Solutions to Brazing Problems

(This is an overview article largely based on customer reports.  Most of this is covered in much greater depth with photographs in other articles and in our book Braze Failure Analysis.) 

Occasionally tips break or come off the saw.  There can be three reasons for this.  First, the bond between the tungsten carbide and the brazing alloy fails.  Second, the tip is hit hard enough to shatter the tungsten carbide or rip the steel.  Third, the bond fails between the brazing alloy and the steel.  There are many articles on other reasons why Brazing Problems may occur, and how to resolve them in our How to Braze section. 

Most common causes of braze failure in order 

1. Improper Plate or Notch Cleaning       
2. Surface Condition of the Tips
3. Wrong Braze Alloy
4. Improper Fluxing
5. Improper Braze Joint Thickness
6. Improper Brazing Temperature
7. Gas Entrapment
8. Overheated Tungsten Carbide
9. A Combination of Things

Other things that have caused problems in no particular order:

1. Cutting Frozen Lumber
2. Cutting Mixed Lumber
3. Different Saw Plate
4. Plate Resonance
5. Torch Problems
6. Cold Anvil Problems
   
7. Too Hot of a Torch
8. Poor Pretinning
9. Chill Lines
10. Plated Tips
11. Drafts
12. Bad Gas
13. Over Testing
14. Solder Burning During Brazing
15. Brazers Were Having a Bad Day
16. Mysterious Saw Tip Performance 

 

Wrong Braze alloy

Wrong braze alloy is the major reason for breakage and loss

In May of 1997 the number one reason for failure is using the wrong braze alloy.   See our article on Brazing Alloy to see the differences and find the right Braze Alloy for your application.  

The most common alloy used to be an AWS Bag- 3 alloy with 50% silver and 16% Cadmium.  The federal government started restricitng the use of Cadmium and leveling large fines.  Two other alloys were used for several years.  They worked but not very well  These were Bag-24 and BAG-7. 

In 1996 we found an alloy descrbed as AWS  Bag-22.  It gives excellent results without Cadmium. If you are having braze problems and are not using this alloy then the simplest thing to do is try this alloy.  We call this alloy “High Impact” alloy.

AWS	Silver	Copper	Zinc	Nickel	Cadmium	Manganese	Tin	Melt pt.	Flow pt.
BAg-3	50	15.5	15.5	3	16			1170 F	1270 F
Bag-24	50	20	28	2				1220 F	1305 F
Bag-7	56	22	17				5	1145 F	1205 F
BAg-22	49	16	23	4.5		7.5		1260 F	1300 F

 

Improper Fluxing

If a change in procedure is going to cause braze problems it will probably be due to a change in flux or flux procedures.  

Black Flux vs. White Flux

We have seen dramatic changes in braze performance just because of a simple change from Black flux to white flux.  Black flux is identical to white flux except that it has extra Boron added.  Flux absorbs oxygen from the air and keeps oxygen out of the joint during brazing.  Because of the extra Boron, black flux will give greater protection during heating.  There is more information on the different types of Flux in our Brazing Flux article. 

The flux people say it shouldn’t happen but we have seen a brazer braze tips on half a saw with black flux and the same tips on the other half of the saw with white flux.  Then the tips were hit with a hammer.  The black flux tips held with absolutely no problems.  The white flux tips broke or came out or both.  

Switching from Black Flux to White Flux can cause tip loss.  The white flux is not as forgiving and does not have as great a capacity to absorb oxygen as the Black flux and this can cause problems. 

Watery Flux

Black flux is water based.  The flux particles are in suspension.  If you get a five-gallon pail of flux it will often be water on top.  Then there is a layer of thick flux.  Then, at the bottom, you have flux cake.  This is just like mud in a farm pond.  It settles out and gets thicker at the bottom.  It is easy to just use the stuff that is soft and near the top.  However this isn’t really true flux.  It has settled out and there is still a thicker layer on the bottom.  You need to mix the flux up until it is an even consistency and then use it. 

Dried or old flux

As flux dries out it forms a crust.  Once again, just like mud from a farm pond.  As it dries it changes. It is very hard to stir the dried flux back into the water.  Even then it is probably not the same material exactly. 

If you use an open flux pot you should keep as little flux in it as possible.  You should keep it stirred. You should change it regularly. 

Improper Braze Joint Thickness

Too thin a braze joint.

We have seen instances where the braze joint was so thin that the tips were essentially resting just against the steel.  It was the fillets that were holding the tip on rather than the proper bond between steel and tungsten carbide. 

