Fastener Joint Failure Evaluation

There have been many papers written on how to design a good fastener joint, what type of fastener is best in a particular application, or the causes of a specific type of failure. These all have good information, but many times, particularly when discussing screw failures, you may feel you need a master’s degree in physics to understand what is being said. If you are the user or the supplier of a screw that is in a joint that has failed, you just want to know what happened, whether it was the screw’s fault, and what to do to prevent it from happening again. You can save a lot of time and money by asking the right questions and obtaining the right samples.

It is a universal truth that screws are the least expensive part of an assembly, and they receive the least amount of attention. But when a joint fails, they are the first thing people assume is to blame. They believe the screw was too weak, too small, too short, too long, too slippery, not slippery enough, etc. Many times, the screw or bolt is not the cause of the failure, but some detective work must be done to find the true cause so it can be prevented. 

When a failure has occurred, it is critical to gather as much information as possible and get samples of the “failed” screw and assembly. This cannot be overstated. Ask questions and get samples. Do not auto – matically fall into the “bad screw” mindset. 

There are two general types of failures. The first failure occurs during installation, when for some reason, the screw can’t be driven or tightened to the required level. In this case, you know there is a problem right away, and it is fairly easy to evaluate and determine what is wrong. 

The second type of failure is much more difficult. In this scenario, the screw goes in fine and tightens to the desired level. Everything looks good, and everybody’s happy, but later the joint fails. This is a far worse situation than the first. In this situation, the product may be totally assembled, and even worse, out in the field. This failure can cost significantly more in time and money than the first type failure. Now you have product that needs to be repaired, downtime, unhappy customers, freight, recalls, etc. This is a very bad situation. In addition, it can be much harder to determine the root cause and fix it. 

What I am going to talk about here is the first type failure. I will cover the delayed type failure in a later article. 

The issue here is that the user cannot assemble their product because the joint won’t tighten. As I pointed out earlier, do not automatically assume it is the screw that is at fault. A successfully assembled joint involves at least four components: the screw, the assembled piece or pieces, the threaded or tapped piece, and the driver. The goal is to drive the screw in and clamp the assembly together. If that goal cannot be met, your job is to determine why. The problem can be with any of these four items. It is possibly a problem with the screw, but what about the tapped hole, or the nut the screw is driving into? 

Internal threads are much more difficult to check dimensionally than the external threads on the screw. It is a mistake to assume the internal threads are correct. I have seen situations in which someone will use a nut as a gage to check the external screw thread. Why so much confidence that the nut is right? As I said, internal threads are much more difficult to check dimensionally than external threads, and tools do wear, and mistakes do happen. Oversized or undersized internal threads can lead to a failure as easily as an oversized or undersized screw thread can.

What about self-tapping screws? In this scenario, a screw is designed to make its own internal threads in the mating pieces of the application. This can be a drilled or stamped hole in sheet metal or a casting of some type. It can also be tapping into a cored hole in a casting, or an extruded hole in sheet metal. We could be talking about driving specially designed screws into molded or drilled holes in plastic. Whatever the scenario, if there is a failure, you have to determine what happened. The hole size, material thickness and hardness, as well as the screw dimensions and heat treat all need to be verified. It is also very possible something has changed with the driver. I have been involved in situations where the customer calls and says the screws are no good because they are stripping, only to find out later they have installed a new air compressor, and their line pressure had jumped, causing the driver guns to over-torque the screws. I have also seen the opposite scenario. The screws have stopped driving, but not because there is anything wrong with the screws. Rather, a new piece of equipment has just gone on-line, and the air pressure has dropped to the point the drivers are not generating enough torque to seat the screws. 

I once made a visit to a customer who complained that our screws were bad because they were stripping. When I got there, I discovered they were driving a 5/16- 18 thread forming screw with a 1/2-inch size impact wrench. They were over-torqueing the screws by more than twice the torque they should have been using, but they were convinced the screws were bad. 

