A selector that filters resistor TCR on the string “≦±100” returns 38,297 Walsin rows. We counted 64,873 that are 100 ppm per degree Celsius or better. The filter hides 26,576 rows, 41 percent of the qualifying set, and the loss differs in every resistance band. This is the count for UK buyers.
Venatech Limited, an authorised Walsin distributor in Hong Kong, holds the chip resistor parametric export published by Walsin Technology (Walsin): 137 files and 234,815 rows. This post is written from the first 68 files alphabetically, which we opened one by one. 37 of them return zero rows and 31 hold 81,249. We grouped the resistive rows by decade of resistance and counted the TCR string on each. Every part number below is real. The question is the one a UK purchasing filter gets wrong: if I ask for 100 ppm or better, what comes back?
How does the published TCR tier change with resistance?
It narrows through the middle decades and widens at both ends. We parsed all 81,249 rows in the 68 files and removed 23 chip attenuators, which carry decibels or a dash in the TCR column, and 18 zero-ohm jumpers, which publish a dash. That leaves 81,208 rows across 12 decade bands, and 12 distinct TCR strings: 10 numeric ceilings and 2 asymmetric ranges. Best published tier is plus or minus 25 ppm per degree Celsius between 100 ohm and 999 kilohm, and plus or minus 50 or worse outside it.
| Resistance band | Rows | TCR strings present | Best tier | Rows at ≦±50 or better | Rows reading exactly ≦±100 | Rows at ≦±100 or better | Match / mismatch of a literal ≦±100 filter |
|---|---|---|---|---|---|---|---|
| 0.1 to 0.99 mΩ | 10 | 3 | ≦±70 | 0 | 1 | 9 | Mismatch — returns 1 of 9 |
| 1 to 9.99 mΩ | 549 | 4 | ≦±50 | 121 | 70 | 549 | Mismatch — returns 70 of 549 |
| 10 to 99.9 mΩ | 569 | 4 | ≦±50 | 431 | 10 | 569 | Mismatch — returns 10 of 569 |
| 0.1 to 0.99 Ω | 480 | 2 | ≦±50 | 60 | 0 | 60 | Mismatch — returns 0 of 60 |
| 1 to 9.99 Ω | 4,114 | 8 | ≦±50 | 228 | 2,152 | 2,380 | Mismatch — returns 2,152 of 2,380; 655 rows hold a range string |
| 10 to 99.9 Ω | 13,160 | 6 | ≦±50 | 3,192 | 7,433 | 10,625 | Mismatch — returns 7,433 of 10,625 |
| 100 to 999 Ω | 14,622 | 5 | ≦±25 | 4,902 | 7,152 | 12,054 | Mismatch — returns 7,152 of 12,054 |
| 1 to 9.99 kΩ | 15,294 | 5 | ≦±25 | 5,802 | 6,924 | 12,726 | Mismatch — returns 6,924 of 12,726 |
| 10 to 99.9 kΩ | 15,192 | 5 | ≦±25 | 5,700 | 6,924 | 12,624 | Mismatch — returns 6,924 of 12,624 |
| 100 to 999 kΩ | 14,527 | 5 | ≦±25 | 5,376 | 6,832 | 12,208 | Mismatch — returns 6,832 of 12,208 |
| 1 to 9.99 MΩ | 2,410 | 4 | ≦±50 | 270 | 795 | 1,065 | Mismatch — returns 795 of 1,065 |
| 10 MΩ and above | 281 | 2 | ≦±100 | 0 | 4 | 4 | Match — returns 4 of 4 |
| All bands | 81,208 | 12 | 26,082 | 38,297 | 64,873 | Mismatch — returns 59.0% |
Below 1 ohm, the four bands hold 1,608 rows and a literal ≦±100 filter returns 81 of the 1,187 that qualify, 6.8 percent. That is current-sense territory, where the tightest tiers are ≦±50, ≦±70 and ≦±75, none of which a string match on ≦±100 will find. For how the current-sense shelf behaves on its own, see Zero of Walsin’s 10,428 thick film current sense parts reach ±50 ppm.
What does a literal 100 ppm filter actually return?
