The 2.5A That Isn’t a Rating: Molex PicoBlade 1.25mm, Explained

Three different numbers for one current specification, a pitch where bigger means less current, and a terminal you have to choose by plating thickness.

The most-quoted number on the Molex PicoBlade is 2.5A.

It is also not a rating. It is reference derating data, published under conditions that Molex states plainly — and the gap between those two things is where designs quietly lose their margin. That single misunderstanding is worth more attention than any other parameter on this connector.

Here is the full system: what “MX1.25” actually refers to, why its current specification is three separate numbers, why the bigger pitch in its own family carries less current, and the ten things to verify before you cross-reference it.

What the PicoBlade 1.25mm system is

“Molex MX1.25” is trade shorthand. Molex does not sell a product under that name — the string is built from the 1.25mm pitch and maps to the Molex PicoBlade 1.25mm connector system.

• Pitch: 1.25mm, single row
• Circuits: 2 to 15
• Configuration: wire-to-board and wire-to-wire, in the same family
• Voltage rating: 125V AC (RMS)/DC
• Rated current: 1.0A at AWG #26, #28 and #30 · 0.8A at AWG #32
• Reference derating: published separately, up to 2.5A at 2 circuits
• Applicable wire: AWG #26 to #32, insulation up to φ1.0mm (50058802* terminal) or φ1.04mm (50079802*)
• Contact resistance: 20mΩ max initially · 5mΩ max crimped portion · 40mΩ max after environmental exposure
• Insulation resistance: 100MΩ min
• Dielectric strength: 250V AC (RMS), 1 minute
• Operating temperature: −40°C to +105°C, including terminal temperature rise
• Durability: 30 mating cycles with gold plating
• Contact design: two-point contact
• Retention: friction lock
• Plating options: gold flash · 0.38µm min gold · tin
• Mounting: through-hole and SMT, vertical and right-angle
• Assembly feature: vacuum caps and a flat pick-and-place area on SMT headers
• Molex series numbers: 51021 receptacle housing · 53048 right-angle plug assembly
• Standards: UL File E29179
• Cross-reference equivalents: KONNRA KR1250 (wire-to-board) · KR1251 (wire-to-wire)

Two design choices define the series. The first is a two-point contact design — each mating pair touches in two places, which Molex describes as providing “a reliable electrical connection” in rugged applications. The second is a friction lock rather than a positive latch, paired with that contact design to keep the mated pair stable under vibration.

The current rating is three numbers, not one

This is the parameter most often quoted wrong, because Molex publishes it at three levels — and only one of them is the rating.

Level 1 — rated current, by wire size. This is the rating, in the ordinary sense: the current the connector is rated to carry continuously.

• AWG #26 → 1.0A
• AWG #28 → 1.0A
• AWG #30 → 1.0A
• AWG #32 → 0.8A

Level 2 — the reference derating table, by gauge and circuit count, at columns of 2, 8 and 15 circuits:

• AWG #26 → 2.5A at 2 circuits · 1.5A at 8 · 1.0A at 15
• AWG #28 → 2.0A · 1.5A · 1.0A
• AWG #30 → 1.5A · 1.0A · 1.0A
• AWG #32 → 1.5A · 1.0A · 0.8A

Level 3 — the sum. Molex adds a note that is easy to skip past: “Do not allow the sum of the current used on several circuits to exceed the maximum allowable current.” The per-circuit figure is not the whole answer, and on a multi-circuit connector that distinction decides whether the assembly is inside its specification.

Why 2.5A is the most dangerous number on this connector. It is real, and it is reference data. Molex states the conditions explicitly: the values are for reference only, the derating is based on not exceeding a 30°C temperature rise, the rise is measured in the barrel area of the crimp terminal, the data is for all circuits powered, and — the one that matters most in practice — PCB trace design can greatly affect temperature rise results.

Read those conditions together and the implication is unavoidable. The 2.5A figure belongs to Molex’s test board, not to yours. If your design needs 2.5A through two PicoBlade circuits, the number to work from is a temperature-rise test on your own layout — not a datasheet row that has travelled into your BOM without its footnotes.

The inversion: the bigger pitch carries less current

Here is the part that surprises engineers who know the family only by reputation.

