1. General Input

Q: We are trying to use SABRE to design a dodecagonal monotube, cantilever traffic mast arm. Two arms are to be attached to the post, one above the other. There doesn’t seem to be a way in SABRE to design/analyze this type of structure. Is this true, or can this configuration be modeled with this program?

A: You may first model it as a one arm cantilever structure, but with the true post height. Once the model is set and mesh is done, all you have to do is to add joints and members for the upper level arm. The new joints and members can follow the current generated ones. You may also add concentrated load(s) to the upper level arm. Once those modeled, don’t press the “mesh” button again. Otherwise, all the added information will be overridden.
The screens under the “Structural generation” are helping to easily generate the sign structure modular model, but are not necessarily the only way. You may use the screens for joints, members and optional joint loads to add or modify the modulated model.

Q: Under ELEMENT DEFINITION, What exactly is a segment? Does it correspond to the number of segments (panels) within the truss with the exception of the connection length? What about number of segments for the posts? Does the output give data at each segment?

A: For truss, segments are where truss nodes locate to create 3-D truss panels. You should test one case and see how the program generates truss span or post for you. For single post, segments are like output points. For truss (double or four leg) post, it’s the same as truss span.

Q: Concerning a VMS’s sign offset, how does the program know how eccentric the VMS is? Does it take the VAMS z-dimension as the offset dimension and then add half the sign thickness in the z-direction? Or is the load applied to the structure directly thru the VAM?

A: Yes, z-dimension is used as the eccentricity.

Q: Under the VAMS menu, the attached sign units and walkway units: If there is no number here, does that mean that the sign is not attached to these VAMS? Will the sign load be applied to the structure thru the VAM member?

A: Yes, the sign (and sign load implied) has to connect the sign structure through VAMs. If not connected, when you click MESH button, warning message will show.

Q: Can we control the Cd (drag coefficient)? If so, on which screen? Our project criteria requires a Cd of 2.0 for the VMS, 1.16 for the static sign, and 1.10 for the tubular members.

A: Yes, you may, by an option screen (Data Type 6012).

Q: On the SIGNS menu, is the z-coordinate the CG of the thickness (i.e. if the sign is 5-ft thick, would the z-dimension be ?ö the chord diameter + the VAM thickness + 2.5-ft ?)

A: When the SABRE was developed, Variable sign is not as popular as nowadays so z-coordinate is directly used for eccentricity. If variable sign unit is used, you should enter the z-coordinate at the center of the load, i.e. ?ö the chord diameter + the VAM thickness + ?ö the sign.

Q: For the cantilever-with-plane-truss-chord example provided with the program, the CROSS SECTION screen gives BLT, BRT and BLB. Why not BRB? Is it defaulted? Or can’t the bottom member be a variable section? Also in this example, there is a Section 3 coded as 999 for a connection. I don’t see where the section 3 is used.

A: The way to generate the input data is quite flexible. BLB is really the section number for the connection member and can be Section 3. I modify the example to two different types: (1) a monotube type with variable sections for both post and cantilever arm, and (2) single post/plane truss with variable sections for both post and cantilever truss, for your reference.

Q: For the double-span example (dbl-span.dat), the CROSS SECTIONS screen has Section 3 for the LBLB, LBRB, RBLB and RBRB. Section 3 is a 999 (connection). Why? I thought that LBLB, for example, is the “Left Beam Left Bottom” meaning the section number of the bottom chord on the left? In general, where would one use the 999 designation?

A: We should change the LBLB, LBRB, RBLB and RBRB to Connections. The program assumes top and bottom chords use the same sections, unless you override them on the “Definition of Member” screen.

Q: I still don’t see what the 999 member does. Do we need one in every run? Should BLB in the cantilever example be a 999 member? Or just the same as BLT and BRT? [The TEST.dat you provided shows same section ID as top (BLT and BRT). That being the case, I don’t see why we would need 999. Would it make any difference in the output if I used BLB = 999 versus BLB = 1 or 5?]. I see in the Input Verification section of the output that the Section Type for connection members is “mm”. What does that mean? Where does the section properties (very high) come from for a 999 member?

