Here we’re focusing on product testing, specifically, drop testing using smartphones as a typical example of a consumer electronic product that are commonly drop tested due to their popularity and regularity of being dropped!
What are drop tests, and why should they be the first test on your to-do list when devising a product test plan? Find out here…
“If you do no other product reliability testing, at least do drop testing as it could save you a great deal of money.”
– Andrew Amirnovin, 2022
3 main types of drop tests
There are actually a few key drop testing options that can be done on a product.
- Package drop test
- Controlled drop test
- Tumble test
Best-known, of course, is the package drop test which is typically done for the purpose of making sure the package is not going to break apart during the transportation process and lead to damaged goods that can’t be sold.
However, there are two other drop tests that are extremely important that you may not even have heard of or considered as a part of your product testing plan, the controlled drop test and random tumble test. They are usually done during the product development process, particularly early on.
Test 1. Controlled drop test
The controlled drop test was initially invented by NOKIA for their mobile phone devices. Later on, it was adopted by many other companies for different versions of the drop test to achieve similar objectives.
The controlled drop test is critical during the product development process to understand what the impact of an extreme drop on the product is, and then typically engineers use the test to understand what happens when you drop it on sides one to six, its corners, and so on. The engineers will see that perhaps there are no issues on side one, but side two houses the display which is cracked. The main objectives of the controlled drop test are to ensure that individual failures and failure modes are detected and fixed.
Hopefully, they will uncover the catastrophic failure modes that can result from dropping the product at a user level drop, a use case environment that would be the worst scenario of an acceptable drop height the user might be dropping their device from. For example, for mobile phones, an acceptable drop height where the device must not break is about one meter, and the logic for that is that a desk or hand height is approximately one meter and users should not lose their expensive device that’s approximately one thousand dollars when dropping it from this common height. So typically it’s not acceptable for a device to break from one meter depending on what it is (we’re using mobile phones as an example in this case) and then using that example as a frame of thought the reliability engineers will think about the use case environment for the product, no matter what type it is, and then they just determine what the normal use case height would be for the controlled drop test. The reliability engineer will use logic to assess what can and cannot be considered acceptable in terms of height. For example, if someone picked up their phone and went to the balcony and dropped it, it’s not acceptable for that product to withstand that height. The same can be said for a user becoming angry and throwing their mobile phone right at the concrete wall. No one expects that mobile phones to cope with such abuse.
However, if a user paid a thousand dollars for a mobile phone and it fell out of their pocket at a height of fewer than six feet while they were bending over to wash their hands, falling to the floor or into, say, a sink, nowadays it’s not acceptable for that device to fail, in fact, the majority of the devices in the phone industry are becoming waterproof and at least one-meter drop-proof.
Test 2. Tumble test
The second important drop test is the random tumble test. This is typically done on smaller products, like mobile phones, however larger tumbler machines to simulate the environment can be used for bigger devices. In the case of mobile phones, tumble testing makes a lot of sense because a dropped phone can tumble and strike the floor at random on different sides and corners during one drop. A phone doesn’t necessarily drop flat on one side, it may tumble and bounce. The tumble machine spins at around nine to twelve revolutions per minute and it simulates a random drop by the user and, as a result, there are a great number of drops that can be done ranging from 2 to 500 drops at either 0.5 or 1 meter heights. This level of random testing gives you a very good idea of how well the product can stand up to drops likely occur in daily life.
Test 3. Package drop test
The package drop test‘s main purpose is to determine what kind of damage could occur to the product during transport and handling using the chosen packaging.
The package is dropped from a pre-determined height on its different sides and corners using a drop machine.
It’s not uncommon for a delivery man to drop the package right on the patio from, say, 0.5 meters. Package drop testing will allow engineers to understand what that kind of impact will have on the package and/or product inside. So, to make sure that the product arrives safely at the user’s home or office the package drop machine is used to test to the ISTA-2a standard which is a simulation of international transportation and handling.
Here’s a picture we took of a driver unloading a truck in the snow in Canada. As you can see, the packaging could be dropped from a height (and is also expected to withstand the cold and wet weather, too):
A key test is the package drop test on its different sides and corners, but it’s also subjected to vibration testing, compression (crush) testing, and atmospheric conditioning testing, too.
This testing will influence the types of packaging the manufacturer will use for the product and it’s recommended to do this early on in the product development process, too, as leaving packaging testing and selection until the last minute can lead to project delays and cost overruns if the packaging is found to be unsuitable.
How to decide which height to test from?
There are two ways to decide what height should or should not be acceptable for the product drop testing.
There are some existing standards for certain products, such as aeroplanes, medical devices, and some military devices. These tend to be more general and make sure there is a certain kind of a standard level of durability for the product so it can withstand a certain amount of drop.
But when it comes to some other products, such as mobile phones and other new electronic gadgets a lot of times the whole team of engineers need to get together and brainstorm the worst-case use environments where this device could be dropped in order to devise a custom test plan.
