TIP TIG Welding is always better quality than TIG and 100 to 500% faster with superior quality than TIG - MIG - FCAW.

• Robots. MIG welding and process control part 1

[1] Weld process ignorance has for decades retarded weld process and consumable evolution and has dramatically increasedd the average weld / manuafcturing companies weld costs and weld rework potential.

[2] Weld ignorance can make SIMPLE ROBOT WELD APPLICATIONS APPEAR COMPLEX.

[3] Weld process ignorance can create a confused weld shop culture and generate numerous weld shop myths.

[4] Weld process ignorance CLOSES THE DOOR TO WELD EQUIPMENT, PROCESS AND CONSUMABLE EVALUATION / SELECTION LOGIC.

[5] Weld process ignorance opens the door to weld salesmanship and to the purchase of less than optimum weld consumables and useless, costly electronic bells and whistles.

[6] Weld ignorance has created a too common frustartingwork enviroment where too many managers and engineers
SEEM TO BE FOREVER CHASING WELD SHOP ISSUES BUT NEVER RESOLVING THEM.

[7] Weld process ignorance hides the real cost of welds and places weld cost focus on invoice items like the cost of weld wires or gas.

[8] Weld process ignorance hides real world WELD DEPOSITION RATE POTENTIAL, DAILY DECREASING WELD PRODUCTION EFFICIENCY POTENTIAL.

[9] Weld process ignorance ties up plant personnel people resources who will during weld team meetings waste valuable man hours trying to resolve weld issues that should take an individual with process expertise minutes to resolve

[10] Weld process ignorance DRAMATICALLY INCREASE A CORPORATION'S WELD PRODUCT LIABILITY.

When those robot weld reject bins fill too rapidly and the production management is trying to decide if the reject bins are half full or half empty, many managers in their weekly reports will tend to blame;
[] the robot operators,
[] the robot programmers,
[] the robot equipment,
[] the robot weld equipment,
[] the robot fixtures,
[] the weld process,
[] the consumables,
[] the parts,
[] the salesman who sold the equipment.
[] Jesus, Mohammed or Buddha.

However the weld reality is simply this. To solve weld process issues, and optimize automated and manual MIG weld applications, both management and engineers should have the ability to recognize the "root cause" of the weld process control issues. The bottom line is they only have to get the robot weld process training resources avaialable at this site.

 

Project: Robot Process Optimization.

Client. Freight Liner.

The following has been implemented at the Freight Liner facility. Major changes were made in the existing welding programs of the ABB 2400 series robots. The process control program changes completed by Ed in the first week of Aug, were designed to improve the weld fusion potential and also to reduce the weld cycle times of both the HD and LD parts. For more info on these Frightliner robot weld, click here.

 

 

• 

  Why is it the welding department seem to always have the most issues?
  

   

  Question. Ed: I have been in this company now for almost 20 years. When I think back, the welding department and it's personnel have always been the area in which we have had most manufacturing and people issues. I know in the last 10 years we have likely gone through 4 people who have supervised the weld department. Why is it we cannot seem to get control in this specific area?

Ed's reply: In reality all weld manufacturing issues work there way down from the plant managers office. It's possible that your management has never correctly identified the root cause of their common issues and are not aware of the caliber of the individual they need to run a highly efficient weld shop. By the way notice how the you focus on the performance of the weld shop supervision and the people in the weld shop, what about the performance and quality of the individuals who hire the ineffective supervision?

Many manufacturing companies employ managers that have little respect for welding and minimal knowledge of the requirements for effective weld process controls. It's therefore logical to assume that the manager's weld process indifference will have it's influence on the weld shop performance. The management indifference is an important part of the plant culture.

Weld quality and productivity issues will be more numerous in weld shops in which confused managers surround themselves with confused engineers who are surrounded by confused employees who get no assistance from their confused job descriptions: 1985. Ed Craig

 

 

A "qualified weld or manufacturing manager" maybe as rare as a bird on the edge of distinction. I would speculate today, that no more than two in twenty individuals who claim they have the necessary management expertise to optimize robot weld production have the capability.

My advice to manufacturing companies who rely on welding automation, when you get the opportunity to get a qualified management individual or weld / robot technician, look carefully at their compensation. How logical is it when you find a "Mr. Right", that this individual should be paid less or no more than some of the ineffective, engineers you already employ, engineers that have been unable to resolve your manufacturing process issues?