A: Section 999 is for a dummy rigid member. You may simulate it with regular member with a relatively large section. The difference is not much. So, you don’t need for every run. Instead of BLB, it should be the connecting member section. So, if there is a section 999, it should be input for BLB. “mm” indicates a dummy section.

Q: For connection members (max length = 2.5-ft) would you recommend an accurate portrayal of the model here? That is, is the connection member length from the CL post to CL of the first vertical? Can we get the program to give us forces/stresses at intermediate points along this member? (in order to check fatigue stresses at stiffener terminus and splice plates which may be at midpoint of the connection member length).

A: Yes, you may let the program to generate members and nodes. If there is a need for additional intermediate node, just use the insert row command on the screen of “Definition of Members” and divide the corresponding member by inserting a node. This node number is the highest node + 1.

Q: If I can change the Cd factors in data screen 6012, what is the format? Is there a way to change the Cd for the VMS sign versus the structural support members? Our client requires a very high Cd for the VMS of 2.00 and a value of 1.10 for the tubular members.

A: (1) The default factors: For single vertical post case, K=2.0; For double vertical post with truss members connected, K=1.2, Chords and truss members, K=0.65 (2) To override factors, use Data Type 06011 as the leading line and follow with 06012 with values. You may specify

  1. Member ID
  2. Wind data, Shield ID (Shield ID = 1 shielding on the -Z direction; = 2 shielding on the +Z direction; = 3 no shielding; = 4 Both shielding)
  3. Wind data, wind drag coefficient
  4. Wind data, Height coefficient
  5. Axial Allowable Fa
  6. Bending Allowable Fb
  7. Shear Allowable Fv
  8. Effective length K Factor

Q: Program assumes k = 0.65 which is for fixed end conditions. Is this the k-value for both in-plane and out-of-plane buckling? Particularly for chord member, can’ t the chord buckle out-of-plane requiring a much larger L in determining allowable compressive stress?

A: See above.

2. Load Input/Output

Q: The sign I am using is 16.0 ft by 4.2 ft. The ice load should be 201.6 pounds. Your output on table 5.1 Shows the ice load on the sign to be .332 kips or 332 pounds. Correct? What am I doing wrong?

A: The program considers not only the sign itself but also the VAM. You re using the wide flange section (section type 9), the program calculates all the surface area and the length, then multiply by no. of VAMs for the additional ice load.

Q: Our criteria requires that Snow Load be added to the top of the VMS sign only. It also requires Ice Load on the top of the VMS (contrary to AASHTO). The snow load and ice load are additive. How best to apply these loads?

A: Our ice loads are applied to one face of the sign and the top. Snow load is not part of the loads defined by AASHTO. In order to count for the snow load, you probably have to use equivalent ice load. The joint load option, up to this point, is considered for all load combinations, so it’s more like a dead load. We consider this as an additional request for future consideration.

Q: Gust Factor?

A: For the 4th edition, the default value for the gust factor is 0.14, but the default value for the 3rd edition is 0.3. Gust factor can be overridden by the optional screen for Parameters.

Q: Is the GUST FACTOR the factor in AASHTO Article 3.8.5? Do I enter 0.14? Or 1.14? It appears that either way, the results are still the same.

A: There is an internal adjustment to make the multiplier, I, 1.14, no matter 0.14 or 1.14 is input.

Q: How can I see what the actual applied wind pressure is? If I can see this, then I can adjust per the client’s project criteria.

A: Tables 4.2, 4.4, 4.6, and 4.8 show the wind loads on each member and the wind force summation on the main (normal) and secondary (transverse) directions. As shown in Table 4.1 near the end of the output, these four (4) tables show two combinations, (1) 100% normal direction + 20% transverse direction and (2) 60% normal direction + 30% transverse direction. You may extract information from there.

