the terminals mate. Shifting the connection spot inside the connector should not affect vehicle performance for better or worse.
Another way to do much the same thing is to disconnect the connector and then plug it back together. This is a great time to look for green corrosion caused by salt water or moisture. As it turns out, unplugging the connector and then reconnecting it almost always results in a change of the position where the actual contact of the terminals occurs. The new connection spot may solve what was a poor or intermittent connection.
Should you discover that you actually have a bad connection, there are three possible solutions. You can replace the whole connector block and re-splice all of the wires, you can remove the individual wire and replace the terminals, or you can jumper around the connector for the affected line.
Probably the best solution is the jumper. Replacing the connector block involves getting a new, correct one and then splicing all of those wires. This is too complicated and may introduce more trouble into the system rather than fixing it. Individual terminals can be removed and replaced, but again it means having a correct source for the terminals and then getting it removed and replaced without affecting the reliability of the other terminals. The jumper leaves everything intact as it originally was. Yes, you will have to be careful to seal up and protect the places where the wire is spliced. Another alternative is to use a contact enhancer such as Stabilant 22. This fluid does two things. It lubricates the mating parts of the terminal, and it also conducts electricity, but only when the terminals cannot make the connection.
What you don’t want to do is to use any kind of pliers on the suspect terminals. The spring part of the terminal depends on the exact, stamped-at-the-factory shape of the brass for its spring effect. Bending the terminal in an effort to “tighten” or improve the connection doesn’t work because it damages the “shape” of the spring.
If the trouble is an intermittent battery drain, a temperature measuring tool can be helpful. Use a non-contact laser-type sensor to measure the temperature of the fuses in the fuse box. Fuses work by having a defined resistance that heats up with the passage of current. When the current gets into the danger zone, the fuse link will melt and the circuit will open.
In the case of a battery draw, the fuse associated with the current draw will be warmer than the ones around it.
You can perform a similar test with a volt meter. The fuse with the voltage drop across it is the one that is flowing current. Once you have narrowed the problem to a certain branch, then it is a matter of using the temperature probe to see which assembly is warm and drawing current when it shouldn’t be. You can also disconnect the suspect assembly from the fused line. If it is the culprit, the voltage drop across that fuse should go back to zero when load is relieved.
The voltmeter has one advantage over the temperature probe. The voltage drop across the fuse can be monitored, while the vehicle is exercised to try to get the intermittent to show up. This works for intermittent key-off drains for things like glove box and trunk lights. Monitor the fuse and push on the door or deck lid to see if the light you can’t see doesn’t suddenly start drawing current through that fuse.
Electronic technicians often have both heat guns and cans of “Freez-It” on their benches to help find intermittents. I have had limited luck in finding trouble this way. There is always the danger of overheating the module/sensor/component and causing damage that wasn’t there in the first place. The freeze spray is good at the electronic component level, but not so useful when you have a whole assembly that is suspect. If the whole assembly is suspect, the best solution is to unplug it, re-plug it and try again. If that doesn’t solve the problem, the next best step may be to substitute a known good unit to see if that solves the issue.
There are flight recorders that are available on the market. The idea is to record data from the ALDL lines and then try to get the vehicle to misbehave. These flight recorders are rather like another piece of gear called a data bucket. Once the data is recorded it still needs someone or some computer to search and analyze the data. These tools are expensive and may be more of an engineering tool than something the technician can really be expected to use.
Clearly, vehicles don’t leave the factory with intermittent problems. If you have one in your service bay, it must mean that something has gone wrong.
We’ve talked about what can go wrong and what sort of problems could cause intermittent issues. If there is any one good piece of advice it would be for you to spend some time with that shop manual. Knowing how the system works is the key to taking the complaint and working it back to what the root cause really is. ●