Youngs Modulus for High Early Strength Concrete 1

[asixth]

Hi guys,

I have been designing some precast/prestressed concrete bridge girders that I have written about on this site quite frequently in the past I have another question that I would like to ask.

I need a formula for the calculation of the Young's modulus for high early strength concrete?

I am an engineer in Australia where the Young's modulus is projected in the concrete code by the following equation:

Ec=?^1.5*0.043?fcm [Metric]

Where ? is the unit weight of the concrete and fcm is the mean compressive strength of the concrete. These terms are metric, so a concrete mix of ?=2,500kg/m^3 (155lb/ft^3) and fcm=40MPa (5.8ksi) results in a Young's Modulus of Ec=34,000MPa (4,930ksi).

Speaking with my local department of transportation, the Young's modulus that they give for this design is 28,600MPa (4,150ksi) but they don't give me a reasoning or reference material as to why the Young's modulus for high early strength concrete is lower than the code projection.

Some material that I have found online for this is the design manual from the Minnesota Department of Transportion which gives the equation for High Strength Concrete as:

Ec=1265*?(f'c)+1000 [Imperial]
Ec=3320*?(f'c)+6900 [Metric]

This formula gives me a Young's modulus of 27,900MPa (4,050ksi) which is similar to the results of my local DOT.

This formula was confirmed in a textbook on High Strength Concrete by Caldarone.

What formula's are other engineers around the world using to calculate the Young's Modulus of High Early Strength concrete?

This is important for both my calculation of camber and elastic shortening losses which I do not want to over/under estimate for these elements because of the spectacular failure possibility?

Any help at all will be appreciated.

 

[ishvaaag]

Aswad in ACI SP-133 for early gain in strength concrete steam cured, p. 309

for a concrete 60 MPa 28 days, 40 MPa at 1 day age, gives at 1 day age the modulus 4618 ksi; however the strength falls outside the recommended ranges.

Branson in ACI would give around 4000 ksi at 1 day.

I have around two dozen evaluations (not all for early high strenght concrete) between at 28 days and at any age. Aggregate, static or dynamic, kind of gain in strength, age, the definition of f'c, fcm, fck, specific weight etc so if you are the most specific I may look for the better or worse you are looking for.

 

[rowingengineer]

asixth,
If you get stuck, Normally your local concrete supplier will be able to give this to you (depending on your mix, because the E will be based on the agg and cement used), as they tend to test any mix that is non-stand and have results for the engineer hidden somewhere. Main problem you will be finding the right guy to talk to, because I’m sure if you say young’s modulus to the first peron that picks up the phone, they will have no an idea of what you’re talking about, I had a similar situation recently, and Wagners were very helpful.


When in doubt, just take the next small step.

 

[asixth]

rowingengineer,

I think they're the same guys I'm dealing with. Did you speak to the guys at Wacol.

 

[rowingengineer]

asixth,
Wacol, dowstress? I take it your designing Super T's??
if so take a look at this for a bit of info that you may wish to know:

http://wyche.com.au/wp-content/uploads/2007/05/Wyche Torsion in AS5100 Austroads Bridge Conf 06.pdf

Returning to topic: unfortunately i know whom you are dealing with and there is a good chance that they will not be able to help. Wagers brains trust is in Toowoomba (the guys you want are in there concrete lab in Toowoomba), unfortunately like any big company getting different divisions to communicate between each other is nearly impossible.


When in doubt, just take the next small step.

 

[asixth]

Nice paper, I did not realize that there are so many debatable clauses in the Australian Concrete Code when it comes to torsion. I have been using the MacGregor textbook written to ACI when designing for torsion and continually switching between metric and imperial units and have noticed that the approach of ACI does vary compared to AS3600. I posted a thread earlier on the Other Topics forum regarding the calculation of torsional stiffness of concrete beams. Ultimate design, bending/shear/torsion is something that I need to get into shortly but I am still in the process of setting up a spreadsheet to do the design at stressing.

I am going to try and get a copy of ACI-363 for projections of Young's Modulus for High Strength Concrete.

 

[rowingengineer]

asixth,
I'm sorry gave you a bum steer on the super t, should have called it a t-roff, The guys at dowstress would have my balls if they know i had said that.

Anyway, while you may end up finding a young modulus i have just re-red your post. and for what you looking for, there is a few things you need to take into account.

1. they steam cure these beams. So they can get 40mpa at 12hrs
2. the beam will go about 75-80 mPa at 28days.
3. Because it is steam cured these if very little drying shrinkage, but there is a large thermal shrinkage.

Taking these all into account the E is a hard answer to give. Wagners again will help, but previous to my past advice, the guy you want is at dowstress, he will give you the full run down as best as a site pereson can. I would also consult the AASHTO code Bridge Design Specifications.


I was involved in your earlier torsion discussion, you will find that these type of t-beams need to be treated completely different to a normal beams for torsion. AS5100 has a bit of a discussion in the code and commentaries, in the commentary see clauses 8.3.7. 8.3.8 and appendix F in the main code. but Wyche has some very god points.




When in doubt, just take the next small step.

 

[rowingengineer]

may want to read this one too

http://wyche.com.au/wp-content/uploads/2006/06/Wenham&Wyche WW Frame ACMSM18.pdf

When in doubt, just take the next small step.

 

[rowingengineer]

Asixth,
Here is the best free PFD I could find on the subject,

http://www.smartec.ch/Bibliography/PDF/P30.pdf

When in doubt, just take the next small step.