Q: Regarding the extra snow and ice loads, I think I’ll just run a separate run with snow and ice as a joint load. Then use one run for Load Cases 1 and 2; and the second run for Load Case 3 with snow and ice. Does that sound reasonable?

A: It sounds ok.

3. Analysis Output

Q: What is the UNIT NUMBER in the output? I see that the Post is Unit Number 1, the Bottom Chord is Unit Number 2, and the Top Chord is Unit Number 3. The diagonals and verticals are Unit 0. What about the “Orientation” and “MEM. OR.”?

A: Unit Number is our own designation for internal management. It is defined in the documentation. Orientation is for section properties. It has the same definition as the orientation defined in any structural program. For doubly symmetric section, this has no meaning.

Q: Table 1.7 in the output lists Coefficients of Height, Ch. Where are these coefficients from? Are they analogous to the Height and Exposure Factor, Kz in AASHTO’s Table 3-5? If so, they do not appear to match. Is the program using an equation? Why are some of the members showing a zero Height Coefficient?

A: The default Ch values are more corresponding to the one defined in Edition 3. You may alter those values in your input under Option then Height Coefficients.

Q: DL4 in the output Table 3.1 – I would expect this to be the sign volume times the density plus the weight of the VAMS. But it appears to be greater than that. What does DL4 include and how is it calculated? To which DL is the joint load added?

A: Tables 3.1 and 8.1 are independently calculated, based on per VAM and its tributary area, with detailed factor included. They have been verified many times. If you have problems obtaining the summation of DL4, send us the data and your calculation.

Q: For the member properties output, there is a “Compressive Area” and a “Tensile Area.” What input differentiates between the 2?

A: After checking the program, I found that they are suing the same arrays, which meant there is no difference between the two.

4. Stress Output

Not available.

5. Code Check Output

Q: We had an existing sign structure that was analyzed using the old version and wanted to compare it to the output produced by the new. I was wondering if you could tell me why the output from the two different analyses varies. If you look at the combined stress ratios and support reactions, they can vary by 30%. I have referred to both the 3rd and 4th edition AASHTO Manuals and it appears that the wind pressure is computed the same way.

A: We are pretty sure that the difference between two codes is wide because of:

  1. G Factor was 1.69 (=1.3*1.3) but it was wrong. The new code pointed out it should be 1.14. The old wind load is 48% higher than the new one. Please read p 3-12 of the 4th edition code w/2002 interim.
  2. % allowable stress was 1.4 but it is 1.33 now (p 3-12). So, the new allowable is 5% lower than the one.
  3. The rest of the changes are trivia. You may see the difference caused by the above two items is 1.48/10.5=1.4. So, there is still 40% difference here.

6. Post-process Design

Q: When you designing splice plate, are you considering pressure distribution between plate into account to determine the bolt size, or just considering only bolts. Also are you designing the bolts for service load or factored loads?

A: When the bolts are designed, no pressure distribution is considered and half of the bolts take tension while under bending. The calculation is exactly shown on the user’s manual. Also, the bolts are designed as service load. There are several ways to design the based plate and the bolts. The built-in design is considered one of the recognized methods. You may establish yours, if so desired.

7. System

Q: My state uses the warlier Specifications for Sign Structures. We now are asking to try the latest version of SABRE. Would you please send me a complete up-to-date copy of SABRE 2001?

A: We combined both AASHTO 4th ed. (2001) and 3rd ed. into one. Since the input pretty much the same, all we have done is to click one of the two executables under the Analysis on the main menu bar.

Q: Are there currently any issues running Dash on Window XP Pro? We keep getting Driver Error when trying to run the program on Windows XP Pro.

A: It might be the Driver is not updated. You may download the new Driver from web http://www.rainbow.com/, then Support, then Download. Get the first one on the list Sentinel System Driver 32 bit MSI installer v5.41.1.