The pains often associated with profiling may be reduced
greatly if certain guidelines are followed and if there
is a strong understanding of the variables that can
be encountered during the reflow process. By David
Suraski
The issue of reflow profiling
has been and continues to be a hot topic. Questions
ranging from Why Profile? to How To Profile? to How
Can I Better Profile? continue to proliferate. However,
the pains often associated with profiling can be reduced
greatly if certain guidelines are followed and if there
is a strong understanding of the variables that can
be encountered during the reflow process. This paper
shall discuss the appropriate guidelines and troubleshooting
methods for reflow profiling, and in particular shall
focus upon the advantages of newer-style reflow ovens
and the benefits of implementing the linear ramp-to-spike
profile.
Delta
() T
Many older-style ovens
were prone to heating different parts of an assembly at
varying rates, often depending upon the color and texture
of the parts and substrates being reflowed. Thus, some
areas of an assembly heated could reach much higher temperatures
than other areas. This variation of temperature is referred
to as the T
(Delta T) of an assembly. The result of a large T
is that, unless eliminated, certain areas of an assembly
may receive excessive heat, while other areas may receive
insufficient heat. This could cause a variety of solder
defects, ranging from solder balling to non-wetting to
damaged devices to voiding to charred residues.
Reflow
Ovens: New and Improved
Most newer-style reflow ovens, known as the forced
convection type, provide heat by blowing warmed air
on and around assemblies. One of the advantages of this
type of oven is that these are able to provide heat
to an assembly gradually and uniformly, exclusive of
part color or texture. Although the absorption of this
heat may vary slightly due to the varying thickness
and component population of an assembly, newer style
ovens generally provide heat in such a manner so that
the T
is not significant. In addition, the maximum temperature
and temperature rate of a given profile can be controlled
strictly with these ovens. That is, these ovens offer
greater zone to zone stability, which provides the operators
of these ovens with a more controlled reflow process.
Soaking:
Why and When
The sole intention of the soak zone is to reduce/eliminate
a large T.
The intended result of the soak is to bring the temperature
of all parts of the assembly to equilibrium before the
assembly reaches the reflow temperature of the solder
so that all parts of the assembly will be reflowed simultaneously.
However, because assemblies reflowed in a more efficient
oven often are at a virtual equilibrium of temperature
throughout the profile, the soak zone can become an
unnecessary step. Therefore, because the soak zone is
unneeded, the profile may be altered into a linear (RTS)
profile.
Fig 3a. Solder beads
next to capacitors
It is important to note that the soak zone generally
is not needed to activate the flux chemistry of solder
paste. This is a common misconception in the industry,
and it bears correcting. Most solder paste chemistries
used today will demonstrate adequate wetting activity
when processed in a linear RTS profile. In fact, as
is discussed later in this paper, using the RTS profile
generally will improve wetting.
Ramp-Soak-Spike
(1)
The RSS profile may be used with RMA or no clean chemistries,
but generally is not recommended for use with water
soluble chemistries, as the soak zone of the RSS profile
may break down the activators of the paste prematurely
and result in less-than-adequate wetting. As stated
above, the sole purpose of using the RSS profile is
to eliminate or reduce a large T.
(1) All profiles pertain to the Sn63/Pb37
alloy, which has a eutectic melting point of 183°C,
and are just an example.
Fig 1. A typical ramp-soak-spike
reflow profile
As seen in figure 1, the RSS profile begins with a
steep ramp up to approximately 150°C within a target
time of 90 seconds at a maximum rate of rise of 2-3°C/second.
Following the ramp area, the profile soaks the PCB assembly
between 150-170°C within a target time of 90 seconds;
the assembly should achieve thermal equilibrium by the
completion of the soak zone. After the soak, the assembly
will enter the spike area, where the assembly will be
reflowed above 183°C for a target time of 60 seconds,
plus/minus 15 seconds
The entire profile should last between 3½ to
4 minutes from 45°C to a peak temperature of 215°C
± 5°C. The cool down rate of the profile
should be controlled within 4°C/second. In general,
a faster cool down rate will result in a finer grain
structure and a stronger and shinier solder joint. However,
exceeding 4°C/second could result in thermal shock
to the assembly.
