Engineering Note #2 — Jiatel Hardware Electronics
A battery holder that survives reflow is not the same thing as a battery holder that works after reflow. Three failures appear after the oven, not before it, and none of them are caught by a visual inspection of the solder joints:
- Housing deformation — the plastic moves, and the outer pads lift off their footprints
- Contact force relaxation — the contact spring loses normal force permanently
- Coplanarity drift — the part was flat before reflow and is not flat after
This note covers what the standards actually govern, where the real process window comes from, and what to do when the window closes.
1. What the standards govern, and what they do not
Two documents matter here, and they answer different questions.
IPC/JEDEC J-STD-020F classifies moisture/reflow sensitivity for non-hermetic solid-state surface mount devices. Its classification profiles put lead-free peak temperature in the 245–260 °C band, with the specific value selected from a table indexed by package thickness and volume. This is the document people quote when they say “260 °C peak”.
ECA/IPC/JEDEC J-STD-075 extends the same idea to non-IC components — which is what a battery holder is. It defines a Process Sensitivity Level (PSL) alongside the moisture sensitivity level, so that a passive or mechanical part can be classified against wave and reflow profiles including lead-free processing.
J-STD-075 contains a sentence that matters more than its title suggests. On the subject of where to place the thermocouple, the standard notes that because connector designs vary widely and process sensitivity varies significantly between suppliers, the connector supplier’s specification should govern the temperature measurement location.
Read that again as a buyer. It means the number on the solder paste datasheet is not your part’s rating. The rating belongs to whoever built the part, because only they know where the thermally weak point is. A battery holder’s limiting location is usually not the solder joint — it is a thin wall, a latch, or the contact spring pocket.
If a supplier cannot tell you where to measure on their part, they have not characterised it.
2. The two-temperature conflict
Reflow soldering a battery holder is a negotiation between two temperatures that pull in opposite directions.
| Requirement | Comes from | Wants |
|---|---|---|
| Joint must reach liquidus and stay above it long enough to wet | Solder alloy | Temperature above alloy liquidus |
| Housing must stay below its heat deflection temperature under load | Housing resin | Temperature below HDT |
For a common lead-free alloy such as SAC305, liquidus is 217 °C, and the profile has to spend meaningful time above it. J-STD-020F allows peaks up to 260 °C depending on the part.
So the housing has to survive a peak that is roughly 30–40 °C above the temperature at which the solder even becomes liquid. Every degree of that margin is consumed by the resin.
The process window is the intersection of the two requirements. When the intersection is empty, no amount of profile tuning fixes it — the part or the process has to change. Section 6 covers the four options.
3. Decision matrix A: housing material × lead-free reflow
The table below is deliberately qualitative. Heat deflection temperature varies substantially within a polymer family depending on glass content, filler type and grade, and quoting a single HDT for “PBT” would be wrong for most of the PBT grades on the market.
| Housing material | Lead-free reflow (peak 245–260 °C) | Dominant risk | Notes |
|---|---|---|---|
| LCP | Passes | Anisotropic shrinkage; thin walls distort | Low moisture uptake; high-flow grades suit thin walls. Verify HDT on the specific grade’s datasheet. |
| PPS | Passes | Low toughness; latches crack rather than flex | Good dimensional stability. Brittleness is the failure mode, not heat. |
| PA6T / PA9T | Conditional | Moisture absorption → blistering at reflow | High-melting polyamides; must be dried to the specified MSL floor life before use. Grade and wall thickness decide pass/fail. |
| PA66 | Usually fails | Melt point close to peak; softens and creeps | Not recommended for parts that see full lead-free reflow. |
| PBT | Usually fails | Softening plus warpage | Glass-filled grades improve stiffness but not enough to clear a 260 °C peak reliably. |
How to decide, rather than guess: get the HDT from the resin grade’s datasheet — at the load your part actually sees, not the standard 1.82 MPa headline figure — and compare it to the peak your profile delivers at the housing surface, not at the joint.
- HDT below peak → the part will fail. Do not run it.
- HDT above peak with less than ~10 °C margin → high risk. Expect lot-to-lot variation and warpage.
- HDT well above peak → the material is not your constraint; look at coplanarity and contact relaxation instead.
4. The moisture trap
Polyamides absorb water from the air. At reflow, absorbed moisture flashes to steam faster than it can diffuse out, and the result is blistering or internal voiding — often invisible from outside until the part fails mechanically.
This is why J-STD-075 classifies moisture sensitivity alongside process sensitivity, and why the floor-life and bake requirements exist. Two practical consequences:
- Sealed packaging is part of the specification, not logistics. A bag that was opened three weeks ago is not equivalent to one opened this morning.
- Baking is not optional recovery. Baking a moisture-sensitive part removes absorbed water, but every bake cycle is additional thermal history on the resin.
LCP and PPS are far less sensitive to this failure mode than the polyamides, which is one reason they dominate reflow-compatible holder designs.
5. Decision matrix B: process window checklist
Run this against the actual profile measured on the actual board, not against the paste datasheet’s recommended curve.
| Check | Criterion | If exceeded |
|---|---|---|
| Peak temperature at the housing | Below the resin grade’s HDT at operating load | Softening, latch creep, contact pocket distortion |
| Peak temperature at the joint | Above alloy liquidus, per alloy spec | Cold joint, incomplete wetting |
| Time above liquidus | Per paste specification — shorter is generally better | Excessive intermetallic growth, brittle joints |
| Ramp rate | Within paste and component limits | Thermal shock; differential expansion between metal and resin |
| Number of passes | Count every pass, including rework | Thermal history accumulates; relaxation compounds |
| Moisture state at reflow | Within MSL floor life, or baked per J-STD-033 | Blistering, popcorning |
| Post-reflow coplanarity | Measured after a full simulated cycle | Outer pads lifted; open joints on the perimeter |
The last row is the one most often skipped, and it is the most expensive.