If you can gather knocked off tips you can measure the thickness of the tip where it rested against the saw and compare that with the thickness of the tip.  Often you can see a gray surface in the middle of the tip with little ridges of solder running from side to side.  This means that all the solder was forced out of the joint.  The only solder that was in the joint was the solder in the grooves created by grinding or gumming the pocket.  This can be because the brazer is pushing the tip into the pocket too hard instead of sort of letting it get sucked into the pocket.  

Uneven solder 

There should be the same depth of silver the whole length of the impression or pocket created by the saw body.  Sometimes the tip can get “cocked” so that there is too much solder between the tungsten carbide and the saw plate at the outside of the saw and there will be no solder at all between the tungsten carbide and the steel at the bottom of the tip or notch. 

The silver solder should have a thickness of .003” to .005”.  When the tip comes off the saw there will be a little pocket made in the silver solder by the saw plate.  The solder thickness between that pocket and the top of the tip is what is important.  You can sometimes determine this by measuring the thickness of the tip where there is no solder and then measuring the thickness of the tip with the solder at the bottom of this pocket.          

We use a standard .010” of material which allows for 60% to 70% of the material to be used to create fillets and still provide enough material to create a proper braze joint thickness.  This amount of solder can be increased or decreased as the customer requires. 

Uneven solder - Poor wiping motion

The tip should be gently wiped in and out of the slot to allow the solder to be distributed evenly.  This eliminates any gas pockets and produces a clearly superior joint.  If the tip has been pre-tinned with wire then this motion will distribute solder over the radius and over the whole curve of the tip.  If the tip that came loose doesn’t have this solder distribution it may be because of poor wiping. 

Tip placement. 

If the strength test is run before grinding then tips that are placed with the plate in the center will outperform those where the plate is severely off to one side. 

Improper Cleaning

Plate cleanliness.  Oil and grease left in the notch will contaminate the braze joint and can greatly weaken the bond.  Anything left in the notch will contaminate the braze alloy.   Because saw plate is often oiled to keep it from rusting this can be a problem.   There are different kinds of oils and greases.  A cleaner that removes one kind may not remove all the others. 

The notch is usually ground or gummed.  Sometimes it is also sandblasted.  If the notch is not cleaned properly  prior to brazing there will be sand or diamond particle left in the notch.  If the tip comes out you will see a sparkly or glittery appearance. 

Improper Brazing Temperature

Underheating the solder

The tendency is to underheat the braze joint to protect the steel plate.   Underheating can result in cold joints.  If you are getting good shoulders or fillets and some feathering then you are probably not underheating.         

Overheating the solder

Cadmium and Zinc both have low boiling points.  Zinc boils at 1664 F and Cadmium boils at 1409.  Silver solder with Cadmium has the 1170 to 1270 range.  That is not too far below the 1409F boiling point.  Cadmium free silver solder has the range of 1250 - 1305F which is not too far below the 1664 of zinc although it is somewhat below it.   if the solder is overheated these metals will fume and boil which changes the chemistry of the braze alloy.  

If the tip comes off you may see gas bubbles.  You may also see these by examining the side of the braze joints. 

You should probably use a magnifying glass to see the gas bubbles.  Maybe a three to five power (diopter).  These will show up looking like balloons with the top sliced off.  These mean that the solder got hot enough that some of the metal actually started to boil out.  Some times you will see gas bubbles on just one side.  This is evidence that the part was heated too fast and the material boiled out of just one side.  

Colors in the Solder

The silver solder should be kind of a rich gold color.  If there is any kind of another color it is a sign that something is wrong.  The different colors all mean something.  Blue-green is a sign that the tungsten carbide is being heated enough to bring the Cobalt out.  Dark pink comes from overheating both the tungsten carbide and the flux.  Copper from the solder and Carbon from the tungsten carbide makes Copper Carbonate which is either yellow or dark green  copper and Tungsten is light green.  Severely burnt flux can yield an orange solid.  In any case if there is any evidence of another color at all it means the solder was not treated properly.  

Gas Entrapment 

Wrong flux

As flux heats it creates gas.  This gas needs to escape out of the braze joint.  The wrong flux can create a shell which traps the gas in the braze joint.  This creates little bubbles through the joint that seriously weaken the joint.  When a tip is knocked off you can see little round half bubbles. 

Underheating

You can also get this same sort of problem by underheating the braze joint.   Generally the proper heat with a little wiggle or short slide of the tungsten carbide tip by the brazer when the parts are at full temperature will very successfully get the flux gas out.   The secret here is to be able to feel the condition of the alloy at temperature. 

