It depends on how you define fragile, glass is fragile, rubber is not easily broken but easily deformed. To say that it is fragile, it is not at all, the key is to see how you want to destroy it. Generally speaking, the synthetic materials of mechanical hard disks are aluminum alloy and stainless steel. In the more than ten years I have been in contact with and helping people repair computers, I have never seen the platters of mechanical hard disks cracked or deformed. The hard disk is mainly composed of a disk body, a control circuit board, and interface components. The disk body is a sealed cavity. The internal structure of the hard disk usually refers to the internal structure of the disk body; the control circuit board mainly includes the hard disk BIOS, hard disk cache (ie, CACHE), and the main control chip. There is a vent hole on the surface of the hard disk, its function is to keep the internal pressure of the hard disk consistent with the external atmospheric pressure. When the disc body is sealed, the vent hole communicates with the disc body through a high-efficiency filter to ensure that the disc body is clean and dust-free. In fact, the most common cause of hard disk failure is that the head touches the rotating disk shaft, rather than the disk being broken or deformed. It is based on the internal structure that we are talking about, if the hard disk vibrates greatly when it is working, the drive will be impacted when the disk is maintained at a speed of 5400 or 7200 rpm, and the head may physically hit the disk. It’s like the 747 broadcast landing on the tarmac, but without the landing gear. At this time, when you check the disc, even if it is not shattered, it will be particularly prone to scratches on the disc due to the aluminum material, and the disc head and disc may be damaged, and may cause permanent damage to the head. Even if the head here is not damaged, the scraped particles need to be filtered out by the internal air filter, otherwise these particles are likely to enter other heads and disks and cause greater damage. Continue to run such a hard disk, it is easy to cause irreversible loss of data. Another common failure mode is that the motor inside the disk cannot operate normally. For example, if the lubricating oil becomes hard or the oil is not regularly lubricated, it will increase the friction between the head and the disk and eventually make our motor unable to operate. In addition, Xiaoming wants to emphasize one point: Mechanical hard disks are not well resistant to shock and vibration. Any mechanical hard disk is recommended to be used or read and written in a non-vibrating state. After all, when the hard disk is running, the internal speed is very high and it is very fragile. The hard disk basically has no effect, so bumps and bumps in the car will not damage the hard disk, but if the frequency is too much or too much force, it may cause the hard disk to be damaged (simply shaking the hard disk by hand, theoretically not impossible). Here, by the way, a mouthful of solid state drives. The physical composition of SSD is basically flash memory + main control + cache + PCB + interface. There are no mechanical parts, and data reading and writing are all electronic signals. There is no bottleneck factor such as motor speed, and it can withstand more than traditional disk drives. Shock and vibration, so in terms of reliability and fragility, SSD is much better than mechanical hard drives. It is for this reason that many of today’s computers are basically solid-state and solid-state.


By zhiwo

0 0 vote
Article Rating
Notify of
Most Voted
Newest Oldest
Inline Feedbacks
View all comments
6 months ago

Winchester hard drives are extremely fragile in operation. As long as you understand how it works, this vulnerability is obvious. Calculated at 7200 revolutions per minute of the disc, the linear velocity at the radius of 3cm of the disc is 2 x 3 x PI x 7200 ÷ 60 = 2160 cm/sec (the pi rate is 3), which is 21 meters per second-or 75km /h. The high-speed rotating disk drives the air near the disk, so that the precision magnetic head does not contact the disk, but floats in the air 0.2~0.5 microns from the disk surface. This working mode protects the delicate and fragile magnetic head so that it will not wear out over time; however, the disk surface must be extremely flat. A single particle of dust may knock the head flying at high speed and then fall heavily on the disk surface. Wipe out a scar, and even directly damage the head. Similarly, accidental vibration may cause the magnetic head to contact the disk surface, causing permanent damage to the disk surface or even the head being scrapped. In the past, the magnetic head would always be placed on the disk; the outermost track was used to park the magnetic head-when the power was off, the magnetic head “landed” there; when the power was turned on, there was the “runway” where the magnetic head took off… …The problem that follows is: because the magnetic head is unavoidably in contact with the magnetic head during start and stop, this track is easily scratched by the magnetic head. Therefore, when important data was stored at that time, it was necessary to avoid opening the stop track; the BIOS also had a setting to change the position of the start and stop track. Later, as the magnetic heads became more and more sophisticated and the speed of the hard disk became faster and faster, the magnetic heads could no longer be parked on the platter. The new hard disk will set a special “start-stop zone” outside the disk, and add a strong magnet to the head arm; once the power is cut off, this magnet will immediately stop the head in the start-stop zone. Some laptops even have built-in sensors. Once weightlessness is detected (indicating that the laptop has fallen from the desktop), the heads will be moved to the start-stop area to wait for the impact. With this protection, this type of notebook can ensure that the hard disk data is not damaged when the height of the drop is not too high. In short, for Winchester hard drives, “long-term work” will only cause wear of the spindle bearing. This wear is extremely subtle. With reliable lubrication, it may work continuously for hundreds of years without problems; as for the magnetic head, it is It’s just floating in the wind, there will be no wear and tear at all. The biggest threat to it comes from the vibration at work. Instead, when the hard disk is powered on/off/sleeping, the head will rub against the track in the start-stop area. This friction is very slight and has little effect on the life of the hard drive, but it still has some impact after all. The reason why the hard disk is often hibernated is to save power and not to extend its service life. Of course, taking into account accidents such as equipment drops, having a drop sensor and powering off in time during a drop is a great advantage. But then you need to pay a certain amount of hardware costs.

