|
|
|
|
|
Important Note – These are descriptions to changes implemented in WoodWorks Sizer for version 10.1 and may not reflect current program behaviour.
The program now allows you to specify the faces of a member that are exposed to fire. Previously, for you could only select from 0, 3 or 4 sides exposed, and the program would assume 3 sides was 2 side faces and top or bottom.
The Fire Design data group has checkboxes surrounding a section of the member allowing you to specify which of the 4 faces are exposed.
For timber or glulam designed using CSA O86 Annex B, any or all of the sides can be selected.
For the NBC fire design method for glulam, only 3 sides and 4 sides are allowed, as before. You can choose which of the smaller faces are exposed, or whether both these faces are exposed. Both larger faces are always exposed.
For CLT floor and roof panels, you can select the top or bottom but not both. Similarly, for wall panels, left or right, but not both.
Fire design is deactivated by deselecting all checkboxes, which is the default condition.
Previously, when an invalid material like built-up lumber members or SCL was selected, the program would allow input of number of exposed sides then revert to 0 when the design button was pressed. Now it disables the input of exposed sides when one of these materials is selected.
For the CSA O86 method, the program reduces the design section by calculating a char depth for each exposed face.
The choice of top or bottom exposed beam or CLT floor panel surfaces does not affect design. Neither does the choice of left or right beam surfaces, or column surfaces perpendicular to applied loading.
For column surfaces parallel to the applied force and CLT wall panels, the choice of left or right surface can have design consequences due to axial load eccentricity.
The choice of smaller exposed faces for the NBC method has no design consequences.
Input fire protection is assumed to apply to each exposed face.
The choice of exposed faces is shown in the materials specification of the Design Check output as follows, as the case may be:
Exposed to fire on [ one [b,d]-face, opposing [b,d]-faces, both [b,d]-faces and one [b,d]-face, all four faces ]
The following problems with the calculation of the effective section modulus Seff from O86 8.4.3.1 for CLT moment design were corrected.
The section modulus Seff,x from O86 8.4.3.1 for the minor strength axis (transverse) CLT design was calculated with a panel depth h which included the outer longitudinal layers, when the depth hx with these layers excluded should have been used.
For a typical example of a 315 mm thick V1 grade floor panel, the section modulus was 5.62 million mm^3 when it should be 7.22 million mm^3 resulting in a bending moment resistance of Mr of 23.26 kN-m when it should be 19.44 kN-m.
The program was not considering the note in O86 B.6.2 regarding the need to calculate the location of the neutral axis when determining the moment of inertia and section modulus fire design bending moment resistance. For a typical example, this problem caused the bending moment capacity to be calculated as 4932 lb-ft when it should have been 3904 lb-ft.
The method for calculating Seff for fire design is given in the FPInnovations CLT handbook, Chapter 8, Eqn. 9, in which the term h/2 in in O86 8.4.3.1 for is replaced by h – y, where y is the neutral axis given by Handbook Eqn. 4 as ∑ yi ti / ∑ ti where yi is the distance to the centroid of each layer, and ti is the thickness of each layer.
Note that since E┴ perpendicular to the direction of loading is E/30, those layers are ignored in the calculation, so this simplified formula is used rather than Eqn. 6 in the CLT Handbook, which includes Ei in the summations in the numerator and denominator.
For the interaction equations for combined axial and bending resistance, for both tension and compression, Sizer now applies the duration factor KD to both moment resistance Mr and the axial resistance Pr or Tr for the shortest duration of loads (highest factor) for either direction of stress. In other words, the program examines a load combination for combined design, and uses the load duration factor corresponding to that combination for both axial and bending resistance, regardless of which load types within the combination contribute to axial and bending stress.
Previously, the program applied KD factors calculated separately for axial resistance and bending resistance using only the loads that contributed to stress in each direction.
For example, for a column under concentric compressive dead load and lateral live load, the program used 0.65 for compression and 1.0 for bending, but now uses 1.0 for both.
These interaction equations are found in O86 6.5.10, 7.5.12, 8.4.6, and 15.3.3.9 for sawn lumber, glulam, CLT and SCL, respectively. The procedure of using the shortest term KD factor is shown in the CWC Wood Design Manual, Section 5.1, Example 1 - Glulam Column and in the CWC’s Introduction to Wood Design 10.1, Example 10.1 Column subjected to snow, wind and dead loads.