Many types of failures can occur during the assembly operation, but generally they can be broken down into just a few categories:

  1. Some dimensional problem prevents the driver from engaging the screw correctly.
  2. The screws will not start in the hole or internal thread in the mating part.
  3. The screws will not torque down and clamp the joint sufficiently.
  4. The screws stripped. This can be a stripped hole or nut, or the threads on the screw are deformed.
  5. The screws broke.

This is your first responsibility: FIND OUT EXACTLY WHAT HAPPENED! “The screws aren’t working” or “the screws break” is not an acceptable answer! Which of the above five things happened? Did something else happen? You have to continue to ask questions until you understand what happened. You may have to talk to an engineer, quality person or production person to get this information, but keep asking questions until you understand what happened. You have got to determine how the joint failed. Without this information, it is impossible to determine what happened, and worse, how to prevent it from happening again. 

Samples of the failed joint are worth their weight in gold. If you or your engineer can look at the failed joint and see what failed, you will be able to come to a conclusion and solution much more quickly. 

Something you need to be aware of is this: A problem can be the best sales tool you have ever had, depending on how you handle it. Put yourself in your customer’s shoes for just a second. You have an assembly (not necessarily a screw) problem. You don’t know what is wrong. You need it fixed ASAP. The guy you bought the screws from asks tons of questions and asks for samples. Maybe he actually visits and watches the screws being installed. He then tells you what happened and how to fix it. How likely are you to be a little more dependent on that supplier? 

A problem in which the driver and the screw cannot mate together well enough to drive and seat the screw should be fairly evident and easy to fix. The wrong driver is being used, the screws have some dimensional problem, or the driver bit or socket has a dimensional problem. 

If a machine screw or bolt will not start threading into a tapped hole or nut, you need to verify the dimensions of both the screw and the hole/nut are dimensionally correct. 

If the screw threads into the hole just fine, but will not seat and clamp correctly, there are a few additional potential causes you need to investigate. The screws may still be too large, the tapped holes or nuts may still be too small, there is not enough torque coming from the driver, or there is not enough lubricity in the joint. 

  1. Verify the screw meets the print. At the minimum, you will need a micrometer and ideally, thread ring gages.
  2. Double check the nut or tapped hole size. If you have a set of thread plug gages, great, but more than likely, none will be available.
  3. Try screwing a different screw that you know is good into the “bad” hole, or try driving the “bad” screw into a different hole that you know is good. If the “bad” screw fits a different hole, then the hole becomes suspect. If a different, good, screw drives into the “bad” hole, the original screw is looking like the guilty party.

If you determine the screw and the hole are dimensionally OK, then start looking at the torque. This becomes a little more difficult because you will need a torque wrench to answer these questions. Use the torque wrench to check how much torque it takes to assemble the joint correctly. 

  1. How does this compare to the recommended seating torque?
  2. Is the gun supplying the correct amount of torque?
  3. Does it take much more torque to install the screw than it should?

Unscrew the fastener and look at it carefully. 

  1. Are the screw’s threads damaged or smeared?
  2. Is some of the material from down inside the hole peeling off or smeared in between the screw threads? If there is damage on either the screw, or the internal thread you need to find out why.
  3. Is the screw the correct hardness?
  4. Are there too few threads engaged? A screw should have a minimum length of threads engaged in the internal threads equal to the diameter of the screw. For example, a 5/16-18 screw should have at least 5/16 of threads engaged in the nut/ tapped hole.

If the joint is stripping, it could be: 

  1. The screw is too soft.
  2. The mating material or nut is too soft (not likely; internal threads are stronger than external threads of the same size).
  3. The screw is too small.
  4. The hole is too big.
  5. There are not enough threads engaged.
  6. There is too much torque being applied.

If all the parts seem to check out dimensionally, the screw hardness is correct, and the driver gun checks out, you may need some lubricity (oil or wax) to lubricate the screw and hole. 