For example, someone might say that the worst case is that a mobile phone will be dropped onto a carpeted floor in someone’s living room, but that’s not true because a user could be dropping it onto their garage’s concrete floor. So the engineers will consider the worst-case scenario on the surface type to be concrete. The worst-case scenario in terms of drop height for a mobile phone would be dropping the phone from ear height, which for most people is around 5 to 6 feet high. So the phone needs to be drop tested onto concrete from about 6 feet, and anything outside of that will be unacceptable for it to pass. The engineers could use the same kind of logic for any other products, too. For example, let’s say you’re testing a TV. Logically it doesn’t need to be able to withstand a one-meter drop like a mobile phone as it is a product which is unlikely to be dropped at all and making it durable for similar drops with upgraded glass, etc, could make its cost prohibitive. However, a package drop would be an ideal drop case for a TV rather than a controlled drop because the packaging protecting the TV during transit is expected to handle that kind of drop and protect the TV because it’s all packaged using foam and all sorts of reinforcement to absorb sharp drops and protect, in particular, the fragile display.
When to conduct the drop testing and how many samples?
Drop testing should take place early on in the NPI process during EVT, certainly, the first functioning prototype must go through at least a very generic drop test (perhaps from just a couple of feet or so at this point) to see if the structure and components, connectors, or soldering joints on the PCB hold up. Later, maybe during EVT 2, we start doing more realistic user-level drop tests on a third build of the prototype. This will result in lots of points for the designers and engineers to fix, as it will probably have anywhere from 60 to 75 per cent failures from that drop test alone. Better to find and fix these early, long before your product gets into mass production, and most design engineers fix the critical areas to make sure that the drop test has no impact on them in the field.
So drop testing starts in EVT, gets more stringent in EVT 3, then the reliability test plan needs to be updated with more and more samples to be tested consistently all the way to validation. For example, if the use case scenario to simulate is a 1.5-meter drop for a mobile phone, then tests throughout EVT3, DVT 1 & 2, and PVT will still be from 1.5 meters because any changes to the tests will result in inconsistent results; you may have too many or too few failures.
How much drop testing to be done is critical. If you don’t test the right amount of samples at the right stress level you will probably not see some of the failures that could truly happen in the field. You may also test from a greater height than your typical use case scenario in order to build a reliability margin into your product, especially if it is meant to be a rugged option. But you need to be cautious here, because if you have a typical device that most people will hardly ever drop from 1.5 meters but you’re testing it for two meters, then you’re spending too much money on reliability testing which could lead to your device having too expensive of a design that, while good, isn’t competitive with your competitors’ prices.
The Product Requirements Document (PRD)
The use case scenarios the reliability engineers use for testing are usually defined by the product manufacturer’s whole marketing team in a product requirements document including at least two to three key reliability requirements.
Use case scenario/s
One of these would probably be the worst use case scenario the device could be dropped or fall from. In the case of a mobile phone, they would most likely mention making sure that if the device is dropped from one meter there are zero failures, but when it drops from 1.5 meters a one per cent display break is acceptable, for example.
Product life expectancy
Another would be the life expectancy for that product. For example, they require this product to be designed to last for at least two to three years. Based on that need, the reliability engineering team will do a reliability assessment and analysis and, based on some formulas, they would come up with some kind of a MTBF and/or failure rates for the product and estimate that every single part in that product must meet at least this level of product reliability so the whole product can meet that goal.
Warranty period
The final key point of the PRD for engineers is the warranty period. The product warranty is important because if you have a product that you have given a one-year warranty including parts and labor and, all of a sudden, it starts failing in under a year, this is a big problem as the repair and rework bill will be footed by your business. After one year then the customer must pay for those repairs, so that’s why typically if your warranty is one year it’s best to make sure your product’s life is at least double so you don’t have any chance of failures happening during the warranty period.
Conclusion: Why drop testing alone can save you a lot of money
Drop testing is just one of a number of tests that reliability engineers will often perform, including HALT tests, life tests, and more. But, pound for pound, the drop test is one of the most critical tests because it has such a huge impact on device design and reliability, especially if the device is fragile and has a display on it like many consumer electronics these days, generally doesn’t require expensive equipment, and is fairly straightforward to do, for instance, it doesn’t require a laboratory environment.
So, if you were only able to perform one type of reliability testing and are wondering which to choose, select drop testing for the product and packaging. This gives you the best results for your money.
Need assistance with reliability testing?
We help our clients develop a test plan that focuses on product quality and reliability and conduct testing in our own testing laboratory. In addition, we provide you with packaging reliability testing, too.
Contact us to discuss your product confidentially and we will advise on how we can help you, potential costs, and an action plan.
*****
In the follow-up post, I’ll explore how you actually design a controlled drop test for your product in more detail.
The Author
Our head of New Product Development, Andrew Amirnovin, is an electrical and electronics engineer and is an ASQ-Certified Reliability Engineer.
He is our customers’ go-to resource when it comes to building reliability into the products we help develop. Before joining us he honed his craft over the decades at some of the world’s largest electronics companies such as Nokia, AT&T, LG, and GoPro.
At Agilian, he leads the New Product Development team, works closely with customers, and helps structure our processes.
