Super Flat Floors & Construction Joints 1

[KootK]

I've got a 250 thick SOG on good soil. It's going to support some serious rack loads and will be travelled by a fancy cherry picker that runs on guided wires saw cut into the slab. Super flat floor requirements apply.

Our original design was rebar and frequent construction joints.

We met with the GC and the specialist slab placing/finishing contractor. The told us that the way to go was to omit the rebar and use BASF steel pins. More importantly, they said that we should pour the slab in a series of strips the full length of the building (200') with no construction joints perpendicular to the strips. This would involve the use of a low shrinkage concrete mix etc.

The specialist slab contractor advised us that control joints fare very poorly when steel pin reinforcement is used. Curling of the concrete at the saw cut joint. They also pointed out that our flatness requirements are not aesthetic, they are for the benefit of these wire guided lifts. Therefore, a random but small crack in the slab may well be less detrimental than a curled control joint.

We went along with all this more or less. We still haven't decided on the construction joints however. On Friday, we spoke with the company that will be providing the testing machine (one of only four in North America apparently) for assessing the flatness of our floor slab in real time. They seem to be quite expert at this super flat slab business. They recommend that substantial rebar be used as well as frequent, well dowelled construction joints. They too seem to be most worried about curling.

So... does anybody have any experience with super flat slabs that they would be willing to share? Control joints? No control joints? Rebar? No rebar? It's getting to be a bit nebulous for my liking.

 

[rowingengineer]

I normally take the easy way out when a super flat slab is required and use a PT slab. However this does create one large joint instead of lots of smaller joints, this joints needs to be appropriately detailed to ensure the flat slab requirement.

Curling is a huge issue with SOG's and the type of vehicle I believe you have. If I was to use a SOG I would go without joints, using a system called joint free slabs, basically they create lots of small cracks in your slab at small spacing’s. I use this system once when no joints were allowed and PT was not allowed.

http://www.jointfreeslabs.com/.

As for the first issue, I would put REO in, a small crack is ok, a large crack will get you in huge amounts of trouble.


Arguing with an engineer is like wrestling with a pig in mud. After a while you realize that they like it

 

[JLNJ]

You might want to take a look at type K cement. No joints required.

 

[rowingengineer]

while type K cement would reduce the joints required, I don't think no joints would be a good idea. With Type K I think the max you could push the joints crts out to maybe 50-100'? all depends on the thermal requirements.

Arguing with an engineer is like wrestling with a pig in mud. After a while you realize that they like it

 

[Ron]

Stick with the reinforcing, forget the pins, use a mixture of standard portland and type K, and make sure the slab gets moist cured for a long as possible, preferably 28 days. Also make sure that you do not have a vapor barrier directl under the slab.

These actions will reduce the curling potential.

Saw cut joints and backfill with semi-rigid epoxy. You can increase joint spacing over standard.

 

[hokie66]

I am not an expert in super flat floors by any means, but I would not discount the advice of the GC and specialist slab contractor. I would want to see an installation of theirs which has been in service for a year or so.

 

[H57]

On a job that we recently designed that had very stringent floor flatness requirements it was decided to heavily reinforce the floor and to place in strips the length of the rack aisles. Control joints perpendicular to the aisles were omitted. Reinforcing and slab thickness was designed based on drag of the slab during shrinkage.

Design to a large degree was based on recommendations by the concrete contractor and floor "specialist".

I wish that I could tell you how this project specifically turned out, but the owner backed out and construction did not happen.

 

[amorrison]

I would visit sites that have had flat floors in for 5-10 years and talk to site personnel about their "experiences".

 

[WillisV]

I designed a 12,000SF superflat SOG with no joints using 4@12 E.W. with Type K and extended moist cure - worked just fine.

 

[KootK]

Thanks for the info guys.

