Appliance Inverter Passive Components: Zero of Walsin’s 4,730 0603 MLCC Reach 500 V

We parsed all 24,026 rows of Walsin’s MLCC export. Not one of the 4,730 rows at 0603 reaches 500 Vdc; the ceiling there is 250 Vdc. For appliance inverter passive components on a 400 V DC link, 0805 is the smallest case size reaching 630 V, and the whole 630 V shelf stops at 470 nF.

---

Venatech Limited, an authorised Walsin distributor in Hong Kong, holds all 203 Walsin Technology (Walsin) parametric export files and 28 Walsin datasheets, 451 pages. This post is measured from 00_ALL_MLCC_parts_combined.xlsx, Thick_Film_Wide_Terminal_Power_Chip_Resistor__1of1.xlsx and Walsin_MOV_ALL.xlsx, plus the one datasheet we hold covering the capacitor line, 04_Middle_and_High_Voltage.pdf. Every part number below came out of those files.

What does a white goods inverter actually need from the passive BOM?

Three populations, and we hold very different amounts of data on each. A washing machine or heat-pump compressor drive rectifies mains to a 320–400 V DC link, switches it, and has to survive whatever arrives on the mains port. We counted what Walsin offers against that: 4,095 high-voltage MLCC rows from 200 V to 4 kV, 4,294 wide-terminal power resistor rows across three case sizes, and 379 varistor rows for which we hold no datasheet at all.

Population	Export rows	Datasheet we hold	What it covers on a drive
High-voltage MLCC, 200 V – 4 kV	4,095	04_Middle_and_High_Voltage.pdf, 13 pages	Snubbers, DC-link high-frequency bypass, DC blocking
Wide-terminal power resistors	4,294	none	Shunts, gate resistors, bleeders, snubber resistors
Metal oxide varistors	379	none	Mains port surge clamping

The documentation gap is the thing to notice first. We checked both chip resistor sheets we hold — Walsin_WR_General_Purpose_Chip_Resistor.pdf and Walsin_MR_Automotive_Thick_Film_Chip_Resistor.pdf — and neither mentions a wide-terminal size. On two of the three populations the export is all we have, and we say so.

How high does Walsin’s high-voltage MLCC shelf go at each case size?

0603 stops dead at 250 Vdc. We grouped all 24,026 MLCC rows by case size and took the highest rated voltage at each. Across the whole export, zero of 4,730 rows at 0603 and zero of 3,990 rows at 0402 reach 500 Vdc. 0805 reaches 1,000 Vdc, 1206 reaches 2,500 Vdc and 1812 is where the real high-voltage inventory lives. For a 400 V bus with any derating discipline, the design starts at 0805 whether the layout wanted it to or not.

Case size	Highest rated voltage in the export	Highest capacitance at 630 Vdc	Rows at 400 or 450 Vdc
0402 and 0603	250 Vdc	none	0
0805	1,000 Vdc	220 nF	0
1206	2,500 Vdc	150 nF	7
1210	2,000 Vdc	150 nF	15
1812	2,000 Vdc (high-voltage series)	470 nF	7

Two counts worth keeping. The 630 Vdc ceiling across all 637 rows at that voltage is 470 nF, on 1812B474K631, and at 1,000 Vdc it falls to 100 nF, on 1812B104K102. And the 400 V and 450 V rungs — the two that map most directly onto a rectified 230 V mains bus — hold 29 rows between them out of 4,095, every one X7R at 1206, 1210 or 1812. We have not seen that counted anywhere. A DC-link design assuming a 400 V-rated ceramic is picking from twenty-nine part numbers, not from a shelf.

Where does the wide-terminal power resistor ladder stop at 1225 and 2512?

At 2 W and 3 W respectively, and nothing in the catalogue goes higher. We measured the Rated Power column at both footprints. All 1,755 rows at 1225 are 2 W — 1,708 wide-terminal thick film plus 47 metal foil — with no 1 W or 3 W rung present at all. At 2512 the ladder has three rungs: 14,039 rows at 1 W, 6,249 at 2 W and 1,866 at 3 W. Those 1,866 are the only 3 W parts anywhere in the 234,815-row chip resistor export.

Footprint	Rated powers present	Rows at each	Where the ladder stops
1225	2 W only	1,755	One rung. No 1 W, no 3 W.
0612	1 W / 3/4 W	878 / 830	Two rungs, split by part-number prefix
1020	1 W only	878	One rung
2512	1 W / 2 W / 3 W	14,039 / 6,249 / 1,866	3 W, and that is the catalogue maximum

So a drive needing more than 2 W in a surface-mount resistor has exactly one footprint available, and it is 2512. The wide-terminal geometry buys better heat extraction into the board at a given area, but it does not buy a higher rated number in this export.

Why do two wide-terminal part numbers share one series name and one footprint?

