NEWS

MLCC Voltage Rating by Case Size: 243 of 400 Cells in Walsin’s Export Are Empty, Mapped for Netherlands Engineers

We counted MLCC voltage rating by case size across all 24,026 rows of Walsin’s MLCC export: 16 case sizes, 25 rated-voltage steps, 400 cells. Only 157 hold a part number. 243 are empty. We have not seen this matrix published anywhere.


Venatech Limited, an authorised Walsin distributor in Hong Kong, works from Walsin Technology (Walsin) data directly: the combined parametric export 00_ALL_MLCC_parts_combined.xlsx, 24,026 rows, plus 26 MLCC datasheet PDFs and the index INDEX_MLCC_series.csv. For this post we parsed the Rated Voltage field on every row, built the full case-size-by-voltage matrix ourselves, and then checked it against five real rail voltages. Every part number below is real.

Which rated-voltage steps actually exist in Walsin’s MLCC export?

25 of them, and they are not evenly spaced. We read the Rated Voltage field on all 24,026 rows. Twenty-four are written in the form NNNVdc. One is not: 1210N103J132 reads 1300V, with the dc missing, and it is the only such row in the file.

Rated voltage Part numbers Case sizes offering it Ratio to previous step
4Vdc 43 8 of 16 —
6.3Vdc 382 7 1.57x
10Vdc 1,468 8 1.59x
16Vdc 2,115 8 1.60x
25Vdc 3,206 10 1.56x
35Vdc 16 5 1.40x
50Vdc 6,582 12 1.43x
100Vdc 3,110 12 2.00x
200Vdc 1,066 10 2.00x
250Vdc 1,875 11 1.25x
400Vdc 11 2 1.60x
450Vdc 26 3 1.12x
500Vdc 1,143 9 1.11x
630Vdc 969 8 1.26x
1000Vdc 861 8 1.59x
1250Vdc 2 1 1.25x
1300V 1 1 1.04x
1500Vdc 61 7 1.15x
2000Vdc 511 7 1.33x
2500Vdc 29 4 1.25x
3000Vdc 316 5 1.20x
3500Vdc 1 1 1.17x
4000Vdc 59 5 1.14x
5000Vdc 163 4 1.25x
6000Vdc 10 1 1.20x

The two widest multiplicative steps in the entire ladder are 50 to 100 Vdc and 100 to 200 Vdc, both exactly 2.00x. Every other step below 1 kV moves by between 1.11x and 1.60x. Both of the 2.00x steps sit directly on top of the rail voltages a power engineer actually designs to, and that is where the trouble is.

How many case size and voltage combinations hold no part number at all?

243 of 400. We crossed the 16 case sizes in the export against the 25 rated-voltage steps and counted the part numbers in every cell. 157 cells hold at least one part number; 243 hold none. That is 60.75% of the grid empty, and the emptiness is structural rather than random — small cases cluster at low voltage, large cases at high voltage, and the overlap is thin.

Case size Rows in export Voltage steps present, of 25 Lowest step Highest step
01R5 608 6 4Vdc 50Vdc
0201 2,544 7 4Vdc 100Vdc
0306 3 1 4Vdc 4Vdc
0402 3,991 10 4Vdc 250Vdc
0505 97 2 100Vdc 250Vdc
0603 4,730 10 4Vdc 250Vdc
0805 4,912 13 4Vdc 1000Vdc
1111 149 6 50Vdc 1500Vdc
1206 3,520 18 4Vdc 2500Vdc
1210 1,370 19 4Vdc 2000Vdc
1808 420 9 500Vdc 5000Vdc
1812 1,020 18 10Vdc 5000Vdc
1825 137 10 50Vdc 4000Vdc
2211 57 2 5000Vdc 6000Vdc
2220 327 14 25Vdc 5000Vdc
2225 141 12 25Vdc 4000Vdc

No case size in Walsin’s MLCC catalogue carries more than 19 of the 25 steps. 1210 carries the most at 19, 1206 and 1812 carry 18, and 0306 carries exactly one — 4Vdc, on three part numbers. The two most heavily populated footprints in the export, 0805 with 4,912 rows and 0603 with 4,730, carry 13 and 10 steps respectively.