If a company employs a technician with extensive expertise in both robotics and welding, any management recruiting company would tell you that this individual will be much harder to replace than an engineer or your corporate accountant. In companies that utilize many robots, when the robot weld quality and productivity is optimized for a manufacturing company the costs benefits to that organization are always substantial, therefore the remuneration for these services should be easy to justify.

If you are one of the unique, few individuals experienced in both weld process and robot programming and you are looking for a job, remember with the dramatic growth of robots in the welding industry, the manufacturing industry will for a long time need you more than you need them. Don't sell yourself short, if at the facility where you work you find the management and supervision are ignorant to the responsibilities of optimizing welding productivity and quality, then make it your responsibility to first try to educate them. Print out appropriate parts of this site and place it in the managers mail box.

If the manager is a logical individual they will get the message. If the management is unresponsive to you, then vist www.monsterboard.com, www.careerbuiler.com, www.worktree.com, (worktree typically has most robot weld jobs, note you have to pay a small fee) www.flipdog.com and www.hotjobs.com. Simply use "welding" as the keyword and you will find many opportunities.

The following add shows why in many instances weld management and HR personnel are out of touch with weld reality.

 

A Typical Weld Industry Want Add: Aug 2008: "Wanted Manufacturing Engineer". Wanted engineer or experienced technician to work for a fortune 500 company with sales over 1 billion annually. This major automotive company requires an engineer to manage and resolve their weld, paint and press shop issues. The engineer will have responsibility to manage the welding production at four plants that utilize over 200 robots. You will be required to understand the soft ware of Panasonic and ABB robots, and manage the laser, plasma and resistance equipment. You will ensure that metal forming, press and paint production is optimized. You will be in charge of capital expenditures of more than five million dollars annualy and be responsible to train over 200 personnel. High energy person required as 60 - 80 hrs hour a week is norm. Salary range: We will pay the same as we did in the 1980s, $55.000.00 to $65.000.00 per year, "depending on experience". Relocation: Not provided, as we don't think to much about this position..

The above add is typical, and also quite remarkable in it's process and manufacturing expertise naivety. The big three auto companies learnt many years ago how specialized the expertise and skills are for process optimization in metal forming, metal joining and metal painting. That's why in many instances these companies have given up on these functions and out sourced the applications to tier one and two suppliers.

Many auto and truck manufacturing companies today have so many quality and productivity issues with painting and welding that they either out source the parts or hire consultants to take over the processes in their plants.With this in mind, its ironic that today you can visit the web and see numerous want adds in which a manufacturing company wants a "multi-skilled" engineer with expertise in welding painting, press and machining. The reality that management needs to come to grips with. Engineers are not taught how to control the processes mentioned. Each of the processes mentioned is unique and can take a life time to master.

MANY BIG THREE EXECUTIVES QUICKLY LEARN WHEN THEY LACK PROCESS KNOWLEGE IT CAN BE DIFFICULT TO CONTROL MANUFACTURING PROCESSES SUCH AS PAINTING AND WELDING, YET MANY TIER ONE COMPANIES STILL TRY TO HIRE ENGINEERS WITH "MULTI-PROCESS EXPERTISE".

 

The naive manufacturing company that looks for the "multi-talented process individual", typically ends up with an engineer that will have limited expertise in one process, and his experiences with the other processes was attained as he walked by them on his way to the coffee machine.

Again we see how the management and HR perception and expertise can have a great influence on why their plants do not get the skills, expertise or training necessary for process optimization.

 

 

THE AUTOMOTIVE ROBOT TIG APPLICATION:
This weld report deals with the robot TIG auto welding issues. The parts required approx. 15 precise small tack welds. The tacked parts were later brazed, The TIG welds were made with a Fanuc Arc Mate 100 robot, and a Lincoln 350 amp "pulsed" square wave power source.

The welding issues at this tier one part supplier were extensive.For more than a year they had struggled to attain a production rate of only 40% of what they desired. The tack welds were frequently missing, arc starts issues were extensive, and the tack welds would leak. After I rectified the problem, I wrote the following report to the plant management.

.

 

THE BIG THREE INSIST THAT THEIR WELDED PARTS ARE ONLY WELDED WITH ROBOTS. It's ironic that on the one hand we have the major auto / truck companies providing a logical mandate which insists that arc welds will only be made by robots. On the other hand, the majority of the companies that use robots will allow "unqualified" robot operators, programmers or maintenance personnel to daily make "unqualified weld parameter changes to pre-qualified welded components. And then of course they will allow unqualified manual welders to make unqualified weld repairs, nice thought though.

