Since being questioned about a previous blog about annealed vs. unannealed wire I thought it might be a good idea to write a short blog about cold heading wire.  To clarify, I am not an engineer or a metallurgist but this is how I understand it from my 35 years in the industry.  Raw material comes from molten steel created from a combination of iron ore and/or scrap melted and poured into moulds.  When it cools, the resultant slab of steel is called a billet.  The billet composition is key to the quality of the wire we use to produce fasteners.  Unwanted contaminants or impurities pose risk to the cold heading process.  Cold heading quality billets are inspected for surface imperfections which can be ground out to reduce the risk of surface defects in the final rod or wire.
  The billets are then rolled using a combination of heat and pressure in order to turn a 4000 lb slab of steel into a 4000 lb coil of rod.  This is called green rod and is the most basic state of coiled raw material.  It can then be annealed and coated to create annealed rod (once referred to as #2 rod).  Or it can be drawn at high speed through a series of draw dies to reach it’s desired finished size and then spherodize annealed to create “spherodized annealed at finished size” or “saafs wire”.  Spherodizing is thermo-processing or heat treating the wire to change the molecular structure to one which is spheroidal in nature.  I like to use the analogy of two bags.  One is full of jacks, the other full of marbles.  The wire starts out like a bag of jacks and ends up like a bag of marbles.  If you nead the bag of jacks the lock up and do not move well.  If you nead the bag of marbles they slide well on one another and move very well.  This is why the wire is spherodized.  In order to allow it to shape easier.  “Saafs” wire still requires a final in-line draw or resizing prior to cold forming as the annealing process can result in wire ovality and thus inconsistency in the diameter.  The most expensive and highest quality wire is “saip” or spherodized, annealed in process.  This process takes the “saafs” wire and runs it through a high speed finishing draw after it is coated.  This wire needs no draw and is the most expensive wire available (with the exception of a few exotic options left out for simplicity).  
    This is a somewhat simplistic look at cold heading wire and meant to provide a general understanding.  Hopefully it has informed the uninformed ever so slightly!

 Obviously, I may be slightly biased on this topic but there are a lot of interesting dynamics that come in to play with this question.  I just came from a Fastener Fair in Mexico City.  As with all of these shows, there is a myriad of fastener suppliers from mainland China and Taiwan.  They often display the same product and unlike 10 or 15 years ago, the quality levels are high and the complexity of capabilities is increasing.  If you talk to just about anyone that buys fasteners in North America they prefer to buy product made in North America but pay the prices from China.

  So why is it that they do not want to buy from offshore?  Lead time is often a consideration.  Reaction time to problems is another.  After sales service and flexibility for design changes will also come into play.  These factors are usually softened by the use of a distributor or middle man that deals with the manufacturer offshore to manage quality and then reacts and services their customer domestically.  The distributor’s roll is critical in this equation as supplier selection can make or break their deal.  Dealing with a supplier half way around the world from a different culture is not an easy task.  Very often, the distributor is caught holding the bill when the customer is not happy with the product but the supplier says it is ok.  Distributors are also left holding the bill when they order 3 months of product or 6 months to get a better price and then the customer discontinues or changes the product and only guarantees their firm 3 or 4 weeks of releases.

  The manufacturer, however, has a different set of problems.  The customer wants to buy their product but wants the best price with the least amount of “pita” (pain in the rear!)  The customer needs the importer around to drive the domestic manufacturers price as low as possible.  So they continue to feed the importer enough orders to keep them around while using their lower pricing as target pricing for the domestic manufacturer.  It is a very delicate balancing act that the tier 2 and 3 purchasers perform.  They do a great job of avoiding paying too much for their product.  Obviously the internet has shrunk the market considerably.  The key as a manufacturer is to define your “wheelhouse” and be very good at it.  