38,297 of the 64,873 rows that meet 100 ppm per degree Celsius, which is 59.0 percent. We counted every row whose ceiling is 100 or lower. The 26,576 that a string match misses are 22,995 rows at ≦±50, 3,087 at ≦±25, 438 at ≦±70 and 56 at ≦±75. A buyer who asks for 100 ppm or better gets the loosest qualifying tier, and the parts that beat the request are hidden.
| TCR string | Rows | Resistance bands where it appears | Where it lives | Match / mismatch against a request for 100 ppm or better |
|---|---|---|---|---|
| ≦±25 | 3,087 | 4 | High precision, one series | Match — qualifies; missed by a literal filter |
| ≦±50 | 22,995 | 10 | High precision 22,383; current sense 612 | Match — qualifies; missed by a literal filter |
| ≦±70 | 438 | 3 | Current sense only | Match — qualifies; missed by a literal filter |
| ≦±75 | 56 | 2 | Current sense only | Match — qualifies; missed by a literal filter |
| ≦±100 | 38,297 | 11 | High precision 23,118; anti-surge 13,817; arrays 571; high power 710; current sense 81 | Match — the only tier a literal filter finds |
| ≦±150 | 115 | 2 | Anti-surge 114; current sense 1 | Mismatch — looser than 100 |
| ≦±200 | 13,171 | 9 | Five of seven groups | Mismatch — looser than 100 |
| ≦±300, ≦±350, ≦±400 | 2,379 | 7 | Arrays only | Mismatch — looser than 100 |
| −200～+400, −300～+500 | 670 | 2 | High precision 585; arrays 85 | Mismatch — a range, not a ceiling, and it will not sort |
We have not seen this counted anywhere: the same request returns 59.0 percent of the qualifying set at 100 ppm, 88.2 percent at 50 ppm (22,995 of 26,082), and 100 percent at 25 ppm. The tighter the request, the more of it a literal match finds, because fewer tiers sit beneath it.
Where does the 25 ppm tier exist at all?
In four resistance bands and one series. We checked every one of the 3,087 rows at ≦±25. All 3,087 sit between 100 ohm and 999 kilohm: 342 in the hundreds of ohms, 1,026 in the kilohms, 1,026 in the tens of kilohms and 693 in the hundreds of kilohms. None sits below 100 ohm or at 1 megohm and above, and all 3,087 belong to the High Precision TC25 series. A selector that offers 25 ppm at 47 ohm or 2.2 megohm is offering a value Walsin does not publish.
| Part number | Resistance | TCR in export | Series | Tolerance | Your incumbent | Match / mismatch against a request for ≦±25 at 100 kΩ, 0603, ±0.1%, 1/10 W |
|---|---|---|---|---|---|---|
WF06J1003BTL |
100 kΩ | ≦±25 | High Precision TC25 | ±0.1% | Match on every field | |
WF06K1003BTL |
100 kΩ | ≦±50 | High Precision TC50 | ±0.1% | Mismatch — TCR 2x looser | |
WF06H1003BTL |
100 kΩ | ≦±100 | High Precision TC100 | ±0.1% | Mismatch — TCR 4x looser | |
SF06H1003BTL |
100 kΩ | ≦±100 | High Precision Anti-Sulfuration | ±0.1% | Mismatch — TCR 4x looser |
All four parts share size, resistance, tolerance and rated power. The TCR is the only field that separates them, which is why we show it and do not tick it.
Where does a TCR selector fail to discriminate?
In 15 of the 46 combinations of series group and decade band, the column holds one string. We crossed the 7 resistor series groups with the 12 bands and counted the distinct TCR strings in each occupied cell. 15 of 46 cells carry one value, covering 1,559 rows. The High Power group holds ≦±100 on all 710 of its rows, and the Convex Type network holds ≦±200 on all 97.
Three more places where the column stops working as a selector:
- Anti-surge, 10 ohm to 99.9 kilohm. Four consecutive bands each hold exactly 2,454 rows at ≦±100 and 168 at ≦±200. The column splits the population identically in every band, so it tells you nothing about resistance.
- Asymmetric ranges. 670 rows hold −200～+400 or −300～+500. The glyph is a fullwidth tilde, and the string will not compare against a ceiling. 15 of them sit at exactly 10 ohm, the boundary the WR and MR sheets use for their low-resistance band.
- Dashes and decibels. 18 zero-ohm rows publish “-“, and 22 of the 23 chip attenuators put decibels in the TCR column.
Does one series name hold one TCR across every resistance band?
