Chip Inductor Part Number Suffix: 29 of 29 P and T Twins Publish Identical Data, and 25 of 25 Second-Letter Twins Do Not

A chip inductor part number suffix is not one code. We parsed all 5,410 Walsin inductor rows: 2,393 end in a two-letter suffix, and 54 part stems appear with two different endings. In all 29 pairs that differ in the first letter the export publishes identical data; in all 25 that differ in the second letter it does not.

---

If you strip the last two characters of a chip inductor part number to “remove the packaging code”, you will sometimes remove a performance variant. Venatech Limited, an authorised Walsin distributor in Hong Kong, holds the inductor export published by Walsin Technology (Walsin), Walsin_Inductor_ALL.xlsx, and this post is written from the ending of every part number in it, set against the other columns of the same row. We hold no Walsin datasheet for any inductor, so nothing below says what a letter means; it says what the export does and does not show when the letter changes. Every part number is real. We will not tell you two endings are the same part. We will show you which columns separate them.

What does the two-letter ending on a Walsin chip inductor part number look like?

Eight different endings, on 2,393 of the 5,410 real rows, 44.2 percent. We excluded the Test123999 QA row and read the last two characters of every part number in the file. Eight two-letter endings occur: PB, TB, TP, PP, PS, PH, TH and TL, across 15 four-letter prefixes, and every one of the 15 begins with W. The other 3,017 rows end differently: the WCA and WCI series end in a tolerance letter, and the MCI, MCB, AIP and CSME families have their own endings.

Both letters are a code, and the export has no column for either. The Inductor file carries 14 columns: series, size, dimensions, thickness, inductance, impedance, unit, tolerance, Q, self-resonant frequency, DC resistance, rated current and saturation current. No packaging field, no termination field, no grade field.

Suffix	Rows	Prefixes using it	Real part	What the export has a column for	Match / mismatch	Your incumbent’s ending
PB	1,243	WLCW, WLFI, WLFW, WLPN, WQCF, WQCW, WQQC, WTBL, WTCF	WLCW1608Z0G62NPB	nothing for either letter	Mismatch — a code with no field to check it against	___
TB	839	WLCM, WLCW, WLFI, WQCM, WQCW	WLCW1005Z0K30NTB	nothing for either letter	Mismatch — same series as PB rows above	___
TP	138	WLBD, WQBD, WQCW	WLBD3216K2U102TP	nothing for either letter	Mismatch — no column	___
PP	39	WLBD, WLPH, WQBD	WLBD3216K2U102PP	nothing for either letter	Mismatch — no column	___
PS	39	WLPH, WQPH	WLPH201610M1R0PS	nothing for either letter	Mismatch — series name differs from PP	___
PH	21	WLBD, WQBD	WLBD3216HCU121PH	nothing for either letter	Mismatch — no column	___
TH	55	WLBD, WQBD	WLBD3216HCU121TH	nothing for either letter	Mismatch — no column	___
TL	19	WLBD	WLBD1005HCU121TL	nothing for either letter	Mismatch — separates from TH by DC resistance	___

Which part stems appear with two different endings?

54 of 2,339. We cut the last two characters off each of the 2,393 part numbers and counted stems that exist under more than one ending. There are 2,339 distinct stems, 54 of them appear under two endings, and every one of the 54 is in the WLBD bead series or the WLPH inductor series. Nothing like it occurs in any other prefix.

The 54 split into four pairs of endings, and the split is the whole post.

Ending pair	Stems	Prefix	Real pair	What differs in the export	Match / mismatch
PP and TP	16	WLBD	WLBD3216K2U102PP and WLBD3216K2U102TP	only how rated current is written	Match on all 12 compared columns after units are normalised
PH and TH	13	WLBD	WLBD3216HCU121PH and WLBD3216HCU121TH	only how rated current is written	Match on all 12 compared columns after units are normalised
TH and TL	8	WLBD	WLBD1005HCU121TH and WLBD1005HCU121TL	DC resistance on 8, rated current on 7, thickness on 2	Mismatch
PP and PS	17	WLPH	WLPH201610M1R0PP and WLPH201610M1R0PS	series name on 17, rated current on 17, DC resistance on 15	Mismatch

Does the first letter, P or T, change anything the export publishes?