Ramp-to-Spike
The RTS profile may be used with any chemistry or alloy,
and is preferred for use with water soluble solder pastes
and difficult-to-solder alloys and parts. If for any
reason a large T
exists on the assembly, such as with processes using
fixturization or those using an inefficient reflow oven,
the RTS may not be the appropriate choice of profile.
Fig 2. A typical ramp-to-spike
reflow profile
The RTS profile has several advantages over the RSS
profile. The RTS profile generally will result in brighter
and shinier joints and fewer problems concerning solderability,
since the solder paste reflowed in a RTS profile will
still contain its flux vehicle throughout the entirety
of the preheat stage. This also will promote better
wetting, and thus the RTS should be used on difficult-to-wet
alloys and parts. Because the ramp rate of the RTS profile
is so controlled, there is much less concern of solder
defects or thermal shock to the assembly resulting from
too high a ramp rate. In addition, the RTS profile is
more economical due to the reduced heating energy used
in the first half of the oven. Furthermore, troubleshooting
the RTS tends to be a relatively simple process, and
operators with experience troubleshooting the RSS profile
should have no difficulty in adjusting the RTS profile
in order to achieve optimum profiling results.
Setting
Up the RTS Profile
As seen in figure 2, the RTS profile is simply a gradual
linear ramp from ambient to peak temperature. The ramp
zone of the RTS profile serves as the preheat zone for
the assembly, wherein the flux is activated, the volatiles
are driven off, the assembly is prepared for reflow,
and thermal shock is prevented. The typical ramp rate
for a RSS profile is 0.6-1.8°C/second. The first
90 seconds of the ramp should be kept as linear as possible.
A simple rule of thumb for the ramp rate of the RTS
profile is that 2/3 of the profile should be below 150°C.
After this temperature, the activation systems of most
solder pastes begin to break down quickly. Therefore,
keeping the front end of the profile cool will preserve
the life of the activator longer, resulting in better
wetting and shinier solder joints.
The spike zone of the RTS profile is the stage where
the assembly reaches the reflow temperature of the solder.
After reaching 150°C, the peak temperature should
be reached as quickly as possible. The peak temperature
should be controlled at 215°C ± 5°C,
with time above liquidus (183°C) at 60 seconds ±
15 seconds. This time above liquidus will reduce flux
entrapment and voiding and will increase pull strength.
As with the RSS profile, the RTS profile length should
be a maximum of 3½ - 4 minutes from ambient to
peak temperature, and the cool down rate should be controlled
within 4°C/second.
Certain board coatings may require an increase in the
profile peak temperature. If soldering to gold over
nickel coated pads, a peak temperature of at least 220°C
should be met; this will prevent post-reflow thermal
reliability issues, as tin and gold form a secondary
eutectic at 217°C. If soldering to pads coated with
an organic surface protectant (OSP), peak temperatures
up to 225°C may be required to penetrate the coating
completely. These profile peak temperature adjustments
are necessary if using either the RTS or RSS profile.
Troubleshooting
the RTS Profile
In effect, the same rules are applied
to troubleshooting both the RSS and the RTS profiles:
Adjust the temperature and/or time at temperature of
the profile as needed to achieve optimum results. Often
this requires trial and error, with slight increases
and/or decreases of temperature made and then the results
of these changes being observed. Following is a summary
of common reflow problems that may be encountered with
the RTS reflow profile and the remedies by which to
resolve them.
Solder
Balling
Solder Balling is recognized by numerous tiny solder
balls trapped along the peripheral edge of the flux
residue after reflow. In a RTS profile, this most often
is the result of too slow a ramp rate, wherein metal
oxidation occurs as a result of the flux vehicle being
burned off far ahead of reflow, resulting in solder
balling. This problem normally can be corrected by slightly
increasing the ramp rate of the profile. Solder balling
may also be a result of too rapid a ramp rate. This
however, is unlikely with the RTS profile, due to the
relatively slow and steady ramp.
Solder
Beading (Satellites)
Often confused with solder balling, solder beading
is a defect recognized by one or a few larger balls,
generally located around chip caps and resistors.