The PicoBlade is 1.25mm at 1.0A. But Molex’s own Pico-EZmate is 1.20mm at 1.5A — a smaller pitch with a higher current rating. JST’s ACH at 1.2mm reaches 2A on 1–3 circuits. And Molex’s Pico-EZmate Plus does 2.8A at 1.00mm.

Do not assume the bigger connector is the stronger one. On this family it is not, and the two Molex systems do not mate: 1.25mm is not 1.20mm, and over a 15-circuit connector the 0.05mm-per-position difference compounds across the entire contact field. KONNRA cross-references them as KR1200 (Pico-EZmate 1.20mm) and KR1250 / KR1251 (PicoBlade 1.25mm).

So what is the PicoBlade’s case? Not current. It rests on three things:

• The two-point contact design. Molex’s stated reason is blunt: “rugged applications can be subject to shock, vibration or rough handling that can dislodge the terminals and cause signal interference.” Two contact points per mating pair means a single disturbed point does not open the circuit. That redundancy is the reason to specify a PicoBlade instead of a simpler single-point connector at a similar pitch.
• The 125V rating. At the same 1.25mm pitch, JST’s GH series is rated 50V. That is a material difference if your circuit runs above 50V — and if it does, GH is not an option at any current.
• One contact system for both connection types. The family covers wire-to-board and wire-to-wire, so both ends of a run can use the same contact system and one qualification.

The friction lock is not your cable retention

Retention on this series comes from friction — there is no positive latch to release. Molex’s own insertion and withdrawal force table shows how modest that friction is:

• 2 circuits — 2.8N minimum at first mating, falling to 1.8N by the 30th cycle
• 5 circuits — 3.8N, falling to 2.8N
• 8 circuits — 5.2N, falling to 4.3N
• 15 circuits — 7.3N, falling to 6.3N

A 2-circuit connector holds with under 3 newtons, and under 2 newtons after 30 cycles. The friction lock keeps the pair mated; it does not carry your cable load.

Molex’s design guidance says the same thing from the other direction: tie the cable at least 50mm from the connector edge, with the tying force applied evenly across all wires, so the terminals can float freely in their pockets. A tie closer than 50mm pre-loads the contacts — which is a subtle failure mode, because the connector still mates and still passes a continuity check.

Four crimp terminals, and the overlap where plating decides

Molex splits the PicoBlade terminal across four part numbers: two wire ranges, each in two plating thicknesses.

• 500798020 — AWG #26–28, gold flash
• 500798025 — AWG #26–28, 0.38µm minimum gold
• 500588020 — AWG #28–32, gold flash
• 500588025 — AWG #28–32, 0.38µm minimum gold

The two wire ranges overlap at AWG #28. At that gauge you have four options and two independent questions — which gauge family, and which finish. Molex makes the distinction explicit in the part number: flash versus 0.38µm minimum. That is precisely the specification a BOM line reading simply “gold plated” fails to record — and it is the difference between a mating interface specified to a measurable thickness and one that is not.

Insulation diameter differs by terminal as well: the 50058802* family allows φ1.0mm maximum insulation, and the 50079802* family allows φ1.04mm.

Two gaps on the second-source side, stated plainly. KONNRA documents one terminal for the KR1250, with a stated wire range of AWG #28 to #32, offered as “Gold flash / Tin over Nickel”. So both the AWG #26 path and the 0.38µm gold option fall outside the published KR1250 documentation. If your design depends on either, raise it on the enquiry rather than assuming coverage.

Circuit counts and the AWG #26 gap

Molex states the PicoBlade is “available in 2 to 15 circuits, Tin- and Gold-plated versions with through-hole and SMT mounting, and vertical and right-angle orientations”, and both the derating table and the force table run to 15. KONNRA documents the KR1250 as 2 to 16 positions — one position beyond the original range.

The wire-range gap matters more than the circuit gap. Molex’s PicoBlade covers AWG #26 to #32, and AWG #26 is the gauge where this family’s capability is concentrated: it carries the highest derated current in Molex’s own table (2.5A at 2 circuits) and the highest crimp pull-out force (19.6N minimum, against 9.8N at AWG #28).