No, in 29 of 29 cases. The 29 stems whose two endings differ only in the first letter are the 16 PP and TP pairs and the 13 PH and TH pairs. We compared them on all 12 columns the file carries: series, size, dimensions, thickness, inductance, impedance, tolerance, Q, self-resonant frequency, DC resistance, rated current and saturation current. After we converted every rated current to milliamps, all 29 pairs were identical on all 12, and the only raw difference was how the unit is written.

That difference is exact. In WLBD3216K2U102PP the rated current reads 500mA; in WLBD3216K2U102TP it reads 500. In WLBD3216HCU121PH it reads 6A; in WLBD3216HCU121TH it reads 6000. In all 29 pairs the T row is the bare number and the P row carries the unit. We have not seen this counted anywhere: 29 of the 71 unitless rated current cells in the whole inductor export are the T twin of a P part that states its unit.

Field	WLBD3216HCU121PH	WLBD3216HCU121TH	Match / mismatch
Series	WLBD(HC) Series	WLBD(HC) Series	Match
Size	3216 (1206)	3216 (1206)	Match
Thickness	1.1 mm	1.1 mm	Match
Impedance at 100 MHz	120 ohm	120 ohm	Match
DC resistance	0.015 ohm	0.015 ohm	Match
Rated current as written	6A	6000	Mismatch in the string, match in the value

The 71 cells were published earlier as a unit defect, with the unit assumed to be milliamps by comparison with adjacent series. These 29 twins now confirm the assumption directly for 29 of them, because the twin states the same current with its unit. It does not confirm the other 42.

Does the second letter change the part?

Yes, in 25 of 25 cases. The 25 stems whose endings differ in the second letter are the 8 TH and TL pairs and the 17 PP and PS pairs, and every one differs in at least one published column. In the WLPH inductor pairs the series name itself changes, from WLPH Series to WLPH(HC) Series on all 17 pairs, and the rated current and DC resistance change with it.

WLPH201610M1R0PP is 1 uH at 60 milliohm and 2.6 A. WLPH201610M1R0PS is 1 uH at 48 milliohm and 3.2 A. The ending is the only difference in the part number, and the series text, the DC resistance and the rated current all differ. On the WLBD beads the second letter separates WLBD1005HCU121TH at 0.15 ohm and 1.5 A from WLBD1005HCU121TL at 0.055 ohm and 2 A, with the same 120 ohm impedance at 100 MHz.

Field	WLPH201610M1R0PP	WLPH201610M1R0PS	Match / mismatch
Series	WLPH Series	WLPH(HC) Series	Mismatch
Size and thickness	2016, 1 mm	2016, 1 mm	Match
Inductance	1 uH	1 uH	Match
Tolerance	±20%	±20%	Match
DC resistance	60 milliohm	48 milliohm	Mismatch
Rated current	2.6 A	3.2 A	Mismatch

The second letter lines up with the series name wherever a high-current variant exists. All 68 WLBD rows end in P as the second letter and all 71 WLBD(HC) rows end in H or L; all 22 WLPH rows end in P and all 19 WLPH(HC) rows end in S. That is an association we measured, not a definition we hold.

Where does the first letter follow case size?

In three series, exactly, and the break point is different in each. In WLCW and WQCW all 283 rows at 1005 end in T and all 1,077 rows at 1608 and larger end in P; in the WQBD beads all 37 rows below 3216 end in T and all 8 rows at 3216 and above end in P. WLCW has 134 T-first rows, all 1005, and 657 P-first rows across five larger sizes. WQCW has 149 T-first rows, all 1005, and 420 P-first rows across five larger sizes.