Overheating

Silver braze alloys have Zinc in them.   Some of them also have Cadmium although this is on the way out.  Both Cadmium and Zinc create fumes before they boil.  This is about the same as water steaming for a while before it boils.  If the braze alloy is overheated these fumes can be trapped inside the solder joint and seriously weaken the joint. 

Surface Condition of the Tip

Some treatments involve a chemical preparation of the tip to clean it and then the plating of a coating or layer on tip of the cleaned tip.  This layer is designed to promote wetting.  The problem can occur when the underlying cleaning doesn’t take and the coating is put on top.  The braze alloy sticks to the coating which does not stick to the tip.   If the tip has had a treatment applied to it can affect the bond strength for either good or bad. 

The ECP process is a process we invented.  It leaves the surface entirely free of oils, greases, free carbon and other contaminants.   It also leaves the surface etched to an even distance of nine microns, which gives excellent wetting and bonding characteristics.  

There is another process based on a high temperature, salt bath treatment that removes individual tungsten carbide grains and leaves the surface cobalt enhanced.  This process has been around thirty years and has a reputation for working well. 

There are three possible approaches.  1.  The tungsten carbide may be prepared in such a manner that there is a very rough surface and the solder will flow into the surface holes and bond physically.  2.  The tungsten carbide may be prepared chemically so that there are electrons available in the tungsten and the silver solder forms silver tungstate during the pretinning.  3.  The tungsten carbide may be activated and then plated with a presolder substance so that the tungsten carbide surface is protected.  In this case the braze alloy can react with the top layer of the plating or the plating will dissolve into the braze alloy and the braze alloy will form a bond directly with the solder.

Overheated Tungsten Carbide

If the heat is applied entirely through the tungsten carbide it can cause overheating of the tungsten carbide.  This is sometimes done to protect the steel plate from overheating.  The plate, braze alloy and tungsten carbide should be brought up to temperature together and then the torch should be drawn away from the joint over the tungsten carbide.  A slow draw has proven to be best.  A slow draw is one where the tungsten carbide is heated for an extra quarter to half second or so.  This apparently gives the tungsten carbide and the whole joint a little more time to adjust to each other. 

Cold Anvil Problems

There can be problems with cold anvils.  This occurs where a large anvil is used to support the whole saw.  The saw is then raised just above the anvil to allow for side clearance.  The plate does not actually touch the anvil. It rests on three or four little shims.  However a cold anvil can make a difference in the brazing.  The cold anvil sucks the heat up.  It can mean that the saw brazer will create beautiful joints all the way around the saw on the top.  The other side will be cold joints. The down side or under side will not have the flow.  Essentially a good joint is created about half way through the tip then the joint turns cold.  

The temperature of the room and particularly the brazing fixture can influence the quality of the brazing.  Tips brazed first thing in the morning, tips brazed Monday mornings and tips brazed after a sudden (colder) temperature change can cause more problems with tip loss than tips brazed later in the day on warmer anvils or fixtures.  The fixture or anvil works as heat sink.  The colder it is the more heat it can and will absorb.  Ice water will take heat out of your hand a lot faster than warm water although both are less than body temperature. 

It is possible to have a braze joint that is perfectly good at the top and that is a “cold joint on the bottom.  You can see this easily by comparing the difference in the fillets. 

This is really critical with torch brazed joints because the bottom of the joint flows differently than the top of the joint does anyway.  The bottom of the joint away from the flame is always colder than the top of the joint where the flame is applied.  

A recommended practice is to make sure the anvil or fixture is at the proper temperature for brazing before starting to braze. 

A Combination of Things

Sometimes a brazing problem cannot be traced to a single cause.  In this case the best thing to do is to examine the whole operation.  You should make sure everything is working right and being done correctly.  Often this will cause the problems to disappear without identifying a specific cause.  

Brazing is a complex process.  Most problems are a bit of this and a bit of that.  You can make a 5 % difference five times and you get a 25% improvement.  It is a lot easier to make lots of little, simple, easy changes than it is to try to make one, big change that will solve all your problems. 

Here is a partial list of things to check if you are having trouble with a brazing operation. 

Cleanliness of the plate

Cleanliness of the tip

Flux clean and stirred 

Flux on the sides of the plate.

Lots of flux inside the joint

Acetylene

Oxygen

Regulators

Hoses

Handles

Handle controls

Tips

Tightness of joints

Temperature of anvils

Temperature of the shop

New furnaces

New fans

New locations

New doors

New brazer

Proper pretinning

Right kind of solder

Temperature of the tips

Kind of flux

Condition of flux 

Amount of flux used

Flux on plate in notch

Flux in contact and protecting the tips

Brazing temperature 

Brazing time

Color of braze joint

Sound of braze joint 

Feel of braze joint

Where is heat being applied?

