Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations TugboatEng on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
When to use washer or not?? 1
- Thread starter brucegr
- Start date
I do not know of any standard that covers this issue.
A washer can provide multiple functions, the two most important ones being: 1) proper bearing area to avoid excessive surface pressure; and 2) hard, smooth, and constant-material-composition surface for good preload control. You must analyze the joint design (geometry, materials, and applied forces/moments) and calculate if the washer is necessary.
Regards,
Cory
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.
A washer can provide multiple functions, the two most important ones being: 1) proper bearing area to avoid excessive surface pressure; and 2) hard, smooth, and constant-material-composition surface for good preload control. You must analyze the joint design (geometry, materials, and applied forces/moments) and calculate if the washer is necessary.
Regards,
Cory
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.
-
1
- #4
As Cory noted, one important use is to reduce the surface pressure to acceptable levels for the material being clamped.
Another is to reduce the portion of the external load carried by the bolt, which is especially critical (because of fatigue) in the case of dynamic loads. It does this by increasing the relative stiffness of the clamped materials (k = AE/L, where A is the effective pressure area, which can be increased by using a hardened washer). In a bolted joint, the preload compresses the clamped members and stretches the bolt. An external tensile load will tend to separate the members, which would tend to stretch the bolt further (increasing the bolt load), but at the same time will also tend to relieve the compression of the clamped members (decreasing the bolt load). In reality, the joint will find some happy medium of compression/extension. The stiffer the members, the smaller the fraction of the external load the bolt will "see".
Shigley and Norton both cover this topic. I prefer Norton, especially for his coverage of calculating the joint stiffness factor. And boltscience.com has a great wealth of information on bolted joints.
Rob Campbell
Another is to reduce the portion of the external load carried by the bolt, which is especially critical (because of fatigue) in the case of dynamic loads. It does this by increasing the relative stiffness of the clamped materials (k = AE/L, where A is the effective pressure area, which can be increased by using a hardened washer). In a bolted joint, the preload compresses the clamped members and stretches the bolt. An external tensile load will tend to separate the members, which would tend to stretch the bolt further (increasing the bolt load), but at the same time will also tend to relieve the compression of the clamped members (decreasing the bolt load). In reality, the joint will find some happy medium of compression/extension. The stiffer the members, the smaller the fraction of the external load the bolt will "see".
Shigley and Norton both cover this topic. I prefer Norton, especially for his coverage of calculating the joint stiffness factor. And boltscience.com has a great wealth of information on bolted joints.
Rob Campbell
As had kinda been stated...not withstanding the clearance issue, we use washers on any stud bolt over 1" to prevent damage to flange, exchanger, nozzle or other bolted surface during the torqueing process. As the metal of the nozzle, flange, exchanger or other attachement gets more and more worn and abrasive due to bolting and unbolting...the coefficient of friction between the surfaces increase significantly irregardless of how much moly lube you place on the bolt, nut, and surfaces. You find yourself torqueing to a given value and assuming that correlates to the correct bolt tention but instead most of the energy was used on the metal to metal contact. Next thing you know, your shutting the plant down at$10,000 an hour to fix a leak because someone saved $20 dollars in washers.
Just remember a standard washer will not correct a bad nut bearing area. This is especially true if the area is cup shaped or an area with proud metal. We use a tremendous number of 4",5",6", and 8" class 1500 C/S flanges that tend have wear and tear in the bearing area. We don't allow the use of washers with the B-16 or B-7 studs. All the fasteners are coated with Dag prior to assembly. If inspection shows that there might be a problem during assembly we have the pipe spool sent to the shop and the bearing area spot faced.
We do use a hardened and ground flat washer under some H-11 SHCS being tightened against some PH S/S
We do use a hardened and ground flat washer under some H-11 SHCS being tightened against some PH S/S
In my experience, washers are used in under the turning element. For example, if you have a bolt and nut, typically the washer is under the nut, which is turned. This is to prevent, as mentioned arto, gauling. Also, the nut and washer surface should be lubricated (if they are not coated bolts being used for the first time, coated being a low friction coating) for flange bolts to reduce the friction between them, and thereby increase the bolt preload achieved by a given torque. Also, as mentioned by uncleysd, the washer should be hardened, note though it should be through hardened, not simply surface hardened. I disagree with rjcj9. If the bolt is suffiently preloaded, it is not subject to fatigue. The cyclic load reduces the contact stress between what is being bolted together. Generally (and there are exceptions), the bolt stress does not change until the length of the bolt changes, which means the contact surfaces separate. So you have to overcome the bolt preload before bolts experience cylic loads. A more detailed assessment would reveal, though, that bolt load can be increased or decreased by external loads in some situations, such as raised face flanges. With a raised face flange, an external tension load can, for example, rotate the flanges and cause a decrease in bolt load because the length of the bolt is decreased by flange rotation. As mentioned by haggis, another reason for washers is for oversize holes.
cb4 wrote:
----------
If the bolt is suffiently preloaded, it is not subject to fatigue. The cyclic load reduces the contact stress between what is being bolted together. Generally (and there are exceptions), the bolt stress does not change until the length of the bolt changes, which means the contact surfaces separate. So you have to overcome the bolt preload before bolts experience cylic loads.