 

[asixth]

rowingengineer,

thanks for the help. The second paper that you posted by Wyche was using 32,000MPa for Ec which I believe is too much. I can't get the equations to work out for the Idriss paper that you posted:

Ec=40,000*(fc)^0.5+10^6

This is giving me values that are way over the mark for 5800psi concrete.

 

[rowingengineer]

Asixth,
Sorry my point wasn’t to give you another place to find the E value from I Just wanted to make sure that what you decided to take into account for "elastic shortening losses" isn’t just E, because this isn’t going to help out with the real life situation. Unfortunately lot of very interesting things happening during the t-roff pouring and unfortunately this isn’t the best understood area of engineering science. Wish i could be a bit more helpful with regards to young's modulus but I am fairly old style and like to use 10% reduced off the code value for all my designs, so for u this would be something like 30,000mpa. If I was using small agg, I would reduce this further (20%), But i will point out, that i am not a t-roff designer, my knowledge of the t-roff system comes from my designs of bridges that are simple but different.

My views on the situation is the code states this can be +/-20%, thus you could have range of possible E's from the code between 25500 and 38000 mPa, for your concrete. Thus there are many possible equations that predict the E (some go high, some go low). Normally the concrete used for t-roff’s is denser than normal concrete so the E should be higher than the AS5100 value, but then again if a small agg is used it will be less. I always ask for the mix design and some tests, even if it isn’t HSC, just so i can see how there agg performs.

But in saying that there are more problems in determining the E, as in t-roffs your test will have different cooling patterns between the test cylinder and the bridge (bridge will be 10% stronger than your cylinder as a rule of thumb), this is one of many items that will influence your final E.


I could go on, but at the end of the day there are a few more variables to take into account than the E value for what i believe is the real question, "shortening losses"? I also believe you want to know the E value at release, not the end E value (these are different! generally release E is about 10-20% less than E final)?

I hope i have talked you into the need to also take a look at temperature effects on shortening of the girder and the effects of storage time on these girders. How does this temperature effect get accounted for in the stress of the steel, during the early ages the steel will be able to expand (not good), but in the later stages this will not be able to happen???

Have i been any help, I think i might be mumbling again?


When in doubt, just take the next small step.

 

[asixth]

AS3600/5100 does say the modulus of elasticity varies by +/-20% so immediately I am thinking of using the -20% case. I will need to read the paper more thoroughly because all that I have been using to calculate "elastic shortening losses" is E, where a larger E will result in less losses and greater compressive stresses at the mid-span of the deck unit. So when considering stresses at release, the lesser the E value the more conservative the answer. I also want to use a realistic value for camber calculations, if I over-estimate the E values than the calculated cambers will be less than what they should be.

Thanks for the papers, I will have a read through them and see what I can get from them.

 

[rowingengineer]

Note the AS3600 and As5100 requirements are different, the as5100 has an effective E defined, this Eeff includes a creep adjustment.


When in doubt, just take the next small step.

 

[asixth]

rowingengineer,

Thanks for the paper, you seem to have a vast collection of technical papers that you have accumulated and aid you in your career.

I have had a chance to read a bit more into the paper and the author agrees that the code predicitions for Young's modulus is appropriate for normal grade concrete but should use the ACI committee 363 equation that I listed my original approach for high performance concrete (I call high strength concrete). This equation doesn't give me the exact value that is used by my local DOT but I am satisfied that is giving me a good estimation without any lab testing of samples to prove otherwise.

Currently I have been putting together a spreadsheet that calculates the losses and cambers at release only, so I haven't considered what the long-term affects will have on the design, but that will be programmed soon.

I am taking the PCI method towards calculating losses at intermediate locations along the beam and then using moment-area to calculate deflections. The paper suggests that this approach over predicts the losses, so the girder may actually camber more and higher stresses from pretensioning may result, which is concerning but out of my control seeing that the paper does not suggest alternate methods or correction factors when calculating losses.

It also made an interesting point that the losses are greater when steam-cured at lower temperatures. I would have thought the opposite, more steel relaxation at higher temperatures and therefore higher pre-stress losses.

 

[rowingengineer]

asixth,
I do have a large collection of papers and books that i have collected (something like 8GB), I have a few more on this subject, but they have a large BS to good stuff ratio. I will post these as well, but i would recommend for the NCHRP report that you look at the graphs mostly and read a little of it, I think these will give you some confidence in what you are doing. John rollers report will give show you that you have to make your own judgements on the report, as often the conclusions can be ambiguous at best. The Barr paper will show that no one has really idea what is going on .

The reason for this hoard of papers and books is a large proportion of my jobs (but not for much longer) are R&D. I also have this rule "if it hasn't been published than it must be tested". Thus I spend time researching papers trying to find answers that can be reference in reports thus that testing is not required, Not that anyone ever bothers to read my reports.

If it helps at all I would use the PCI method for design.

As for your spreadsheet hope, you have reviewed this thread770-218057






When in doubt, just take the next small step.

 

[rowingengineer]

barr

When in doubt, just take the next small step.

 

[rowingengineer]

roller

When in doubt, just take the next small step.

 

[rowingengineer]

and this has some good pointers for young players

http://www.sciaust.com.au/pdfs/supertstandards.pdf

When in doubt, just take the next small step.

 

[asixth]

Just looked over the draft Australian concrete code DR05252 and it gives the following estimation for Young’s modulus:

?^1.5*(0.024*sqrt(fcm)+0.12)

This still gives a much higher approximation than ACI-209. Does anyone know if the committee is available for contact?