Walsin Technology (Walsin) specifies an MLCC reflow profile with a maximum ramp-up rate
of 3°C/s, a 150–200°C soak of 60–120 seconds, 60–150 seconds above 217°C, at least 30
seconds above 245°C, a 260°C peak, and a maximum ramp-down rate of 6°C/s. Total time to
peak must not exceed 480 seconds. Two reflow passes are the recommended limit.
Source: Walsin MLCC Application Guide, section 5-2,
January 2022, page 8.
What is the recommended reflow profile for a Walsin MLCC?
Six numbers define it. Ramp up no faster than 3°C/s. Hold 150–200°C for 60–120 seconds.
Spend 60–150 seconds above 217°C. Spend at least 30 seconds above 245°C. Peak at 260°C.
Ramp down no faster than 6°C/s. The whole run to peak must fit inside 480 seconds.
The profile is drawn for SnAgCu solder paste. The two rate limits matter more than the
peak. A ceramic chip cracks from a temperature gradient across its body, not from absolute
temperature, and the gradient is set by how fast the oven changes.
| Profile stage | Walsin limit | What it controls |
|---|---|---|
| Ramp up | 3°C/s max | Thermal gradient into the chip |
| Soak, 150–200°C | 60–120 s | Flux activation, body temperature equalisation |
| Above 217°C | 60–150 s | Time above liquidus for SnAgCu |
| Above 245°C | 30 s min | Wetting of the termination |
| Peak | 260°C | Upper process limit |
| Ramp down | 6°C/s max | Thermal gradient out of the chip |
| Start to peak | 480 s max | Total thermal exposure |
Walsin publishes one profile for the MLCC range. It is not restated per series, so the same
six numbers apply whether the part is a general-purpose
0603B101K500
or an AEC-Q200 automotive
MT18B104K250.
How many times can an MLCC go through reflow?
Two. Walsin’s guide states a two-time limitation for reflow soldering as the
recommendation. A third pass is not prohibited outright in the document, but it is outside
what Walsin recommends, and a double-sided board that reflows twice has already used the
budget before any rework is attempted.
Plan the board around this. Parts on side one have seen both passes once side two reflows,
so any later touch-up on them is iron rework under the tighter conditions below. Ask Walsin
for written concurrence before a third oven pass on a qualification-critical build.
Can I wave solder an MLCC?
Only in three case sizes, and only below 1 mm thick. Walsin recommends wave soldering for
0603 (1608), 0805 (2012) and 1206 (3216) at a thickness below 1 mm. Everything else is
reflow or hot-plate. The thickness qualifier is the part that gets missed, and it
disqualifies a large share of parts in the three permitted sizes.
Venatech Limited, an authorised Walsin distributor in Hong Kong, holds an extract of 24,026
Walsin MLCC part numbers. Counting the three wave-permitted case sizes in that extract:
| Case size | Parts held | Thickness < 1.00 mm | Thickness ≥ 1.00 mm | Wave soldering per Walsin |
|---|---|---|---|---|
| 0603 (1608) | 4,730 | 4,672 | 0 | Match — all held parts qualify |
| 0805 (2012) | 4,912 | 3,579 | 1,258 | Mismatch — 1,258 parts too thick |
| 1206 (3216) | 3,520 | 1,832 | 1,615 | Mismatch — 1,615 parts too thick |
| Total | 13,162 | 10,083 | 2,873 | 21.8% of the three sizes excluded |
Unique to Venatech’s data. 2,873 of the 13,162 Walsin MLCC parts in the three
wave-solderable case sizes are 1.00 mm thick or more, and so fall outside Walsin’s own
wave soldering recommendation. The case size alone does not tell you. This count is not
published anywhere; it comes from measuring Walsin’s own parametric export.Flagged for review. 206 of the 13,162 rows carry no thickness figure in Walsin’s
export and are excluded from both thickness columns above. The percentage is calculated
against the 12,956 rows that do carry one.
Two concrete cases. An 0805
MT21B224K500
is 1.25 mm thick — right case size, wrong thickness, not for the wave. A 1206
MT31B105K500
is 1.60 mm thick and is further outside the rule. By contrast a general-purpose 1206
1206B102J500
is 0.80 mm thick and does qualify. Same case code, opposite answer.
The wave profile itself: preheat to 120°C in under 100 seconds, a ramp limit of 150°C/sec
into the wave, 3–5 seconds in contact at up to 260°C, then natural cooling over more than
60 seconds.
What land pattern should I use for an 0603 MLCC?
