One Part Number In, Six Candidates Out: Reading a Cross Reference Result Honestly

A cross reference result is rarely one part. We paste one real position — 0603, 10 kΩ, ±0.1% — into our own export and get six Walsin part numbers back, not one. Median across the whole chip resistor catalogue: three candidates per result, and 73% of results are not unique.

---

Venatech Limited, an authorised Walsin distributor in Hong Kong, writes this from Walsin Technology (Walsin)’s own chip resistor parametric export: 137 .xlsx files, 234,815 rows, read and combined by us row by row. Every part number below is real and checkable against that export. This post is written for a design engineer anywhere a cross comes up against a lead-time problem — France included, where sourcing teams are increasingly running the same second-source check we show below.

How many Walsin candidates does a typical cross reference result return?

A median of three, not one. We grouped all 234,815 rows of the chip resistor export by the three fields a buyer most often supplies when crossing a value — case size, resistance and tolerance — and counted the distinct part numbers landing on each combination. That gives 37,173 distinct three-field positions, with a median of 3 candidates per position, a mean of 6.3, and a worst case of 36.

What we measured	Count
Distinct (size, resistance, tolerance) positions across the catalogue	37,173
Median Walsin candidates per position	3
Positions resolving to exactly one part number	10,160 (27.3%)
Worst-case candidates at one position	36

We have not seen this counted anywhere. It is the arithmetic sitting underneath every “which part is this?” question a cross reference result raises, and it is why a result that returns more than one candidate is normal, not a tool fault.

What does a six-candidate cross reference result actually look like?

Six Walsin part numbers, for one of our own case-size, resistance and tolerance positions: 0603, 10 kΩ, ±0.1%. We pulled every row matching that exact position from the combined export and got SF06H1002BTL, WF06K1002BTL, WF06H1002BTL, WF06J1002BTL, WF06H1002BTR and WF06K1002BTLJ — six real part numbers, none invented, all present in the same 0603, 10 kΩ, ±0.1% bucket. All six live in one source file, High_Precision_Thick_Film_chip_resistor_0_1_1_10W.xlsx — proof the candidate count is a feature of the catalogue, not an artefact of combining several files together.

Part number	TCR	Series	Match / mismatch vs 0603, 10 kΩ, ±0.1%, 1/10 W
WF06J1002BTL	≦±25 ppm/°C	Thick Film High Precision, TC25	Match on size, resistance, tolerance, power; tightest TCR of the six
WF06K1002BTL	≦±50 ppm/°C	Thick Film High Precision, TC50	Match on size, resistance, tolerance, power; mismatch on TCR against the TC25 row
WF06K1002BTLJ	≦±50 ppm/°C	Thick Film High Precision, TC50 (Automotive)	Match on size, resistance, tolerance, power; mismatch — automotive grade, different qualification path, suffix carries a J
WF06H1002BTL	≦±100 ppm/°C	Thick Film High Precision, TC100	Match on size, resistance, tolerance, power; mismatch on TCR, double the TC50 row
WF06H1002BTR	≦±100 ppm/°C	Thick Film High Precision, TC100, LF 100 ppm variant	Match on size, resistance, tolerance, power; mismatch — same TCR number as the row above, different series label and a R suffix
SF06H1002BTL	≦±100 ppm/°C	High Precision Anti-Sulfuration	Match on size, resistance, tolerance, power; mismatch — anti-sulfuration grade, different series prefix entirely

All six share size, resistance, tolerance and rated power (1/10 W), and all six are packed on a 7″ reel with the same Sn-base lead-free termination. We will not tell you a part is interchangeable with another. We will show you where it differs — and here the difference is temperature coefficient of resistance and qualification grade, neither of which a bare resistance-and-tolerance request mentions.

Which fields do all six candidates share, and which field splits them?

Four fields match on every one of the six rows — size, resistance, tolerance, and rated power — and one field, TCR, splits them into four distinct values: ±25, ±50 (twice), ±100 (twice) and the anti-sulfuration variant. Series name is the second splitting field, because two rows share a TCR of ±100 ppm/°C but come from different series (TC100 standard versus TC100 LF 100 ppm) and two share ±50 ppm/°C but split on automotive grade.

So a request that specifies only size, resistance and tolerance has, by construction, told us nothing about temperature stability or grade — the two fields that decide which of the six is right for a given circuit. That is not a gap in the tool. It is a gap in the request, and it is the same gap every cross reference result on this catalogue exposes once you look past the headline count.

Why do three of the six part numbers end in a different suffix?

Because the final one or two characters carry packaging and grade information that the rest of the part number does not. WF06J1002BTL, WF06K1002BTL and WF06H1002BTL all end TL. WF06H1002BTR ends TR, matching its LF 100 ppm (low-force, 100 ppm) series variant rather than the plain TC100. WF06K1002BTLJ carries an extra J, matching its automotive-grade series label. In the 234,815-row export we checked, these suffix differences track the Series column exactly — they are not noise.

That matters for a cross reference result because a reader who strips the suffix to “simplify” the part number for a BOM line erases the one signal that distinguished it from its siblings. Keep the full orderable code until you have confirmed which of the six candidates you actually mean.

How often does a cross reference result collapse to exactly one part?

27.3% of the time, at the three-field level we measured — 10,160 of 37,173 distinct size, resistance and tolerance positions. The other 72.7% return more than one candidate, with a mean of 6.3 and a median of 3. That median is the honest headline: most cross reference results on chip resistors are not single-answer lookups, they are short lists that still need a decision on TCR, grade, or both.

