NEWS

Walsin Parametric Search Limits: 15 of 80 Columns Select Nothing

We re-measured every column in Walsin’s parametric exports across seven product lines and 359,517 rows. 15 of the 80 declared columns carry the same value on every row or no value at all. The hard limit on parametric search is not how you filter. It is how many of the fields can discriminate.


Venatech Limited, an authorised Walsin distributor in Hong Kong, holds all 203 Walsin Technology (Walsin) parametric export files. This post re-counts every column in seven of those lines — chip resistor, thin film, 00_ALL_MLCC_parts_combined.xlsx, Walsin_MOV_ALL.xlsx, Walsin_Inductor_ALL.xlsx, Walsin_DiscCap_ALL.xlsx and Walsin_Antenna_ALL.xlsx — plus the seven RF files, and names each column that cannot narrow a result set. Every part number below came out of those exports.

Which Walsin export columns look like selectors and select nothing?

Fifteen of eighty. We checked all 80 declared columns across the seven product lines, covering 359,517 rows, and counted the distinct values in each. Fifteen columns hold exactly one value on every row or are empty on every row. That is 18.8% of the selectable surface area of the catalogue doing no work at all, and the pattern is not spread evenly — the disc capacitor export loses five of its eleven columns, while the inductor and antenna exports lose none.

Product line Rows Columns Columns that select nothing Which ones
Chip resistor 234,815 10 1 Termination
Thin film 93,914 9 2 Packaging, Termination
MLCC 24,026 16 3 Packing, SPQ, Thickness
Varistor (MOV) 379 10 4 Tolerance, Packing, SPQ, MOQ
Disc capacitor 929 11 5 T.C., Lead Spacing, Packing, SPQ, MOQ
Inductor 5,410 14 0 —
Antenna 44 10 0 —
Total 359,517 80 15 —

Two of these we had not published before. Thin film Packaging reads 7" reel taping on all 93,914 rows, and the varistor Tolerance column reads +/-10% on all 379. We have not seen either counted anywhere. A selector that offers you a packaging dropdown on the thin film line is offering you a dropdown with one option in it.

How many Walsin chip resistors are identical on every selectable field?

193,724 of 234,815. We stripped the two free-text columns — Series and Series Group — and the part number itself, leaving the seven fields a parametric search actually filters on: size, resistance, tolerance, TCR, rated power, packaging and termination. Those 234,815 part numbers collapse into 94,139 distinct signatures. 82.5% of the catalogue shares its signature with at least one other part number, and the largest single group holds 15.

Part number Series as exported Size Resistance Tolerance TCR Rated power Match / mismatch vs WR06X1002FTL
WR06X1002FTL Thick Film General Purpose Chip Resistor 0603 10,000 Ω ±1% ≦±100 1/10 W The reference row
WR06X1002FTLJ General Purpose (Automotive) 0603 10,000 Ω ±1% ≦±100 1/10 W Match on all seven selectable fields; mismatch only in the series text — automotive
SR06X1002FTL High Grade Anti-Sulfuration (Automotive) 0603 10,000 Ω ±1% ≦±100 1/10 W Match on all seven; mismatch — anti-sulfuration construction
MR06X1002FTR General Purpose LF 100 ppm (Automotive) 0603 10,000 Ω ±1% ≦±100 1/10 W Match on all seven; mismatch — different family and finish code
WK06V1002FTL Thick Film High Voltage Chip Resistor 0603 10,000 Ω ±1% ≦±100 1/10 W Match on all seven; mismatch — high-voltage series
WA06X1002FTL Chip Resistor Array 0603*4, Convex type 0603 10,000 Ω ±1% ≦±100 1/10 W Match on all seven; mismatch — this is a four-element array, not a discrete resistor

Thirteen Walsin part numbers sit on that one signature. Six of them are in the table. A parametric selector set to 0603, 10 kΩ, ±1%, ≦±100 ppm/°C and 1/10 W returns all thirteen and has nothing left to rank them with — including the fact that one of them is an array. The discriminating information is in a free-text series string that no dropdown reads.