For the case that O86 5.3.2.3 is used to determine a KD for long-term loading, the program applies the highest factor so calculated to both directions.
In the Factors table of the Additional Data, the program was showing the load combination number for the loads contributing to axial stress design for both the axial and bending lines in the table. This load combination corresponded to the KD factor used in this check, however, due to the change for Bug 3385, above, the same load KD factor is now used for both axial and bending components in the combined equation, and the critical load combination will be in fact the same for these components.
Please note that the fact the same load combination number was shown may have misled users to believe we were considering
The following problems affecting columns and walls loaded on the d-face entered the program for version 10 and have been corrected.
The built-up bending strength for No. 2 Grade members that is used for glulam weak-axis design using O86 7.5.3, was being used to calculate the weak-axis lateral support factor KL (O86 7.5.6.4.4) for sawn lumber materials, instead of the published bending strength for those materials.
When built-up column materials were set to Ignore in Database Editor, designing any sawn lumber or glulam column in Sizer caused a crash.
The warning messages shown when built-up column database files were missing or disabled in Database Editor so that glulam weak-axis glulam design according to O86 7.5.3 was not possible, have been clarified and improved, and the same message now appears for both beams and columns.
The calculation of the critical notch length of 0.25d in O86-14 7.5.7.4.1 was including ½ the min. req’d bearing length, when it shouldn’t have. Beyond this critical length the shear strength is based on residual depth rather than full depth, and it is measured between the member depth d from the inner edge of the support and the beam end. This has been corrected.
The program was not applying the 20% allowable vibration span increase from O86 A.8.5.3 for CLT floor panels due of the effect of non-structural elements when this option was selected in Beam view. This has been corrected.
Starting with version 10, a point of interest was added to a wall stud or column in Column Mode, the program crashed when member design was invoked. It did not happen for beams. This has been corrected.
If the Design setting Satisfies lateral support conditions and d/b for KL= 1 indicating that lateral support conditions from O86 6.5.4.2 are met, and the checkbox in the Supports for bearing design data group indicating that interior multi-span supports are not laterally supported was unchecked for any interior support of a multi-span beam, the program now longer over-rides the KL = 1 setting to apply the lateral support factor KL based on 7.5.6.4.
The setting was overridden because of the requirement in 6.5.4.2. that lateral support be provided at points of bearing to prevent lateral displacement and rotation. This has been reinterpreted to mean end supports only, as support in two places is sufficient to prevent displacement and rotation of the beam. KL = 1 is now applied regardless of lateral support for interior supports.
When a new span is added to create a multi-span beam, the Laterally supported at support checkbox for interior supports is now unchecked by default. Previously it was checked, but in most common situations lateral supports are not provided to interior supports.
The following changes have been made to the explanatory text that appears in the Lateral support spacing section of Beam view under certain circumstances.
The text when the KL = 1 Design setting is set has been revised to remove the explanation that KL can be overridden if there are unrestrained interior lateral supports, as described in the previous item.
For multi-span beams, text now appears indicating whether interior supports are restrained, as the input for this is under the Supports for Bearing and Notch Design and not immediately evident in this section of the Beam view.
The explanatory text now appears when the setting Calculate KL using 7.5.6.4, is selected, indicating that the use of the spacing input depends on d/b > 4 as per O86 6.5.4.2.1 (a). Previously it only indicated that the inputs only apply when d/b > 9 when KL = 1 was selected as per 6.5.4.2.1 (f), which it still does.
The explanatory text now appears in all cases for glulam, indicating that the use of the spacing input depends on d/b > 2.5 as per 7.5.6.3.1.
The following changes were made for the Support for bearing design input for CLT wall panels.
The Type choice Bottom plate has changed to Sill plate. Bottom plate is relevant to framed walls only.
The Bearing length Lb choices have changed from Column width and Column depth to Panel width and Panel depth.
When Sill plate is selected as the type, the Bearing length Lb input is now disabled and shows Panel width. That is, we assume they are continuously supported and show the calculation for the 1m or 1ft standard width.
When Panel width is selected as the Bearing length Lb, the lower support will be set to None and disabled. This will always be the case for Sill plate support type.