What is the worst-case cross reference result in our own catalogue?

36 candidates, at 0603, 100 kΩ, ±1%. We checked this position directly: it spans 26 different series, from general-purpose and high-precision thick film through automotive, anti-sulfuration, high-voltage, UL-safety-certified and four-gang resistor arrays sharing the same footprint. A buyer who crosses only on size, resistance and tolerance at this common value is choosing from 36 real part numbers, most of which differ on grade, TCR, termination style or whether the part is a single resistor or an array element.

Position	Candidates	Span
0603, 10 kΩ, ±0.1%	6	High-precision thick film, 4 TCR grades, 1 anti-sulfuration variant
0603, 100 kΩ, ±1%	36	26 series — general purpose, automotive, anti-sulfuration, UL safety, resistor arrays

This is why the free cross reference tool carries a grade filter and special-requirement filters rather than returning a single answer: at the high-volume values, a single answer would be a guess dressed up as a match. The tool is matched against Walsin’s published catalogue, which is exactly the export we counted by hand above.

How should a buyer in France read a six-candidate result honestly?

By treating every candidate as a stated match on the fields you gave us and an open question on the fields you did not. A sourcing engineer in France checking a second source against a 26-to-32-week lead time on an incumbent resistor gets the same six-candidate result we show above for 0603, 10 kΩ, ±0.1% — the catalogue does not change by country, only the urgency does. Paste the position into the free cross reference tool, read the TCR and grade columns before the part number, and confirm the winning candidate against Walsin’s own datasheet before it goes anywhere near a BOM.

For how the matching itself works field by field, see Which manufacturers cross to Walsin, and how the matching works. For what we will and will not say once a candidate is chosen, see Why we publish mismatches instead of equivalents. Confirm what is physically in Hong Kong on ready stock, a dated snapshot as at 23 September 2026, not a live feed. The full chip resistor range sits on the chip resistor page.

The short version

A cross reference result is a short list, not a single answer, on most of the catalogue we checked: median 3 candidates across 37,173 distinct size, resistance and tolerance positions, only 27.3% unique, and a worst case of 36 at one common value. At 0603, 10 kΩ, ±0.1% the six real candidates share size, resistance, tolerance and power and split entirely on TCR and grade — the two fields a bare resistance request never states.

Paste your position into the free cross reference tool and read past the headline count: the field that splits six candidates into one right answer is almost never the field you started with.

---

Frequently asked questions

<!– wp:rank-math/faq-block {"questions":[{"id":"faq-question-01","title":"How many candidates does a typical chip resistor cross reference result return?","content":"A median of three. We grouped all 234,815 rows of Walsin's chip resistor export by size, resistance and tolerance and found 37,173 distinct positions. Only 10,160 of them (27.3%) resolve to exactly one part number; the rest return more than one candidate, with a mean of 6.3 per position.","visible":true},{"id":"faq-question-02","title":"Why does one case size, resistance and tolerance position return six different part numbers?","content":"Because the position only fixes three of the fields Walsin publishes. At 0603, 10 kΩ, ±0.1% six real part numbers share those three fields and the rated power, and split on temperature coefficient of resistance (TCR) and grade — ranging from ±25 to ±100 ppm/°C, plus an anti-sulfuration variant.","visible":true},{"id":"faq-question-03","title":"What is the worst-case cross reference result you have measured?","content":"36 candidates, at 0603, 100 kΩ, ±1%. We checked the position directly and it spans 26 different series — general purpose, automotive, anti-sulfuration, UL-safety-certified thick film and four-gang resistor arrays sharing the same case size, resistance and tolerance.","visible":true},{"id":"faq-question-04","title":"Does a tight suffix difference like TL versus TR actually matter?","content":"Yes. In the position we checked, WF06H1002BTL and WF06H1002BTR share size, resistance, tolerance and TCR but belong to different series — standard TC100 against the LF 100 ppm variant. The suffix is not decoration; stripping it to tidy a BOM line removes the signal that told the two parts apart.”,”visible”:true},{“id”:”faq-question-05″,”title”:”Can I trust a cross reference result without checking the datasheet?”,”content”:”No, and we do not ask you to. A cross reference result tells you which Walsin part numbers match the fields you gave us. It does not confirm qualification grade, TCR class, or fit in your circuit. Confirm the winning candidate against Walsin’s own datasheet before it goes into a bill of materials.”,”visible”:true},{“id”:”faq-question-06″,”title”:”Does Venatech have an office in France?”,”content”:”No. Venatech Limited is based in Hong Kong, with operations in Shenzhen, Singapore and Jiangmen and a Chengdu location in preparation. Nothing else. A French buyer uses the same free cross reference tool and the same dated Hong Kong stock snapshot as any other market.”,”visible”:true},{“id”:”faq-question-07″,”title”:”Why does a cross reference result matter more now than it used to?”,”content”:”Because second sourcing has to happen faster when an incumbent part is on a long lead time. The trade press has reported MLCC lead times of 26 to 32 weeks on high-grade series through 2026, and the same pressure is showing up on chip resistors. A six-candidate result read correctly in one pass beats three partial checks spread over a week.”,”visible”:true}],”listStyle”:””,”titleWrapper”:”h3″,”titleCssClasses”:””,”contentCssClasses”:””,”sizeSlug”:””} –>

---