That is the whole point of this post. We will not tell you a part is equivalent. We will show you where it differs — and on 82.5% of this catalogue the difference is in a field no selector filters on.

Does the Termination column ever tell you anything?

Not once, on either resistor line. We read the Termination column on all 234,815 chip resistor rows and all 93,914 thin film rows: 328,729 rows, one value — Sn base (Lead free). It is the single largest dead column in the catalogue by row count, and it is exactly the field a buyer checking solderability or whisker risk would reach for first.

The MLCC export inverts the problem. There the Termination column holds 14 distinct values, which looks useful until you count the rows: it is populated on 1,106 of 24,026, so it is blank 95.4% of the time. A filter on it silently discards 22,920 parts rather than narrowing them. Packaging on the chip resistor line is the mild case — seven distinct values, but 220,532 of 234,815 rows carry 7" reel taping, so 93.9% of the catalogue answers that question the same way.

What do the RF and antenna exports fail to discriminate?

Impedance, on every RF line we checked. We opened all seven RF files. Impedance (Ω) reads 50 on all 17 rows of Walsin_Coupler.xlsx; Walsin_Balun.xlsx reads 50 on all 72 unbalanced impedance rows; Walsin_Multiplexer.xlsx on all 116; the BAW file on all 5; the SAW file on all 3. Across 213 RF parts the impedance field never once separates two candidates. It records a convention, not a choice.

Walsin_Coupler.xlsx loses a second column in the same ten: Category reads LTCC on all 17 rows, and on Walsin_Filter.xlsx it reads LTCC on 233 of 235. On the antenna line the near-miss is Voltage Standing Wave Ratio: 32 of 44 rows read 2.0 (max.), eight are blank, and two of the remaining four carry an embedded newline and a per-band breakdown, so they will not sort against the other 32 at all.

Which columns are declared and completely empty?

Eleven, across three lines. We counted non-null values rather than distinct values, which is the test that catches a header with nothing under it. In 00_ALL_MLCC_parts_combined.xlsx, Packing, SPQ and Thickness are empty on all 24,026 rows. In Walsin_MOV_ALL.xlsx, Packing, SPQ and MOQ are empty on all 379. In Walsin_DiscCap_ALL.xlsx, T.C., Lead Spacing, Packing, SPQ and MOQ are empty on all 929. Eleven headers, 25,334 rows, not one value between them.

The disc capacitor case is the one that matters technically. T.C. is temperature coefficient — the field that decides whether a safety capacitor’s capacitance holds across its operating range — and it is declared and blank on every one of the 929 rows. Thickness on the MLCC line is the same story: the export carries L, W and T dimension strings on 23,679 rows but the column actually named Thickness is empty on all 24,026. If your height restriction is 1.25 mm, the field named for your problem is the one field with nothing in it.

Ordering data is the other casualty. SPQ and MOQ are empty across all three lines, so nothing in 203 export files tells you a reel quantity or a minimum order. That has to come from us, and the dated snapshot does carry an SPQ column — which is why we quote it from there and date it.

Does the Size column tell you whether a part is an array?

Usually not. We counted all 10,022 chip resistor array rows: 9,079 of them carry a plain discrete case code in the Size column — 0603(1608) on 5,076 and 0402(1005) on 4,003 — and only 943 use array notation such as 0602*8, 0201*4 or 0201*2. So on 90.6% of arrays, the field you would use to spot a four-element part reports the footprint of one element.

One family does both at once, and the split is governed by a capital letter. The series “Thick Film Chip Resistor Array 06028, Convex Type” has 555 rows whose Size reads 0602*8; the series “…Convex t*ype” has 555 rows, the same 555 resistance values at the same 1/16 W, whose Size reads 0603(1608). The part numbers differ only by a trailing N — WA06W10R0FTLN against WA06W10R0FTL. We have not seen this counted anywhere, and it is the cleanest example in the catalogue of a selector returning a confident, wrong answer.