How is heat being applied?

Does everybody have the same problems?

Does the problem occur at a certain time of day?

Does the problem occur with a certain kind of plate?

Does the tip manufacturer have the same problem?

Does anyone else have the problem? 

What tips failed? 

 

You may never know the answer.  Quite often the problem goes away while you are looking for it.  When you tune up a car you can’t point to one thing and say this made the car run better and everything else was a waste of time and money.   

Over the years it seems that companies that do the best are those that work the hardest and the smartest.  They are companies that look at their own shop first when they have a problem.  They are companies that have a definite program to identify and solve problems.  Once they solve a problem they make sure it doesn’t happen again.  They concentrate on every part of their operation and truly work to make every part as good as it can be. 

Cutting Frozen Lumber

Frozen lumber can cause problems if you do not make allowances for it.  When the water in a log freezes into ice there are three significant changes.  Ice is a lot harder to cut.  Ice is also a lot harder to push out of the way than water is.  Finally the ice and the wood fibers act to create separate composite material. 

Water has different viscosities at different temperatures.  Technically, viscosity is a measurement of the resistance to change of form.  This means that water wants to move slower as it gets colder just as oil does although not nearly as much because it is a different material. 

Viscosity is measured in units called “poises”.  A poise is dyne seconds per centimeter squared, which is sort of like pounds per square inch. 

Example 1

At 16 degrees F water has a viscosity of 2.549 centipoises.  At 32 degrees F water has a viscosity of 1.787 and at 68 degrees F water has a viscosity of 1.002.  At -15 degrees F it takes 2 1/2 times as much energy to push water out of the way.

Example 2

We took an eight-ounce aluminum weight.  It sinks in water.  It does not sink in ice. We used a little 3/4-inch thick ice cube and tried to get through it by dropping a weight on it.  This would be sort of like trying to cut through ice.  We had to drop the weight from nine inches to chip the ice and we had to drop the weight from 12 inches to crush the ice. 

It gets worse.  Logs are not all water.  They are also wood fiber.  When the log freezes the water sets up like concrete and the wood fiber acts like rebar. 

Cutting Mixed Lumber

A lot of breakage can result from running warm and cold lumber together.  This is similar to running green and dry lumber together in the summer.  If you are set up for the green lumber to run at a maximum rate then mixing in a hard, dry board can really impact the saw.  The same thing can happen with cold weather.  Lumber out of the wind can be warmer than lumber exposed to the wind and can cut differently. 

Different Saw Plate

No one is really sure why different plate brazes differently but it has been reported several times.  

Different saw plate from different manufacturers can braze differently.  Tips from the same box will stay on one plate beautifully and be easy to break off a different plate.  This is particularly true with people who “gum and go”.  That means that they just gum out the notches and then braze the parts in.  This is less likely to happen if the shop follows the steps some shops do.  

These steps are a follows:

1. Gum the notch
2. Soak the plate in the Oakite or Sodium Hydroxide bath to clean it
3. Wire brush the notch
4. Wipe the notch with a clean rag
5. Make sure you put flux in the notch as well as on the sides of the plate.  

 

Plate Resonance

There can be a problem with different plates.  It is not widely researched but apparently the fact that plates are steel means that they all have natural resonance.  Plates are tensioned as part of making good saws.  The tension can establish a resonance in the plate so that the plate vibrates a certain amount in different circumstances.  The amount of resonance or vibration under stress can help contribute to tip loss and breakage.  Once again, this appears to be true but we do not have hard data on just how important it is. 

I am not really sure that this is a problem with resonance but if we treat it that way then the problem gets solved. 

Using the correct braze alloy can usually eliminate this problem.  Braze alloys with Cadmium used to be used before the laws on Cadmium use got so tight.  Now braze alloys with Manganese give excellent results.  

Torch Problems

We have seen situations where the torch caused braze joint failure.  An oxygen rich flame can cause oxidation in the joint.  A bad valve can fail intermittently and will not be readily apparent however it can change the composition of the gas. 

Too Hot of a Torch

A torch that is too hot can overheat the tungsten carbide and leave the plate relatively cool.  The tungsten carbide and the plate both grow and shrink as they are heated and cooled.  The plate grows about 2 to 3 times as much as the tungsten carbide.  At about 1250 degrees F the solder gets hard but the plate and the tungsten carbide keep cooling down to room temperature.  This can cause the tungsten carbide to be stressed.  As the plate shrinks it wants to bend the tungsten carbide tip like a string bends a bow.  This stress can make the tungsten carbide much more susceptible to bowing. 

