Mill Scale

Date 08-07-2007 20:29

According to AWS D1.5, Section 3.2.1 (under the heading of Preparation of Base Metal):

...Mill scale that can withstand vigorous wire brushing, a thin rust-inhibitive coating, or antispatter compound may remain except that all mill scale shall be removed from the surfaces on which web-to-flange welds are to be made...Mill scale shall be prohibited from remaining in the joint boundary of groove welds subject to calculated tensile stress.

Charles

ps - it says something similiar in D1.1, section 5.15.

By Kix

Date 08-07-2007 20:31

What type of flux core wire are you using? If it were my call i would say to remove it. Your weld quality consistency goes way up when you remove the mill scale. A good welder that keeps a proper gun angle and doesn't run to big of a pass at one time will produce acceptable results on millscale. When you start getting into tight spots out in the field and gun angle consistency and visibility goes down you may not get an acceptable weld. Key word may not, but with the mill scale off your chances of getting a good fillet in with good fusion across your fillet legs lengths goes way up. I just did some macro's on welding with the scale on and with it removed. With the scale left on there was no fusion about an 1/8" in from the toe's of the fillet. The weld starts were really bad and basically had no fusion. With the scale removed results were spot on.
So again i say if your guys may have a chance of getting into an uncomfortable position it would be good practice to remove the mill scale for better quality consistency. Mill scale plays havoc on bead appearance in out of position welds also. Not sure about what part millscale plays in the mechanical properties of a weld, but thats why i asked what wire are you using because some fluxes have all kinds of goodies in it to help get that stuff out of the weld.

By thcqci

Date 08-07-2007 21:25

I agree with Charles above. Further, we remove mill scale whenever we CJP weld. I see the point that Kix is making if he is getting substandard fillets. In a bridge shop I can see more need for that kind of care. But on the other hand, working in a structural shop, I cannot picture removing mill scale everywhere I want to put a clip, shear tab, etc. It just would not be practical or economical and is not required by code.

By aevald

Date 08-07-2007 21:19

Hello hggbecker, if your VP is hard and fast on his non-prep stance you might at least see if you can use a nice aggressive needle scaler to knock the majority of the loose scale and even some of the not-so-loose stuff off prior to making the welds. Have the fabricators tack in the parts and then give all of the surfaces that are going to be welded a good going over with the scaler before welding.
As Kix and chall have stated, scale and rust can have a detrimental effect on the finished weld performance and as Kix added and stated his testing showed the issues that can come up if scale is ignored. I have noticed many times that as the plate thicknesses increase so too do the mill scale deposits and they definitely affect the quality of the weld deposits. You mentioned that you are looking to weld 1/2" fillets all-around on these stiffners, that either means multi-pass welds with smaller diameter wires or large diameter wire with very large single fillets, the latter of which could compound the issues of lack of fusion on the lower toe or undercut at the tops of the fillets, depending on the skill levels of your welders. Most FCAW wires don't deal very well with thick scale, rust, or other surface contaminants, they tend not to wet out well all the way to the ends of the toes of the weld and thus exhibit a condition similar to cold lap, although not caused by the same thing and certainly affected by gravity and weld position(by this I am referring to the top toe of the weld versus the bottom toe of the weld in the case of a horizontal fillet, it is common to have undercut on the top toe even though you may have a form of cold-lap on the bottom toe, especially if you are trying to travel slowly to increase the size of the fillet). I hope some of this has made sense even though I may not have done as good a job of explaining as I could have. Keep us posted on your situation. Regards, aevald

By Kix

Date 08-08-2007 12:40

Well, if your going by D1.1 be careful because you can only do 1/2" fillets in one pass in the 1f and 3f. So i think your going to fall in the multipass category. Again better consistency with multipass also.

By swnorris (****)

Date 08-08-2007 13:28

Welding over thick mill scale can cause porosity, however, under certain conditions, the gases formed by welding have time to escape before the weld freezes. D1.1 requires the removal of all mill scale on cyclically loaded members on which flange to web welds are made.