Head cracks are taking the industry by storm!  Why is it that head cracks are so hard to control, inspect for, or even eliminate?  First we should talk about what causes them.  Sometimes head cracks are a result of asking the metal to do more than it is capable of (plastic deformation).  This can result from poor material selection.  Sometimes cracks are a result of a flaw in the raw material (often called a seam and sometimes called a lap).  These flaws can be from the billet that the wire originally came from or they can be from the wire processor that draws and anneals the wire or they may be from the cold heading operation in the form of tool marks that slightly break the grain of the steel and create a starting point for the crack.  If it is caused from tool marks at the cold heading operation, the defect will most likely be on all parts produced.  Similarly if it is plastic deformation ( exceeding the maliability limits of the steel) the defect will most likely be on all of the parts.  When all of the parts have the defect it is more easily caught by in process visual inspection.  Rarely, however, thermal processing will open up a crack that might have been difficult to see with the naked eye prior to thermal processing.

  By far the most difficult head crack to detect is the one which is caused by a flaw in the raw material.  This is often a short section or sections of a 4000# coil of wire that has a defect in it.   Unless that defect drops into the inspector’s hands at the precise time that he is catching parts for his in process inspection, it will not be seen until it is caught during 100% post production inspection or until it is caught during installation at the assembly plant.  The fastener producer is then caught in the middle of a raw material supplier that offers replacement value as a guarantee and a tier one or two manufacturer that demands sorting costs, assembly costs, administration costs etc. etc.

  While varying technology exists to catch cracks through 100% inspection machines, each technology has it’s limitations.

Eddy current coil – limited to large obvious defects

Eddy current probe – very effective but requires 100 degree rotation of every part resulting in slow sort speed, limitations on part shape, and higher cost.

Back lit camera – effective when the crack is open and light penetrates through crack.  Not as effective on defects that are more visible from one side.

Front light and conical mirrors – very effective for a specific size range of product but currently limited to smaller parts.

  Customer demands for “certified” shipments then place the fastener supplier into this difficult spot to try and find a solution.  Often when dealing with very high volumes, a dedicated solution can be found and afforded.  When volumes do not support a dedicated solution the interim solution is usually 100% visual sort.  This is labour intensive and while costly can be relatively effective but still carries the high risk of eventual human error.  Moving forward, they are now faced with the challenge of making a profit on a job that has added unforeseen cost.  Got to love the fastener industry!!  Never a dull moment!!!

  Many had predicted that as the baby boomers grew closer to retirement, there would be a shortage of young workers to take their place.  Today, many manufacturers, not just fastener manufacturers see this problem as their biggest challenge in the years to come.  It seems that the advent of the computer and the exodus of many manufacturing plants to places like China and Mexico have led to a youth that has no interest in manufacturing jobs, particularly ones that require getting your hands dirty!  The youth of today are heavily influenced in their career choices by their parents who don’t see a career in manufacturing as a viable option for their kids.

  Because the labour pool is so small, one needs to become very good at assessing potential.  I have tried several things over the years and although it is very early to tell, I believe that I have stumbled onto something that I think will help us moving forward.  We have had a written mechanical aptitude test that we have been playing with for years now but it hasn’t really proven to be the best indicator of potential.  I have toyed with producing a training video that I believe may be very helpful once you have found the best possible candidate for training.  Our problem in the past has been spending 3 months trying to train someone finally to decide that they just aren’t going to cut it.  We usually start the training with some whiteboard discussions on how a cold header works and a brief explanation of the machine and the process.  Then we would pick our best set-up guy and do shadowing for weeks on end.  This is not totally ineffective, again, if you have the right trainee.  But it does place a heavy strain on your best operator.  And, it requires that your best operator is a very good communicator (not always the case) and very good coach (not always the case) and a very good judge of potential (not always the case).

  I guess what I am getting at is that we need to find out very early on if we have a good candidate.   How do we do that?  Recently, out of necessity, I have had to put on my shop coat and return to the shop floor to train some new candidates.  My production manager had been screening resume’s and picked out three guys for me to train at the same time.  I arrived to work ready to go and before we even started we were down to two.  My production manager had called all three the night before to confirm their start time the next day when one of them decided that he couldn’t make it for 8:00 am “could he come in for 10:00?” he asked.  To which, wisely, my guy replied “I don’t think this is going to work out”.  And then there were two!

