Thin Film Resistor TCR by Resistance Decade: ±3 ppm/°C Reaches 868 of 1,181 Values

Thin film resistor TCR by resistance decade is not flat. We counted all 1,181 resistance values in Walsin’s thin film export: 868 can be had at ≦±3 ppm/°C and 313 cannot. ≦±25 ppm/°C reaches all 1,181. The tight grades start at 4.70 Ω and stop at 120 kΩ.

---

Your selector shows six TCR grades and lets you tick the tightest one at any resistance. At 2 MΩ that tick returns nothing, and at 2.2 Ω it returns nothing either. The grade menu is real; it just does not apply across the whole resistance range.

Venatech Limited, an authorised Walsin distributor in Hong Kong, holds all 20 thin film parametric export files from Walsin Technology (Walsin). This post is measured from the eleven non-empty ones — the eight Thin_Film_Precision_Chip_Resistor_WF_Series_*.xlsx sheets, Tantalum_nitride_Thin_Film_Chip_Resistor_MF_Series_1.xlsx and two Anti_Sulfuration_Thin_Film_Chip_Resistor_SF_Series_*.xlsx sheets, 93,914 rows in total. We parsed the Resistance column on every row and grouped it into decades. Every part number below is real, and every count is a count of part numbers rather than pieces.

Which TCR grades exist at each resistance decade?

All six grades reach four of the seven decades, and the two extreme decades break. We grouped all 93,914 rows into decades from 1 Ω to 10 MΩ and counted how many distinct resistance values in each decade carry each grade. From 10 Ω to 100 kΩ, every grade reaches every value. Below 10 Ω and above 1 MΩ, the tight grades thin out hard.

Resistance decade	Rows	Distinct values	Values at ≦±3	at ≦±5	at ≦±10	at ≦±25
1 Ω – 10 Ω	2,518	108	32	32	32	108
10 Ω – 100 Ω	18,136	205	205	205	205	205
100 Ω – 1 kΩ	20,065	205	205	205	205	205
1 kΩ – 10 kΩ	20,070	205	205	205	205	205
10 kΩ – 100 kΩ	17,257	205	205	205	205	205
100 kΩ – 1 MΩ	14,304	205	16	159	205	205
1 MΩ – 10 MΩ	1,564	48	0	0	19	48

We have not seen this counted anywhere. Four of the seven decades — 10 Ω to 100 kΩ, 820 of the 1,181 values — are complete: every TCR grade, every value. The other three are not, and the failure is at both ends rather than only at the top. The column that matters adds up to 868 of 1,181 values at ≦±3 ppm/°C, which is 73.5% of the resistance ladder, not all of it.

Can I get ±25 ppm/°C at the value my divider needs?

Yes, at every value in the catalogue. We checked all 1,181 distinct resistance values against the ≦±25 ppm/°C grade and found a part number at every single one, across 32,889 rows. The same is true of ≦±50 ppm/°C. If your divider tolerates 25 ppm/°C, resistance range stops being a constraint in this catalogue entirely.

That is the useful half of the finding. Many instrumentation front ends are specified at 25 ppm/°C, and at that grade the whole 1 Ω to 3 MΩ span is open. The constraint only appears when the requirement tightens: ≦±10 ppm/°C covers 1,076 values, ≦±5 covers 1,011, and ≦±3 covers 868.

Where exactly do the tight TCR grades stop?

≦±3 ppm/°C exists only between 4.70 Ω and 120 kΩ. We sorted every row by resistance inside each grade and read off the first and last value. Each tighter grade is a shorter ladder than the one below it, and all four tight grades share the same bottom rung at 4.70 Ω while their ceilings differ by a factor of twenty-five.

TCR grade	Rows	Lowest value	Highest value	Distinct values	Of 1,181
≦±3 ppm/°C	3,064	4.70 Ω	120 kΩ	868	73.5%
≦±5 ppm/°C	6,007	4.70 Ω	597 kΩ	1,011	85.6%
≦±10 ppm/°C	11,812	4.70 Ω	1.50 MΩ	1,076	91.1%
≦±15 ppm/°C	11,808	4.70 Ω	1.50 MΩ	1,076	91.1%
≦±25 ppm/°C	32,889	1.00 Ω	3.00 MΩ	1,181	100%
≦±50 ppm/°C	28,334	1.00 Ω	3.00 MΩ	1,181	100%

The 120 kΩ ceiling on ≦±3 ppm/°C is reached by exactly 16 part numbers above 100 kΩ, all of them 1206, all at ±0.01% tolerance and 1/4 W. WF12B1003TTL is the 100 kΩ row and WF12B1203TTL is the last one on the ladder. We have already published the 1206 side of that ceiling in our precision divider post; what the decade cut adds is that no other case size gets anywhere near it.