By hggbecker Date 08-08-2007 14:43

I have found in ASME Section VIII Div 1 UW-32(a) that states the following: "surfaces to be welded shall be clean and free of scale, rust, oil, grease, slag, detrimental oxides, and other deleterious foreign material." I appreciate your help, but he has the attitude of "it's always been done that way, so why change" I do agree that to make a sound attachment weld, the highest care should be taken, and since this is being built to the code, the required prep will be done regardless of his mentality. Thanks again for your help.

Chuck

By swnorris (****)

Date 08-08-2007 16:19

Chuck,

"it's always been done that way, so why change" is the kiss of death. We've all heard it. Henry Ford once said "I am looking for a lot of men who have an infinite capacity to not know what can't be done".

By Greg G. (**)

Date 08-09-2007 13:45

I agree with you on the metal prep. These people that call them selfs upper management need to stay upstairs and keep looking at magazines. Only thing is they have to much pull. That's the problem with getting people into this industry just because they have a 4 year degree, the only thing they are concerned with is production, screw doing the job right. If you show some of them in the book where it says that this needs to be done, then look out, because they are going to be after you. Be careful in your approach to him as he can make your life hell.

By Fredspoppy (**)

Date 08-09-2007 12:20

Removal of mill scale is very important in virtually all welding applications. I have a great demo on the effect that mill scale has on a weld. Take a piece of hot rolled steel plate or flat bar, ~8" long. Thoroughly grind 1/2 of the surface to remove all mill scale. Using a 1/8" E7018, at about 110 amps, make a simple weld bead, without weaving, starting on the unground end and finishing at the other end (ground end). After removing the slag from the weld, you will see a very graphic example of the effect of mill scale. The weld bead, on the unground section will be "humped up and ropey looking" and the minute you transistion to the ground surface, the weld will wet out and have a much better appearance. Give it a try and see for yourself.

Can you get away with welding over heavy mill scale.....probably in most cases. Why would you take a chance? For welding to "advance into the 21st Century" we need to get out of the "well, we have always done it that way" mindset and do it the right way.

By Greg G. (**)

Date 08-09-2007 14:22

My old company thinks that way and buddy let me tell you they are stuck in that mind set for good. I had to go through all kinds of hell when a salesman saw a person properly preparing the metal for welding. " That takes to much time, I did not allow any time for that. How am I going to explain that to the customer." " We are going to lose money". I would tell them that they needed to rethink the way the quoted the job in the future. If your are experienced as you say you are in the industry then you should know what needs to be done the material for a proper welding job to be done on it. Oh, and then just spray some paint on it. Well the paint will come off if we don't prep the surface for it. The reply was " we don't do that here, we just wipe it down with xylene and paint. I don't have any time in there for painting, but we need to make it look good. I replied to them, " but the paint will come off and look really bad." Their reply to me, I don't care, and the customer doesn't care.
There are to kinds of shops in business today. The ones that take the time and do it right and the ones that are JUST GET IT OUT ASAP. We will blame the Supervisor for it. After all isn't that why he is here to put all the blame on for our short comings.
Thank you, I had to get that off of my chest. I was getting heavy.

By new tito (***)

Date 08-09-2007 15:27

No, that's why QC is there!! haha

Cuz when the supervisor starts getting the blame, his little grubby index finger starts pointing to the inspector while saying "well, he looked at it. He should have caught it"!! :)

By swnorris (****)

Date 08-09-2007 16:50

I totally agree. Quality should be built in, not inspected in.

By Sourdough (****)

Date 08-09-2007 20:53

Muriatic acid...........

By swsweld (****)

Date 08-10-2007 02:10

A 4 1/2" grinder with a wire wheel works good and fast for most scale.

By Fredspoppy (**)

Date 08-10-2007 12:52

A wire brush is not a real good way to remove mill scale. The worst of the mill scale is tightly adhering to the surface and can only be removed by abrasive methods (grinding or grit blasting give best results). Wire brushing will only polish up the mill scale.

A better way (notice I didn't say the best way) is with a flapper disc. Grinding with a solid grinding wheel can induce gouges into the base metal surface that, in some cases where wall thickness is critical, can take you below minimum wall. The flapper disc allows for pretty agressive material removal, without as much chance of gouging the surface. For larger areas, grit blasting is the more efficient way.