  We started out my usual way with a whiteboard description of the machine and process.  That was approximately 45 minutes.  We then went right to the floor.  We were fortunate in that we had a machine that had just finished a production run and needed change over.  We reviewed the workings of the machine observing the things we talked about in classroom and then went straight to work.  I alternated them through simple tasks as we cleaned the machine and removed tools cleaning them and placing them in the tool package box.  I would stop them after each task and offer suggestions on more efficient or effective procedures such as replacing the wrench back on the shadow board rather than laying it in front of you on the machine.  As time went on, I rapidly got a sense of each individual’s ability to retain information, process advice, and even their ability to handle a wrench.  By the end of the first day we had removed all of the tools of a two die three blow machine and put the new dies in and quill and cutter and aligned and re-aligned the quill and cutter 4 times.  I already new that I had one good candidate and one that was not so good.  I gave the not so good candidate one more day to show me that he was capable of retaining and understanding instructions and by the third day we were down to one good candidate.

  To me, the key exercise of the assessment was the alignment of the quill and cutter.  This exercise while simple showed me whether the candidate understood simple mechanical alignment which is the primary task of any cold header set-up whether it be alignment of the quill and cutter, the fingers to the die, or the punches to the dies.

My challenge now, and I will keep you posted on future blogs is to create a fixture that mimics the quill and cutter alignment exercise.  If we can use this for a mechanical test at interview, we can eliminate that 2-day window that it took me to rule out the initial candidate.  It also eliminates the time and money spent on orientation and setting the individual up as a new employee.

This is a repost from our original website!

I get it!  With the cost of personal injury lawsuits and class action lawsuits driving not only risk up but costs up it is incumbent upon the design engineer to ensure the safety of the product through design.  I believe the disconnect happens in the engineering schools where not enough is taught about the benefits of forming vs cutting and the limitations of both.  For instance, there is a distinct advantage to screw machining when volumes are low (depending on the size of the product, “low” could mean 5,000 pieces per year or 50,000 pieces per year.  This is primarily because there is little or no tooling cost involved with screw machining.  On the flip side of that screw machining is a much slower process which involves much higher amounts of scrap (waste or “mudda” in lean terminology) and results in “apples to apples comparison with cold heading” a weaker product.  Let’s break our comparison down to two separate discussions; piece price and design limitations.  For simplicity we can assume we are discussing cylindrical parts.

Piece price

-raw material – It is obvious that in order to create a part by screw machine your raw material must be at least as large as the largest diameter.  Different diameters are created by removing material through cutting.  The removed material, while still a part of the cost of the part, is scrapped.

          - to cold head a similar part you are moving material into cavities by the use of pressure and displacement.  Subsequently, the only scrap produced is created at set-up or when bad parts are made.  Bottom line, unless the product is a straight pin with no chamfers and no diameter changes, cold heading material costs will always be less expensive.

-set-up-  I must first admit that I have no experience with screw machines.  My experience is with NC (numerically controlled) secondary operation lathes.  For the most part with todays technology, set-up time is almost an insignificant cost to the screw machine process.  Often times the set-up is a program that can be initially created while the machine is running another part and tweaked at initial set-up then locked in for future runs.

            -  cold heading set-up is definitely a significant cost issue.  Machine manufactures have been designing servo-motor driven machine adjustments in order to reduce set-up time.  These advancements have definitely reduced the impact of set-up time  however, the increased cost of these features on the more basic machines results in increase charge out rates for both set-up and run times offsetting much of the reduction benefits from the machine design.  Bottom line, the higher cost of set-up for cold heading vs. screw machine is one of the factors that makes screw machining more economical for shorter runs.

-tooling-  Tools for screw machining include the holding device (a collet or chuck) which is not necessarily dedicated to the part but can be used across a series of parts the same diameter.  The cutters are typically holders with inserts that once again are not necessarily dedicated to one particular part.  Tools for screw machining, once again are usually an insignificant part of the cost of the product.

             -cold heading tools conversely can cost thousands of dollars and are often used only for the part they were designed for.  Tool costs will vary from a simple rivet at $1000 to a complex 5 die part at $30,000.  This is initial cost tooling.  Perishable tools add another chunk to the piece price as more complex parts involving more pressure invariably result in more perishable tools and more costs.  Bottom line, the higher cost of tooling for cold heading vs. screw machining is another factor the makes screw machining more economical for shorter runs. 