MANAGEMENT AND WELD PROCESS LIABILITY

 

Does today's corporate, automotive manufacturing management, seriously believe that the weld integrity of the MIG welded components on their cars or trucks should be left in the hands of robot operators or personnel that daily make unqualified weld process changes.

Does the same management believe that the weld process is in control when its controlled by unqualified" engineers, maintenance electricians or the millwrights on the third shift?


Its not a big deal. To attain weld process knowlege can be as simple as reading and digesting the process books available at this site.

"PEOPLE CONTROL" IS THE MOST CRITICAL PART OF BOTH MANUAL AND ROBOT WELD PROCESS CONTROL: Its ironic in the automotive industry that the rare experienced robot programmer who knows what it takes to manage the arc welding robot line, is rarely allowed to manage the operators running the robot cells. You may be familiar with this scenario. You have a finely tuned robot, controlled by an operator or maintenance person who rather than work with the available welding instructions would rather make their unique unnecessary process changes and play around with the robot program or settings. In many manufacturing plants the robot operator or maintenance person not suited to work with robots, will typically report to a shift or maintenance supervisor who knows less about weld process controls than the robot operator does. Lets face it, the prime responsibility of most supervisors and lead hands in many manufacturing plants that use robots, is meet the shift production demands and get the just in time parts out of the lean door.

 

 

 

Question. Ed. We have had robots for a few years, I don't think my department or manufacturing managers are aware of all the issues you bring up.

Answer: Ignorance of the resolutions for shop floor weld issues, are the reasons why so many managers spend a great portion of their time fighting fires without ever putting out those fires.

Of course process awareness will depend on your companies management and engineers expertise. It will depend on their ability to see through the weld smoke, comprehend the weld shop issues and measure the real robot or automated weld equipment performance potential. Unfortunately at this time too many managers are happy just to see the robots running. After these mangers arrive home in the evening, they are so exhausted fighting fires that few of them will visit this site or review the process control resources.


Weld Fact: A company does not have robot weld issues:

• If the robot weld reject rates are less than two percent every shift.
  
  
• If the total robot down time, "per robot per shift" is less than twenty minutes.
  
  
• If the weld data or program data set on your robots is so fine tuned its impossible to increase the robot speed or the weld speed.
  
  
• If your organization has achieved this, then this web site will offer you little. The weld reality, maybe 10 robots per 100 in North America, Japan and Europe are operating at their peak performance potential.

Anyone can set a welding robot to run at 40 to 75% efficiency. Few can optimize the robot so it can consistently achieve optimum weld quality, with optimum weld productivity at plus 97% robot efficiency.

Who should run the robots?

Question. Ed, at our company we normally take production welders and let them run the robots. Is it necessary to use welders as robot operators, and what is the criteria for a good robot operator?

Answer . Why use the skills of a manual welder when the robot is designed to provide these skills. When hiring a MIG welding robot operator, it's logical to get a worker that has a good work ethic. You want a person who will stay at the robot station, a team player with a good work attitude. You want a person who will follow process control instructions, a person who will take pride in maintaining their work area, and take pride in achieving optimum weld quality and production each shift.

The operator should be given basic MIG process training so they can recognize "what a change in an arc sound means", They should also receive training on how to inspect a weld, and understand what an acceptable weld is.

For a trainee robot programmer, select an individual that likes working with computers, CNC or electronic controls. An individual who has an interest in welding. My robot MIG weld process training book and video is all they need to control the welds, and the robot companies "robot training program" will do the rest. The bottom line, an individual requires no weld experience to eventually become a good welding robot operator.

Why so many weld issues in auto/ truck plants

Question: Ed, why are so many robot weld issues found in plants that produce auto / truck parts?

Eds Reply. The welding process issues discussed in this site are prevalent throughout the total welding industry, even in those manufacturing plants that work with stringent ASME or AWS specifications. However automotive plants deal with more robots than most other industries and work with high volume applications. Also it's a sad fact, that for decades in contrast to other metal manufacturing industries, when it comes to welding the auto /truck industry has been notorious for not hiring experienced, qualified weld personnel.

• Automotive / manufacturing management. It's common in the plants to find the auto managers
  tied up with demanding just in time inventory, ISO requirement and spread sheet issues. Few managers get involved in seeking the root cause of specific manufacturing and process issues.
  
  
• Automotive / trade unions. The automotive industry in manufacturing years is old. Many of the trade union rules and regulations in use today were established fifty years ago when manual shop floor weld practices were very different.
  