By CWI555

Date 08-16-2007 18:46

"quality should be built in, not inspected in." You got that right. It never ceases to amaze me how much blame quality gets. They didn't make the damn thing, they are only verifying the presence of or lack of quality per contract documents and code requirements.

By billvanderhoof (****)

Date 08-10-2007 05:12

The reason you waste the time is so when the lawyers show up with a picture of that bridge in Milwaukee (as an example and not meaning to imply that I have any knowledge of what actually caused that) you get to say "we used best practices".
Bill

By ssbn727 (*****)

Date 08-10-2007 05:48

Hi Bill!
You mean Minneapolis, Minnesota and, the only reason why I noticed that was because I know that Milwaukee is off Lake Michigan from swimming in it off the shores of the city, and the bridge collapsed over the Mississippi River... There sure are alot of places starting with the letter "M" in this paragraph :) :) :)

Respectfully,
Henry

By 803056

Date 08-10-2007 14:06

It is like the fireman's triangle, you need oxygen, fuel, and heat to have a fire. Remove one of the three elements and a fire will not occur. Quality welds start with good design, good materials, good preparation, and good welders. If you remove any one element, quality is going to suffer.

I've read many comments that were square on the mark. Management issues, corporate culture, minimum acceptable level of acceptability, etc., have been mentioned. Quality work requires a team approach where all the responsible parties recognize the value of quality work and are not satisfied with merely "good enough".

Customer's, like you and I, vary in their attitudes on quality. Some of us recognize the value of quality and we are willing to pay the difference in price to obtain the longevity that comes as a result of the quality "built" into the product. Others are only interested in getting the job completed in the shortest time and at the least cost. They are not interested in whether the object will last 20, 30, or 100 years, only that it will last "long enough" to get the job done. Each has valid points of view. As long as the object lasts as long as it is needed, as long as safety isn't compromised, and as long as the intended purpose is served, each will be satisfied and neither will change their attitude or philosophy.

I find it interesting that the fabricators I know that have been around for many years are those that may charge a little more than the competitor, but they turn out high quality work. Most of the "bottom feeders" that are always the least expensive don't usually stay in the game for the long term.

Likewise, those customers that recognize quality work gravitate to those companies that provide the level of quality expected. They will utilize the services of the quality fabricator when high quality is required and pay the premium charged. When "good enough" is adequate for the intended purpose, they utilize the less expensive supplier. The trick is to recognize when to use which supplier. A mistake can cost the company their reputation and their business. For that reason, those companies that are most successful are those that insist on quality work, both from a purchasing standpoint and for the products they produce.

In this case, management's philosophy of "good enough" is a short term vision and one which will be recognized by their customers. They will attract those customers that find the quality of their products "good enough". The duration of the relationship will be relatively short, either the customer will find the quality inadequate and search out a new supplier or they will not survive and they will no longer be around to be a customer.

Best regards - Al

By billvanderhoof (****)

Date 08-11-2007 02:55

Right, Minneapolis, it's a good thing i'm just typing and not flying a plane. At least they both start with an M.
Bill

By bkoz

Date 08-14-2007 20:32

Sorry to rain on the parade, but I generally agree with the VP. This assumes we're only talking about AWS D1.1 or similar structural Code and not talking about flange-to-weld connections of girders.

Tightly adhearing mill scale, blue mill scale, is not easily removed by needle guns, wire brushes, or flap wheels. This kind of mill scale gets burnished by grinding processes unless considerable effort is expended. AISC says the cost for this kind of preparation is 20% greater than normal structural preparation when common hand tools are used. We are not talking about rust or loose scale in this discussion.

Flux-cored filler metals are built to tie-up this blue mill scale and, according to the manufacturer's recommendations, some rust and oil, too. Even some solid wires will do this and SMAW cleans up the weld, too. Please see the information on ER70S-6. This cleaning action is documented in Lincoln, TriMark, and other mild steel electrode manufacturer's literature.

Running an 1/8" E7018 at 110 amps is at the bottom of the amperage range for this electrode diameter. The wetting action described is probably accurate. I suspect the wetting difference would not be so apparent if the electrode were either run in the 3F up normally used with this amperage or at 130 - 135 amps normally used for flat or horizontal.