Runtime charge out-  let’s make an assumption here for simplicity.  We will assume that a cold heading shop and a screw machine shop have similar overhead costs and similar labor costs per hour.  Here is where it gets interesting!  A screw machine (depending on part complexity) will run a simple part at a speedy 20 parts per minute.  A cold header will run the same part at  400 parts per minute.  Even if you put 20 machines in a plant and were able to run them with one operator, you can’t compete on high volume as the shear cost of floor space would kill you!  If on the other hand you only need 2000 pieces. Now you run them in 5 minutes on the cold header but have to charge in $200 for set-up and $1000 for tooling for a piece cost of maybe 70 cents each.  The screw machine costs are $300 for material and $200 for set-up and run and it takes less than two hours at a piece cost of 25 cents each.

Design Limitations Cold Heading vs. Screw Machining

Tolerances- Truth is, screw machining can most likely hold tighter tolerances than most cold heading process.  This is particularly true with Suisse type screw machines.  The types of things that require these tighter tolerances typically are lower volume parts. From a cost perspective screw machining is the most cost effective way to produce these parts anyway.  Most engineers with little knowledge of cold forming are surprised to find out just how accurate and repeatable cold heading can be.

Typical cold heading/forming tolerances

Contained Diameter  +/- .04mm (.0015”)

Free form diameter  +/- .25mm (.010”)  typical used for large heads and washers.

Length       +/-.10mm (.004”)  up to 25mm ( 1”) long

                   +/-.20mm (.008”)  up to 50mm (2”) long

Geometry- This is where the debate can get complicated.  Let’s try and keep it simple by first assuming normal cold heading tolerances are acceptable for the application.  If this is the case, and the annual requirements are say 100,000 pcs or more it is wise as an engineer to design the part for cold forming. 

Undercuts-   these are typically created to allow strength at diameter changes without a radius restricting clearances.  When turning operations are  used these undercuts are usually located on the smaller diameter at the transition from small to large.  This is not feasible for cold forming.  In fact cold forming typically requires that diameters go from largest to smallest in either direction but not from large to small and then back to large again.  Without expensive collapsible tooling, this is virtually impossible.  If a “0” radius condition is required and the strength of a reasonable radius is also required the undercut can be designed into the surface above the smaller diameter at the transition from small to large.  This does not restrict the “release” of the material from the heading die.

In conclusion, as a design engineer, you need to consider the subtle differences between screw machining and cold forming and the subsequent cost implications of each when estimating eventual annual usage.  

Recently, I had a customer ask me for more detail regarding why their tooling cost got so much more expensive after a slight print change.  The following was my explanation:

When determining the process for a particular part we always have to balance the starting wire (stock) diameter with the upset ratio (the amount of material required for upsetting into the remaining cavity space).  For a single die process, the starting wire diameter is usually determined by the smallest diameter (in this case 9mm on the original rev).  We then do a calculation of volume to determine if we are able to upset 9mm wire into the cavity required for the finished product.  The volume calculation is expressed as the number of diameters that is required to complete the upsetting process.  For example, if we determined by volume calculation that 36 mm of 9mm diameter wire are required to fill the cavities then it is expressed  as an upset ratio of 4 diameters.  The rule of thumb limit for this type of process is 4.5 diameters so therefore we can produce in a single die process.  When the smallest diameter drops to 8mm but the volume required for the balance of the part does not change significantly this upset ratio rises drastically.  In the example, the upset ratio would go from 4 times to 5.7 times which is well above the acceptable maximum.  The reason for a maximum upset ratio lies in a principle called "material flow control".   In order to be successful jamming a bunch of material in a cavity it is critical that the flow of material is even, and consistent from the centerline out.  If it is not "buckling" occurs and we call this "losing control".  When you lose control of the material flow you start to get uneven fill and flashing.  By increasing the starting wire size and using the first die in a multi die process for extrusion down to the smallest diameter, we are then able to reduce the upset ratio to an acceptable level.  The multi-die machine takes longer to set-up, has more tooling and more expensive tooling (extrude dies take longer to make and require a lengthy polishing process) and may run slightly slower.  All of these factors contribute to piece price.  I hope I haven't confused you with this.  Please feel free to call or have anyone call for further clarification.

