Steel Support Frame Details for Mechanical Unit Support

[Deadblow]

Hello,

Please refer to the attachment. I am designing a support structure for three mechanical units to sit on it. The structure will span over the roof of a walk-in cooler and the units will be bolted directly to the W16x67 beams. I am thinking that these long beams will be connected to the columns with single plate shear connections to handle the shear and then have full penetration welds at the top and bottom flanges to resolve the moment. Does that connection sound like the best approach? Can the top flange weld be made to the column cap plate if the cap plate is sufficiently welded to the column?

I haven't figured out my wind and seismic loads just yet but hope to do so today. The structure will basically be constructed of two 50'-0" long moment frames. I am generally looking for advice on the best way to connect the two 50'-0" long frames together so the structure can resist lateral loads transverse to the 50' frames. I will ask specific questions below:

1. What is the best way to connect the columns together as shown in "ELV. 2" to create a 7'-0" long frame transverse to the 50' frames? Should I just put a beam at the top and design it for moment connections, or is "X"-bracing with angles or rods a better design? There will be no gravity loads that need supported at this location.
2. What is the best way to connect the two 50'-0" long beams together in order to get transverse loads over to the two 7'-0" long frames? Should I use angles welded to the underside of the top flanges at approximately 45 degree angles like my sketch shows?
3. If I do connect the 50'-0" long beams together as I propose in question #2 above, would it be appropriate to design the beam with the compression flange braced at those intervals? Right now I have conservatively designed the beam as completely unbraced. If I can consider it braced somehow, I could use a taller and lighter beam with less deflection.
4. All of this steel will be exterior and will be exposed to the elements. Is hot-dipped galvanizing, HDG, my best option for corrosion protection? Does HDG hinder the ability to make welded moment connections in the field?
Thank you in advance for your input!

EIT

 

[SlideRuleEra]

IMHO, W16s for a 50' span... span to depth ratio of 37.5:1 (600" / 16"). Not a good start: I'm sure the calcs show this is "strong" enough and deflection is probably "ok", but the assembly will be excessively "springy". In general, mechanical equipment should not be bounced around while operating. Also, not good for any piping or electrical supply connected to the the equipment.
My advice, look at a light weight, deeper beam to get span/depth ratio down to say, 25:1... and brace the frame, no moment connections.

http://www.SlideRuleEra.net

 

[Deadblow]

SRE,

Thank you for your input and I agree 100% with wanting a deeper beam to span the 50'. One of the problems I am having is that I am treating the compression flange of the beam as completely unbraced as I am unsure if the diagonal bracing I am proposing between the two beams can be relied upon as bracing. Therefore, the shorter stockier beams are actually stressed less in bending than a taller beam of similar weight, or even a W24x76, due to lateral torsional buckling. If I can consider the compression flange braced, than it's a whole new ball game.

What are your thoughts on how to brace the frame? I do have a 10'-0" tall x 48'-0" wide walk-in cooler beneath the frame, but I'm definitely open to suggestions.

EIT

 

[JAE]

Agree with SRE above.

Also:
1. Yes you need beams on all four sides.
2. You can do either moment connections or X-braces.
3. Moment connections, with thin little beams and columns like you show, probably aren't enough for lateral stiffness....thinking PDelta effects here.
4. Be sure to include PDelta effects.
5. Diagonal angles as you've shown work...we've done that before - connected to the bottom of the top flanges.
6. HDG coatings OR - use a high build epoxy coating - similar to TNEMEC N69. (our notes read: : ALL STRUCTURAL STEEL SHALL BE CLEANED PER SSPC SP 6 BLAST CLEANING AND SHOP PAINTED WITH ONE COAT (2.5 MILS MIN.) OF ZINC RICH PRIMER (SSPC PAINT 12.01 TNEMEC SERIES 90 97 OR EQUAL). FINISH PAINT SHALL BE TWO COATS (4 5 MILS MIN. PER COAT) OF EPOXY POLYAMIDE PAINT (TNEMEC SERIES N69 OR EQUAL).
7. The size of the units on top of the beams will not only cause lateral shear forces from wind or seismic, but also there will be overturning vertical forces on each beam in each direction....you must include those.
8. Besides the diagonals, using perpendicular cross beams goes a long way to helping to brace these beams.



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[SlideRuleEra]

Deadblow Yes, I would brace the compression flange. The top flange bracing shown in your sketch appears to have been proposed without consideration to how the equipment is mounted on the beams. Integrate the equipment mounts with the bracing plan. This will probably allow much less, and possibly no additional top flange bracing. Also, I would tie the beams together near the columns, as shown on my sketch, to help the two frames work together. With attention to top flange bracing, deep beams can likely be very light.

http://www.SlideRuleEra.net

 

[Deadblow]

JAE & SRE,

Thank you both for your input and advice! JAE, thanks especially for the heads-up about the P-delta effects and the notes on the epoxy coating! I'm curious, what software do you use for structural steel frame analysis and design that considers the P-delta effects? Or are you doing it by hand?

EIT

 

[JAE]

RISA 3D

Check out Eng-Tips Forum's Policies here:
faq731-376