Most people have no idea that without some type of lubrication, 80 to 90 percent of the seating torque used to install a screw or bolt is eaten up in friction; this is the friction between the underside of the head and the surface of the joint and the friction between the screw threads and the internal threads in the hole. That means only 10 to 20 percent of the torque you are applying is actually generating any clamp load in the joint. When you tighten a bolt to 50 foot pounds, 40-45 foot pounds of that torque is being used just to overcome the friction; only 5-10 foot pounds is being used to actually clamp the joint. 

Lubricity has a tremendous effect on how much torque is required to tighten a joint. Too much lubricity can actually cause a screw to strip out that otherwise would be working fine. A simple rule is: “If it won’t go in and seat, lubricate it. If it strips, take away some lubrication.” 

If the screw is a self-tapping screw and it will not pick up a thread and start tapping into the hole, you need to verify: 

  1. Is the hole too small?
  2. Is the point on the end of the screw too large?
  3. Is the screw too soft?

Any of these problems can prevent the starting threads of the screw from digging in and beginning the tapping operation. In these cases, the screw will just spin in the hole without tapping and eventually burn up. 

If the screw is a self-tapping screw and it stalls the gun or will not seat and clamp correctly, there are additional things to check. 

  1. Is the screw too big for the hole?
  2. Is the hole too small for the screw?

Were the screw threads deformed as the screw was driven in? A tapping screw must be very hard on the outside in order to form the mating threads in the application. If they are not hard enough, they will flatten out, or deform. If they deform, then they will not make clean, correct threads in the material they are being driven into. If this happens, there can be all kinds of strange results that will cause joint failure. 

The twist in this story is that the lubricity question becomes even more important than it is with machine screws or bolts. As stated previously, 80 to 90 percent of the seating torque is used up in friction. It is even worse with the driving torque with self tapping screws, but the problem is greatest when the screw is forming the mating threads. It takes a tremendous amount of force for a screw to form its mating threads, and if there is excessive friction, it can cause the torque to spike to a point that damage to the threads, or breakage of the screw, is very possible. If there is not enough lubrication between the screw threads and the material it is tapping into, there can be galling that will destroy both the screw and the hole being tapped. Here, adding a little wax or oil can solve a driving problem instantly. 

What happens if the screw “strips out?” Now is when samples become even more critical. 

Screw stripping can be subdivided into two possibilities. Either the external threads of the screw failed, or the internal threads in the application failed. By failed, I mean they deformed during installation. This is critical information. Self-tapping screws need to be hard on the outside so they can tap the hole, but they need to be soft on the inside for strength and ductility. 

If the threads on the screw deform instead of tapping the mating piece, it may be: 

  1. The screw is not hard enough.
  2. The material being tapped is too hard. Just because a screw is a self-tapping screw, does not mean it can drive into every material on the planet. A standard self-tapping screw cannot form threads in hardened steel or in many stainless steels.
  3. Is the hole too small?

If the threads on the screw are fine, but the hole has stripped out, you will normally see slivers of the mating material that have been ripped out of the hole, trapped between the screw threads. If you see that, there are four primary causes: 

  1. The screw is too small.
  2. The hole is too large.
  3. The torque is too high.
  4. The material is too thin.

As companies strive to improve their profitability, materials are becoming thinner and thinner. Thinner material means less thread engagement. The way to counteract that loss is by decreasing the hole diameter, extruding the holes to gain more length, and adding features to the screw that help stop it from rotating when it seats. These help a little, but at some point, you will no longer be able to effectively clamp the joint. 

This condition will increase the demands on the screw to do its job with less room for any variation, and at some point, you have an undependable joint. Unfortunately, the screw will still be looked upon as the problem. 

It must be noted: If the failure is a stripping type failure, lubrication is your enemy. 

In an effort to summarize, when there is a joint failure problem, you MUST get all the information you can and obtain samples, or visit the line and watch the screws being driven. You need to check: 

  1. screw size;
  2. hole size;
  3. screw hardness;
  4. mating material hardness;
  5. driver torque; and
  6. lubricity.

Find out what failed and how. In the majority of these joint failure occurrences, the screw is not the source of the failure.


View this article in the American Fastener Journal published on BlueToad (p.30-34)