If the inspector has any doubt concerning the ability of the welder or the process outside of the Code minimum requirements, I suggest running a simple test most welders use to check their heat input. Tack two pieces of plate together, similar in thickness and surface condition to the actual work piece. Run a 2" or 3" single weld pass joining the two pieces. With a hammer, bend the stem plate back and forth until the stem breaks off. Inspect the broken surfaces. If the base metal tears out, ductilly, the production welds will probably be OK, at least good enough for production fillets. Fillets are designed at 33% of electrode strength anyway and todays electrodes are typically 40% stronger than 50 ksi yield base material. I say "probably" because VT is a surface examination and anything can happen. The Code anticipates this.

If there is a concern that this test will show false positives, run the same test using GMAW-S with no oscillation. The stem will peel off every time with no penetration on the stem if the plate is thicker than 1/4".

Even ASQ says there is an economic limit to quality. Or as Lincoln Electric says in the Procedure Handbook of Arc Welding, "Why put in a $5 weld when a fifty-cent weld will do?"

Bkoz

By ssbn727 (*****)

Date 08-14-2007 20:42

Another "Ringer" in disguise - huh??? :) :) :)

Respectfully,
Henry

By bkoz

Date 08-16-2007 16:58

A Ringer? No. Maybe too many years in the business and too much education ;}

Recent work by ASCE (Schafer, Ojdrovic, Zarghamee) discounts the importance on weld "defects" on structural performance and places the burden on structural design. Blodgett has been talking about the importance of welding design for decades. The new "Fitness for Service" Codes like BS 7910 and API 579 offer a rational means of assessing weld discontinuities and their actual impact on the performance of a weldment rather than painting with the broad brush of consensus Codes. The U of Utah performs full section load tests and the results would surprise you. They have a 2,000,000 lb. ram there and they can push the world. We have seen visually & UT rejectable CJP welds survive while the structural sections buckle. That says a lot about good design. The structural tests routinely exceed the test requirements set by AISC and ICC, even with known weld flaws.

The designers using the U's structural lab are testing minimum designs to keep their structures light and inexpensive. They design in accordance with the minimum requirements of the Steel Construction Manual. But today's structural shapes and weld metal are actually stronger than the minimums established by ASTM or AWS. A992 permits as much as 30% greater yield than designed. AWS has no limit to maximum yield, but we're typically measuring weld metal at more than 40% stronger than the minimum designed yield strength. This helps explain why the structures exceed performance expectations.

I am not suggesting that the welding Codes are too tight, nor do I suggest we not enforce the Codes to the fullest extent as we understand them. I believe AISC and AWS have developed their Codes with diligence. This diligence considers the imperfect knowledge and abilities of the qualifed welders and inspectors working in our country. We shouldn't leave the Codes behind, however, because of speculative would be's and could be's. I recommend when we go outside the minimums established by the Codes, we do so with consideration and with input by other members of our fabrication and erection teams. There is nothing worse than a team member with the true religion, regardless of what hat they wear.
Amen ;)

Bkoz

By CWI555

Date 08-16-2007 19:35

I don't mean any offense, but your putting the cart before the horse here. "We have seen visually & UT rejectable CJP weld survive while the structural sections buckle".
Yep, if the design sucks, no amount of good welding or bad is going to make it better.
However; the inverse of that is true as well. A good design can be weakened by bad welding. Any design is only as strong as it's weakest link. That has been and will always be true.

When pushing the edge of mimimalistic design, care must be taken to remember the human cost of failure. Sometimes a little extra strength is better.

"imperfect knowledge and abilities of the qualified welders and inspectors working in our country". Odd how you focus on that. The best welder, and inspector cannot overcome poor design and or sub standard materials. Just being in management or engineering doesn't make someone "perfect" contrary to what many of them would have you believe.