6. When the window closes: four options
If the housing cannot survive the profile and the joint still needs to form, you have four levers. They are ordered by how much they disturb the rest of the design.
Option 1 — Change the housing resin. Move to LCP or PPS. Highest confidence, requires a tooling change if the part was designed around a different resin’s shrinkage.
Option 2 — Change the solder alloy. Low-temperature alloys such as Sn42/Bi58 melt near 138 °C, which moves the joint requirement roughly 80 °C downward and reopens a window that was closed. Trade-offs are real: bismuth alloys are more brittle and have lower impact strength, and the joint is not suitable for high-vibration or high-service-temperature environments. This is a legitimate engineering choice, not a shortcut — but it must be made deliberately, with the mechanical consequences accepted.
Option 3 — Selective or hand soldering. Take the holder off the reflow profile entirely. Slower and harder to control, but it removes the thermal constraint from the holder and puts it only on the joints.
Option 4 — Change the mechanical design. Move the constraint away from the plastic: a metal frame carrying the contact springs, with the resin doing only locating. This is the answer when no resin can meet the thermal requirement at acceptable cost.
The mistake to avoid is trying to solve a closed window by slowing the conveyor and calling it “profile optimisation”. A closed window does not reopen because the belt moved slower; the peak is set by the alloy and the mass, not by belt speed.
7. Coplanarity: the failure that scales with part size
A coin cell holder is a relatively large, relatively flat part with a thin base. That geometry is close to a worst case for warpage, because warpage is driven by differential shrinkage between regions that cooled at different rates.
Why it matters: the solder joints at the centre of the part and the joints at the perimeter do not see the same gap to the pad. If the part bows, the perimeter pads lift. The joint there may still look formed — it may even have wet — but it is a thin, stressed joint that will crack under thermal cycling in the field.
Two rules that follow:
- Measure coplanarity after a full thermal cycle, not before. A part that is flat at incoming inspection tells you nothing about the part that comes out of the oven.
- On large holders, check the perimeter joints first. That is where warpage shows up, and it is where an automated optical inspection focused on the centre will miss it.
Gate location, wall thickness uniformity and the number and position of hold-down features all move this number. It is a tooling and design conversation, not a profile conversation.
8. Contact force relaxation after reflow
The contact inside a battery holder is a spring. Springs held at temperature lose force permanently — this is stress relaxation, and unlike elastic deflection it does not recover when the load is removed.
For a coin cell holder the consequence is direct: lower normal force means higher contact resistance, which shows up as voltage droop under load rather than as an open circuit. A device that works on the bench at low current can brown out at operating current, and the holder will be the last thing anyone suspects.
Practical check: measure contact resistance or contact force after reflow, on a sample from the actual production profile. A pre-reflow measurement is a measurement of a different part.
Parts that see two reflow passes — double-sided assembly — accumulate two thermal histories, and the second pass reflows the first side’s joints while the holder on the bottom side sits through the whole profile again.
9. How Jiatel specifies reflow compatibility
We state reflow compatibility per part and per resin grade, including where to place the thermocouple on that specific holder — because J-STD-075 says that location is the supplier’s to define.
We do not publish a single site-wide “260 °C compatible” claim, because that number belongs to a profile and a part, not to a product family.
What we will give you for a specific holder:
- The resin grade and its datasheet HDT at the relevant load
- The recommended thermocouple location for that geometry
- The maximum number of reflow passes we support
- Post-reflow coplanarity expectation and how we measure it
- Contact resistance before and after a simulated reflow cycle
If you are choosing between two holders and neither supplier can give you items 2 and 4, you are comparing two uncharacterised parts.
10. What to do next
If you are specifying a holder now, the sequence that avoids the expensive mistakes is:
- Fix the solder alloy first — it sets the peak you must survive
- Check the resin HDT against that peak with margin, not against a nominal
- Confirm the thermocouple location with the holder supplier
- Measure coplanarity and contact resistance after a full simulated cycle
- Only then release the tooling
→ SMT Battery Contacts and Holders — reflow-compatible contacts, characterised per part.
→ Battery Holders and Contacts — full range, including coin cell, cylindrical and low-profile formats.
→ Request a 2D Drawing — send your board footprint and reflow profile, and we will confirm the thermocouple location and the pass count for the specific holder.
References
- IPC/JEDEC J-STD-020F — Moisture/Reflow Sensitivity Classification for Nonhermetic Solid State Surface Mount Devices. Lead-free classification peaks in the 245–260 °C band, selected by package thickness and volume.
- ECA/IPC/JEDEC J-STD-075 — Classification of Non-IC Electronic Components for Assembly Processes. Defines Process Sensitivity Level (PSL) and references J-STD-020 for base reflow conditions; states that for connectors the supplier’s specification governs the temperature measurement location.
- IPC/JEDEC J-STD-033 — Handling, packing, shipping and use of moisture/reflow sensitive surface mount devices (floor life and bake).
- Alloy liquidus figures (SAC305 ≈ 217 °C, Sn42/Bi58 ≈ 138 °C) are properties of the alloys named, not of any Jiatel product.
Material verdicts in matrix A are qualitative by design. HDT varies by grade, and the governing number must come from the specific resin’s datasheet.