Because the prefix carries a specification the Series column does not. We checked all 4,294 wide-terminal rows. At 1225, 830 part numbers beginning WF and 830 beginning WK cover the identical set of 830 resistance values at the identical 2 W, and both say “Thick Film Wide Terminal Power Chip-Resistor” in the Series column. The only exported difference is temperature coefficient: 738 of the 830 WF rows are graded ≦±100 ppm/°C, while all 830 WK rows are ≦±200.

Part number	Footprint	Resistance	Tolerance	TCR	Rated power	Match / mismatch vs 1.00 Ω ±1% 2 W at 1225
WF25L1R00FTLJ	1225 (3264)	1.00 Ω	±1%	≦±100	2 W	Match on all five; best TCR at this footprint
WK25L1R00FTL	1225 (3264)	1.00 Ω	±1%	≦±200	2 W	Match on resistance, tolerance, size, power; mismatch — twice the TCR, same series name
WK20L1R00FTL	1020 (2550)	1.00 Ω	±1%	≦±200	1 W	Mismatch — 1020 footprint, half the power
WK12L1R00FTL	0612 (1632)	1.00 Ω	±1%	≦±200	3/4 W	Mismatch — 0612 footprint, 3/4 W
WF12L1R00FTLJ	0612 (1632)	1.00 Ω	±1%	≦±100	1 W	Mismatch — 0612, but 1 W where WK is 3/4 W
WW25LR100FTL	1225 (3264)	0.100 Ω	±1%	0 ~ 350	2 W	Mismatch — sensing part, TCR stated one-sided

At 0612 the two prefixes mean something different again: WF is 1 W and WK 3/4 W on the same 830 values. At 1020 only WK exists. So the prefix means “better TCR” at 1225 and “higher rated power” at 0612, and the Series column does not say which rule is in force. We will not tell you a part is equivalent. We will show you where it differs — here, a single letter.

What does Walsin publish about varistors, and what do we actually hold?

A 379-row parametric export, and no datasheet whatsoever. We checked our library: zero varistor PDFs in 451 pages. Every number here is quoted as the export states it and nothing more. The export shows 17 part numbers at a maximum working voltage of 275 Vac, the usual rung for a 230 V mains port, and their stated peak current spans 400 A to 80,000 A — a 200-to-1 spread on one voltage rating.

Part number (275 Vac rung)	Series as exported	Size	Stated peak current	Match / mismatch vs a 230 V mains port clamp
SR431K14D	Standard (D Series)	14 mm	4,500 A	Match on working voltage; mid-range energy
SR431K20E	Super High Energy (V Series)	20 mm	13,000 A	Match on working voltage; mismatch — V series, but the code ends E
SR431K10E	Super High Energy (V Series)	10 mm	4,000 A	Match on working voltage; mismatch — smaller disc, lower energy
SR431K34RD	Industrial	34 mm square	80,000 A	Mismatch — industrial disc, far beyond a mains port
SR471K05D	Standard (D Series)	5 mm	400 A	Mismatch — 300 Vac rung, lowest energy in the family

One decode trap we measured and will not assert. All 60 part numbers Walsin’s export labels “Super High Energy (V Series)” end in the letter E, and not one of the 379 varistor part numbers ends in V. The series letter and the suffix do not agree, and the only way to tell a V-series part from an E-series part is the Series column plus the disc size. For the specifications themselves, go to Walsin’s own product pages — that is the only source either of us has.

What does the high-voltage MLCC datasheet settle that the export gets wrong?

The one specification that decides whether these parts can sit across the mains at all. We read 04_Middle_and_High_Voltage.pdf, 13 pages, revision ASC_Middle & High Voltage Sep.2026. It states an operating temperature of −55 to +125 °C, an X7R capacitance range of 100 pF to 1.0 µF, and this, verbatim: “Capacitors with X7R dielectrics are not intended for AC line filtering applications.” That sentence rules the whole 1,672-row X7R population out of the X-capacitor position on a mains input stage.

The sheet also publishes the ordering key, and the key disagrees with the export on five rows. It defines J as ±5%, K as ±10% and M as ±20%. We checked all 2,906 high-voltage rows whose code carries one of those three letters: 2,901 agree with the Tolerance column and five do not, among them 1206B473M631, 1210B274M451 and 1206N101K631. A 0.17% defect rate is low enough to trust the key and high enough to check your exact part number rather than assume it.

Two further defects sit in the same 4,095 rows. 1808B681M302 carries 680pFG in the Capacitance column — the Symbol column’s G concatenated onto the capacitance — and is marked a customised product. And 1210N103J132 states its rated voltage as 1300V where all 4,094 other high-voltage rows write Vdc. Neither will parse. Both are real parts.

What should an appliance manufacturer in Italy or Poland check before design-in?

Check the voltage ceiling at your footprint first, then the documentation, then the stock. A compressor-drive programme in Lombardy or Lower Silesia usually starts from a working BOM and asks whether a second source exists. The binding constraint is earlier than that: zero rows at 0603 reach 500 Vdc, so a 0603 position on a 400 V node cannot be crossed to Walsin at all, and that is better learned at schematic review than at first article.