KONNRA’s stated KR1250 wire range is AWG #28 to #32, which does not cover #26. If your harness uses the heaviest wire in the family, say so explicitly — it is the one gauge where the published cross-reference does not reach.

Cross-referencing to the KONNRA KR1250

Where the documents agree is worth stating, because on a cross-reference it is not a given: the pitch, the single-row configuration, the 125V voltage rating, the 20mΩ contact resistance, the 100MΩ insulation resistance, the −40°C to +105°C operating range, the two-point contact design, and the friction lock. On this series, the agreement is broad — including the two defining design features.

Where they differ. Ten points, in the order they tend to bite:

1. Rated current at AWG #32 — 0.8A versus a flat 1A. At AWG #26, #28 and #30 the two sides match at 1.0A, so this is specifically an AWG #32 issue — and at that gauge the published KR1250 figure is 25% optimistic.
2. The applicable wire range does not reach AWG #26. Molex covers #26 to #32; the published KR1250 range is #28 to #32. AWG #26 is where the family’s current capability and crimp strength are concentrated, so this is the gap most likely to matter on a real design.
3. The insulation window is narrower on the KONNRA side. Molex allows up to φ1.0mm on the 50058802* terminals and φ1.04mm on the 50079802* terminals. KONNRA documents 0.4 to 0.8mm. A wire with 0.9mm insulation passes Molex’s specification and falls outside KONNRA’s stated maximum. The KONNRA figure does publish a lower bound, so check both ends against your cable.
4. Dielectric withstanding voltage differs by a factor of two — in KONNRA’s favour. Molex specifies 250V AC (RMS) for one minute; KONNRA documents 500V AC / minute. A higher figure on the replacement is not automatically a problem, but it is a different test claim — confirm what is being tested and between which points before relying on it. The rated voltage, 125V, is the same on both sides.
5. Mating durability is published by Molex and not by KONNRA. Molex specifies 30 cycles with gold plating. KONNRA publishes no cycle count for the KR1250.
6. Plating thickness is specified by Molex and not by KONNRA. Molex offers three finishes — gold flash, 0.38µm minimum gold and tin — and the thickness is part of the terminal part number, so it is unambiguous at ordering time. KONNRA states “Gold flash / Tin over Nickel” with no thickness figure.
7. The KONNRA page contradicts itself on plating. The specification table states “Gold Flash/Tin over Nickel”, while the advantages text states that “the contact area is tin-plated.” Those describe different finishes on the same mating interface. On a two-point contact interface the finish is what determines long-term contact resistance stability, so this is worth resolving before quoting.
8. Circuit counts — 2 to 15 on the Molex side, 2 to 16 on the KONNRA side. One position beyond the original range. Confirm the count you intend to order.
9. Wire-to-board versus wire-to-wire. KONNRA’s KR1250 Overview states the series supports both wire-to-wire and wire-to-board, but KONNRA’s own 1.25mm listing categorises KR1250 as wire-to-board — and the wire-to-wire part in the family is KR1251. If your design is an inline connection, the cross-reference you want is KR1251, not KR1250.
10. Applicable wire construction. Molex specifies that the applicable wire is, in principle, tin-plated copper stranded wire, and that other constructions require consultation and evaluation. KONNRA does not state an applicable construction. If you are using silver-plated, solid core or a high-flex construction, raise it before qualification.

Design cautions — and the contact-resistance figure that doubles

Molex publishes an unusually detailed set of usage constraints for this series. They are not marketing; they are the failure modes the product is known to have, and they are the most useful content in the specification. KONNRA does not publish an equivalent list.

On electrical and mechanical limits:

1. Do not mate or unmate under load. Molex states the product is not designed for mating and unmating under an active electrical circuit — doing so can cause a spark and product damage. If your service procedure disconnects this connector live, the procedure needs changing.
2. Respect the per-circuit rating, and the sum. The sum of current across several circuits must not exceed the maximum allowable current, on top of the per-circuit figure.
3. Tie the cable at least 50mm from the connector, force applied evenly across all wires, so the terminals float freely.
4. No continuous pulling or stress force when mated — it can damage the contact area or the crimp.
5. Expect contact resistance to rise after environmental exposure. Molex specifies 20mΩ max initially, but 40mΩ max after vibration, mechanical shock, temperature cycling, heat, cold, humidity, salt spray, and SO₂ and NH₃ gas exposure — all specified with conditions in the datasheet. If your specification quotes a single contact-resistance figure without saying which condition it applies to, it is incomplete. The two numbers are a factor of two apart, and the difference is the one that appears after the product has been in the field.