Walsin publishes two different footprints for the same case size — one for reflow, one for
wave. They are not interchangeable. The wave footprint is longer, adds dummy tracks to
shadow the part from the solder flow, and demands tighter placement.
| Dimension | 0603 reflow | 0603 wave | Match / mismatch |
|---|---|---|---|
| A (overall pad span) | 2.30 mm | 2.40 mm | Mismatch — wave is 0.10 mm longer |
| B (pad length) | 0.70 mm | 1.00 mm | Mismatch — wave pad is 43% longer |
| C (gap) | 0.80 mm | 0.70 mm | Mismatch — wave gap is tighter |
| D (pad width) | 0.80 mm | 0.80 mm | Match |
| E | 0.20 mm | 0.20 mm | Match |
| F | 2.55 mm | 3.10 mm | Mismatch |
| G | 1.40 mm | 1.90 mm | Mismatch |
| Dummy tracks | none | 1 × (0.20 × 0.80) mm | Mismatch — wave only |
| Placement accuracy | ± 0.25 mm | ± 0.10 mm | Mismatch — wave is 2.5× tighter |
All figures from the Walsin MLCC Application Guide, section 3,
pages 4–5. Walsin gives reflow footprints for 17 case sizes and wave footprints for only
four: 0603, 0805, 1206 and 1210.
Note the placement accuracy asymmetry. A wave-soldered 0603 must be placed to ± 0.10 mm,
against ± 0.25 mm for the same part reflowed. The tightest figures in the whole table are
± 0.05 mm, for 01005 and 0201.
What placement head pressure should I set?
1 to 3 N of static weight. Set the bottom dead centre of the mounting head just on the
board surface, not pressing into it. Support the underside of the board to absorb impact
energy. Walsin names a worn centering jaw as a specific crack cause and asks for preventive
maintenance on it.
This is the cheapest crack to design out. A head set 0.1 mm low will not show on any
inspection step and will not fail the electrical test, but it produces part-cracked chips
that fail in the field months later. If a line is producing intermittent MLCC shorts,
measure head height and jaw wear first.
For adhesive-mounted parts, Walsin gives a fillet height of 0.2 mm minimum and an adhesive
thickness of 70–100 µm, with the adhesive not touching the solder land. The guide’s
examples are drawn for 0805 (2012) and 1206 (3216).
How much preheat does an MLCC need before the solder joint forms?
Enough that the temperature difference between the board and the solder never exceeds the
limit for the case size. Walsin sets ΔT ≤ 150°C for wave soldering at 1206 or less. For
reflow and for manual soldering the limit is ΔT ≤ 190°C at 1206 or less, and tightens to
ΔT ≤ 130°C at 1210 or more.
| Process | Case size | Maximum ΔT |
|---|---|---|
| Wave soldering | 1206 (3216) or less | 150°C |
| Reflow soldering | 1206 (3216) or less | 190°C |
| Reflow soldering | 1210 (3225) or more | 130°C |
| Manual soldering | 1206 (3216) or less | 190°C |
| Manual soldering | 1210 (3225) or more | 130°C |
Cooling has its own limit. Natural air cooling is recommended, and if chips are dipped in a
solvent for cleaning the temperature difference must stay below 100°C.
The 1206 / 1210 boundary splits the catalogue unevenly. Of the 24,026 Walsin MLCC parts
Venatech holds, 20,305 sit in case sizes at 1206 or below and take the 190°C allowance,
while 3,471 sit at 1210 or above and are held to 130°C.
Flagged for review. 249 parts do not map cleanly onto a rule written for rectangular
case codes: 97 at 0505 and 149 at 1111 are square microwave sizes, and 3 at 0306 are
reverse geometry, 0.8 mm long by 1.6 mm wide. Walsin’s guide does not say which side of
the 1206 / 1210 boundary these fall on. Treat them as the tighter 130°C case until Walsin
confirms, and ask Venatech to raise the question if the build depends on it.
Can I rework an MLCC with a soldering iron?
Yes, within a tight window, and never touching the ceramic. Walsin allows a tip temperature
of 350°C maximum at 1206 or less, and 280°C maximum at 1210 or more, with the board
preheated above 150°C in both cases. Typical dwell is under 5 seconds at 300°C, under 3
seconds at 350°C, and as short as possible.
Direct contact between the iron tip and the ceramic dielectric can crack the part. Touch
the pad and the termination only. The 280°C ceiling on larger parts is deliberate: a 1210
or larger chip has more ceramic mass to develop a gradient across, so the tip has to be
cooler and the board hotter.
Cleaning has its own constraints. Wait at least 5 minutes after reflow before cleaning.
Use mildly activated rosin flux below 0.1 wt% chlorine. If ultrasonic cleaning is used,
stay within 20 W/L, 40 kHz and 5 minutes.
Which Walsin case sizes have no published land pattern?
