SAK123
Structural
- Oct 1, 2006
- 23
Thank you for reviewing this question.
I have a concrete wall foundation with embedded concrete anchors placed every 12 inches. The embedded concrete anchors follow ASTM specification have 1554-07a. Placed over the concrete anchors are structural steel angle members. The structural steel angle members are in place to hold down structural fabric along with 1/8 inch thick gasket material. The fact that the joint has structural fabric and gasket material in the grip classifies this joint as a soft joint. The implications of the soft joint are as follows, the total load on the anchor is equal to the pre-load ADDED DIRECTLY with the bolt load induced from wind forces. This is not the same as a rigid joint where a high preload is desired to eliminate any possibility of cyclic loading, as the preload level itself if high enough will eliminating any cyclic stress. So the first part of my problem, is that I cannot arbitrarily choose to preload the anchor bolt to such a high torque hoping to eliminate any induced fatigue loading from wind loading. The higher preload when added to wind induced load, will result in a much higher total bolt load.
The second part of the problem, which is really the core of my question, is how do we determine the number of cycles that are appropriate to apply given the fatigue analysis procedure as described in the AISC. manual? There is a section in the steel manual (bolt / anchor fatigue) that describes the appropriate procedure for determining the allowable stress cycle range given the number of cycles for that particular load. What makes this issue so complex is I do not have any deterministic means of estimating the number of wind load levels (namely wind speeds) that will occur over the life of the structure. For example, the maximum wind speed the structure is designed for is 180 mph. The question is how often does this wind speed occur during the 12 year life of the structure? What about a condition where the wind is a sustained 100 miles an hour with only a few cycles reaching 120 miles an hour? In this case the increase in stress would be due to stress increased from 100 miles an hour to 120 miles an hour, basically a relatively benign stress increase. The only site information I have for the wind is a plot of wind speed velocity versus return frequency. I'm not sure exactly how to convert this linear probability plot into data that corresponds to the number of cycles that will occur at any given wind speed over the 12 year design life of the structure. Other complexities arise when one considers sustained winds and gusts that can occur and induce additional cyclic loading.
Does anyone have any experience or idea on how to convert the probability of occurrence versus wind speed velocity into useful number of cycles that I can use the equation provided in the steel manual?
I have done some preliminary calculations assuming a storm duration of one hour and that the cyclic loading occurs from a zero when condition up to the wind velocity of interest. For example, for a wind speed of 180 miles an hour, I can estimate the induced stresses on the anchors. The method I use first determines the number of one hour storms that can occur over a 12 year life structure. This calculates to be 105,120 one hour timeslots over 12 years. Given the information that the 180 mph wind speed has a probability of occurrence of one in 10,000 (from the plot I have), when multiplied times the number of timeslots available comes to be (1/10,000 * 105,120) 10.5 occurrences of the 180 mile an hour wind over the 12 year life span. Now that I know the number of cycles is 10.5 and I know the stress level goes from zero to the maximum induced stress at 180 miles an hour, I can calculate the design fatigue range permitted by the steel code. The problem is, using this procedure I described results in a fatigue life degradation that is excessively high. Given that for 180 miles an hour I have a very high fatigue life portion used, repeating this for all winds (0 thru 180 mph) will degrade the fatigue life more. Using this conservative approach is resulting in an estimated fatigue life being used up long before the 12 year life of the structure, i.e. adding all the applied stress/allowable stress range for each of the wind speeds, results in a sum greater than Miners rule maximum of 1.0.
Any assistance would be appreciated. Thank you for your time.
If you have any questions please post them as I will be watching this discussion thread quite closely anxious for some response.
I have a concrete wall foundation with embedded concrete anchors placed every 12 inches. The embedded concrete anchors follow ASTM specification have 1554-07a. Placed over the concrete anchors are structural steel angle members. The structural steel angle members are in place to hold down structural fabric along with 1/8 inch thick gasket material. The fact that the joint has structural fabric and gasket material in the grip classifies this joint as a soft joint. The implications of the soft joint are as follows, the total load on the anchor is equal to the pre-load ADDED DIRECTLY with the bolt load induced from wind forces. This is not the same as a rigid joint where a high preload is desired to eliminate any possibility of cyclic loading, as the preload level itself if high enough will eliminating any cyclic stress. So the first part of my problem, is that I cannot arbitrarily choose to preload the anchor bolt to such a high torque hoping to eliminate any induced fatigue loading from wind loading. The higher preload when added to wind induced load, will result in a much higher total bolt load.
The second part of the problem, which is really the core of my question, is how do we determine the number of cycles that are appropriate to apply given the fatigue analysis procedure as described in the AISC. manual? There is a section in the steel manual (bolt / anchor fatigue) that describes the appropriate procedure for determining the allowable stress cycle range given the number of cycles for that particular load. What makes this issue so complex is I do not have any deterministic means of estimating the number of wind load levels (namely wind speeds) that will occur over the life of the structure. For example, the maximum wind speed the structure is designed for is 180 mph. The question is how often does this wind speed occur during the 12 year life of the structure? What about a condition where the wind is a sustained 100 miles an hour with only a few cycles reaching 120 miles an hour? In this case the increase in stress would be due to stress increased from 100 miles an hour to 120 miles an hour, basically a relatively benign stress increase. The only site information I have for the wind is a plot of wind speed velocity versus return frequency. I'm not sure exactly how to convert this linear probability plot into data that corresponds to the number of cycles that will occur at any given wind speed over the 12 year design life of the structure. Other complexities arise when one considers sustained winds and gusts that can occur and induce additional cyclic loading.
Does anyone have any experience or idea on how to convert the probability of occurrence versus wind speed velocity into useful number of cycles that I can use the equation provided in the steel manual?
I have done some preliminary calculations assuming a storm duration of one hour and that the cyclic loading occurs from a zero when condition up to the wind velocity of interest. For example, for a wind speed of 180 miles an hour, I can estimate the induced stresses on the anchors. The method I use first determines the number of one hour storms that can occur over a 12 year life structure. This calculates to be 105,120 one hour timeslots over 12 years. Given the information that the 180 mph wind speed has a probability of occurrence of one in 10,000 (from the plot I have), when multiplied times the number of timeslots available comes to be (1/10,000 * 105,120) 10.5 occurrences of the 180 mile an hour wind over the 12 year life span. Now that I know the number of cycles is 10.5 and I know the stress level goes from zero to the maximum induced stress at 180 miles an hour, I can calculate the design fatigue range permitted by the steel code. The problem is, using this procedure I described results in a fatigue life degradation that is excessively high. Given that for 180 miles an hour I have a very high fatigue life portion used, repeating this for all winds (0 thru 180 mph) will degrade the fatigue life more. Using this conservative approach is resulting in an estimated fatigue life being used up long before the 12 year life of the structure, i.e. adding all the applied stress/allowable stress range for each of the wind speeds, results in a sum greater than Miners rule maximum of 1.0.
Any assistance would be appreciated. Thank you for your time.
If you have any questions please post them as I will be watching this discussion thread quite closely anxious for some response.