Which case sizes carry a usable spread of voltage steps, and which carry almost none?

Four of the sixteen are effectively single-purpose. 0306 holds 3 rows at one voltage. 0505 holds 97 rows at two steps, 100Vdc and 250Vdc. 2211 holds 57 rows at 5000Vdc and 6000Vdc only, all safety-certified S2 parts covered by WTC_MLCC_Safety_Certified_S2.pdf. 1111 holds 149 rows over six steps from 50Vdc to 1500Vdc, which looks broad until you see the counts: 3 rows at 100Vdc, 1 row at 250Vdc.

Now take five real rail voltages with a 2:1 derating rule and ask where the ladder’s own step is missing.

Rail Step a 2:1 rule lands on Case sizes with a part at that exact step Case sizes with nothing at that step
12 V 25Vdc 10 of 16 0306, 0505, 1111, 1808, 1825, 2211
24 V 50Vdc 12 of 16 0306, 0505, 1808, 2211
48 V 100Vdc 12 of 16 01R5, 0306, 1808, 2211
60 V 200Vdc 10 of 16 01R5, 0201, 0306, 0505, 1808, 2211
100 V 200Vdc 10 of 16 01R5, 0201, 0306, 0505, 1808, 2211

Two readings of that table are both honest and they differ. Filtered on the exact step — which is what a parametric selector does — a 60 V rail loses 0201 and 01R5 entirely. Filtered on at or above the derated figure, which is what the circuit requires, a 60 V rail loses 0306, 01R5 and 0201: three of the sixteen footprints, because 0201 tops out at 100Vdc across all 2,544 of its rows and 01R5 tops out at 50Vdc across all 608. A 48 V rail loses 0306 and 01R5 only, and keeps 0201. Everything at 200Vdc and above is governed by WTC_MLCC_Middle_and_High_Voltage.pdf, whose published range INDEX_MLCC_series.csv records as 200 V to 4000 V; everything at 100Vdc and below is in WTC_MLCC_General_Purpose.pdf at 6.3 V to 100 V.

What does a 60 V rail cost you in capacitance if you derate two to one?

Between 2.2x and everything, depending on the footprint. We took the deepest capacitance available at 100Vdc and at 200Vdc for every case size that offers both, decoding each capacitance from the part number itself. At 0402 the ceiling falls from 4.7 nF to 100 pF — a factor of 47 — for the same board area.

Case size Deepest capacitance at 100Vdc Part number Deepest at 200Vdc Part number Match / mismatch
0201 220 pF 0201N221J101 nothing — 0 of 2,544 rows reach 200Vdc — Mismatch — the footprint has no 200Vdc option at all
0402 4.7 nF 0402B472K101 100 pF 0402N101J201 Mismatch — 47x less capacitance, and NP0 replaces X7R
0603 100 nF MT18B104K101 10 nF 0603B103K201 Mismatch — 10x less; match on size, X7R and tolerance band
0805 1 µF 0805A105K101 100 nF 0805B104K201 Mismatch — 10x less; also X7S to X7R, so the dielectric changes
1206 2.2 µF MT31B225K101 220 nF 1206B224K201 Mismatch — 10x less; AEC-Q200 MT part replaced by a general part
1210 10 µF SH32B106K101 680 nF 1210B684K201 Mismatch — 14.7x less; soft-termination replaced by standard
1812 2.2 µF 1812B225J101 1 µF 1812B105K201 Mismatch — 2.2x less; match on size and thickness, 2.50 mm both
2220 22 µF 2220B226J101 2.2 µF 2220B225K201 Mismatch — 10x less; tolerance moves from ±5% to ±10%

We will not tell you a part is equivalent. We will show you where it differs — and across this whole table the only field that matches cleanly is the footprint. Note the 1812 row: it is the mildest penalty in the catalogue, 2.2x rather than 10x, and 1812B105K201 holds 1 µF at 200Vdc in a 2.50 mm thickness, a figure we read from the T column of 00_ALL_MLCC_parts_combined.xlsx.