6 months ago

Xiaomi’s statement is undoubtedly correct. The principle of the warm disk technology is that the airflow generated by the rotation of the disk lifts the magnetic head to float on the top of the disk (without contact) for radial movement to achieve the purpose of reading and writing data. When the hard disk is shut down or hibernation, the head is placed on the head rack located at the edge of the disk. In the inherent thinking of many people, letting the hard disk hibernate can better protect the hard disk. It looks like this in theory, but it is not necessarily true in practice. The factory has a technical indicator during production-the number of cycles of magnetic head loading. Simply put, there is a limit on the number of times the head is released from the head holder and then returned to the head holder. Because the magnetic head is in 100% physical contact with the magnetic head carrier, the magnetic head will be worn microscopically every time the load is released. The theoretical value of enterprise-level hard disks is 300,000 times, which is about half of that of ordinary disks. Don’t forget that this is a theoretical value, in actual applications this value will be lower. Every time you release the load, there is a risk, bad luck will lead to damage to the head, probability problems. Like some large data centers, server hard drives work almost 24 hours a day. As long as the quality of the power supply is guaranteed, the continuous operation of the shock-proof, moisture-proof and dust-proof hard disk has a longer life than the repeated load. The only drawback is that power consumption will increase, but what is this compared to data security? Therefore, do not think that hard drive hibernation protects the hard drive more than continuous operation. Add that there are friends in the comment area who hold the opposite opinion. It is a good thing to have a dispute, and you are welcome to speak freely. I will elaborate on this further. In theory, less use is definitely longer than more use. But the problem should be divided into two. Hard drives are consumables, no matter how much you care for, there will always be a bad day. To reduce the usage or sacrifice the experience for the sake of it is not bad is putting the cart before the horse. As long as it is used, it will be aging, even if it is not used, it will also be oxidized. According to the tens of thousands of hard drives I have handled, the proportion of damage caused by the aging of the motor and the aging of the pcb is less than one percent. The most fragile part of the hard disk is the magnetic head, because it is a high-precision and micro-volume component. When the hard disk is dormant or the magnetic head is broken, it will instantly bounce back to the head holder. If the magnetic head does not bounce back correctly due to power supply, external forces, or spontaneous reasons, even if there is a slight difference, the consequences are absolutely devastating. In short, the greater the number of rebounds, the greater the probability of encountering a failure. There is no hard drive that is 100% reliable. Of course, this bad situation is still only a small probability event, but who can guarantee that it will not be encountered? Therefore, it is correct to reduce some unnecessary operations as much as possible. The problem of background reading and writing occurs when the hard disk continues to run. But each hard disk has a maximum annual load capacity, enterprise hard disk 550tb/year, ordinary disk 80tb/year. The amount of data read and written in the background is usually not large, and the overhead of the hard disk is generally not considered.

6 months ago

In fact, mechanical hard drives can live longer without hibernation. It sounds counter-intuitive. But it is true, especially for server hard drives. I used to have an HP thin client, I changed the file print server, there was a bug, and the writing load was not large. It only took more than 1,000 hours to power on, and the disk 004 started and stopped counting 79,000 times, and then the disk hung up. For example, you have a bulb with a lifespan of 20W hours, and the average failure rate of normal working MTBF is 2.5 million hours, but the bulb’s switching life is only 50,000 times. From this perspective, frequent switching of the lamp is more likely to cause the bulb Broken.

