SUPERPAVE Digest 324
Topics covered in this issue include:
1) Quality Assurance and Complete MP1 Testing
by
- To:
- Subject: Quality Assurance and Complete MP1 Testing
- From:
- Date: Wed, 8 Mar 2000 16:47:32 -0600
Wanted to get your thoughts on doing complete MP1 testing on our regular QA
construction project samples.
Currently: We sample at the the plant site on a daily basis. A production
day is considered a lot. Lots are divided into sublots when the grade of
binder is changed or the supplier changes. Out of these daily samples, we
do a complete test (full range of MP1) every 5th sample.
Your thoughts on whether this is too much or not enough complete testing.
Reduced Sampling Schedule: With our participation in a Combined State
Group, consisting of 6 states receiving binder from the same supplier, we
are able to reduce our sampling. Sampling will be 1 sample (1 quart) per
7500 tons of HMA. Quite a reduction in samples.
Your thoughts on how often complete testing should be done. Our thoughts
is to go to every other sample or maybe every third. Is that too much or
not enough?
We will, obviously take more samples and do more testing when non-compliant
testing occurs.
How often are you doing complete testing? (to those that this may apply
to)
SUPERPAVE Digest 325
Topics covered in this issue include:
1) Re: Quality Assurance and Complete MP1 Testing
by "Kenneth Hobson" <>
- To: <
- Subject: Re: Quality Assurance and Complete MP1 Testing
- From: "Kenneth Hobson" <
- Date: Thu, 9 Mar 2000 20:13:01 -0600
Our quality control agreements have worked out fine. Qualifying samples is
our main focus for full testing. We do an orginal DSR on a days run. Even
that is probably too much. It will capture an inconsistent product or wrong
shipment though. A better binder shipped is not always better. Better
(higher grade) is always acceptable providing it is designed that way in the
mix design and at no additional cost. That is why we put a limit on the
maximum for G*/Sin(delta). I'll present these modifications at the
Government Engineers Forum at AAPT in Reno, NV on Sunday.
We do a 3rd level testing (liaison samples) which is less frequent than
daily runs. Depends on tonnage laid but is generally one per weeks run or
so and should be at least one per project and two if at all possible. We
run RTFO DSR's on all liaison samples and also BBR's on the 90+ binders for
the liaison samples.
I'm pretty happy with the way we do it but the daily DSR's are probably a
little redundant. We seldom get failures and when we do we require a new
refinery sample for full testing if a trend is seen. Basically, we do just
about as much testing as we did prior to the PG system. I knew when we
started this that we would be overloaded and we did specify a reduced
frequencey but old ways are hard to break.
My only concern with this method is that we only hold the refinerys
responsible. The shipment could be contaminated in the transport, in the
asphalt mix plant tank, overheated in the mix plant, separate in the mix
plant. This doesn't even address the contribution of an anti-strip additive
or not. Our 2nd and 3rd level testing does address the asphalt mix tank
issue to a degree. So basically, the producer of the mix is held
accountable for the mixture properties. The only long term testing on the
mixture is our moisture sensitivity test OHD L36 or AASHTO T 283 for
Superpave mixtures. Hopefully, we will be able to use the APA (Asphalt
Pavement Analyzer) to some degreee in the future. Most likely, we will use
it on the higher traffic designed roadway mixtures. As some of you may
know, our 1999 construction season saw significant problems with regards to
stripping test results both in the field and the lab. I'll leave that
discussion for Sunday.
See you at AAPT in Reno, NV.
Kenneth Hobson
Bituminous Branch OKDOT
405-522-4918
405-522-0552 fax
SUPERPAVE Digest 326
Topics covered in this issue include:
2) Re: Quality Assurance and Complete MP1 Testing
by "Darren Hazlett" <>
- To: <>, <
- Subject: Re: Quality Assurance and Complete MP1 Testing
- From: "Darren Hazlett" <
- Date: Fri, 10 Mar 2000 15:23:55 -0600
At Texas DOT, we generally operate by "sample and approve at the source". Our current PG spec does allow manufacture of the PG binder on the project (such as injection of SBR latex). When this is done, the spec requires a preconstruction sample at the point of mix design verification and construction samples taken from the project. For project samples, one sample is required per day. The first day's and random of one every 9 days call for complete MP1.
We experimented with a QC/QA binder spec for all binders that called for the same sampling schedule and MP1 testing, but additionally required the contractor to test RTFO/DSR every day and TxDOT verify RTFO/DSR every three days.
When we were developing this QC/QA spec we looked at the reduction in confidence when you got the wrong binder. If you select based on 98% confidence and get one grade lesser performing on the high or low temp designation the confidence level you get is 30% and 80% respectively. This told us that we needed to emphasize the high temp tests. This is also the property most influenced by the addition of polymer modifiers. We felt comfortable in testing for low temp properties at a reduced frequency.
One could even rationalize this drop in confidence level as a basis for penalties. Due to the definitions of the high and low temp designations, the weather database has low standard deviations on high temps and larger standard deviations for low temps. (STDlow is about 3 times the STDhigh). This means that the precipitous drop in high temp confidence when supplied with a lower PG binder is not confined to Texas, but exists throughout the whole SHRP weather database (I checked).
Back to the sped described above, we decided that if we got started "right" we would be pretty sure of maintaining the low temp properties we desired. This also factored into the reduced low temp testing frequency.
Finally, after our QC/QA experimentation, we are still approving binders at the source. Our current spec requiring project sampling and testing for jobsite manufacturing of binder (SBR injection) has had the effect of drying up the market for injected latex. No one wants to do it if pre-blended binders require no jobsite sampling and testing. Too much risk.
I hope this is helpful.
Darren G. Hazlett, P.E.
TxDOT
Construction Division, Materials Section, Asphalt Branch