Between 250Vdc and 500Vdc the catalogue thins to almost nothing: 37 part numbers in total, 11 at 400Vdc across 1206 and 1210 and 26 at 450Vdc across 1206, 1210 and 1812. For a 230 Vac mains input stage, which sees roughly 325 V peak, that is the band where the export has least to offer and the 500Vdc class with its 1,143 part numbers is the practical shelf.

Which voltage steps are held by one or two part numbers?

Three of the 25. We counted part numbers per step: 1250Vdc holds 2 rows, 3500Vdc holds 1, and the malformed 1300V entry holds 1. Between them, four part numbers out of 24,026 occupy three of the twenty-five rungs on the ladder. A selector that lists 25 voltage options is telling you the catalogue is three times deeper in choices than it is.

Step Part numbers Case sizes offering it
1000Vdc 861 8
1250Vdc 2 1210 only
1300V 1 1210 only, on 1210N103J132
1500Vdc 61 7
3500Vdc 1 1812 only

35Vdc is the same story at the other end: 16 part numbers across five case sizes, against 6,582 at 50Vdc. Walsin’s own ordering key puts the voltage in the last three characters, so 500 is 50Vdc and 201 is 200Vdc — see how to read a Walsin MLCC part number for the whole twelve-character scheme.

What should a Netherlands engineer do with the gaps in the voltage ladder?

Pick the step before you pick the footprint. Three steps. The matrix above says the voltage class decides which case sizes remain available to you, not the other way round, and at 200Vdc it removes 0201 and 01R5 from the conversation across 3,152 rows.

  1. Fix the rated-voltage step first, then read which footprints survive it. At 200Vdc that is 10 of 16.
  2. Cross the whole list in one pass. Paste up to 500 part numbers or drop your Excel or CSV into the free cross-reference tool, filter by commercial or automotive grade, and read every stated difference rather than a claimed match.
  3. Check what is physically in Hong Kong on ready stock before committing a build slot — a dated snapshot as at 23 September 2026, not a live feed.

For how capacitance behaves once the DC bias is applied on top of the voltage class, see MLCC DC bias and voltage derating. The full line is on the Walsin MLCC page.

The short version

Walsin’s MLCC export carries 25 rated-voltage steps across 16 case sizes. We counted every cell: 157 hold a part number and 243 do not. No case size carries more than 19 of the 25 steps. The two widest steps in the ladder, 50 to 100 Vdc and 100 to 200 Vdc, are both exactly 2.00x and both land on real rail voltages. A 60 V rail derated two to one has to sit at 200Vdc, which has no part at 0201 across all 2,544 of its rows and none at 01R5 across all 608, and which costs 47x the capacitance at 0402 against the 100Vdc shelf. Three of the 25 steps are held by four part numbers between them, and one row, 1210N103J132, is the only one in 24,026 whose Rated Voltage omits the dc suffix. Qualification stays with you.

Send us the rail voltages off your Netherlands design — not the part numbers, the rails — and we will tell you which Walsin case sizes survive your own derating policy, what the capacitance ceiling is at each one, and what was physically in Hong Kong on 23 September 2026. And if your selector offers you 25 voltage options, ask it how many part numbers sit behind the one you just ticked.