  
  
  
  

Most union weld manufacturing rules and regulations were established at a time when manual weld skills had more meaning than the complex requirements of weld automation and process controls. In other words there is time for engineers to be involved with a process and there is a time they are not required. In the case of robots they should be involved and of course they rarely are.

The automotive industry competes in a highly competitive global manufacturing market. To remain competitive, a company has to use "when effective", the latest state of the art, automated welding equipment. In the last decade the weld manufacturing world has turned a corner, it's become a world in which manual welding skills are secondary to the unique skills required to optimize "automated welding equipment" yet in the auto / truck plants few robot job descriptions are clearly defined.

Correct union job descriptions are absolutely necessary for controlling automated welding equipment.

• Automotive management and Inexperienced weld process engineers. There is too much "hands off engineering" going on in automotive manufacturing plants. An engineer's role has to be more than the person who gets someone else to provide advice, or someone who will come to his plant to fix the plant's equipment. Complex automated equipment needs more than a passing interest from the maintenance department, it needs the involvement of "hands on engineers" who can optimize both the equipment and processes critical to the plant.
  
  
• Automotive double manufacturing standards. Why would any plant manager be tolerant of manufacturing engineers who allow gaps in the robot welded parts which are greater than those allowed in the design requirements.
  
  
• Automotive and weld part design. Few designers of welded parts understand the weld process requirements for those parts. Designers can have a dramatic impact on weld cycle and weld quality issues. However to achieve this, they would have to develop an interest in the weld process used on their parts.
  
  
• Automotive effective weld process controls. Lack of focus by management on establishing and maintaining effective weld process controls. This has to be a number one management priority.
  
  
• Automotive and ineffective process training programs. "Our workers are this corporations greatest asset". This is a common management theme, yet if the workers are the companies greatest asset shouldn't the quality of the training provided reflect the importance of that asset? In reality, the weld process training provided over the last three decades to workers in the metal forming and auto industry has been either none existent or so poor and ineffective it's had minimal impact on the weld quality or production.
  
  
• Automotive unique production demands. We need 20% more products each year for 20% less costs. Poor component weld design, sloppy workmanship, frequent design changes in combination with poor fixtures, lack of process control expertise, it all takes its toll.
  
  
  For many companies who lack weld process expertise, it can take them twelve to twenty four months to get the weld production issues resolved for a part with a three year life.
  
  
  A Weld Reality. If all the robot weld variables are controlled up front, it should never take more than two days to program and optimize the robot weld quality and production for the most complex, robot welding application. Most "automotive" robot MIG applications should require a few hours to optimize the weld program. For those plants that are taking longer, the reasons and resolutions are in this site.

LETS HOPE YOUR SHOCK WELDS ARE OK.

Monroe, a tier one, USA shock manufacturer could not get the robot welded shock brackets to qualify for a Chrysler "weld bracket load test" and they contacted Ed to get to the root cause.

The Chrysler load test weld specification required that the robot MIG bracket welds on the shocks absorb at least a 13,000 lb test load. After welding with a robot and using the high energy spray transfer mode, the monroe bracket welds on the 2 to 3mm gage steel shocks would fail typically in the 7000 to 9000 lb range.

What was ironic about this application was the Chrysler weld spec for the parts required a minimum test load of 13,000 pounds on the bracket welds. On evaluating the welds, it took me two days of manual MIG welding and testing the parts to reveal that any test load of less than 19,000 lbs indicated a lack of weld fusion in the shock bracket welds.

The reason the Monroe could not meet the test load requirements was;

[1] The robot "weld speeds were set too high" for the round cold rolled parts,

[2] The required shock brkt "weld lengths" were "too small".

After I identified the weld issues I made the following;

[a] changed the robot weld parameters,

[b] slightly slowed the weld speed,

[c] changed the gun angles,
[d] slightly increased the weld length.

With the changes the shock bracket welds on average failed at 21,000 lbs. The sligh decrease in the robot weld speed did not impact the weld production as I compensated with an increase in the robot speeds between the welds.

HOW FAST, HOW SLOW SHOULD A WELDING ROBOT GO?

Many robots today are under utilized and not welding as fast as they should be. While other robots are welding at speeds which have a negative impact on the weld integrity. The robot weld speeds can be influenced by;
[a] the wire size or gas selected,
[b] the shape of the part, the part thickness, or the part surface,
[c] the parameters selected are not optimum,
[d] using pulsed when they should be using spray.
[e] using short circuit or globular when they should be using pulsed or spray,
[f] perhaps its the fixture, part design, joint type, part thickness or ridiculous gaps,
[g] maybe its the weld size and weld length that could be changed.