Either side of the fence can point to one or the other study for support. Very often statistical data is used to buy off something as was the I 35 bridge. My professor informed all of us the first day that there are only two kinds of statisticians, liars and damn liars. That has been borne out since many times over. There are some values to be had with it, but relying on it over common sense and good sound judgement based on experience is a fools errand. Sooner or later the odds catch up to you. Unfortunately, these events tend to cost human lives, not to mention millions. I'm sure the engineers who keep the I 35 bridge open had good numbers, as did the engineers from the NASA, Marriott St Louis, Sonatrach in Algeria (LNG), Union carbide plant in India, TMI in PA, bridgestone, just to list a few notables.

"Good enough" is not good enough when children lose their fathers, or whole families cease to breath because some engineer somewhere said it's "good enough".
Are you going to accept "good enough" when you have a home built? Two nails will hold that truss down, after all what are the odds of a heavy wind here? Somehow I don't think that mentality would go over to well with most people.

Even with the best designs, highest quality, and strictest adherence to the relevant codes, there are sometimes failures for reasons no one could or did foresee. In all other cases, someone saw the potential and said "it's good enough".

"I am not suggesting that the welding codes are to tight". That is exactly what you are suggesting with " Recent work by ASCE discounts the importance on weld "defects".

By CWI555

Date 08-16-2007 18:48

Those pictures and reports have already surfaced. Your example is spot on.

By bkoz

Date 08-17-2007 16:28

Hi Folks,

Let's remember. We are talking about a "thin layer of mill scale" on a stiffener-to-wide flange connection. The talk about "life threatening" is a red herring and we should all know that. The stiffener is a secondary connection, most likely placed there to prevent local buckling at a hole. I would be more worried about the micro-cracks in the hole periphery from the punching operation or if the hole gets reamed out in the field to fit misalignment during assembly. I would really be more concerned if the welder ran the weld into the "K" zone of the wide-flange instead of the correct hold-back now required in most designs and AWS.

The buckling I refered to in the structural tests is good design. The failure mode was the way the designer intended it to fail and the structure failed at margins way above the the AISC test protocol limits. These are data-based limits not consensus limits.

To talk about the I-35 bridge collapse is premature, but we are finding the lack of maintenance to be a more pervasive problem for our bridges than weld failures. Concerning babies and mothers dying in a collapse or failure of any kind, we've all been making that decision with our tax dollars for years. MinDOT and the Minnesota State Legislature only a few months before the collapse determined the bridge was safe enough for the public. It would be naive to think there aren't any other bridges or other structures teetering on the edge of "safety".

Do you really think we found all of the cracks in Northridge? Today's data indicates the majority of the Northridge cracks were due to design. Poor welding was not commonly observed in those failures.

I was involved with the Challenger disaster. I was QA Mgr for the company that cut the 3rd O-Ring groove on the booster rockets. That disaster was caused by a failure of management at Thiokol under pressure to launch from NASA. They ran the boosters at temperatures outside of the proven design range. Neither welds, inspection, nor design had anything to do with it. It is a red herring to the discussion of blue mill scale.

The Hyatt Catwalks failed because of a field modification to the bolted connections. This has been described in depth in Structures magazine. The revised connection put the load of all of the catwalks on two channels welded toe-to-toe. Those channels were not designed to take that much load and the rods pulled through. An unchecked design change caused that failure, not a weld discontinuity. It is a red herring to the discussion of blue mill scale.

Bhopal and the rest, well, others know more about that than I. But I do know those failures were not due to the remote possibility of a lack of fusion in a secondary weld from blue mill scale.

Read and follow the manufacturer's data and recommendations. Welding electrodes are built to handle minutia like blue mill scale. One can run that destructive T-joint test if they have any concerns about the data offered by Lincoln Electric and other experts. One shouldn't climb to the top of the mountain to show the rest of us their hair is on fire.

Live within the parameters of the Code. Or I'll eat you with backcharges ;)

Bkoz

By Fredspoppy (**)

Date 08-17-2007 17:59

Lots of good discussion here regarding welds that are "fit for service". I do get nervous when I hear things like "Flux-cored filler metals are built to tie-up this blue mill scale and, according to the manufacturer's recommendations, some rust and oil, too. " Reminds me of the days in the Maintenance Welding business when it was promoted that you could "weld pass over pass without chipping the slag". Then one day a guy called in, who was repairing a big old crane, and said, "your salesman said I could weld "pass over pass without chipping off the slag"". After I came to, I promptly set him on the right path.