6 months ago

When the mechanical hard disk is working, the magnetic head does not work closely to the disc, because the frictional resistance is large and the disc will be worn, but it will not be far away from the disc, because the increase in distance means that the recognition ability of the magnetic head is required. It is strong enough so that its power will accelerate the aging speed of the magnetic head if its power is too high. After the hard disk is energized, the spindle motor drives the disc to rotate at high speed, generating a strong air flow above the disc, which generates upward buoyancy. The lifted magnetic head is suspended in parallel on the disc. The height of the suspension is 0.02 microns, which is 20 nanometers. One percent of. Enumerate several situations that will cause the hard disk to fail (1) The hard disk falls during operation. The head hits the disc and is damaged, and the magnetic information of the disc cannot be recognized. The magnetic head hits the disc and deforms, and the deformed magnetic head is in close contact with the high-speed rotating disc for a long time and wears the disc. For a multi-disc, large-capacity hard disk, if it is dropped at an upright or sideways angle, the weight of the disc will be unevenly pressed onto the spindle motor, causing the motor to lock and fail to rotate. (2) The hard disk falls when the power is off. With the current hard disk, the landing zone is moved to the edge of the hard disk, just like the taxiway of an airplane, ensuring the safe take-off and landing of the head every time. After the hard disk is powered off, the heads are parked here. At this time, if the hard disk is severely impacted, the magnetic head may hit the landing gear and cause huge deformation. When the power is turned on again, the deformed magnetic head may run close to the disc and wear the disc. This is the reason why mechanical hard drives are not resistant to drops or shocks. Because the distance between the magnetic head and the disc is too close, any violent shock may cause it to hit the disc and be damaged. Also because the distance between the head and the disc is so close, any slight deformation may make it contact the disc. If the demand for mobile office is great, the mechanical hard disk can be replaced with a solid-state hard disk.

6 months ago

The automatic sleep use of the mechanical hard disk is not necessarily longer than the continuous use. In the mechanical hard disk, there is a Load Cycle Count parameter, which records the number of loading times of the head. Each time the mechanical hard disk activates the automatic hibernation function, a new loading record will be generated. When the number of times reaches a certain level, the possibility of abnormal noise and damage of the magnetic head will increase. One of the main reasons why the old mechanical hard disk was detected abnormally by Master Lu. The internal of the mechanical hard disk is very stable, but the external use environment is more demanding. Drive shaft and reaction force return spring device, etc., of course, circuit board and aluminum protective shell. The internal structure is quite complicated, but the compact space design also gives it a very strong stability. Since the disks storing data on Baidu Encyclopedia are of glossy and fragile material, slight scratches will lead to the normal read and write support of regional data, so mechanical hard disks should be protected against shock and dust in advance when using or storing, and avoid internal magnetic heads. Contact with the disk or the influence of dust. When the mechanical hard disk is in normal operation, the airflow generated by the rotation of the disk will lift the magnetic head to float above the disk for radial movement, and the disk and the magnetic head will not come into contact. In addition, the high temperature environment will also cause certain operation of the internal parts of the mechanical hard disk. Due to the impact of burden, it is recommended to select the mechanical hard disk preload position fixed in the chassis during installation to ensure the use in a ventilated and dust-proof environment.

6 months ago

Is the mechanical hard drive fragile? It should be said to be quite prudent. I have been using mechanical hard drives for 30 years, and I have seen two broken ones in the school dormitory, work unit, and my own. One is that the power cord is plugged in the wrong way (or the big power jack of IDE hard disks back then, I admire that buddy’s hand strength); the other is that the hard disk at home died of bad sectors last year, and it took 10 years. I have never seen a broken hard drive again.

6 months ago

Vibration and vibration, as well as the high temperature, which is basically the hard disk on hand. One is the high temperature, the highest is more than 60 degrees, and it is hung up. There is vibration, and the rest does not matter what it is waiting for. The longest-lived hard drive in my family’s NAS has been around for at least six or seven years since I bought it. The cache disk is read and written at full capacity throughout the day. It took 4 years for the 2.5-inch notebook to hang. And it’s still at a high temperature, around 42 degrees. You can understand the hard disk as a brushless motor. There is a bearing on the motor and it keeps rotating. This kind of basic wear is very small (more than 20 years in the industry, I have never seen a hard disk because the spindle is broken). Part of it is that the head is hung up first.

