Slab-On-Grade - Parallel-to-Cut Cracks 1

[dbnerds]

I have an incident of a slab-on-grade that has been placed that is exhibiting cracks in a pattern that I have not witnessed before and was hoping I could get some feedback/ideas from all here. Below is the conditions:

-6" slab on grade, 4000 mix, reinforced with WWR 6x6-W2.9xW2.9
-two "plates" were placed: #1 was 65'x520', #2 was 65'x455' (building bays are roughly 65'x65'). The plates are independent of each other, but placed within the same day. Neither are fully restrained edges. #1's long edge is against an existing building. #2's long edge is against a new foundation wall.
-only #1 plate saw the main area of cracking. The area occurred at the very beginning of the day's pour and within the first (2) bays
-the cracks exhibited "pattern" is that the cracks run parallel to only the sawcut lines running the long distance and the sawcuts are adjacent at about 1" to 12" to the sawcuts. All stay very close and parallel to the saw cuts.
-saw cuts were performed with sofcut at spacing roughly 16'3". At the area in question, the cuts were made approximately 12 hours after the concrete was placed. The cracks in question are parallel to these first (12 hr) cuts. Cuts in the perpendicular direction were cut later (13-14 hours) but did not exhibit this "parallel" pattern.
-there are other smaller/random areas, where the same parallel pattern runs the perpendicular direction, but still near/adjacent to the final saw-cut lines.
-crack width varies: hairline to 3/32"
-the cracks were not noticed when saw-cutting began. They were first noticed less than 1 day later.
-the concrete was pumped.
-minimal water was added.
-slumps averaged 6.5" and were consistant.
-breaks came back good - all passing and none WAY over.
-no super admixtures added.
-"story" from the site was when the day's pour began, the beginning (where these cracks are found) of it saw a lot of "commotion" e.g. the crew trying to rush and keep up with the trucks pumping the concrete in. Later, the timing of unloaded concrete with the crew placing it feel into a better timing/rhythm.

The field superintendent has provided us a drawing showing all of the cracks. Its uncanny how close these cracks marry up and hug next to the final saw cuts. Some designers may say this proves that the saw-cut pattern we prescribed is correct since its supposed to "mimic" where the slab would naturally crack. Thats good in theory, but isn't a hard science. These cracks are almost dead-on (relatively speaking). I can't figure the reason why, but my gut tells me the saw-cutting caused it.

There may be more info I can provide to help enlighten this case, just ask.

I appreciate any feedback.

Thanks

 

[dbnerds]

I should add, saw cut depth was approx 2" deep.

 

[OHIOMatt]

A few issues / questions

1. 6 1/2" Slump with no admixture, what was the water cement ratio?
2. What were the environmental conditions? Temperature, humidity, wind speed, sun exposure, etc.
3. What was the condition of the base? Did concrete trucks or screeds drive on and rut the base creating soil restraint?
4. When did curing start? If not until after the saw cuts were made, it may have been too late.
5. Was the pour complete when saw cutting began? If so, you had a slab with an aspect ratio of 8:1.
6. Did all of the concrete come from the same batch plant? Given a 24 hour day and a 10 yard truck capacity, your would have needed to have a truck batched out every 12 minutes.

Sounds to me that the cracks started before saw cutting began. This project had over 63,000 square feet of slab using 1,200 cubic yards of concrete. Was the contractor truly capable of placing, finishing, cutting and curing so much?

 

[FixedEarth]

Did they place a 1/2 inch felt between the old concrete and the new pour?

http://www.soilstructure.com/

 

[jgailla]

Sounds like the joints were cut too late.

 

[dbnerds]

answers to some questions posed:
-w/c = 0.45
-environment was idea: low breeze, 60's
-base was ideal: subgrade had been conditioned with a lime mix, and the concrete was pumped in (hence trucks never touched the subbase)
-when did curing start? meaning?
-was the pour complete? meaning had all the plates been placed before cutting began? yes, but why does that matter?
-ratio was not 8:1 - that was the "plate'. the sawcut pattern was ~16'x16'
-all concrete did come from one plant. However, the two plates weren't placed the same day.
-expansion material was placed against the foundation.

 

[OHIOMatt]

I will use my original numbering to keep myself straight

1. Not sure how you would get a 6 1/2" slump with a .45 W/C ratio and no admixtures. Those two conditions do not align.
2. Any chance that the sun could hit one portion of the slab and be shaded in other areas. We have seen this before on projects with where walls have been erected and the roof is off, or there is an adjacent structure.
3. That sounds good.
4. I assume that some curing technique was used / specified to prevent the concrete from drying. Was it a chemical compound, water, sheets?
5. It matters because concrete starts shrinking and therefore cracking as soon as it starts to harden. So, for some period of time, you did have a slab with an 8:1 aspect ratio.

For a slab like this, of which we have done several hundred, I would expect to see the saws on the first portion of the slab before the pour is finished. As soon as the saw is finished, a crew is dispatched to place moist curing sheets down to prevent excessive drying and shrinking.

It really sounds as if the saw cutting took place too late. The cracks may not have been visible, but they had likely started. I would guess that the joints that were sawn, did not activate as a weak plane had been established at the crack.

 

[dik]

6-1/2" slump without and superplasticiser is way too high... should be looking for 4" max. Sawcutting should have started approx 7 or 8 hours after finishing at most, earlier if a sofcut type saw used. Why no admixtures? AEA, WR, etc.

For a 6" slab, you should be looking at about 15'x15' sawcut panels.

Dik

 

[rittz]

we used saw-cut pattern 20x20 25 y years ago. We have joined the crowd that uses 10 x

​

10 squares. AND Cut as soon as possible (soft cut

 

[Ron]

Your cuts were made too late. The cracks had already occurred, though you likely couldn't see them at the time.

Sof-Cut is a process that should be used almost immediately after finishing...not 12-14 hours later. While thickness variations can cause "random" cracking after sawcutting, the fact that the cracks run parallel to the sawcuts is very telling. This is indicative of late sawing or sawing at insufficient depth.

Temperature, humidity, wind, W/C ratio and placement techniques all will exacerbate the conditions, but bottom line....too late.

 

[dik]

Yup 10x10 is better... should have noted 15x15 max...

Dik

 

[dik]

Just noticed a couple of things...

Why would you add minimal water to a 6-1/2" slump? and who added the water? the concrete supplier or the contractor?

I didn't realise you had a sofcut saw...12 hours too late... after 7 or 8 hours, the microcracking has already determined the location of the cracks... as you've found out...

Dik

 

[dbnerds]

appreciate the feedback everyone.

Seems the favored verdict is that the cutting happened too late. That would makes sense. However, there is an curveball to that theory. The majority of these cracks occured parallel to the long axis of the plate. These cracks are adjacent to the FIRST sawcutting of the plate. Sawcutting in the perpendicular direction happened later, but no parallel cracks are exhibited in this direction..???

Some additional info from previous comments:
- sun wasn't a problem since the pour started at 3:00am
- more feedback from the field: "water was added, but never more than 10 gallons".
- got the batch tickets: 27 oz of MRWR was in the mix. It also denote the mix should only give a 3" slump.
- our (conservative) rule of thumb is 2 to 2.5 times slab thickness = sawcut patterrn. The 16'8" was stretching this rule, but the column line locations lead the reasoning.

Thanks for everyone's time for responding.

 

[dik]

Longitudinal cracking is a bit of a puzzle, in particular if it was done first. I would have anticipated transverse cracking. Is there any transverse cracking?

Sawcut depth is good. Granular base compacted smoothe? Was the concrete placed in the longitudinal direction? Cracking indicates that sawcutting was too late...

Dik