A325 Bolt Inspection 1

[pattimelt]

Is there a definitive "right" amount of bolts to inspect, be it a number, percentage, or whatever when checking a building? I can't find anything in my paperwork, and I have been doing 10% of each connection. Is this overkill? Should it be 10% of the total number of bolts? We are using DTIs, but I am not a big fan of a squirt washer-sometimes it is a week before I get to inspect, and the silicone has disappeared.

 

[connectegr]

RCSC provides instructions for pre-intallation tests. DTI's are expensive and are not preferred by most erectors. But, if you use DTI's, Squitters are a good choice. They are basically DTI washers with the addition of the silicon gel. If inspected during installation the gel is a visual indication. But if time has past, the DTI can be check with a traditional gauge. If pretension is required all of the DTI's should be inspected Other methods such as tension control bolts will require only the preintallation tests and supervision of the installation to verify that the procedures are followed.

http://www.FerrellEngineering.com

 

[vanmorrison]

In my opinion the determination of the use of a sample plan should to be documented in the inspection plan and criteria and the QA program. The use and approval of a sampling plan should be based on an engineering assessment and judgment that will provide reasonable assurance that the sample is representative of the lot inspected.

I am not familiar with your business of this bolt inspection but do know there are factors to be considered when using a sampling plan. Factors that come to mind may be; risk associated with failure, track records, installers qualification and experience, confidence in the bolt, difficulty of installation, lot and homogeneity, etc.

The company I work for adopts a sampling plan guideline published by Energy Power Research Institute's (EPRI)for procurement inspections.

Here is a copy

http://mydocs.epri.com/docs/public/TR-017218-R1.pdf

It provides some rational on a sample plan that you may be able to draw from.

CYA. If you take it upon yourself and decide to inspect to a sampling plan without a basis or documented approval for its use and the building falls down and kills a bunch of people you will need a good lawyer.

 

[connectegr]

I agree with vanmorrison with regard to using a sampling. 10% sounds like a small number. Unless used in combination with other installation method, DTI washers only verification of pretension is inspection of the final washer condition.

http://www.FerrellEngineering.com

 

[Ron]

I have typically started with 10 percent of the connections, checking all bolts in each of those connections.

If failures occur, I increase the frequency, usually by 10 percent in each case.

 

[vanmorrison]

If a failure is detected on a 10 % sampling plan what is the assumption made on the previous 90% that was not inspected?....

The sampling plan that I am familiar with is that when 1 is rejected this means confidence has been lost on the ones that were not inspected before the rejected one. The entire batch will require 100% inspection.

Sampling is kept at 100% thereafter until confidence is re established as per procedure. Rejections after that will force 100% inspection and sampling will not be permitted.

 

[Duwe6]

I ALWAYS use a Sample Plan. If the bolting requires more than 10-15% inspection, the Erector is 'failing to perform' and needs to be dealt with by the General Contractor. What I tell the Ironworkers is "There is Good News & Bad news. The Good News is that I don't check many bolts. The Bad News is that I don't check many bolts"

Each erection 'Sequence' is a different 'Lot' to be sampled. Pick widely separated joints in this area and check several bolts in each joint. I 'Reject' when I find two joints/connection points with undertorqued bolts. Entire Sequence is rejected. Ironworkers now retorque all connections.

Sample size should similar to that in vanmorrison's linked document. Sample percent is high with a small lot size, and tapers off as the lot size [# of connections in Sequence] increases. Sampling works. I have several exhausted Ironworker boltup crews than can testify to that fact.

 

[bridgebuster]

The NYSDOT procedure is to check 10% of the bolts at each connection, but no less than two. If any bolt fails, then everything has to be checked.