What happens to TCR below 10 Ω?

It collapses, and it collapses in a way the row count hides. The 1 Ω to 10 Ω decade holds 2,518 rows, which looks healthy. We counted the distinct resistance values inside it: 108, of which only 32 reach better than ≦±25 ppm/°C. Those 32 all sit at 4.70 Ω and above. Below 4.70 Ω there is no thin film part at ≦±3, ≦±5, ≦±10 or ≦±15 ppm/°C at any case size.

Inside that decade sits a band we have not seen described. From 1.74 Ω to 4.64 Ω — 51 consecutive E96 values — the export holds 408 rows and every one of them carries ±0.01% tolerance and nothing else, paired with ≦±25 or ≦±50 ppm/°C and nothing else. All 408 are WF precision parts, split exactly 204 at 1206 and 204 at 2512. WF12Q1R74TTL is the first of them. So in that band Walsin sells the tightest tolerance in the catalogue welded to the two loosest TCR grades, with no option either way. It is the inverse of what a selector implies, and if your current-sense leg sits at 3.3 Ω it is the whole menu.

Does the anti-sulfuration SF series offer the same TCR menu?

No. It offers three grades of the six, and one of those three is four parts. We re-counted all 15,080 real SF rows: 9,418 at ≦±25 ppm/°C, 5,658 at ≦±50, and exactly 4 at ≦±10. There is no SF part at ≦±3, ≦±5 or ≦±15 ppm/°C anywhere in the series, at any resistance, at either of its two case sizes.

The four are SF12W10R5FTL, SF12W20R5FTL, SF12W30R5FTL and SF12W50R5FTL — 10.5 Ω, 20.5 Ω, 30.5 Ω and 50.5 Ω, all 1206, all 1/8 W, and all at ±1% tolerance, which is the loosest tolerance in the whole thin film catalogue. The tightest TCR the anti-sulfuration series reaches comes bundled with its worst tolerance, on four part numbers, in one decade. We have not seen this counted anywhere.

Sulfur resistance and low drift are both environmental requirements, so a designer tends to assume they stack. In this catalogue they do not. The tolerance letter map behind those codes is one we established earlier and it still holds across all 15,080 rows with no exception: A is ±0.05%, B is ±0.1%, C is ±0.25%, D is ±0.5% and F is ±1%. The nine-of-twenty empty thin film files are covered in our post on the empty SF A-series file.

What does one real 200 kΩ requirement return?

Seventy-seven part numbers, and none of them at ≦±3 ppm/°C. We filtered the whole export to 200 kΩ, a common top leg in a 10:1 instrumentation divider, and read every row. The tightest grade present is ≦±5 ppm/°C, available only at 2010 and 2512, because 200 kΩ sits above the 120 kΩ ceiling on ≦±3.

Part number	Series	Size	Tolerance	TCR	Rated power	Match / mismatch vs 200 kΩ, ≦±3 ppm/°C
— no row —	WF precision	any	—	≦±3	—	Mismatch — the ≦±3 ladder ends at 120 kΩ, so 200 kΩ has no row in that grade at any case size
WF20Z2003TTL	WF precision	2010	±0.01%	≦±5	3/4 W	Match on value and tolerance; mismatch — TCR 1.7× looser, and 2010 is four footprints larger than 0603
WF25Z2003TTL	WF precision	2512	±0.01%	≦±5	1 W	Match on value and tolerance; mismatch — TCR 1.7× looser, largest footprint in the line
WF12W2003TTL	WF precision	1206	±0.01%	≦±10	1/4 W	Match on value and tolerance; mismatch — TCR 3.3× looser
MF12W2003BTL	MF tantalum nitride	1206	±0.1%	≦±10	2/5 W	Mismatch on both — tolerance 10× wider and TCR 3.3× looser; different film material
SF20U2003BTL	SF anti-sulfuration	2010	±0.1%	≦±25	1/2 W	Mismatch — TCR 8.3× looser and tolerance 10× wider; adds sulfur resistance

