Flat slab bridge thickness 1

[tmalik3156]

Good day.
I was told to look into a conceptual design of a simply supported flat slab bridge.
That is, a bridge consisted of normally reinforced slab only, and no girders.
It can not be a portal type frame, it must be simply supported. It can not be prestressed / post tensioned.
The span is 20.3 m (66.6 feet).
How do I estimate a thickness of this slab? Is span divided by 20 a reasonable assumption? But that will be a huge thickness. On the other hand, a smaller thickness will give unacceptable deflection at the mid-span.
I have never worked on a girder-less bridge before. So any suggestion will be helpful.
Thank you

 

[rapt]

Type of loading might affect it.

And type of design, RC , PT?

If you are a bridge designer you should be able to some quick numbers in about 5 minutes!

L/20 might be too thin for many cases!

 

[BridgeSmith]

There really is no rule of thumb for solid slab bridges of that length. Generally, they're considered uneconomical, both in material costs and labor costs. A voided slab or box girder bridge would be much lighter, much less expensive to build, and probably not any deeper than the solid slab. A CIP concrete T girder bridge would only be slightly deeper.

Just curious, why no girders, prestressing or postensioning?

Rod Smith, P.E., The artist formerly known as HotRod10

 

[tmalik3156]

@ rapt: Thank you for your reply.

@ BridgeSmith:
Why no girders - local NU girder manufacturer seems to supply span larger than 25 m. This bridge is only 20.3 m

Why no post tensioned voided slab - local jurisdiction suggest solid section for small spans.

 

[hokie66]

Girders don't need to be precast. Form them, conventionally reinforce, and cast in place. A uniform thickness slab spanning 20 metres is wasteful.

 

[tmalik3156]

@ hokie66: Yes you are correct. For 20 m span, ribbed / haunced / T-beam slabs are more appropriate than a uniform thickness slab.

 

[rapt]

Or put void formers in the slab if you do want to simplify formwork and not use to cast in place beams to reduce self weight.

 

[hokie66]

Keeping void formers in position while placing the concrete is not easy. A way to reduce mass, but not necessarily better labor wise than forming beams. And if you use polystyrene voids, I think it is as expensive as concrete, volume wise.

 

[BridgeSmith]

For bridges in that span range, we've used prestressed decked bulb tees, and steel girders with CIP concrete deck. We've considered prestressed double tees, and if deflection wasn't critical, we'd even consider a prestressed tri-deck superstructure. We wouldn't even consider a slab, or any CIP superstructure, for that span - too much shoring and formwork.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[tmalik3156]

Indeed, normally reinforced slab wouldn't be a good option for that span range.
It's a road over train track grade separation bridge, so there is train clearance height issue. That's one reason why girders and bulbs were initially not considered.
It's only at conceptual level at this time, and a posttensioned slab might work. Girder options give about 1200 mm of superstructure depth, while normally reinforced and posttensioned slabs give about 1100 mm and 750 mm respectively.

 

[BridgeSmith]

We consider a girder depth to span ratio of .04 to be optimal for composite steel girders, and we routinely go as low as .032 where clearance is limited. With a 200mm CIP concrete deck, and a 900mm girder, for a total superstructure depth of 1100mm, your depth to span ratio would be .044.

I did a 42" precast prestressed decked bulb tee for a simple span of 65'. We had to vary the asphalt overlay thickness from 2" (at centerline) up to 4" (at the abutments) to compensate for the camber, but it worked. If you can accommodate a little bit of a crest vertical curve, you could probably eliminate the overlay.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[la belle vie]

Another issue with an non-prestressed concrete slab is that it will creep something like 2x the elastic deflection.

 

[hokie66]

Creep can be more a problem with prestressed than non-prestressed.

 

[la belle vie]

Yes- true, creep happens with prestressed too, but usually hogging up. With non-prestressed the bridge will keep sagging until the total final deflection is about 3x the elastic dead load deflection. This would need to be considered in low headroom situations and could be unsightly, cause ponding, and other problems.

 

[BridgeSmith]

Yes, concrete creeps, but it's fairly predictable. So prestressed girders can be balanced so that the girder starts out with a slight crest and the creep results in the girder being the right shape after a couple of years, and very little additional deflection in the decades that follow.

Postensioned slabs and girders can be adjusted for creep throughout the service life, in order to maintain the desired shape.

Anyway, the OP's query was regarding the use of a CIP slab bridge. I think there have been plenty of reasons given why it would not be an efficient or economical choice for the span proposed, especially given the depth available for the superstructure.

 

[hokie66]

Bridge,

You are assuming non-bonded posttensioning? Bonded posttensioning can't be adjusted, but you wouldn't want to use non-bonded in a bridge. Too much risk of corrosion.

The main problem with creep in bridge decks has been rideability due to upward creep with time. Ever drive across the Chesapeake Bay Bridge, US 13?

 

[BridgeSmith]

Hokie, I get the impression we may need to define some terms.

Post-tensioning in the vernacular we use, involves pre-casting concrete with ducts for high-strength strands or threaded rods, to be installed and tensioned after the structure is in place. Grouting of the ducts is common, but takes place after the strands or rods are tensioned. As far as I know, this precludes the strands from being bonded. After further reading, it appears I was wrong; the grouted post-tensioned system can be considered bonded.

When the strands, either bonded or unbonded, are tensioned, and then the concrete is cast around them, we call that prestressed concrete.

With that out of the way, yes, long term camber due to concrete creep can be an issue with prestressed girders. I'm not sure about grouted post-tensioned, since I don't know how relaxing the strands after grouting would affect the integrity of the grout. Seems that can and does work, but you're likely correct that this type of system can't be adjusted for the effects of creep. As I said, ungrouted post-tensioned girders could be adjusted pretty much at anytime, and yes, they can and are used in bridge superstructures. With proper sheathing of the strands or rods, and detailing to allow water to escape from the ducts, ungrouted post-tensioned girders can last the service life of the bridge. They can also be replaced if necessary.

The amount of camber can be reduced significantly by increasing the time to release of the prestreessed strands, or increasing the age of the concrete at the time of post-tensioning.

 

[BridgeSmith]

Btw, I haven't been anywhere in the US east of western Kentucky, except for one conference in Miami, and a few other stop-overs in the Miami airport.

 

[hokie66]

The bridge I referenced has pretensioned deck panels which have caused a problem over the years due to upwards creep.

 

[BridgeSmith]

Yeah, pretensioned (AKA prestressed) deck panels would seem to be a bad idea. especially if there's not going to be a variable-thickness overlay that can be easily replaced, such as asphalt.

At least with a simple span prestressed girder bridge, it's one curve over the whole length of the bridge, which can sometimes be made to fit the vertical profile of the roadway, but even if it can't, is less obtrusive than a series of little bumps.