In my process control books and training CDs you will find optimum robot weld speeds
for all your applications.

,

WHICH MANUFACTURING METHOD SHOULD WE USE THIS MONTH? In their quest to resolve weld shop issues, many automotive manufacturing engineers get caught up with trendy manufacturing methods. The steel weld manufacturing industry is becoming notorious for it's reliance on confusing, unnecessary manufacturing practices and standards. These practices will change as soon as a new author has a book published on a new approach on how to get the welded parts from the east end of the plant and out the door at the west end.

IF YOU DONT ASK THE RIGHT QUESTIONS YOU
CANNOT ADDRESS THE WELDING ISSUES.

The manufacturing manger has directed the HR person to look for a manufacturing engineer to work with the MIG welding robots. The HR requested that the individual must have experience with some of the following. HAVE YOU GOT EXPERIENCE WITH THE FOLLOWING? Dynamic Control Plans. TOPS. DOE. FMEA. LEAN MANUFACTURING. KAIZEN. KANBAN. MRP. J.I.T. T.Q.M. SPC. POKE. YOKE. During the interview no one asks the engineer about his WELD PROCESS expertise and ability to provide WELD PROCESS CONTROLS for the robot weld lines.

It's not rocket science. The bottom line in most plants steel comes in
the doors at one end of the plant you then;

[a] cut the steel,
[b] clean the steel,
[c] shape the steel,
[d] tack then weld the steel,
[e] examine and test the welded parts,
[f] clean and paint the parts,
[g] assemble the final product,
[h] test the final product,
[I] package and ship the product out of the door at the other
end of the plant.

When working through [a to I], as a manager responsible for manufacturing, its logical to employ practical individuals who know how to 'minimize material handling" The individuals will put "practical equipment" in place that will allow the "highest possible robot weld duty cycle each shift". The individuals will have the "ability to optimize the robot performance", and to "train the employees to optimize the equipment and processes required".

What manufacturing companies really need for robot weld process optimization, is to hire employees with the "necessary process expertise" and make sure the employees have a good dose of the best manufacturing asset an employee can have "practical common sense" .

 

Weld shifts and inconsistent weld results

Question. Ed, we find that with the second or third shifts the robot weld issues often increase. Many times the operators on these shifts cannot resolve the robot programming issues. The robot down time and weld rework from these two shifts can be extensive. How would you address these problems?

Ed's Reply. It's typical in many plants, that those with the most programming or process experience work on the first shift. It's rare to find auto / truck plants which have highly qualified individuals on all the shifts. The reality therefore in plants without effective weld process controls, is someone on the second or third shift is going to get into the robot program to try and rectify a welding issue. Once in the robot program its easy for someone without in-depth programming or process expertise to create more robot issues than they are trying to resolve.

The lack of in-depth, robot programming expertise is one reason companies should improve and extend the robot training provided. Also consider the use of custom PLC controls for simplifying the operation of a robot, so the operator does not have to go into the complex robot pendant. For example, the robot stops in the middle of the weld and your robot operator on the second shift has one of the following robot program options. The operator;

[a] wants the robot to continue from the welding point it stopped,
[b] wants the robot to go on to the next weld,
[c] wants the robot to go to the previous weld,
[d] wants the robot to go to the home position,
[e] wants the robot to go to the gun station for a nozzle clean then return to the next weld,
[f] wants the robot to go to the automatic tool center point for a TCP check and then start at the next weld,
[g] wants the robot to go to the next part.
[h] wants the positioner to rotate and start welding the next part.

Examine the above robot program options for the robot operator who does not have the expertise of the robot programmer, it can get complex. Here is a simple solution. Examine all the common daily programming issues that could occur for an operator on the second or third shift. I mentioned eight (a-h). Lets say you find thirty fundamental robot programming issues. If you do not have a high level of robot programming expertise on each shift, and you want the robot operators to have more control of the robot program, you could attain a PLC control, connect the PLC to the robot control and have the cutomized PLC keys address the primary issues.

With the PLC control, the next time the robot operator has a robot program issue they likely no longer need the intervention of the highly qualified programmer. For example, the robot has stopped in the middle of a weld and the operator wants the robot to go on to the next weld, sometimes a difficult robot command especially if the robot has to go to another part or the fixture positioner has to move. The operator walks up to the PLC control and simply presses the PLC [b] key, which denote robot go to next weld.