You might be surprised!  Is 100% laser inspected good enough for you?  It sure sounds good! Reality may be somewhat different from perception.  There really are a lot of different sorting techniques and machine styles resulting in a wide variation of risk based on how comprehensive the 100% sort really is.  After all 100% inspection is simply a risk management tool.  The degree of risk can be measured objectively by weighing a number of factors including part simplicity, degree of process control in the manufacturing process and most importantly sort criteria.

Part simplicity
The more features that a part (rivet, screw, stud, etc.) has, the higher the risk of out of specification dimensions or tool wear or breakage causing assembly affecting deformations.  Part simplicity may also be affected by tolerancing on the print.  A very simple straight pin is not a simple straight pin if it has a .002mm diameter tolerance over a 50mm length!  The same pin with a .2mm diameter tolerance is a much simpler part.  A less simple part may require a less simple sort or inspection process. 

Degree of process control
Process control can be done through in process inspection, spc documentation, and/or electronic process monitors.  If none of these steps are being taken, the risk of bad product getting into the good parts bin is fairly substantial.   That is why the majority of today's cold heading machines are equipped with short feed detection, and electronic process controllers.  Even the controllers themselves depending on the make and model can pick up the slightest defect if properly set.  Having a quality process monitor and short feed detection reduces risk and, depending on the part, may reduce the complexity required at sort/inspection.

Sort Criteria

Many customers demand 100% inspection or 0 defect.  The question that needs to be answered is "What are you sorting for?"  A single camera machine can sort dimensionally for numerous defects but can only see one side of the part.  A four camera machine can sort 360 degrees for numerous dimensions but if you did not set it up for length under the head it will still accept under and over length parts.

We have discussed some of the things that we know from a customer print might determine sort technique or machine required.  What about information that you don't always get from the customer print.  For instance, the degree of safety built in to the assembly (safety critical usually shows up on the final rev pre-production), the feed technique used in assembly or the fail safes built into the customer’s assembly.

Degree of safety built into the assembly
If the assembly holds a child safety restraint and the part requires heat treat to grade 10.9 a simple pin point laser inspection may not be suitable to ensure that every part is the right hardness.  Failure here leads to possible loss of human life and legal exposure is extreme.  Eddy current sorting should always be built into safety critical components requiring hardening.  It is critical to gather information such as this at the apqp (advance product quality planning) meeting prior to establishing the sort portion of the bill of materials.  Safety critical components require extra attention when determining the sort/inspection process.

Feed technique used in assembly
While this is a lesser point in determining sort technique it does merit consideration.  Product that is bowl fed vs. hand fed (at assembly) is typically more susceptible to foreign material issues.  1 foreign part in 1000 pcs on a bowl fed machine will cause frequent jam ups and downtime instantly raising red flags.  1 foreign part in 1000 pcs on a hand feed assembly is likely to be felt and discarded instantly without disruption to the line and less urgency for action. 

Fail safes built into customer assembly
Believe it or not your customer's management of risk in assembly reduces or increases your level of risk depending on how good of a job they do.  If they put into place fail safes that catch a potential defect from you, your exposure does not go beyond their plant.  If, however, your bad part gets to their customer the **** hits the fan big time!  While the process FMEA is designed to address this, it is rarely (probably never!) done in conjunction with the supplier as a team working to anticipate any potential problem.  You should take some time to see how your customer is working to reduce your risk.

Finally, while a determination of risk will help in selecting the right sort/inspection process, the simple geometry of the part will often limit the choices  by the nature of how well or stable the part is introduced to the sorting station. Very short parts with head diameters close to their overall length will not slide well down an inclined v-track machine without tumbling.  They must be run on glass rotating tables, indexing tables or magnetic belt sorters.  Similarly, studs that have a large washer in the middle often wobble down the v-tack causing instability and a high good part rejection rate.  These parts can't rest on their head for a glass table top and are typically unstable on a magnetic belt so they must run on an indexing type table machine.

Conclusions for fastener manufacturers
You have probably figured it out already if you are in this business!  One sorting machine isn't enough and no matter how many machines you do have there is always one or two or a series or parts that you can't do a proper sort on.  The tendency is to take the "that's the best we can do" approach.  Remember when you take that approach to measure the risk!  Find a good outside sorter with complimenting capabilities and pay the piper!  Once your volume of a specific type of parts warrants bringing the sort inside, it will be very easily cost justified and your capabilities will be broadened.