I know there are many welding applications where removal of mill scale will not make a difference in the serviceability of the weld, but I bet Henry (SSBN 727) wouldn't want to hear "Dive, Dive" if he knew that all of the welders at EB were slugging the welds and welding over mill scale, grease, oil, etc. Here in the offshore Oil & Gas industry a defect 0.5mm high x 15mm long is a reject in a 12" diameter x 1.5" wall pipe for SCR applications. Push just a little over 1/2" of lead out of your 0.5" mechanical pencil to get a real idea of what a small defect that is.

In general, cleanliness is next to Godliness in welding. Let's make sure that all those failures that bkoz recalls continue to NOT be remembered as weld failures.

By ssbn727 (*****)

Date 08-17-2007 18:36 Edited 08-17-2007 18:51

I concurr!!! :) :) :)

That's why anytime I had to deposit welds to join any members of the pressure hulls, I made darn sure that this welder was'nt going to leave any defects or discontinuities which could potentially turn into defects later - on due to the wide variety of cyclical loads submarines will experience!!!
So far, none of "My Submarines" had any problems to speak of in that department!!! :) :) :)

When one takes a pressure vessel as deep as the submarines we built back then or today, You better believe that there will be NO MILL SCALE WHATSOEVER IN OR AROUND THE JOINTS, WHETHER THE JOINTS ARE PART OF THE PRESSURE VESSEL OR THEY'RE LOCATED WITHIN THE PRESSURE HULL - PERIOD!!!
NO EXCEPTIONS!!! :) ;)

Respectfully.
Henry

By 803056

Date 08-17-2007 20:10

Most catastrophic failures are the result of a confluence of minor problems that are unlikely to act in unison. However, given the correct conditions the additive affects result in failure.

Each player in the construction of a machine, building, or bridge makes certain assumptions. The engineer hopes the loads he plans for are those the structure will experience only in the worst of scenarios. Did he consider the effects of corrosion over time? Did he listen to the assurances of the politician that maintenance would be performed on a regular basis? Were extra inspections or additional NDT methods specified for the design containing non-redundant members?

The welder tries to make the best welds he can given the conditions of the design of the joint, process, position, etc. Did he do his best, or was he tired from the 60-hour workweek required to ship the material on time and within budget? Did management count on the design safety factor to allow for a certain number of undetected defects to pass through the system?

The inspector employs the test methods specified by the engineer. Hopefully the probability of detection is high enough that any critical defect is not missed and the indication is interpreted and evaluated correctly. Was the inspector qualified to operate the ultrasonic testing machine? Was the electrode properly stored or was the correct electrode used on the completed weld that the inspector is asked to visually inspect?

Did the ironworker put in all the bolts? Everyone knows the connections are over designed. That 48th bolt is hard to reach and the hole is misaligned. No one will ever see the missing bolt or the four or five bolts that were not fully tightened. Where problems detected and reported by the workers? As the engineer on one project told me, "Son we pay you to put the building up, not to analyze the design" (at the time I was an apprentice Ironworker on the first Hartford Civic Center and I was commenting on the flexibility of the roof structure). God forbid that management listen to a worker!

Then there is the politician that usually reacts after the accident happens and shouts "how could this happen?"

How, because the necessary maintenance required to mitigate the slow self-destruction of the structure was not performed because of budget considerations. Or maybe it was the idea that full time on-site inspection by a qualified inspector is a waste of money, after all, the contactor's foreman of two weeks is an expert. He wouldn't cheat!

Who is to blame? Where will the finger point to when the inquisition starts?

It is something to think about when you and your family go to watch that next basketball or football game in some super stadium.

Al

By CWI555

Date 08-17-2007 19:26 Edited 08-17-2007 19:31

"Let's remember. We are talking about a "thin layer of mill scale" on a stiffener-to-wide flange connection"
Yes this is a true statement. So where does your statement of "discounts the importance on weld "defects" come from?
As far as a red herring, and I 35 bridge, I've read the reports myself, fracture critical members were reported as cracked and let go.