Frequently asked questions

<!– wp:rank-math/faq-block {"questions":[{"id":"faq-question-01","title":"How many rated-voltage steps does Walsin's MLCC range actually have?","content":"25, measured across all 24,026 rows of the combined MLCC export. Twenty-four are written as NNNVdc. The twenty-fifth, on 1210N103J132, reads 1300V with no dc, and it is the only row in the file formatted that way. Three steps are held by four part numbers between them.”,”visible”:true},{“id”:”faq-question-02″,”title”:”Why is there nothing between 100 Vdc and 200 Vdc?”,”content”:”Because the ladder steps by a factor of two there, and so does 50 to 100 Vdc. Those are the two widest multiplicative steps in the whole range; every other step below 1 kV moves by 1.11x to 1.60x. Nothing in the export sits at 120, 150 or 160 Vdc.”,”visible”:true},{“id”:”faq-question-03″,”title”:”Can I get a 200 Vdc MLCC at 0201?”,”content”:”No. We checked all 2,544 0201 rows in the export: the highest rated voltage at that case size is 100Vdc, on 101 part numbers, the deepest of which is 220 pF. 01R5 tops out at 50Vdc across all 608 of its rows. A 200Vdc requirement starts at 0402.”,”visible”:true},{“id”:”faq-question-04″,”title”:”What does moving from 100 Vdc to 200 Vdc cost in capacitance?”,”content”:”At 0402, a factor of 47: the ceiling falls from 4.7 nF to 100 pF. At 0603, 0805, 1206 and 2220 it is a factor of ten. At 1812 it is only 2.2x, from 2.2 µF to 1 µF, which makes 1812 the mildest place in the catalogue to take the voltage class up.”,”visible”:true},{“id”:”faq-question-05″,”title”:”Which case sizes are nearly single-purpose in this export?”,”content”:”Four. 0306 holds three part numbers at one voltage step. 0505 holds 97 rows at 100Vdc and 250Vdc only. 2211 holds 57 rows at 5000Vdc and 6000Vdc, all safety-certified. 1111 spans six steps but with 3 rows at 100Vdc and 1 at 250Vdc.”,”visible”:true},{“id”:”faq-question-06″,”title”:”Where does the catalogue thin out for a 230 Vac mains input stage?”,”content”:”Between 250Vdc and 500Vdc. We counted 11 part numbers at 400Vdc and 26 at 450Vdc, 37 in total, against 1,875 at 250Vdc and 1,143 at 500Vdc. Confirm the working voltage against the series datasheet rather than the export, and qualify in your own circuit.”,”visible”:true}],”listStyle”:””,”titleWrapper”:”h3″,”titleCssClasses”:””,”contentCssClasses”:””,”sizeSlug”:””} –>

How many rated-voltage steps does Walsin’s MLCC range actually have?

25, measured across all 24,026 rows of the combined MLCC export. Twenty-four are written as NNNVdc. The twenty-fifth, on 1210N103J132, reads 1300V with no dc, and it is the only row in the file formatted that way. Three steps are held by four part numbers between them.

Why is there nothing between 100 Vdc and 200 Vdc?

Because the ladder steps by a factor of two there, and so does 50 to 100 Vdc. Those are the two widest multiplicative steps in the whole range; every other step below 1 kV moves by 1.11x to 1.60x. Nothing in the export sits at 120, 150 or 160 Vdc.

Can I get a 200 Vdc MLCC at 0201?

No. We checked all 2,544 0201 rows in the export: the highest rated voltage at that case size is 100Vdc, on 101 part numbers, the deepest of which is 220 pF. 01R5 tops out at 50Vdc across all 608 of its rows. A 200Vdc requirement starts at 0402.

What does moving from 100 Vdc to 200 Vdc cost in capacitance?

At 0402, a factor of 47: the ceiling falls from 4.7 nF to 100 pF. At 0603, 0805, 1206 and 2220 it is a factor of ten. At 1812 it is only 2.2x, from 2.2 µF to 1 µF, which makes 1812 the mildest place in the catalogue to take the voltage class up.

Which case sizes are nearly single-purpose in this export?

Four. 0306 holds three part numbers at one voltage step. 0505 holds 97 rows at 100Vdc and 250Vdc only. 2211 holds 57 rows at 5000Vdc and 6000Vdc, all safety-certified. 1111 spans six steps but with 3 rows at 100Vdc and 1 at 250Vdc.

Where does the catalogue thin out for a 230 Vac mains input stage?

Between 250Vdc and 500Vdc. We counted 11 part numbers at 400Vdc and 26 at 450Vdc, 37 in total, against 1,875 at 250Vdc and 1,143 at 500Vdc. Confirm the working voltage against the series datasheet rather than the export, and qualify in your own circuit.


Share This, Choose Your Platform!