This is because we assume continuous upper wall panel support for the standard 1 m or 1’ panel width, in which case the lower support of the sill plate or CLT floor becomes irrelevant, because O86 8.4.7.3.2 for loads at the support reduces to O86 8.4.7.3.1 for all other conditions when one bearing length is greater than 1.5 times the other.
For exterior supports of CLT floor and roof panels, the program was always using a value of 1.5" or 38 mm as the minimum bearing length, instead of the one input in the Design settings. This value is used as a lower limit on the design bearing length and appears in a note under the Bearing design table when used as the design bearing length.
The following problems pertaining to the CLT long-term deflection creep adjustment factor Kcreep from O86 A.8.5.2 were corrected
For floor panels, the default Kcreep that appeared in Load Input view for new files was 1.5, but this value should have been 2.0, as per O86 A.8.5.2. This has been corrected.
The input for Kcreep in Load Input view was available only for floor panels, and Sizer used the default value of 2.0 O86 A.8.5.2 for roof panels. It is now available for roof panels. Previously Sizer used the default value of 2.0 for roof panels.
For fire design of CLT wall panels with doubled outermost parallel layers, the calculation of shear rigidity (GA)eff from 8.4.3.2 now treats the doubled outermost parallel layers on either side as a single, doubly thick layer. For fire design, (GA)eff is used in the resistance to combined axial and bending from 8.4.6.
There are no standard CLT layups with doubled outermost layers, but it is possible to create doubled custom layups using Database Editor.
The input of preservative or fire-retardant treatment has been activated for CLT materials based on O86 8.3.3.
The user input factor is used for KT due to fire-retardant treatment.
For preservative treatment, O86 6.4.3 for lumber is used, as there is no guidance specifically for CLT. The member thickness used in Table 6.4.3 to select the factor KT is the thickness of the smallest CLT layer because 6.4.3 indicates that the treatment must be applied before gluing.
In practice the factors 0.90 for modulus of elasticity and 0.75 for shear, bending, axial compression/tension and bearing design are used.
The following changes have been made to the modification factors used for fire design.
For beams you designate as having load sharing, the program was assigning a system factor KH as it would for non-fire design. It now assigns a factor of 1.0 as per O86 B.3.4.
For members you designate has subject to wet service, the program was assigning a service factor KS as it would for non-fire design to both shear and moment design. The program now applies these wet service factors for glulam only, but uses KS = 1.0 for sawn lumber, as per the CWC Wood Design Manual, 10.5 Determining Fire Resistance Ratings, CSA O86 Annex B Method, Beam Example 1.
Dashes appear in the Factors table output in the KS column for sawn lumber service factors.
CLT does not allow for wet service.
For members you designate has having preservative or fire-retardant treatment, the program was assigning a treatment factor KT as it would for non-fire design. For preservative treatment, the program now program assigns KT = 1.0 for both sawn lumber and glulam. For fire-retardant treatment, it assigns the factor you input for glulam only, but uses KT = 1.0 for sawn lumber. These procedures are also from the CWC Wood Design Manual, 10.5, Beam Example 1.
Dashes appear in the Factors table output in the KT column for preservative treatment and sawn lumber fire preservative treatment factors.
CLT does not allow for chemical treatment.
For design sections having analysis values, e.g. Mf , that are greater than design resistance values e.g. Mr, by an amount that is less than ½ of 1% of the value, the Analysis vs. Design table in the Design Check showed a 1.00 design ratio (when in decimal format) and showed a passing section note instead of failure warning.
Now, the program considers a design to be failed if the ratio is greater than 1.0005, and outputs the ratio with an extra digit of precision, e.g. 1.003. For example, a member with Mf = 20295 lb-ft and Mr with 20205 would show a ratio of 1.00 and pass, but now it shows 1.004 and fails.
When percentage is chosen as the design ratio output in the Preference settings, a greater tolerance in determining design failure is possible. In this case, the program currently considers a design to be failed if the percentage is greater than 100.05% (1.0005). Now the program considers design to be failed with a ratio greater than 1.00005, and outputs the percentage with a digit of precision, e.g. 100.03%. For example, a member with Mf = 20215 lb-ft and Mr = 20205 will show a ratio of 1.00 and pass when decimal is chosen, but when percentage is chosen, it shows 100.04% and fails.