Fig 3b. Solder beads
next to resistors
Although this normally is the result of an excessive
paste deposit during printing, it sometimes can
be resolved with a profile adjustment. As with
solder balling, solder beading which occurs during
the RTS profile is normally a result of too
slow a ramp rate.
In this case, the slow ramp rate causes capillary
action to draw the unreflowed paste away from
the pad on which it was deposited to a place under
the component. The paste reflows there forming
a bead of solder that comes out to the side of
the component.
As with balling, the solution to solder beading
occurring during the RTS profile is to raise the
ramp rate until the problem is resolved.
Poor
Wetting
Poor wetting often is the result of time and temperature
ratios.
Fig 4. Poor wetting
The activators contained in solder pastes consist
of organic acids, which degrade with time and
temperature. If a profile is too long, the wetting
of the joint can be compromised. Because paste
activators normally survive longer with the RTS
profile, poor wetting with this profile is less
likely than with RSS. If poor wetting is experienced
with the RTS profile steps should be taken to
ensure that the first 2/3 of the profile occur
below 150°C. This will extend the paste activators
life and will result in improved wetting.
Solder
Deficients
Solder deficients are often the result of uneven heating
or an excessive heating ramp, which causes component
leads to get too hot, which will result in the solder
wicking up the leads. The leads will appear thicker
after the profile and the pads will have an insufficient
amount of solder on them. Reducing the ramp rate or
otherwise ensuring the even heating of the assembly
will help to prevent this defect.
Tombstoning
Fig 5. tombstone
Tombstoning normally is the result of non-equal
wetting forces, which causes a component to stand
on end after reflow.
In general, the slower the heating, and more stable
a board is, the less this will occur.
Reducing the ramp rate as it passes through 183°C
will help to remedy this defect.
Voiding
Voiding is a defect found commonly with an
X-ray or cross-section inspection of a solder
joint. Voiding is recognized by the appearance
of tiny "bubbles" in the joint. These may be
air or flux entrapment. Voiding is generally
caused by one of three profile errors: insufficient
peak temperature, insufficient time at temperature,
or excess temperature during the ramp stage.
As the ramp rate of the RTS profile is so controlled,
voiding normally is the result of the first
and/or second error, which generally results
in non-volatized flux entrapment in the joint.
To correct voiding in this case, a profile should
be taken at the point where the voiding is occurring
and adjusted appropriately until the problem
is resolved.
Fig 6. Voiding
Dull
& Grainy Joints
A relatively common reflow defect is dull and grainy
joints. This may be a defect of only aesthetics, or
it could be the sign of a weak joint. To correct this
defect in the RTS profile, the two zones before the
spike zone each should be reduced by 5°C; the peak
temperature then should be raised by 5°C. If this
is not successful, then the temperature should continue
to be adjusted slightly in this manner until the desired
results are achieved. These adjustments will prolong
the life of the activator of the paste, thus reducing
the paste's exposure to oxidation and improving the
wetting ability of the paste.
Charred
Residue
Charred residues, although not necessarily a defect
of functionality, may be experienced using the RTS profile.
To correct this, the temperature and/or time of the
spike zone may have to be reduced. If a recommended
RTS profile is being followed, this normally is a simple
matter of a slight temperature (5°C) decrease.
Conclusion
The RTS profile is not a cure-all for every reflow
profile related soldering issue. Nor can the RTS profile
be used with all ovens or all assemblies. However, the
implementation of the RTS profile can reduce energy
costs, increase efficiency, reduce solder defects, improve
wetting, and simplify the reflow process in general.
This is not to state that the RSS profile has become
obsolete, or that the RTS profile never can be used
with an older-style oven. However, engineers should
be cognizant of the fact that there may be a better
reflow profile style available for the processes.
References
DAVID SURASKI,
may be contacted at AIM, 25 Kenney Drive, Cranston,
RI 02920; (401) 463-5605.
E-mail: dsuraski@aimsolder.com
(*) All profiles
pertain to the Sn63/Pb37 alloy, which has a eutectic
melting point of 183°C. and are just an example