Three gaps sit between the application guide and the parts Walsin actually lists. The 0306
reverse-geometry parts have no footprint in the guide. The three-terminal FT feed-through
series has no three-pad footprint. And the ultra-small parts are called 01005 in the guide
but exported as 01R5.
| What the catalogue lists | What the application guide gives | Match / mismatch |
|---|---|---|
| 0306, 3 parts, e.g. 0306B225K4R0 | No footprint at 0306 | Mismatch — no land pattern published |
| FT feed-through, 8 parts, e.g. FT21B103M500 | Two-terminal footprints only | Mismatch — no 3-pad land pattern |
| 01R5 ultra-small, 317 parts, e.g. RF02N0R2B250 | Footprint listed under “01005” | Match, different name |
| 0508 and 0612 footprints in the guide | — | Mismatch — no parts returned by Walsin’s search |
| C Array 0603×4 and 0402×4 footprint | — | Mismatch — no MLCC array parts in the data |
Unique to Venatech’s data. Walsin’s MLCC application guide publishes a capacitor
array footprint for 0603×4 and 0402×4, and land patterns for 0508 and 0612. Venatech’s
extract of Walsin’s own parametric search returns zero MLCC parts for any of them. The
footprints exist; the parts, as far as the public search tool is concerned, do not.
Meanwhile 0306 and the three-terminal FT series are real, orderable parts with no
footprint published in the guide at all.
The 0306 parts are worth a second look because the geometry inverts. 0306B225K4R0 is
0.8 mm long and 1.6 mm wide — terminations on the long sides, current path shortened. It
behaves differently on the board from an 0603 and it cannot use an 0603 footprint. Ask
Venatech for the drawing before laying one out.
FAQ
What is the maximum reflow temperature for an MLCC?
Walsin’s recommended MLCC reflow profile peaks at 260°C, with at least 30 seconds spent
above 245°C and 60–150 seconds above 217°C. Total elapsed time from start to peak must not
exceed 480 seconds. The profile is drawn for SnAgCu solder paste and applies across
Walsin’s MLCC range.
How fast can I ramp an MLCC through reflow?
Ramp up at no more than 3°C per second and ramp down at no more than 6°C per second.
Ceramic chips crack from the gradient across the body rather than from the peak
temperature, so the two rate limits are the numbers to hold. Confirm the oven’s measured
profile on a thermocoupled board, not the recipe setting.
Can an MLCC be reflowed twice?
Yes. Walsin recommends a two-time limitation for reflow soldering. A double-sided board
that reflows both sides has already used both passes for the parts on side one, so any
later touch-up on those parts should be iron rework under the separate temperature limits,
not a third oven pass.
Can you wave solder a 1206 MLCC?
Only if it is under 1 mm thick. Walsin permits wave soldering for 0603, 0805 and 1206 at a
thickness below 1 mm. Of the 3,520 Walsin 1206 MLCC parts Venatech holds, 1,615 are 1.00 mm
thick or more and fall outside that recommendation. Check the thickness, not the case size.
What soldering iron temperature is safe for an MLCC?
350°C maximum at 1206 or smaller, 280°C maximum at 1210 or larger, with the board preheated
above 150°C. Typical dwell is under 5 seconds at 300°C and under 3 seconds at 350°C. Never
let the tip touch the ceramic dielectric — contact the pad and termination only.
What pick-and-place head pressure should be used for MLCCs?
1 to 3 N of static weight. Set the head’s bottom dead centre just on the board surface
without pressing into it, and support the underside of the board. Walsin also names a worn
centering jaw as a distinct crack cause, so include jaw wear in preventive maintenance.
Does a soft termination MLCC change the reflow profile?
No. Soft termination addresses board flex and bend cracking, not thermal profile. The
reflow numbers are the same. If the concern is cracking from board bend or depanelling
rather than from the oven, the soft termination series are the relevant selection, not a
profile change.
How long after reflow can a board be cleaned?
Wait at least 5 minutes after the reflow process before cleaning. Use a mildly activated
rosin flux below 0.1 wt% chlorine. If ultrasonic cleaning is used, keep within 20 W/L
power, 40 kHz frequency and 5 minutes of washing time, and if chips are solvent-dipped keep
the temperature difference below 100°C.
Next step
If a board is cracking MLCCs, check the part against the process rather than the other way
round. Send the BOM lines and Venatech Limited, an authorised Walsin distributor in Hong
Kong, will return the case size, the thickness, whether the part is inside Walsin’s wave
soldering recommendation, and which ΔT limit applies to it.
Send your BOM to Venatech · Browse the
Walsin MLCC range · Read
Soft termination MLCC: SH, SG, ST and VT and
How to read a Walsin MLCC part number.