Doing it right, is always the quickest way. Of course when it comes to welded parts, with the high daily production demands of an auto / truck plant, the plant focus is typically placed on weld productivity rather than on the weld quality. This focus can have a tremendous negative impact on the a shop floor welding culture. All managers should recognize the following fundamental fact, and hammer it home to all plant welding personnel. It takes less time to produce a robot welded part that does not require weld rework, than it does to weld a part that ends up in the scrap or rework pallet.

 

Effectve weld process control training

Question: Ed, what about training, my company spends millions annually on training yet the extensive weld manufacturing issues continue?

Es's reply: The management approach to weld training and the effectiveness and quality of the training programs provided are key manufacturing issues. As an example, there is a big three company that spends over 40 million annually on training. At one plant it manufacturers truck frames. The truck frame plant uses several robot lines that require both robots and manual welders. The manual welders are on the robot line to repair or do welds the robots miss. The frame parts moving along the robot line are poorly manufactured and have excessive gaps in the weld locations. The majority of the robot made welds on the poor joints will require extensive weld rework at a later date.

The robot truck line moves at a fast pace. With the short time allowed, the manual welders placed along side the robots are using "oversized MIG wires", poor techniques and cannot do a proper weld repairs. The frame gets to the end of the line, the majority of frames are rejected, then shipped to another part of the plant. The manual welders who work on the robot line are paid overtime to repair the frames on the weekends. This was and still is a common frame weld scenario in a Michigan plant.

Instead of addressing the part gaps, the poor consumable size selection, the inadequate robot weld process settings and the lack of robot weld process controls, the misguided manufacturing managers and engineers at the big three frame plant would daily vent their spleen and threaten the company that provided robots and the fixtures. The big three company decided that the solution to the weld issues was to provide training for the manual welders on the line.
|
Training "manual workers" will have minimal impact on this automation problem. The management, engineers and union at this frame plant will live with this problem till the management learns how to address the root cause of all the welding issues.

Addressing the poor part fit, selecting correct size weld consumables and setting optimum robot weld data that compensates for "acceptable weld gaps" this along with the implementation of robot weld process control will lead to eliminating the need for manual welders on the robot line and will greatly reduce the amount of weld repairs required.

Its worth repeating. For a management to solve manufacturing process issues the management and engineers require some expertise with the processes utilized. So many "hands of" auto manufacturing engineers and managers direct their energy at placing blame on sources outside their company.

Manual MIG process training is only applicable when welders have time to operate the weld process in a "controlled manner" Manual MIG process training is applicable when welders have time to make different weld parameter settings for the different welding problems. When the time is not provided for the welder to control the process, MIG training will not be affective and management and engineers should look for other ways to resolve the welding issues. This issue is one of the prime reasons MIG training at the big three frame plant has not been effective.

 

Effectve weld process control training

WELD TRAINING IN THE AUTO INDUSTRY:

I believe that the majority of training programs available in North America provide little value for robot weld process controls. On a weld training program evaluation, you would find that many weld instructors typically focus on "weld skills" rather than on simplifying the weld process parameter requirements.

In the automotive industry, thin gage applications with weld gaps create unique welding techniques, skills and practices that are not traditional. The gaps would not be accepted in manufacturin plants that take pride in the parts and welds they produce. The gaps require unique solutions and specific parameters for the robot program.

When welders come back from a MIG weld training course they are frequently taught weld skills they cannot use. Correct skills will have little meaning when the production line speed does not allow the use of those skills, or the MIG wire selected is too large for the thin gage parts. Few automotive manual welders will have the time to adjust their welding machines and make MIG process parameter changes. After going through most training programs that emphasize "skills" the welders still usually still end up "playing" with their weld controls.

The reasons the typical skills focus MIG training is often ineffective.With most automotive companies, is the prime weld issues that result in the plants are caused by factors outside the welders control.

 

Can you Relate to the Following?

Robot welded parts with unacceptable "weld gaps".

Thanks to apathetic manufacturing managers and engineers, this is the number one problem for "robot" MIG welds. Management and designer and engineers frequently forget that manual MIG welders can use "adjustable skills" to compensate for weld gap variations. With robot welds, it takes sophisticated programming and complex joint sensing controls to provide solutions to weld gaps.