6 months ago

As a non-employed semi-professional, after several years of research in school, he was forced to write countless reports, and he still knew a little about hard drives. After reading the answer to this question, I feel that it is necessary to re-read the current state of the hard drive…Everyone’s knowledge of hard drives seems to have been stuck at more than 20 years ago, although hard drives have indeed not developed qualitatively in the last 5 or 6 years. ……First of all, the hard disk is really afraid of bumping. I give a schematic diagram of the structure of the hard disk [1]. It is very simple to look at the cross section. The upper part is the read-write head, and the lower part is the hard disk body. In actual work, the two sides do not touch, and the air-bearing surface is formed by the special design and the high speed of the hard disk. This principle is designed for aerodynamics without much explanation. In short, there will be a layer of air between the read-write head and the disk body under high-speed operation. Then there is a layer of diamond-like carbon on both surfaces, which seems to be translated as diamond-like carbon film. In short, it is very hard carbon. An extra layer of lubricant is added to the hard drive. Then it is also introduced in the figure. Now the surface of the general read/write head and hard disk is only 5nm. According to the 0.35nm standard of one layer of graphene, 14 layers of atoms can also be plugged. So the biggest enemy of mechanical hard drives is indeed vibration. But it does not mean that the hard drive will be broken in a flash. The air-bearing itself is designed to protect the hard drive. Now the hard disks are generally 5400 and 7200RPM, and the high ones are 15000RPM. Slight disturbances at such high speeds are unlikely to cause problems. This is why you can run around with a notebook with a mechanical hard drive. Then the carbon layer coating on both sides, part of the role is also for accidental collision. It is generally believed that there are two main factors for hard disk surface damage, one is start and stop, and the other is collision. The collision is easy to understand, that is, the hard disk is spinning well, the read/write head is flying well, and suddenly it bumps. As long as you hit hard enough, such as hitting the concrete floor, your read/write head will come together with the hard disk. Intimate contact. If the impact is relatively light, the lubricating layer and carbon film may help you block the next blow. However, if the impact is too heavy, the read/write head may scrape up a layer of carbon, and anything that is scraped off at a distance of 5nm is a disaster. This can cause rapid damage to the hard drive. This part is unavoidable, because you cannot prevent people from dropping the hard drive. But one thing is very important. The impact generally refers to the hard disk when it is running, and the hard disk is not running. Unless you fall apart, it shouldn’t be a big problem. Start-stop is a problem that has basically been solved now. Users of ancient hard disks should remember that the original hard disk cannot be unplugged directly, and must be ejected, otherwise the hard disk is likely to be damaged. The reason is that the power to the read/write head is still hovering on the surface of the hard disk. As the hard disk no longer rotates, the air cushion caused by the rotation will disappear, resulting in friction between the read/write head and the disk surface. However, today’s hard drives basically have a built-in emergency power supply to ensure that the read-write head can be safely evacuated from the disk surface when the power is suddenly cut off, and there is a start-stop area. And this part is a very mature technology, I believe that few hard drives are damaged due to excessive startup and shutdown. Back to the problem, the main cause of hard drive damage is vibration. But what is better depends on the usage of the hard disk. I don’t know how the hard disk in the router mainly works. If you want to read information frequently, it is really not good to start and stop frequently. The probability is low, but there is still a probability. But if you only need to read and write once a day or a lower frequency, I think it will increase the possibility of the hard disk being hit when it is running. Reuben Jueyuan Yeo. “Ultrathin carbon-based overcoats for extremely high density magnetic recording.” Springer, 2017.

6 months ago

The fragility of the mechanical hard disk actually comes from the vibration of the mechanical hard disk when it is running. When the mechanical hard disk is in operation, the data storage disk rotates at high speed (commonly 5400 rpm, 7200 rpm, sici hard disk speed can reach 15000 rpm or more), and the magnetic head reads the data on the disk in a non-contact manner . However, if the mechanical hard disk receives vibrations when the disk is rotating, the head will contact the disk, causing the disk to be physically damaged. This is why most people think mechanical hard drives are fragile. But I think that mechanical hard drives are not fragile at all. As long as they are used well, they are not touched when reading and writing data, and the power is not abnormally cut off. The lifespan of mechanical hard drives is very long. My first 13g hard drive is still good after more than ten years. Even if the power is not turned on for a long time, the data will not be lost. I really can’t read it, and I can save it by reading the data directly after opening the market. The solid-state hard drives that look very solid are actually not that good. Chip reads and writes have a lifespan, and the lifespan of frequently read and write SSDs is not as long as you think. And if the ssd is broken, it is really broken, and there are no remedial measures. It just means that the SSD has no moving parts and is not afraid of vibration. In terms of data security, it is actually not very good.

6 months ago

Industry-related. I have done storage security business for a period of time, experienced Huawei Symantec storage, and the start of H3C (now Univision) security business. I have a lot of dealings with manufacturers. At the beginning, servers are generally not When it was turned off, the hard disk was broken very quickly in the first two years, and it would still drop. This is what the manufacturer explained. The enterprise-level hard disk is damaged mainly due to two reasons. Hard disk, the hard disk life is worn out quickly, the other is that the power management is not good, the hard disk interface is easy to break; because the servers are fixed, there is generally no vibration problem. After a year or two of failure, the product is stable, and the hard drive is not so easy to break.

Would love your thoughts, please comment.x