On assembly and handling:

1. Withdraw slowly, axially and straightly. Angled or rough withdrawal can damage the connector.
2. Mate along the mating axis; if diagonal mating is unavoidable, touch the external faces at under 10° and push from there.
3. Use a new housing after extracting a terminal. Extracting a crimp terminal with a jig can deform the housing lance and, in Molex’s words, “reduce the terminal retention force enormously” after re-insertion.
4. Do not wash the connectors — Molex states this may damage the product’s function.
5. Avoid sympathetic vibration between the wires and the PCB: fix both to the chassis, because sustained relative movement wears the contact area.

Two notes on the board-mounted side, which KONNRA specifies as six orderable components — four wafer types plus a housing and a terminal:

• KONNRA states that its SMT headers carry solder lugs that act as strain relief as well as securing the connector — which addresses the solder-joint cracking risk Molex’s own caution list warns about.
• On an SMT line, Molex’s PicoBlade SMT headers provide vacuum caps and a flat placement area for industry-standard pick-and-place nozzles. Confirm the equivalent placement feature on the specific wafer you order.

The 1.25mm class: where the PicoBlade sits

1.25mm is the most crowded pitch in the connector industry — three manufacturers, at least eight systems in the neighbourhood, and no two of them mate.

• Molex PicoBlade — 1.25mm, 1.0A (0.8A at AWG #32), 125V, AWG #26–32 → KR1250 · KR1251
• Molex PanelMate — 1.25mm, 1A, 125V → KR1253
• Hirose DF13 — 1.25mm, 1A, 150V → KR1256
• Hirose DF14 — 1.25mm, 1A, 150V → KR1255
• JST GH — 1.25mm, 1A, 50V AC/DC, AWG #30–26 → KR1259
• Molex Pico-EZmate — 1.20mm, 1.5A, 50V, AWG 28–30 → KR1200
• JST ACH — 1.2mm, 2A (1–3 circuits) · 1.5A (4–5 circuits), 50V AC/DC, AWG #32–28 → KR1201
• JST SH — 1.00mm, 1A, 50V AC/DC, AWG #32–28 → KR1000

Three things this settles.

First, “1.25mm” identifies a pitch, not a connector. Five systems sit at exactly 1.25mm here, from three manufacturers — KONNRA cross-references them as KR1250/KR1251 (PicoBlade), KR1253 (PanelMate), KR1255 (DF14), KR1256 (DF13) and KR1259 (JST GH). Getting the pitch right is only the first step.

Second, going smaller can mean more current, not less. Molex’s own Pico-EZmate at 1.20mm is rated 1.5A, and JST’s ACH at 1.2mm reaches 2A on 1–3 circuits — both above the PicoBlade’s 1.0A. So the PicoBlade’s case does not rest on current. It rests on the two-point contact design, the 125V rating, and one family covering both connection types.

Third, the voltage rating is where the PicoBlade separates from JST’s 1.25mm option. 125V against 50V is a material difference if your circuit runs above 50V — and if it does, JST GH is not an option at any current.

Engineer’s pre-release checklist

Run this before releasing a drawing for a PicoBlade-family connector.

