Corner 90 degree butt glazed glass thickness calculation

[hans_freek]

Hi everyone,

I'm new to this community and would first like to thank you all for the incredible work you're doing in sharing your experience, expertise, and knowledge for the benefit of the engineering community. It's truly appreciated!

I'm currently working on a project involving a uPVC window with a 90-degree corner butt-glazed glass unit, and I’d love your input on the approach I’ve taken.

Window Specs:​

• Corner width: 72" x 72" (forming the 90° corner)
• Common height: 107"
• Wind load: 37.33 psf (1.78 kPa) .....factored wind load
• Glass requirement: Laminated from the inside
• Deflection limit: 40 mm (based on CAN/CGSB 12.20-M89)

My Calculation Approach:​

Using ASTM E1300, I analyzed:

• Interior glass: 4mm annealed + 0.030 PVB + 4mm annealed laminated
• Exterior glass: 6mm tempered
• Airspace: 7/16"
• Support condition: Three-side supported assumption (due to butt-glazed corner)

The combinations were selected based on graphs for 8.0 mm PVB laminated and 6 mm monolithic glass in ASTM E1300.

I do understand that I’m intersecting two different standards here (ASTM E1300 and CAN/CGSB 12.20-M89), but I’ve attempted to balance the stress resistance per ASTM E1300 with the deflection criteria from CGSB, especially considering this is a residential application.

​

My Questions:​

1. Does the glass combination I’ve selected seem adequate in terms of stress and deflection under the stated conditions?
2. Is it acceptable in practice to combine ASTM E1300 and CAN/CGSB 12.20-M89 in this way for a project?
3. Are there any other reference documents, formulas, or calculation methods that can be used (excluding FEA software)?
4. Can this method also be applied to 180° or other angular butt-glazed units, assuming similar or different edge support conditions?

Looking forward to hearing your thoughts, critiques, or recommended improvements. I’m always eager to learn from those with more field experience.


Thanks in advance!

 

[jjl317]

I guess in my mind, the first question is what do expect to happen at the butt-glazed corner? Will each half of this joint move under wind load (opening and closing the corner joint between the lites), or will you try and somehow connect the corner together? For insulated glass units, if I remember correctly (10 years plus), we would generally use a small tube in the corner (which architects loved - haha), attached to the back of each lite with structural silicone, to tie them together. Without connection, my concern with the approach you described is keeping the glass stiff enough to prevent it hitting the adjacent lite - the 40mm (over 1.5") deflection limit you describe is pretty high (though unclear if for a three sided support if it would be at the unsupported edge).

 

[RenHen]

2. I don’t work with the Canadian codes, but I would think they probably have a referenced standard for glass design. As long as you're meeting the requirements of your applicable code, I'd think pulling additional information or running additional checks per some other code you want to reference is fine though.

3. Not really that I'm aware of. The NGA’s GANA Glazing Manual is a great resource in general but I don't know that it would have specific guidance on this case. I will note that a 90 degree glazed butt joint may meet the ASTM’s criteria for a supported edge; check out section 5.2.4. For example, if you have a glazed joint that meets the requirements to be considered structural & WL on the glass puts the joint in compression and that puts the adjacent/perpendicular glass into compression & passes that load into a stiff support on the vertical edge of that glass, you may end up with a stiffer support than a vertical framing member spanning over the height of the glass. If it's less clear cut than that, assuming that the glazed butt joint does nothing is conservative.

4. For 180 degree butt joints I typically assume 2 side support for distributed load cases unless there’s framing on one side. I would be more careful with the detailing of glazed joints around atypical corner geometry but as far as determining how many sides are supported, it’s still conservative to assume the joint is not contributing to supporting the glass.

 

[hans_freek]

I guess in my mind, the first question is what do expect to happen at the butt-glazed corner? Will each half of this joint move under wind load (opening and closing the corner joint between the lites), or will you try and somehow connect the corner together? For insulated glass units, if I remember correctly (10 years plus), we would generally use a small tube in the corner (which architects loved - haha), attached to the back of each lite with structural silicone, to tie them together. Without connection, my concern with the approach you described is keeping the glass stiff enough to prevent it hitting the adjacent lite - the 40mm (over 1.5") deflection limit you describe is pretty high (though unclear if for a three sided support if it would be at the unsupported edge).

Thank you for your response.

Typically, in corner butt-glazed glass assemblies, a backer rod is inserted before the joint is sealed with silicone. Given the geometry of the joint—where the external lite extends slightly beyond the internal lite for both panes—this setup helps provide lateral resistance to wind loads. I assumed this configuration could be considered somewhat four-side supported, though I acknowledge this interpretation may not be entirely accurate.

Regarding the 40 mm deflection limit, I referred to the Canadian standard CAN/CGSB 12.20-M89, which specifies this allowable deflection for guard and railing glass. These systems are generally three-side supported, and I extended this interpretation to our current condition based on that reference.

Please let me know if you have any thoughts or corrections on this approach.

 

[jjl317]

I am unfamiliar of the Canadian standard, but I do know in the US the allowable deflection can be much higher than people often think - but these limits are often defined for the center of a 4 sided lite. And guardrails are a completely different condition. I would be hesitant to blindly follow these allowables, without fully vetting what will happen at the corners. Especially if the outside lite extends out past the inside lite - could be prone to breakage

and I agree with RenHen about the 180 degree butt joints - supported top and bottom - though I would add that I might consider a lateral load from occupants at handrail height on one lite, to make sure the deflections are compatible, and the joint wouldn't want to open (or sealant fail)

 

[RenHen]

I don't think I've seen any specific center of glass deflection criteria in the US, except for glass used as guardrails (which is very permissive & I assume is that way because they are often 1-side supported cantilevers). If you're familiar with a reference for that I'd love to check it out jj.

& agree on considering a concentrated load on one lite to check out the differential deflection behavior for 180 degree butt glazed joints. I believe deflection criteria for that case is explicitly required in my applicable codes for non-glazed/open joints.

 

[jjl317]

From memory, I thought the industry standard for center of glass deflection was L/60 (L being glass span), and maybe with a maximum (3/4", or 1"?), though I am not finding it published - maybe it was/is addressed in the specifications on a project by project basis. I do remember seeing the glass physically deflect in performance mockups, and being surprised how noticeable it was.

 

[hans_freek]

Thank you, "RenHen" and "jj," for the information. I estimate that adding an aluminum cover cap on the joint would not only protect the exposed edge but also enhance resistance to wind load.