For what it's worth, I toured another facility in town with a similar slab and a wire guided lift system. The building is about six years old. The stats are as follows:

Thickness = 6"
Reinforcing = 30 kg/m^3 steel pins only (no rebar)
Control joints = none down the lift isle
Racking = 40' high furniture storage (moderate load)
Mix = low shrinkage mix with relative little cement or fly ash
Placing = non-pumped concrete

The slab looks alright for the most part. Most isles only have one or two spots that required grinding to meet FL65. Cracks tended to be small and spaced at about 15' o/c. As was purported, no curling was evident at these cracks.

There was, however, one large crack that formed with a width of about 2mm. This crack did in fact exhibit some curling.

Does anyone have an explanation for why curling seems to occur at cracked control joints but NOT at randomly formed cracks? That's still something of a mystery to me.

 

[rowingengineer]

KK,
can I query what you mean by pins, I am getting confused by the kg/m3 because I thought that pins were dowels.

I like Hokies idea of seeing some previous jobs, but I would want to know more than just the stats as you have reported.

I would want to know:
Was the slab poured with or without the roof?

what was the bedding material? (as Ron pointed out the vapor barrier should be under 4" of gravel or similar).
What was the machine used for finishing? (Lazer level ect)
did they lazer level the bedding?
what do they do different from then to now.

as for the crack, I would assume the crack occurred due to thermal movement, thus the curling due to differential shrinkage between the top and bottom of the slab was relived due to the self weight of the slab around. I have linked two PDF's that are worth the read.

http://www.concrete.net.au/publications/pdf/Curling.pdf

http://www.nrmca.org/aboutconcrete/cips/19p.pdf

Arguing with an engineer is like wrestling with a pig in mud. After a while you realize that they like it

 

[rowingengineer]

One more quick note, When I say gravel I mean gravel, not sand. The sand base while it will help with curling it will reduce you CBR significantly (some times from 15 to 3).

Because while we are discussing the slab, The base is also important. but I'm sure your well aware of this fact.

Arguing with an engineer is like wrestling with a pig in mud. After a while you realize that they like it

 

[KootK]

Rowing:

Still pins in my market are steel fibers of sorts. They resemble flattened out paper clips. They're about 50 mm long and have simple deformations at the ends to facilitate anchorage.

Of the extra data that you mentioned regarding the existing slab, all that I know so far is:

1) The finishing machine was lazer level.
2) The big difference between the existing and proposed slabs is that the proposed will shrink more. The original slab concrete was not pumped. The proposed slab concrete will be pumped. As such, it requires a higher cement content (at least according to our supplier).

What is "CBR"? A compaction measure? Thanks for the PDF's. I'll check 'em out tonight. I don't quite get your explanation of the larger crack. Maybe it'll come to me after I read up on it.

 

[rowingengineer]

KK,
Even i don't understand what I have written half the time .

CBR is the California bearing ratio,

http://en.wikipedia.org/wiki/California_bearing_ratio

I don't know what you use in the UK (I think your in the uk but most probably your not) to spec the subgrade. Here in Aust we use CBR's and level of compaction.

I don't know much about steel pins, Ron from memory has a large knowledge of this type of reinforcing. Maybe he can shed some light on using these pins in a pump situation, from my knowledge this can be problematic.

Arguing with an engineer is like wrestling with a pig in mud. After a while you realize that they like it

 

[KootK]

Ahhh... California. I'm in Canada. What the heck are you doing using California's bearing ratio in Australia?

When you look at the pins, it is quite surprising that they can be pumped at all. They're going to be mixed in at the plant no less.

 

[rowingengineer]

I don't know why we use CBR, maybe Hokie can shed some light on the mater he has been around longer than me.

Arguing with an engineer is like wrestling with a pig in mud. After a while you realize that they like it

 

[hokie66]

CBR has been around a lot longer than I have, and is used extensively around the world, not just in California. It is an indicative test of the the strength and deflection characteristics of a given soil.

The use of "pins" in this threads is new to me too. I call them steel fibres. Steel fibre concrete can be pumped, although it can cause problems for the uninitiated.