1. Fix the case size against the voltage ceiling before the value. 0805 is the entry point at 630 Vdc; 1812 holds the 470 nF ceiling; 29 rows exist at 400 or 450 Vdc.
2. Ask for the paperwork you will be audited on. 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 it. REACH, SVHC, PFAS, PPAP, IMDS and a Declaration of Conformity appear nowhere in the 451 pages we hold.
3. Cross the whole BOM 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.
4. Then check what is in Hong Kong on ready stock, a snapshot dated 23 September 2026, not a live feed. We checked it: 732 of 1,125 codes are MLCC, 254 chip resistors, and exactly one a varistor — SR471K05D, which the snapshot files under “Chip resistor”.

See middle and high voltage MLCC, 200 V to 4 kV, wide terminal power resistors and how to choose an MOV varistor. The range is on the Walsin MLCC page.

The short version

An appliance inverter BOM hits three different data realities inside one Walsin catalogue. The high-voltage MLCC line is the best documented: 4,095 rows, a 13-page datasheet revised September 2026, and hard ceilings at 250 Vdc for 0603, 470 nF at 630 Vdc and 100 nF at 1,000 Vdc. The wide-terminal resistors are 4,294 rows with no datasheet we hold, a single 2 W rung at 1225, and a WF/WK prefix that changes meaning between footprints while the Series column stays identical. The varistors are 379 rows with no datasheet, a single-valued tolerance column, and 60 part numbers whose suffix contradicts their series name.

Send us the inverter BOM with the DC-link voltage and the footprint you are committed to, and we will come back with Walsin candidates field by field, every difference stated, every unconfirmed figure flagged — plus what was in Hong Kong on 23 September 2026. Which position on your board is still assuming a 0603 can hold 500 V?

---

Frequently asked questions

<!– wp:rank-math/faq-block {"questions":[{"id":"faq-question-01","title":"Can a Walsin 0603 MLCC be used on a 400 V DC link?","content":"No. We parsed all 24,026 MLCC rows and the highest rated voltage at 0603 is 250 Vdc, on zero rows above that. 0402 behaves the same way. 0805 is the smallest case size that reaches 630 Vdc, and 1812 is where the highest capacitance at high voltage sits.","visible":true},{"id":"faq-question-02","title":"What is the largest capacitance Walsin offers at 630 V?","content":"470 nF, on 1812B474K631. We measured the capacitance on all 637 rows at 630 Vdc. At 1,000 Vdc the ceiling falls to 100 nF on 1812B104K102. The datasheet 04_Middle_and_High_Voltage.pdf states an X7R range of 100 pF to 1.0 µF across the whole series, so the 630 V ceiling is well below the series maximum.”,”visible”:true},{“id”:”faq-question-03″,”title”:”Can Walsin’s high-voltage X7R MLCC be used as a mains X-capacitor?”,”content”:”Not according to Walsin’s own sheet. Page 2 of 04_Middle_and_High_Voltage.pdf states that capacitors with X7R dielectrics are not intended for AC line filtering applications. That rules out all 1,672 X7R rows in the series for the X-capacitor position. Ask us about the safety-certified series instead, and request the certificate for your part number.”,”visible”:true},{“id”:”faq-question-04″,”title”:”What rated powers exist at 1225 in Walsin’s chip resistor catalogue?”,”content”:”One: 2 W. We checked all 1,755 rows at that footprint — 1,708 wide-terminal thick film and 47 metal foil — and every one is 2 W. There is no 1 W or 3 W rung at 1225. At 2512 there are three rungs: 1 W, 2 W and 3 W, and the 1,866 parts at 3 W are the only ones in the whole 234,815-row export.”,”visible”:true},{“id”:”faq-question-05″,”title”:”What is the difference between a Walsin WF and WK wide-terminal part number?”,”content”:”It depends on the footprint, and the Series column does not say. At 1225 both prefixes cover the same 830 resistance values at 2 W, but 738 of the 830 WF rows are graded ≦±100 ppm/°C against ≦±200 for all WK rows. At 0612 the WF codes are rated 1 W and the WK codes 3/4 W. We hold no datasheet confirming either reading.”,”visible”:true},{“id”:”faq-question-06″,”title”:”Does Venatech hold a datasheet for Walsin varistors, and why do the V-series codes end in E?”,”content”:”No datasheet at all — we hold 28 Walsin sheets, 451 pages, and not one covers varistors. So every varistor figure we publish is the export’s own value, flagged as unconfirmed. We also measured that all 60 codes labelled \”Super High Energy (V Series)\” end in E, and none of the 379 ends in V. We flag that rather than decode it.”,”visible”:true}],”listStyle”:””,”titleWrapper”:”h3″,”titleCssClasses”:””,”contentCssClasses”:””,”sizeSlug”:””} –>

---