The honest resolution is a grade change plus a footprint change, not a part swap. At ≦±5 ppm/°C both legs of a 200 kΩ over 20 kΩ divider do exist in one grade and one size: WF20Z2003TTL with WF20Z2002TTL at 2010, or WF25Z2003TTL with WF25Z2002TTL at 2512. At ≦±3 ppm/°C the bottom leg WF06B2002TTL is available at 0603 and the top leg is not available at all. We will not tell you a part is equivalent. We will show you where it differs.

Is any of this AEC-Q200, and does the export say so?

No, and the export contains a trap for anyone who checks by search. We read all nine column headers in each of the 20 thin film files and all 93,914 data rows. Not one row carries an AEC-Q200 label in any column. The file named Thin_Film_Professional_AEC_Q200_Chip_Resistor_WF_Q_S.xlsx opens to zero rows, as does Anti_Sulfuration_AEC_Q200_Thin_Film_Chip_Resistor_SF.xlsx, and both are 6,289-byte header-only workbooks.

Here is the trap. We searched every cell in the thin film export for the string Q200 and got 59 hits. All 59 are false. Each one is a part number — MF12Q2003ATL is a clear example — in which Q is the TCR and power code and 200 is the first three digits of the resistance code. Grep the export for AEC-Q200 compliance and you will find 59 parts that make no such claim. We have not seen this reported anywhere, and it is the kind of thing that puts a wrong line in a qualification matrix.

What should a precision designer in the United Kingdom do with the decade map?

Fix the resistance decade before the grade. United Kingdom instrumentation, metrology and test-equipment work runs on precision dividers and bridge legs, and the decade decides which grades are on the menu at all. If every leg sits between 10 Ω and 100 kΩ you have all six grades at all 205 values per decade. Push one leg to 200 kΩ and you lose ≦±3 ppm/°C; push it to 2 MΩ and you lose ≦±5 as well.

1. Write the decade, the grade and the footprint as one requirement. “±3 ppm/°C” alone is not buyable above 120 kΩ. “±3 ppm/°C at 1206, under 120 kΩ” is.
2. Scale the value if the grade is binding. A 20 kΩ over 2 kΩ divider has the same ratio as 200 kΩ over 20 kΩ and sits entirely inside the complete part of the ladder.
3. Cross the whole BOM in one pass. Paste up to 500 part numbers or drop your Excel or CSV into the free cross-reference tool. It decodes 18 manufacturers’ part numbers, is matched against Walsin’s published catalogue across 800,000+ crossable parts, and states every difference from your original rather than hiding it.
4. Check what is physically in Hong Kong on ready stock. That page is a dated snapshot as at 23 September 2026, not a live feed, and thin film has zero representation in it — ask us directly instead of assuming.
5. Ask us for the series datasheet. We hold none for thin film today, and your request is what puts it on our list to Walsin.

For the grade-against-tolerance view of the same data, see the thin film TCR and tolerance grid. The full range sits on the Walsin thin film range page.

The short version

Thin film resistor TCR by resistance decade is a staircase, not a plane. Four of seven decades — 10 Ω to 100 kΩ, 820 of 1,181 values — carry all six TCR grades at every value. ≦±3 ppm/°C lives only between 4.70 Ω and 120 kΩ and reaches 868 of 1,181 values; ≦±5 reaches 1,011; ≦±25 reaches all 1,181. Below 4.70 Ω nothing better than ≦±25 exists, and the 51 values from 1.74 Ω to 4.64 Ω carry ±0.01% tolerance and only the two loosest grades, on 408 rows split evenly between 1206 and 2512. The anti-sulfuration SF series offers three grades of six, and its ≦±10 grade is four part numbers at ±1%. No thin film row carries an AEC-Q200 label, and a text search for Q200 returns 59 false hits.

Send us the resistance value, the TCR grade and the footprint together, and we will tell you whether that cell holds a part number, which ones, and exactly where each differs from what you asked for. Qualification stays with you — we will not pretend otherwise.

So: when your selector offered you ±3 ppm/°C at 2 MΩ, what did it think it was selling you?

---

Frequently asked questions

---