• Pitch measured, not assumed. 1.25mm is Molex PicoBlade, Molex PanelMate, Hirose DF13, Hirose DF14 or JST GH. 1.20mm is Molex Pico-EZmate or JST ACH. None of them mate.
• Wire gauge confirmed inside the range both sides cover — Molex documents AWG #26 to #32; the published KR1250 range is AWG #28 to #32. If you are on AWG #26, raise it.
• Current checked per gauge, not per family — 1.0A at AWG #26, #28, #30, but 0.8A at AWG #32.
• Derating position established from your own board, not the datasheet table. The 2.5A-at-2-circuits figure is reference data for a 30°C rise with all circuits powered on a test board, and Molex states that PCB trace design greatly affects the result.
• Sum of current across circuits checked against the maximum allowable, not just the per-circuit figure.
• Insulation outside diameter confirmed — up to φ1.0mm or φ1.04mm per Molex terminal, against KONNRA’s stated 0.4–0.8mm.
• Plating finish and thickness specified — flash or 0.38µm minimum gold. Note that KONNRA’s own page states both “Gold Flash/Tin over Nickel” in its specification table and “the contact area is tin-plated” in its advantages text.
• Mating durability considered — Molex specifies 30 cycles with gold plating; KONNRA publishes no figure. Design so this connector is not the one being cycled in service.
• Retention sized against reality. Withdrawal force is 2.8N minimum at 2 circuits, falling to 1.8N by the 30th cycle. The friction lock is not a cable retention device.
• Cable tie placed at least 50mm from the connector, with even force across all wires.
• Service procedure checked for hot-mating. Molex states this product is not designed for mating or unmating under an active circuit.
• Terminal extraction procedure written — fit a new housing after extracting a terminal, because the lance deforms and retention drops sharply.
• Mating axis alignment in the build instruction — straight, or under 10° if diagonal mating is unavoidable.
• Wire construction confirmed as tin-plated copper stranded, or raised for evaluation if it is not.
• Wire-to-board or wire-to-wire confirmed — KR1250 for wire-to-board, KR1251 for wire-to-wire.
• Wafer orientation and mounting type matched to your layout — DIP or SMT, straight or right-angle — and the pick-and-place feature confirmed if the board runs through an SMT line.
• Contact resistance quoted with its condition. Specify whether the figure you are qualifying against is the initial value or the post-environmental value; they differ by a factor of two.

Getting a cross-reference check

Most MX1.25 enquiries stall on the same thing: the buyer is not sure what information the supplier needs, so the enquiry never gets sent. The list is short.

1. The original part number, if you have it — a housing such as 51021**00, a terminal such as 500798025, or a plug assembly such as 53048**60
2. Circuit count — the Molex range is 2 to 15
3. Wire gauge, and specifically whether it is AWG #26. Molex covers #26 to #32; the published KR1250 range is #28 to #32
4. Insulation outside diameter, not just the gauge — Molex allows up to φ1.0mm or φ1.04mm depending on terminal; KONNRA documents 0.4–0.8mm
5. Plating finish required — flash or a specified gold thickness. 0.38µm minimum is a distinct Molex option, and the KONNRA documentation states no thickness
6. Mounting and orientation — through-hole or SMT, straight or right-angle
7. Wire-to-board or wire-to-wire — the wire-to-board cross-reference is KR1250; the wire-to-wire part is KR1251
8. Your load current, so the AWG #32 0.8A question and the derating position can be settled rather than assumed
9. Application and annual volume, so configuration, tooling and packaging can be matched to your programme
10. A drawing or photo, if the part number is unreadable or the design has been reverse-engineered

When you send a part number, four things get confirmed against Molex’s own documentation: the terminal that matches your wire gauge and insulation diameter, the gold thickness you need, the wafer orientation and mounting type, and whether your design needs the wire-to-board or wire-to-wire variant. Those four are where an MX1.25 cross-reference most often goes wrong.

Connector lead time is typically 2–3 weeks, wiring harness lead time typically 3–4 weeks, and complete connector set samples can be delivered within 45 days. Where crimp pull-out force or contact resistance needs validating before commitment — both of which Molex specifies in detail on this series — sample harnesses can be supplied from the same process used in production.

Full technical write-up: Molex MX1.25 Connector Complete Guide

Disclosure: I work with KONNRA, which manufactures the KR1250 and KR1251 cross-references to the Molex PicoBlade 1.25mm system. The Molex figures in this article are taken from Molex’s published Product Specification 510211000-PS; the KONNRA figures are from KONNRA’s published KR1250 documentation. Where the two sources disagree — or where one source contradicts itself — I have flagged the difference rather than averaged it, and where a parameter is stated by neither, I have said so instead of filling the gap.

https://konnra.com/molex-mx1-25-connector-complete-guide/
Dongguan Konnra Electronics Co., Ltd