Portal Frame Vertical Deflection

[SteynvW]

Hi All

I am aware this has come up in previous discussions, my
intuition is just fighting with all the literature i am finding.

I Have a portal frame with 35m clear span and 6,5m eaves height.
I have calculated the deflection as 134mm, this is including nominal
rotation stiffness at the base (20%). This comes down to span/260.

My gut feels like 135mm is just to much, but on the other hand 135mm
over 35m is not that much.

Curious to know what other engineers use as their limit.

 

[Tomfh]

There is no hard limit, but >100mm is pretty large, and L/260 is not insignificant for a large span.

It all depends on the circumstances L/180 wind deflection is common in Australia for warehouses etc, but there’s so much subjective judgement involved, and so much depends on the circumstances. Can people see the deflection? Are there fixed references? Will it disrupt ceilings? Services? Are there stiff lines (Eg end walls) that will highlight the deflection? Etc.

Dead load deflections are much more problematic too, because you will get it, as opposed to a hypothetical wind deflection which might happen once in the buildings life. I’ve gotten into trouble with 35mm dead load deflection on a 14m span. I figured L/400 would be fine. Nope. The client squealing that the roof is sagging.

I’ve seen people get in trouble with 25mm on a ~6m span. Other times a big deflection doesn’t matter.
It all depends.


What deflection case are you talking about?

 

[HTURKAK]

Is this the total deflection ? Typical beam vertical deflection limits are for LL , L/360 and for total service load L/ 250..
If supporting brittle elements L/500.

If the portal frame is for industrial bldg without suspended ceiling etc, and 135 mm deflection does not hinder roof slope for gravity flow ,IMO , 135 mm deflection is acceptable for portal frame with 35m clear span..

 

[1503-44]

Will ponding be an issue?

 

[SteynvW]

Hi All

It is a industrial building without ceilings with a 15 degree slope.
So no brittle elements and ponding will not be an issue and there is no
reference to which the deflection can be seen against.

This deflection is for dead load and live load.

 

[Tomfh]

What about dead alone? Have you checked deflection in the perpendicular direction along the ridge?

 

[SteynvW]

With Deadload alone it just more 60mm.

Tomfh
Do you mean along the length of the building?
The building is 75m long with 2 braced bays,do mean the deflection of
the horizontal bracing truss?

 

[Tomfh]

Yes I mean the deflection along the building. You need to check the relative deflection of adjacent bays. 70mm might be fine over 35m, but it will look terrible if the next bay along deflects 0mm (Eg an end wall).

 

[rowingengineer]

I would consider the following
1.a preset for the dead deflection, don't do 100 precent of calc def, use a reasonable percentage.
2. For the live load assuming you have taken into account the max area of 200m2, If so I would consider the likely service use and redo the Cal's with likely service use.
3. Redo the semi rigid cal using a higher order method for you base connection.

 

[JoshPlumSE]

The deflection check is all about serviceability, right? The questions I ask when considering deflection:
1) Is there any deflection sensitive equipment?
2) Are there architectural materials connected to the frame that are deflection sensitive. Drywall, cladding, et cetera.
3) Any strength or stability issues (i.e. ponding) associated with deflection?
4) Are people going to be walking on this area? If so, then LL deflection (and possibly vibration) criteria (like L/360) should probably be enforced.

If the answer to all of these questions is no, then I don't see an inherent problem with the large deflection (and it is certainly large). I don't like it. But, it doesn't sound like a critical issue to me.

Personally, I'd start to look at the relative cost of reducing it. If it's reasonably inexpensive to increase the Moment of Inertia by 30 or 40% then I'd do it. If I got push back from the client, then I would tell him/ her that I could save some $$ by decreasing the member size. However, I'd want to document in writing the idea that they're okay with a total deflection on the order of 135mm.

 

[SteynvW]

Thank you all for your input. I have refined my loads and have gotten the deflection down to span/310.

Josh
That is a no for all those questions. I will have a conversation with the client.