It's a fact that in most automotive / truck plants, weld part dimension tolerances are not in accordance with the part design. The acceptable dimensional weld gap tolerances for optimum, gage MIG robot welds is typically 0.060. This dimension is acceptable even for thin gage metals up to 0.060. For less than 0.060 the weld gaps should be no larger than the gage thickness. For fillet welds on parts > 3/16 the tolerance should be no more than 0.010. What is ironic is the dimensional weld tolerance which are different for the gaps and parts are rarely known by the designers or manufacturing managers.

To compensate for the common oversize weld gaps on the thin parts, the automotive manual welder who rarely touches their power source controls, develops "reactive welding skills." The welder will dramatically increase the welding wire stick-out to compensate with a weld current reduction. In contrast the robot typically needs process data to compensate for the gaps.

You will frequently see "long MIG wire stick outs" (> 25 mm) used in automotive / truck part plants. Welders working with thin exhaust parts with excess gaps will often use a MIG wire stick out of one to three inches (25 to 75 mm). However the reality is that any company that allows manual MIG welders to weld with more than >18 mm wire stick out is a company waiting to be sued for poor weld quality. When welding with long wire stick outs the weld integrity is no longer an issue as the welds will have more in common with poorly placed chewing gum.

The practical solution to welding gaps. Place a little focus on providing training for manufacturing engineers and supervisors so they can build the welded parts in accordance with the part design dimension tolerances.

Design Issues that effect robot welds.

With most automotive welded parts, the designers weld expertise and weld responsibility frequently ends with him or her "placing a welding symbol on the blue print". Typically automotive part designers will have no idea of the MIG weld process limitations. How many designers ask?

• How thin can we design the part, so there are no concerns for weld burn-through?
• How can we design a part so the robot weld time is at it's fastest?
• How can we design a part to ensure no weld distortion issues arise?
• How many designers are aware that a robot can produce a 3/16 fillet weld at two to three times the speed of a 1/4 fillet weld?
• How many designers are aware that dramatic improvements in weld quality and beneficial reductions in weld cycle times are attained by placing one 100mm long weld rather than two 50mm welds.
  
  

When it comes to welding for robotics, the designers too frequently don't realize the impact of their design decisions on the potential weld quality or productivity, and what should be a major concern to any engineering manager, "they rarely ask or seem concerned".

How many engineering managers have considered sending designers for MIG process control training? I provide this type of training and the point it gets across to the designers is this, . " optimizing robot weld quality and production starts with the part and weld joint design".

Poor robot welding fixtures.

How many robot fixtures have you seen which do not allow optimum MIG gun access or optimum gun angles?

How many fixtures have you seen that have clamps that don't hold the parts with sufficient rigidity?

How many fixtures have you seen in which require a crow bar or hammer to open a clamp.

How many fixtures have you seen that need rework after being in operation for a few days?

How many fixtures have numerous manual operated clamps when a simple pneumatic control would open all?

For thin gage welds, you could reduce weld burn through issues, if the fixture designer would think "heat sink" and add highly conductive alloys to the clamps and fixture in the weld vicinity. Again who trains fixture makers in designing to minimize traditional MIG welding issues? Do weld part designers communicate with the part fixture designers? do both these individuals communicate with the robot programmers or with person who understand the weld issues?

It saddens me when I see designers on the way to their office, walk past the weld production robots and do not even glance at their parts in the scrap or rework bins. Weld part design should never be considered complete till all the weld manufacturing and production issues are resolved.

A common management influenced issue in many manufacturing plants, "ineffective communication between all the parties involved in the robot welding project"

Issues that can effect robot welds.

• Incorrect size welding consumables selected.
  In too many instances, welders and robots are using MIG wire diameters which are too large and require too high weld current for the part. This is a common problem in automotive plants, and the bottom line it's a reflection of the weld process expertise and of the engineers who too frequently make poor weld process decisions.
  
  
• Incorrect robot weld data selected.
  The robot weld data is frequently selected by individuals who lack in-depth MIG process expertise. The weld data while sufficient to get the robot up and running rarely provides optimum quality or productivity.
  
  
• Unnecessary, sophisticated electronic pulsed welding power sources and other unnecessary electronic equipment placed in the robot cell. This equipment is typically recommended by salesmen and purchased by people who believe in salesmen. Less than two percent of MIG welds will require expensive and complex pulsed welding equipment. The prime reason this unnecessary welding equipment is purchased, the purchaser is frequently over optimistic that more costly equipment will compensate for a plants general lack of manufacturing controls and MIG process expertise.
  