The rule does not carry. WLCM is all T-first, 559 rows across 0603, 1005 and 1608. WLFI has 37 T-first rows at 1608 and 2012 and 7 P-first rows at 2012. In WLBD all 79 rows below 3216 are T-first, but from 3216 up both letters occur. A decoder that assumes the first letter is a case-size marker is correct on three series and wrong on the others, and it would need a different break point for each.

Series	Smaller sizes	Larger sizes	Pattern	Match / mismatch
WLCW	1005: 134 T-first, 0 P-first	1608 and up: 0 T-first, 657 P-first	split at 1005	Match with a size rule
WQCW	1005: 149 T-first, 0 P-first	1608 and up: 0 T-first, 420 P-first	split at 1005	Match with a size rule
WQBD	1005 to 2012: 37 T-first, 0 P-first	3216 and up: 0 T-first, 8 P-first	split at 3216	Match with a size rule, at a different size
WLBD	1005 to 2012: 79 T-first, 0 P-first	3216 and up: 30 T-first, 30 P-first	mixed above 3216	Mismatch — the larger sizes carry both letters
WLCM	0603 to 1608: 559 T-first, 0 P-first	none	T-first at every size	Mismatch — no split to read
WLFI	1608: 22 T-first, 0 P-first	2012: 15 T-first, 7 P-first	mixed at 2012	Mismatch — both letters at one size

Where do the 71 unitless rated current cells sit?

On T-first endings only. We counted every rated current cell by its unit across the 2,393 rows. Of 1,051 T-first rows, 71 carry a rated current with no unit, 6.8 percent; of 1,342 P-first rows, none do. The 71 are 41 on TB, 16 on TP and 14 on TH, and 29 of them are the bare twin in the pairs above.

The other 42 are 41 rows of WLCM and one row of WLBD, and they have no twin to compare against. A script that casts rated current to a number will read them correctly only if it knows to treat a bare value as milliamps, and that knowledge comes from the 29 twins, not from the export.

What should a cross tool do with an ending it cannot decode?

Keep it, flag it, and do not collapse stems. That is how our free cross-reference tool handles any suffix it cannot read: the string is carried through to the result and a difference from your original is stated as a mismatch. Three rules, each measured above:

1. Never strip the last two characters to deduplicate. It merges 25 stems whose two endings publish different data.
2. Never read the second letter as packaging. In 25 of 25 pairs it separates parts that differ in published data.
3. Do not match endings across manufacturers by letter. Fill the incumbent column above from your own datasheet. Other makers use their own conventions and we do not reproduce them. Confirm against both datasheets and qualify in your own circuit.

For the prefix side of the same files, see how the AIP, MIP and WIP inductor prefixes decode, and for the tolerance letters inside the same part numbers, inductor tolerance codes. We hold 28 Walsin datasheets, 451 pages, and none of them covers an inductor or bead, so everything above is read from the export alone. What is in Hong Kong is on ready stock, 1,125 part numbers as at 23 September 2026, a dated snapshot and not a live feed. The inductor range sits on the inductor page. Our address is Unit 207, 2/F, Sun Ling Plaza, 30 On Kui Street, Fanling, N.T., Hong Kong, (852) 2628 4777.

The short version

Eight two-letter endings sit on 2,393 of 5,410 Walsin inductor and bead rows, and the export has no column for either letter. 54 of 2,339 stems appear under two endings. Where the first letter differs, P against T, 29 of 29 pairs publish identical data and differ only in how rated current is written; where the second letter differs, 25 of 25 pairs differ in published data. The first letter follows case size in three series, at a different break point in each. All 71 unitless rated current cells sit on T-first endings. We will not tell you two endings are one part. We will show you the columns that differ. Qualification stays with you.

Send us the inductor lines on your BOM with the endings your tool strips, and we will tell you which twins it merged. And a question back: when your deduplication collapsed two inductor part numbers into one, did anyone compare their DC resistance?

---

Frequently asked questions

---