Reflow Soak Zone Control in Solder Profiles: 6 Rules

The soak zone is the part of a reflow profile between the initial ramp and the ramp to peak, where the board temperature is held roughly level for a period. Its purpose is to let the whole assembly equalise in temperature and to let the flux activate before the solder melts.

It is also the part of the profile most often shortened when a line needs more throughput, because it looks like time that does no work. Removing it moves the defect from one place to another: the paste that used to slump into a bridge instead spatters, and the joints that used to form cleanly instead show voids or cold fillets.

What the Soak Zone Does

Three jobs are done at the same time during the soak. The flux wets the surfaces, the volatile part of the paste is driven off, and the board and its components reach a common temperature before the solder becomes liquid.

Each of those jobs has a time constant. Volatiles need time to leave the paste body, and if the ramp is too fast they leave violently and carry paste with them, which produces solder balls and small voids at the joint.

Flux activation is a chemical reaction and it also needs time at temperature. A profile that runs straight to peak may activate only the outer layer of the flux and leave the rest to burn off, which leaves a residue that is still active after the joint has formed.

Equalisation is the third job and the one that protects the components. A large thermal mass next to a small one reaches temperature at a different rate, and the soak is what allows the two to converge before the solder melts.

Ramp Rate Into the Soak

The ramp into the soak is usually controlled between one and three degrees per second, and the limit is set by the paste and by the components rather than by the oven. Too slow a ramp costs throughput; too fast a ramp causes spatter, tombstoning and thermal shock to the smaller parts.

The ramp should be measured on the board, not taken from the oven display. A thermocouple attached to a pad near a large component shows a different slope from one attached to the leading edge of the panel, and the profile that matters is the slowest of those.

Where the assembly carries a large connector or a metal shield, the ramp into the soak is often reduced specifically for that part, because its thermal lag is what would otherwise set the difference across the board.

Dwell and Flux Activation

The soak dwell is usually expressed as the time spent between two temperatures, for example from 150 to 180 degrees, rather than as a fixed duration. Expressing it that way ties it to the chemistry, because the flux does its work in that band.

Too short a dwell leaves unactivated flux and un-evaporated solvent; too long a dwell oxidises the surfaces and consumes the activator before the solder melts. The window between those failures is what the paste supplier characterises, and the profile has to sit inside it. The activation temperature notes describe how that band is established.

A useful check is the residue left after reflow. A pale, dry, non-tacky residue indicates that the flux was activated and then consumed, while a sticky or dark residue usually means the dwell was wrong in one direction or the other.

Delta T Across the Board

Delta T is the difference in temperature between the hottest and the coldest point on the assembly at the same moment. It is measured during the soak, because that is where it is smallest and where the equalisation work happens.

A large delta T at the end of the soak means that some parts will reach the peak later than others, so a profile that produces a good joint on one component may overheat another. Reducing the ramp rate into the soak and extending the dwell are the two levers that close the difference.

Fixtures and pallets are part of the thermal path. A carrier that shields the underside of the board, or one that adds mass at one edge, increases the delta T and has to be considered when the profile is qualified.

Soak Zone Problems and Their Symptoms

A soak that is too short shows up as solder balls around the joint, as tombstoned chips and as voids in the bulk of the joint. Those defects are all consistent with volatiles leaving the paste after the solder has already melted.

A soak that is too long shows up as poor wetting, as dark residue and as a dull joint, because the activator has been consumed or the surface has reoxidised before the solder could wet it. The copper surface notes describe the surface condition that the profile assumes.

Where the delta T is too large, the symptom is a defect that always appears in the same location. That pattern is the signature of a thermal problem rather than a paste or stencil problem, and it is worth checking the profile before changing anything else.

Profiling, Verification and Records

Profiles are taken with thermocouples attached to representative points: a small chip, a large component, a ground plane, a connector and a corner of the panel. The board is run through the oven in the production orientation and with the production carrier.

The result is compared with the paste supplier window for ramp, soak band, time above liquidus and peak. Any point outside the window is either a profile change or a design issue with the thermal mass, and the thermal pad voiding notes describe one of the consequences of an uneven thermal profile.

Records should include the profile curve, the paste batch, the oven settings and the date, because a profile that is not dated cannot be compared with a later one. The wetting behaviour the profile is trying to achieve is described in the wetting check notes, and the residue that should remain afterwards is covered in the no clean residue review.

Where the oven has several zones, each zone has a setpoint and the profile is the sum of them. A change to one setpoint moves the whole curve, so the profile is verified after any change rather than assumed to be unchanged.

Oven Zones and Profile Drift

A multi-zone oven is set by zone, but the board sees the sum of the zones and the conveyor speed. Changing one zone setpoint moves the whole curve, so a profile verified today is not necessarily valid tomorrow unless the settings and the speed are recorded together.

The commonest drift is a slow loss of heat transfer as the oven ages. Elements lose efficiency, blowers lose balance and the recirculation paths accumulate flux condensate, and the effect appears first in the soak zone because that is where the process has the least margin.

Thermocouple attachment is another source of apparent drift. A probe held with tape measures the tape, while one bonded with a high-temperature adhesive measures the board. Comparing profiles taken with different attachment methods produces a difference that has nothing to do with the oven.

For a hard-to-heat product an inert atmosphere changes the profile as well, because the gas has a different heat capacity and the transfer rate falls. The profile is therefore requalified whenever the atmosphere is changed.

FAQ

Can the soak zone be removed to increase throughput? On a small, thin board with a simple paste it can sometimes be shortened, but it cannot simply be deleted. The volatiles still have to leave the paste and the flux still has to activate, and both take time that the oven has to provide somewhere.

How is the soak band measured? By attaching a thermocouple to a representative pad and reading the time between the two temperatures that bound the band. The measurement is only valid if the thermocouple is attached the same way each time.

Why does the same profile give a different result in summer? Because the ambient temperature and the humidity change the board temperature at the entry of the oven and the behaviour of the paste. Profiles taken in the same week are comparable, and those taken in different seasons may not be.