Poor Pretin

Good pretinning means that the surface is cleaned and etched.  Then solder and flux are put on the surface and melted.  The solder forms both a physical bond and a chemical bond.   We do this in several different steps.  We have a special processing line to clean and etch the parts.  We have special ovens with digital feedback controls to heat the parts exactly enough without overheating them. 

You can also pretin with just a torch.  The problem with that is that you can very easily overheat the parts.  When the solder is overheated even slightly the chemical composition starts to change dramatically.  

Good brazers can detect a difference of a couple per cent in braze alloy composition.  They can tell the difference in our pretinning and our competitors because of the way it flows and generally handles. 

Torch pretinning can leave the parts with color changes.  You can see a pretty rose red color.  You can also get some brown and some blue - green color.  This depends on the torch flame composition and how much of what metal got burned or oxidized.  You can also get burnt solder that does not have a detectable color change or a detectable surface condition.  Tips pretinned with a torch or a ceramic kiln can look good but still cause problems.   It is a lot cheaper to clean the tips up after doing them wrong than it is to do them right in the first place.

Chill Lines

To get a good braze; the tungsten carbide, the solder and the steel plate all have to be at the same temperature.  If the solder is too cold it will not melt fully and flow fully.  If either the steel or the tungsten carbide is too cold then the solder will not bond to it. 

When the saw plate is heated the crystalline structure is changed.  A chill line is created.  If this chill line is too far back then the shoulder can rip off.  The natural tendency is to keep the steel as cool as possible. This can mean that the steel is too cool.  If the steel is too cool, then the composite joint of the silver solder, steel and tungsten carbide is not formed properly. 

Ideally all the components should reach the same temperature at the same time.  NASA recommends a temperature about 95 degrees above the flow point of silver solder.  50% Silver solder with Cadmium has a range of 1170 - 1270 F.   50% silver solder without Cadmium has a range of 1250 - 1305F. 

Plated Tips

Some tips have a plating or coating on them to make them braze better.  This coating always makes the solder flow well over the tip.  The problem can come from the fact that this coating is put on over the tungsten carbide.  If the coating doesn’t adhere to the tungsten carbide then the coating can come off and the tip will come off the saw.  You can see this on the knocked off tips.  

The tungsten carbide saw tip is activated and then plated with copper and nickel. The copper protects the surface and gives a nickel surface that will wet very readily.  Theoretically the nickel and underlying copper will melt into the braze alloy and the new braze alloy containing the additional copper/ nickel will be bonded well to the tungsten carbide.  This does take a certain amount of time at temperature for the nickel and copper to be absorbed and the intermediate silver tungstate to form.  

Drafts

A draft can cause a brazing problem.  It can be from a fume collector or the furnace coming on.  It can be from a door opening or from a new fan in the summer.  The operator may not notice it because the fume collector is right at the work surface.  He may even like it because it makes him more comfortable. 

Bad Gas

This is highly unlikely but we have had reports of it happening.   This can be true with either oxygen or acetylene.  This can also be problem with shops using natural gas from the pipeline to braze. The pressure can drop and the torch ratio and temperature both change. 

Over Testing

It is common in better brazing operations to test the braze quality by hitting the tip with an oak bat or a plastic hammer.  We discussed different kinds of strengths in “How Solder Works”.  Basically this is a pretty good test because it is easy, cheap, simple to understand and clearly superior to no test at all. 

A common form of testing is to hit the saw from the top with an oak stick.  We use a 16” oak 2”x2” stick.  Some use bigger.  A good man beats on the saw harder and harder until he is beating as hard as he can and if the tips stay in then the saw is good. 

Solder Burning During Brazing

You can also burn the surface of the solder with the torch while you are brazing it onto the saw.  The braze alloy on the surface of the sawtip needs to be protected with flux the whole time it is being heated.  Generally a good brazer will put enough flux in the notch to do this.  You do not need to flux the tip before brazing.  You can just push it into the flux that is in the notch and then start heating. 

Brazers Having a Bad Day

Sometimes a brazer has a bad day and the tips either fall off or break.   A Brazer can be doing a beautiful job in the morning and a poor job after lunch.  A really excellent brazer with years of experience can start having problems.  He will step back and let his assistant try and the assistant will do a beautiful with the same, identical set up.  The situation can also happen in reverse.

Mysterious Saw Tip Performance

There are always those exeptions to the rule and sometimes, despite having everything right, tips just don't work and nobody knows why.