----------
That simply isn't true. Maybe the key to the midunderstanging is the mistaken belief that the bolt does not change length until separation. The bolt, which is under tension in the first place, does indeed lengthen. The clamped members, which are in compression, relax by the same amount, so there is no separation as long as there is some of the initial compression still in the members.
The portion of an external load, P, carried by the bolt is:
Pb = kb*P / (kb+km)
Where kb and km are the stiffness of the bolt and the clamped members, respectively. Therefore, increasing the stiffness of the clamped members relative to the bolt decreases the portion of the external load carried by the bolt. But it does not eliminate it.
One of the figures commonly used to illustrate this concept is a block of wood wedged between a force pull gauge (essentially, a compression spring). In this case, the stiffness ratio is virtually infinity - try compressing a block of wood between your fingers. So pulling down on the block does not increase the load on the spring until the preload is exceeded and the pieces actually separate. The stiffness ratio in a bolted joint is much less than infinity.
Rob Campbell
----------
If the bolt is suffiently preloaded, it is not subject to fatigue. The cyclic load reduces the contact stress between what is being bolted together. Generally (and there are exceptions), the bolt stress does not change until the length of the bolt changes, which means the contact surfaces separate. So you have to overcome the bolt preload before bolts experience cylic loads.
----------
That simply isn't true. Maybe the key to the midunderstanging is the mistaken belief that the bolt does not change length until separation. The bolt, which is under tension in the first place, does indeed lengthen. The clamped members, which are in compression, relax by the same amount, so there is no separation as long as there is some of the initial compression still in the members.
The portion of an external load, P, carried by the bolt is:
Pb = kb*P / (kb+km)
Where kb and km are the stiffness of the bolt and the clamped members, respectively. Therefore, increasing the stiffness of the clamped members relative to the bolt decreases the portion of the external load carried by the bolt. But it does not eliminate it.
One of the figures commonly used to illustrate this concept is a block of wood wedged between a force pull gauge (essentially, a compression spring). In this case, the stiffness ratio is virtually infinity - try compressing a block of wood between your fingers. So pulling down on the block does not increase the load on the spring until the preload is exceeded and the pieces actually separate. The stiffness ratio in a bolted joint is much less than infinity.
Rob Campbell
PAN,
It depends on the design - geometry, materials, and forces. The washer can provide increased bearing area and higher/more consistent preload, which may be necessary. However, the design may not require extra bearing area or more consistent preload, therefore it won't need washers.
Regards,
Cory
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.
It depends on the design - geometry, materials, and forces. The washer can provide increased bearing area and higher/more consistent preload, which may be necessary. However, the design may not require extra bearing area or more consistent preload, therefore it won't need washers.
Regards,
Cory
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.
Speaking of pressure vessels and piping.
Flat washers are a bad habit to get into, always try to fix the bearing surface where the nut lands. As stated previously we don't use washers on any flange bolting. I seen too many problems with flanges and leaks caused by the use of washers to correct a crippled flange. This has come in the form of cupped washers, washers as bad, proud metal, as the problem they are trying to fix, thinned, worn out washers, stacked washers, and so on.
If a washer is used on a flange due to a problem, real or otherwise, this flange is corrected the next outage.
Having said all that, if you deem a washer is needed make sure you get the proper one for the job.
Flat washers are a bad habit to get into, always try to fix the bearing surface where the nut lands. As stated previously we don't use washers on any flange bolting. I seen too many problems with flanges and leaks caused by the use of washers to correct a crippled flange. This has come in the form of cupped washers, washers as bad, proud metal, as the problem they are trying to fix, thinned, worn out washers, stacked washers, and so on.
If a washer is used on a flange due to a problem, real or otherwise, this flange is corrected the next outage.
Having said all that, if you deem a washer is needed make sure you get the proper one for the job.
- Status
Similar threads
- Locked
- Question
- Locked
- Question
- Locked
- Locked
- Question