FCM "Fracture critical members or member components are tension members or tension components of bending members (including those subject to reversal of stress), the failure of which would be expected to result in collapse of the bridge." D 1.5.

I don't know what world you live in but a bridge collapse I would classify as a threat to life and limb for any who find themselves on it when it goes. Are you suggesting that we let a problem go just because thats how it's always been?

Your right about bad maintenance, it does play a large roll in it. Which touches on another point, even a good weld and a good design is subject to stress over time. Again, engineers made an analysis, only in this case they got it wrong.

"Today's data indicates the majority of Northridge cracks were due to design". Yep, I've read that as well. All of the analysis is again based on statistics derived from a model developed from sampling. This again is subjective. If I want a positive result to a sampling plan for say who makes the best cheese, where do I pull that data from? Wisconsin?
Any data pulled from Wisconsin will be skewed for obvious reasons. If I want to shuffle the blame off of any given event such as a bridge collapse, Do I design my model around the findings from site, or do I throw in design data to muddy the waters? Hind site is 20/20 for anyone, and having the benefit of that, a case can be made to fit the particular need of the party concerned with CYA.

Yes you are correct that the challenger was not a weld failure, and no it was not a red herring, it was an example of engineering analysis being influenced by monetary concerns and schedule pressure.

Hyatt cat walk a red herring? Again it was an schedule and monetary pressure concern. Had they stuck to the original design, there would be not problem. The entire structure was reported as inspected in 1 hour. You think that was sufficient time given the nature of that structure?

Taking a question in regards to welding over mill scale and twisting it to this statement "discounts the importance on weld failures" Who is using Red herrings here?

You began with quoting a study, those studies inherently contain copious amounts of statistical data designed by someone or group who is trying to prove a theory they came up with. There is an inherent bias in these studies.

Therein was the purpose of my post. If the statistical data was always correct and unbiased; multiple failures throughout history would not have occurred. There is a limit to the worth of statistics. This is recognized by the codes by margins deliberately built into it for that purpose. However; when schedule pressure, the bottom line, and or some engineer wishing to prove a pet theory, encroach upon these margins and discount the advice of experienced welders and inspectors, that is where the threat to life and limb begins. To say that concern for the well being of people utilizing bridges, buildings, and other structures, is a red herring speaks volumes for your perspective.

I dunno, maybe I am wrong, but it's my belief that if your first concern is measured in dollars and cents, rather than the lives and well being of those who would use the end product, then I think it's time for that individual to find a new career.

By DaveBoyer (*****)

Date 08-18-2007 04:36

So while We are here on the net arguing over what might be *best*, what are the actual practices in the field?

By bkoz

Date 08-20-2007 17:23

For steels:

We sandblast pressure vessel Cat A & B welds and steel girder webs and flanges to SP 6 commercial blast to remove all rust, mill scale, and old paint completely, except slight streaks, shadows, or discolorations caused by rust stain, mill scale oxides. Joint bevels and 1" beyond are ground to bright metal on both sides of the joint. We blast to save time on grinding and for coating later. Cat C & D welds are often in the same condition, but we will grind off the producer's paint finish for fittings. Reinforcing pads may or may not be blasted before welding. Supports and attachments are never prep'd unless to facilitate coating.

Structural D1.1, D1.8, and D17.1 CP & PP joints are prep'd by burning and ground to bright metal at the joint bevels and 1" beyond on both sides of the joint except one-side PP joints. Fillet, and flare-bevel welds have no surface prep unless the connection is rusty, oily, or coated. Square butt joints of secondary members such as toe-plates, ladders, handrails, etc have no surface prep unless rusty, oily, or coated.

D1.4 joints are prep'd by burning with no further surface preparation. Our welding procedures are developed and welders qualified to weld through galvanizing. Rebar is readily available galvanized.

D1.3 joints are cleaned of oil and grease. Our welding procedures are developed and welders qualified to weld through galvanizing and coating. Metal deck is almost always galvanized or coated.

Did any of you guys run that destructive test on a tee-joint yet? Mine worked fine!
See you on another thread.

Bkoz