Conclusions for the OEM's and assembler's
I know that I have been talking for the most part about the fastener manufacturer managing risk but let's face it.  The more risk your fastener manufacturer takes on, the more risk you take on!  Get together with your fastener manufacturer (as a team not a grumpy dictator!) and do a part by part risk assessment.  If it seem’s  like your fastener manufacturer is trying to pull the wool over your eyes then you should be the grumpy dictator.  Real big job?  Then start with the parts that are most important to you and your customers.  If you sweep the dirt under the rug someone will come along eventually and pull the rug right out from under you!  

We have been producing mild steel rivets and shoulder rivets for over 30 years now and over those 30 years I have seen more than my fair share of stress relieving callouts that were simply wasting the customer’s money! I truly believe that there is not enough practical teaching on the topic of metallurgy and cold forming in post secondary engineering curriculum.  I am going to take a stab at simplifying the topic.  As I sit here thinking and typing and reading and erasing it's not that simple! Let's start by trying to understand what stress is, and why it can become a problem.  The word "cold forming" is used somewhat loosely as, while there is no heat introduced intentionally, "heat" is simply a by-product of moving metal.  The more we try and move the metal, the more heat we generate.  Heat changes the molecular structure of the steel and this in turn changes the mechanical properties of the steel. The heat generally concentrates where most of the metal moving is happening.  When we make a simple rivet, we usually start with stock that is quite close to the shank diameter.  (1/4" rivet = 1/4" stock)  When we squash the 1/4" rivet to make a head (hence the word "heading") the heat is generated primarily in the head area.  If enough heat is generated we end up with a head that has different mechanical properties from the shank.  This can be a serious problem as it weakens the transition from the head to the shank.  The result is rivet heads that pop off either during assembly or under load in the assembly.  The solution is to stress relieve the rivets by taking the whole rivet to the same high temperature and slowly cooling them down to get consistant mechanical properties
from head to shank.  The trouble is that stress relieving, normalizing and annealing all similar processes are not always required and add cost to the product.  Let's take a look at the underlying factors.  The amount of stress created by the cold forming process is a function of part geometry, material grade and processing condition and the cold forming process used to make the part.

Part geometry
As mentioned above, the more we move the metal, the more heat we generate, the more stress we create.  The larger head parts typically are the problematic parts requiring stress relieving.  The question is, how do you determine where the cutoff point is?  It is actually a calculation based on the head diameter, the head height and the stock size.  We call it the upset ratio.  When the upset ratio creeps above 80% we recommend stress relieving.  The "rule of thumb" is actually 85% but we use the 80 as a bit of a safety cushion.  Because I am not an engineer, I don't know the actual formula as I cheat with a forging calculator called the "Dr. Forge".  They were developed and sold by Asahi Sunac of Japan.  I am pretty sure there is an online version available but I still like my "Dr. Forge"!

Material Grade and Processing Condition
Since stress and heat go hand in hand, we need to understand how different steel grades react when forming pressures are applied to them.  As a general rule of thumb, the higher the carbon content in the steel, the more heat is generated from the forming process.  This is called work hardening.  This becomes more of a factor when high carbon and alloy steels are involved.  The state of the raw material is another factor.   Spherodised raw material greatly reduces heading stress vs. non spherodised raw material.  Today, with the exception of only a few specific jobs most cold headed product is made from spherodized raw material while years ago when some of these stress relieving callouts were made spherodized wire was not always the norm.

Forming Process Used

Some rivets and shoulder rivets can be made successfully with more than one heading process.  Shoulder rivets made with a two die process typically produce less head stress than single die produced rivets.  This is because we start with a larger diameter raw material in the case of a two die process (using an extrusion in the first die).  This results in a lower upset ratio and thus lower heading stress.

In conclusion, if you are an engineer designing a fastener for an assembly, feel free to reach out and ask an opinion on the need for stress relieving.  If you are a buyer and looking for cost savings, ask your supplier if eliminating stress relieving is an option.