No. The High Precision TC50 series ships ≦±50 on 13,703 rows and ≦±100 on 1,094. We checked it band by band. TC50 is ≦±50 from 1 ohm to 999 kilohm on all but 866 rows, and at 1 megohm and above it splits 254 rows at ≦±50 and 228 at ≦±100. A series name is a family label, not a TCR guarantee.
| Part number | Series | Resistance | Tolerance | TCR in export | Your incumbent | Match / mismatch against a request for ≦±50 |
|---|---|---|---|---|---|---|
WF06K1000BTL |
High Precision TC50 | 100 Ω | ±0.1% | ≦±50 | Match | |
WF06K1003BTL |
High Precision TC50 | 100 kΩ | ±0.1% | ≦±50 | Match | |
WF06K1004BTL |
High Precision TC50 | 1 MΩ | ±0.1% | ≦±50 | Match | |
WK06K1504DTL |
High Precision TC50 | 1.5 MΩ | ±0.5% | ≦±100 | Mismatch — same series name, TCR 2x looser | |
WF04A1003FTL |
Thick Film High Power | 100 kΩ | ±1% | ≦±100 | Mismatch — one value on all 710 rows |
What should a UK buyer filter on instead?
On a number, with the original string kept beside it: parsed, a 100 ppm request returns 64,873 rows and not 38,297. Three rules, each measured above:
- Parse the ceiling to a number and filter “at or below”. A request for 100 ppm or better then returns 64,873 rows, not 38,297.
- Treat the two range strings as their own class. 670 rows will not parse, and all of them sit at or below 10 ohm.
- Read TCR with resistance, never alone. The 25 ppm tier exists in four bands of twelve, so a TCR request without a resistance value is half a specification.
Our free cross-reference tool states differences from your original part rather than hiding them behind a tick. For the structural reason tolerance and TCR codes do not join blind, see tolerance, TCR and dielectric codes decoded. What is in Hong Kong is on ready stock: 1,125 part numbers as at 23 September 2026, a dated snapshot. The range sits on the chip resistor page.
The short version
A literal ≦±100 filter returns 38,297 of the 64,873 Walsin rows in 68 files that are 100 ppm per degree Celsius or better, 59.0 percent, and the loss runs from 93 percent below 1 ohm to none at 10 megohm. The 25 ppm tier exists only between 100 ohm and 999 kilohm, on 3,087 rows in one series. 15 of 46 group-and-band cells publish one TCR string. 670 rows hold a range that will not sort. A series name is not a TCR: TC50 ships ≦±100 on 1,094 rows. We will not tell you a part matches on TCR. We will show you the string the file holds. Confirm against both datasheets and qualify in your own circuit.
Send us the TCR line and resistance range off your specification and the part numbers your filter returned, and we will tell you what it excluded. And a question back: when your selector returned nothing at 25 ppm, did you check the resistance value before you blamed the part?
Frequently asked questions
How many Walsin chip resistor rows meet 100 ppm per degree Celsius or better?
64,873 of 81,208 resistive rows in the 68 files we opened, 79.9 percent. A string filter on the text ≦±100 returns only 38,297 of them. The other 26,576 publish ≦±50, ≦±25, ≦±70 or ≦±75 and are hidden by a literal match.
Where does Walsin publish a TCR of 25 ppm per degree Celsius?
On 3,087 rows, all in the High Precision TC25 series, and all between 100 ohm and 999 kilohm. We found none below 100 ohm and none at 1 megohm or above. Request the datasheet or confirm with Walsin before specifying 25 ppm outside that range.
Does the TCR change with resistance in Walsin’s export?
Yes. The best published tier is ≦±25 between 100 ohm and 999 kilohm and ≦±50 elsewhere, and the widest asymmetric ranges at 1 to 9.99 ohm. The number of distinct TCR strings per band runs from 2 to 8, with the widest spread between 1 and 9.99 ohm.
What does −200～+400 mean in the TCR column?
It is a range, not a ceiling: the resistance change per degree Celsius may fall anywhere between minus 200 and plus 400 parts per million. 670 rows carry a range string, 585 in high precision and 85 in arrays. It will not sort or compare against a plus-or-minus ceiling.
Is a series named TC50 always 50 ppm?
No. In the High Precision TC50 series, 13,703 rows publish ≦±50 and 1,094 publish ≦±100. At 1 megohm and above the series splits 254 to 228. Read the TCR column for your exact part number and do not infer it from the series name.
Do the exports have a datasheet for these resistor groups?
No. We hold 28 Walsin datasheets, 451 pages, two of them for chip resistors, and those cover the WR general purpose and MR automotive series. None covers the high precision, anti-surge, high power, array, network or current sense groups, so the export is the only source we can show you.