Field What a selector implies it does What we measured it does
Size on an array Reports the array footprint Reports one element’s footprint on 9,079 of 10,022 rows
Termination on resistors Narrows by finish One value on 328,729 rows
Impedance (Ω) on RF Narrows by system impedance 50 on all 213 parts across five files
T.C. on disc capacitors Narrows by temperature coefficient Empty on all 929 rows
Thickness on MLCC Narrows by height Empty on all 24,026 rows

What should a buyer in the United States or Germany do instead of filtering?

Cross the part, do not search for it. An engineer in Texas or North Rhine-Westphalia filtering a 0603 10 kΩ position gets thirteen Walsin answers and no basis to choose between them, because the seven filterable fields are identical on all thirteen. The useful operation is the opposite direction: start from the part you already have and ask what differs.

  1. Send the incumbent part number, not a specification. A specification lands you in a signature group of up to fifteen. A part number lets us tell you which of the group matches on construction, grade and finish — the things living in the free-text column.
  2. Cross the whole list in one pass. Paste up to 500 part numbers or drop an Excel or CSV file into the free cross-reference tool. It is matched against Walsin’s published catalogue and states every difference from your original rather than hiding it, with a commercial or automotive grade filter and special-requirement filters for soft termination, high temperature, safety certified, high Q / low ESR and low profile.
  3. Ask what is not in the data. No export column anywhere states RoHS status, REACH, lifecycle, lead time, operating temperature or qualification status. UL file E182369 covers Walsin’s safety capacitor line and is searchable in UL’s own public database in about thirty seconds — check us on that, then ask us what else we hold. REACH, SVHC, PFAS, PPAP, IMDS and a Declaration of Conformity appear nowhere in the 451 pages on file, and we would rather say so now.
  4. Then check Hong Kong stock on ready stock. It is a snapshot dated 23 September 2026, not a live feed, and 254 of its 1,125 codes are chip resistors.

For what these files can and cannot answer line by line, see what Walsin’s 203 export files can and cannot answer. For the part-number formatting that breaks a BOM import, see why Walsin part numbers do not paste into your BOM. For the column-offset defect behind the multiplexer figures above, see the multiplexer export offset by one column. The precision line is on the Walsin thin film page.

The short version

A parametric selector is only as sharp as the number of columns that vary, and on this catalogue fifteen of eighty do not vary at all. Termination is one value across 328,729 resistor rows. Thin film Packaging is one value across 93,914. Varistor Tolerance is one value across 379. Eleven columns on three lines are declared and entirely empty, including T.C. on every disc capacitor and Thickness on every MLCC. RF impedance is 50 Ω on all 213 parts in five files. And when you strip the free text, 193,724 of 234,815 chip resistor part numbers share their full parametric signature with at least one other part — thirteen of them on the single 0603 10 kΩ position used above, one of which is an array the Size column does not flag.

Stop filtering and send us the part number. We will tell you which Walsin candidates share its signature, which of them differ in construction, grade or finish, and which was in Hong Kong on 23 September 2026 — then show you the differences rather than hide them. Which field on your selector have you been trusting that has only ever had one value in it?