  A lack of management awareness of the complexity of the welding problems. For example.
  
  [a] A management that places the responsibility for the robots and weld process controls with mechanical or electrical engineers, these individual typically have minimal weld process expertise. If the individual responsible cannot operate a process and equipment at its highest efficiency potential, then they should not manage it till they receive adequate training.
  
  [b] A management that gives maintenance personnel a free hand to make weld process changes which negatively effect the weld quality or productivity. This is a plant which is ignorant to the product liability consequences of inconsistent weld quality. Maintenance should be in on Saturday and Sunday doing preventive PM. They should not be allowed to make any unqualified process changes which will impact on the weld quality. If maintenance has to make changes, the solution is simple ensure the maintenance individuals have effective MIG process training.
  
  [c] A Management that does not implement and adhere to weld process controls. Weld quality and production consistency only comes when all personnel doing a task "repeat" that task in accordance with the instructions and training provided.
  
  
  
  

At a tier one Canadian (Simpson) plant which makes bumpers for GM trucks, I was asked by the management to resolve some robot welding issues which were causing costly rejection by GM.

When I got to the plant I was asked by the management to first get the union rep's blessing to work with the robots. I have to admit at this point I started to wonder who I should invoice for my consulting services, the plant union rep or the management.

I quickly resolved the robot bumper weld problems, then arranged a training program which would ensure the welding issues would not repeat. To ensure all three shifts received the process training, I provided the training over a four day period. Now keep in mind the welding problem at this tier one supplier was a major concern for GM, yet In the four days I provided the training not one of the plants managers or one of the plants engineers attended the process control training, yes they were invited.
Two major management issues need addressing at this plant,
[1] Management needs to get back control of the shop floor.
[2] Management needs to send a strong message to its engineers about their manufacturing responsibilities.

Robot Weld Calibration: Are the robots in your plant calibrated? The majority of the robots sold before 2000 were not. Few robot manufacturers or integrators bothered or were aware of the necessity of determining if the weld data in the robot welding program correlates with the real weld data used. You cannot create process controls in a plant in this situation . If the measure robot wire feed rate is 600 in./min but the robot pendant data shows 700 in./min, process controls become a joke.

A MANAGEMENT STEP TOWARDS PROCESS CONTROL. Identify the real root cause of robot welding production issues then to correct the issues put the responsibility and accountability on the right shoulders.

[a] Put the responsibility on the correct shoulders. If a designer has specified a part tolerance and the manufacturing department does not meet that tolerance, then its logical that the management reproach should be with the design and manufacturing engineers, not with the production personnel or the robot programmers

[b] Put the responsibility on the correct shoulders. If a robot operator does not adhere to the robot operating requirements and he reports to a shift supervisor, managers don't find fault with the welding team or robot programmer, find fault with the shift supervisor.

THERE ARE FEW AUTO APPLICATIONS THAT BENEFIT FROM LASER WELD TRACKING SYSTEMS. In the automotive industry, money for manufacturing equipment sometimes appears to come from a bottomless pit. In this environment it's not difficult to understand why every conceivable bell and whistle is pushed by those selling the weld, laser and robot equipment. Also its easy to comprehend why unnecessary, costly equipment is purchased by those who believe that the more money you spend the less weld problems you will have in the robot cell.

I remember at a Ford truck plant in Detroit, that with the purchase of a multi-million million dollar robot line they also purchased every possible robot option including a costly laser vision system for each robot. The engineers figured they needed the laser tracking systems becasue they were not qualified to control the dimensions of the parts provided to the robots. Within twelve weeks all the lasers were removed (costing > $300.000.00). The high tech lasers were dumped in cardboard boxes and today they are still likely gathering dust in the plant's store room.

The reason the lasers were dumped, no one in the plant could handle the highly technical issues that were derived when problems or corrections were generated by the lasers.

Good parts, good fixtures, good weld data, good process controls, through arc tracking, TCP controls, trained personnel and knowledgeable managers and engineers, these are the keys to weld automation.

A Professional Approach to Weld Process Control.
There are typically many personnel involved in robot weld production decisions. Designers of the robot welded parts, quality control personnel, maintenance personnel, supervisors, department mgrs, weld technicians, engineers, and operators. None of these individuals are taking a professional approach to robot weld process control unless they understand the robot / process requirements for optimum weld productivity and quality. Ed's CD training resources will help.


Back to www.weldreality.com home page.