The behaviour of the program regarding the compressive size factor Kzcp for sill plates vs. beam supporting members for columns was inconsistent and error-prone.
As per CSA O86 6.5.7.5, sill plates should have a Kzcp of 1.15 in most cases because they are on the flat, whereas beams are assumed to have a Kzcp of 1.0.
In some cases, when the support was changed to a beam then back to a sill plate, the Kzcp changed to 1.0, the value for beams.
In another case, there were two files with identical input and a sill plate support, with one showing a 1.0 factor and the other showing a 1.15 factor.
In all these cases, the incorrect Kzcp factor was used in design, resulting in an erroneous Qr value in the Analysis vs Design table. This has been corrected.
The following problems that occurred when Simpson Hangers were used as the supporting member, a feature introduced in version 10.
When there was no Simpson hanger in the Simpson database for the combination of program inputs, the program designed the hanger with zero bearing length, zero capacity and issued a bearing failure warning message. Now the program treats the hanger as a non-designed hanger and performs main member bearing design only.
After a design check of a member with an unknown Simpson Hanger support type, the program now selects the designed Simpson hanger model and fasteners in the Hanger options upon return to Beam view.
For I-joists with SPF flanges as main member and Simpson hanger as support, the program designed the Simpson hanger using load and uplift resistance for I-joist with Douglas-Fir (DF) flanges rather than Spruce-pine-fir (SPF) flanges. This has been corrected.
The resistance shown under Hanger options for Simpson hangers displayed in pounds regardless of the Format setting for Force. It now shows the value in kips if that is selected.
In those cases that there were two Simpson hangers with identical model name and fastener specification on all three flanges, the program lists only the first of these in the selection dropdown and considers only that hanger for unknown fastener design.
This is because it is impossible to distinguish between these two hangers when querying the Simpson database.
Sizer was not including joist materials with the same material name as beams in its selection list for headers supporting the hanger and main member. Now lumber, rough lumber, MSR, MEL, and proprietary SCL joists are included as possible hanger selections, with the word "Joist" beside the material name to distinguish them.
When SCL materials were set as the main member, the program often passed the wrong species of the header member to the Simpson database, returning incorrect resistance information. This has been corrected.
Sizer now displays the model number, fastener information and any special information in the status bar when a hanger is selected.
The default glulam fire design method in the Design Settings is now CSA O86-14, Annex B, which is a mechanics-based approach, instead of NBC, Appendix D-2-11 which is an older, empirical approach.
In Beam mode, Design Summary for unknown design, for all sections except the first one listed, the program was showing a shear design ratio for a load combination other than the critical one used to select the section.
This was a display issue only and did not affect the determination of the section to be shown. If that section was then selected for a design check, the design ratio in the Design Check output was correct.
On rare occasions, incorrect load duration factors program found their way into the initialization file and project files, causing incorrect design results or making design impossible. The program now checks for and prevents this condition.
For versions before 10.1, it is possible to correct this problem by clicking Reset original settings and Save as default settings buttons in Loads view, then saving the project file.
For extremely large beam members, such as a 25 m long 2 m deep member, Sizer could detect a small negative moment due to accumulation of numerical rounding when no such moment exists. In this case, the program would report weak axis design results when it shouldn’t.
Although the bending moment was extremely small, the program could mistakenly fail the design due to an invalid slenderness ratio using the lateral support conditions on the side of the beam with the negative moment.
Due to the size of the members, it is very unlikely to have occurred in practical situations. It has been corrected
The design note regarding continuity for beam and stringer grades for multi-span beams based on CSA O86 6.5.3 has been removed, as this provision was removed from the CSA O86-14 update 2.
A design note has been added for CLT floor panels to give the vibration span adjustment entered in the CLT floor vibration (O86 A.8.5.3) Design setting.
The program now shows the design note for Simpson Hangers when they are specified as supports for joists. Previously the note appeared for beams only.
For steel beams, the program referred to the 2009 edition of the CSA S16 in both the Force vs. Resistance table and in a design note, but the program was using the 2014 edition and indicated so by CSA S16-14. The number 2009 is removed, as stating the year here is unnecessary.