Frequently asked questions

<!– wp:rank-math/faq-block {"questions":[{"id":"faq-question-01","title":"How many columns in Walsin's parametric exports carry only one value?","content":"Fifteen of eighty, across seven product lines and 359,517 rows. We re-counted the distinct values in every column. Four more are near-single: chip resistor Packaging at 93.9% one value, antenna VSWR at 32 of 44, inductor impedance at 100 MHz reading zero on 4,801 of 5,410 rows, and the MLCC Termination column which is blank on 95.4%.”,”visible”:true},{“id”:”faq-question-02″,”title”:”Does the Walsin chip resistor export distinguish parts by termination finish?”,”content”:”No. We read the Termination column on all 234,815 chip resistor rows and all 93,914 thin film rows — 328,729 rows, one value, Sn base (Lead free). It cannot narrow a result set by a single part. We hold no thin film datasheet, so that single value is unconfirmed against any Walsin document.”,”visible”:true},{“id”:”faq-question-03″,”title”:”Why do several Walsin part numbers come back for the same specification?”,”content”:”Because only seven of the chip resistor export’s ten columns are filterable, and 193,724 of 234,815 part numbers share their seven-field signature with at least one other part. There are 94,139 distinct signatures in total and the largest group holds 15 part numbers. The difference between them sits in a free-text series string.”,”visible”:true},{“id”:”faq-question-04″,”title”:”Which Walsin MLCC columns are empty?”,”content”:”Packing, SPQ and Thickness, on all 24,026 rows. We counted non-null values, not distinct values. The export does carry L, W and T dimension strings on 23,679 rows, so height data exists — just never in the column named Thickness. Termination is populated on only 1,106 of 24,026.”,”visible”:true},{“id”:”faq-question-05″,”title”:”Can I filter Walsin RF parts by impedance?”,”content”:”Not usefully. We opened all seven RF files and the impedance field reads 50 on all 17 couplers, all 72 baluns, all 116 multiplexers, all 5 BAW filters and all 3 SAW filters. Across 213 parts it never separates two candidates. Category also reads LTCC on all 17 couplers and 233 of 235 filters.”,”visible”:true},{“id”:”faq-question-06″,”title”:”Does the Size column show that a Walsin resistor is an array?”,”content”:”On 943 of 10,022 array rows, yes. On the other 9,079 the Size column reports a plain discrete code — 0603(1608) or 0402(1005) — so a four-element array looks like a single 0603 part. One 0602*8 family appears both ways, split by whether its series name capitalises \”Type\”.”,”visible”:true},{“id”:”faq-question-07″,”title”:”What should I send Venatech instead of a filtered list?”,”content”:”The part number you already have. A specification puts you in a signature group of up to fifteen Walsin parts with nothing to choose between them. A part number lets us match on construction, grade and finish, state every difference, and tell you what was in Hong Kong on the dated snapshot.”,”visible”:true}],”listStyle”:””,”titleWrapper”:”h3″,”titleCssClasses”:””,”contentCssClasses”:””,”sizeSlug”:””} –>

How many columns in Walsin’s parametric exports carry only one value?

Fifteen of eighty, across seven product lines and 359,517 rows. We re-counted the distinct values in every column. Four more are near-single: chip resistor Packaging at 93.9% one value, antenna VSWR at 32 of 44, inductor impedance at 100 MHz reading zero on 4,801 of 5,410 rows, and the MLCC Termination column which is blank on 95.4%.

Does the Walsin chip resistor export distinguish parts by termination finish?

No. We read the Termination column on all 234,815 chip resistor rows and all 93,914 thin film rows — 328,729 rows, one value, Sn base (Lead free). It cannot narrow a result set by a single part. We hold no thin film datasheet, so that single value is unconfirmed against any Walsin document.

Why do several Walsin part numbers come back for the same specification?

Because only seven of the chip resistor export’s ten columns are filterable, and 193,724 of 234,815 part numbers share their seven-field signature with at least one other part. There are 94,139 distinct signatures in total and the largest group holds 15 part numbers. The difference between them sits in a free-text series string.

Which Walsin MLCC columns are empty?

Packing, SPQ and Thickness, on all 24,026 rows. We counted non-null values, not distinct values. The export does carry L, W and T dimension strings on 23,679 rows, so height data exists — just never in the column named Thickness. Termination is populated on only 1,106 of 24,026.

Can I filter Walsin RF parts by impedance?

Not usefully. We opened all seven RF files and the impedance field reads 50 on all 17 couplers, all 72 baluns, all 116 multiplexers, all 5 BAW filters and all 3 SAW filters. Across 213 parts it never separates two candidates. Category also reads LTCC on all 17 couplers and 233 of 235 filters.

Does the Size column show that a Walsin resistor is an array?

On 943 of 10,022 array rows, yes. On the other 9,079 the Size column reports a plain discrete code — 0603(1608) or 0402(1005) — so a four-element array looks like a single 0603 part. One 0602*8 family appears both ways, split by whether its series name capitalises “Type”.

What should I send Venatech instead of a filtered list?

The part number you already have. A specification puts you in a signature group of up to fifteen Walsin parts with nothing to choose between them. A part number lets us match on construction, grade and finish, state every difference, and tell you what was in Hong Kong on the dated snapshot.


Share This, Choose Your Platform!