48 V Bus MLCC: Zero of Walsin’s 24,026 Rows Are Rated 63 V or 80 V

We counted every rated voltage in Walsin’s MLCC export: 25 distinct classes across 24,026 rows, and the ladder jumps straight from 50 Vdc to 100 Vdc. A 48 V bus MLCC voltage rating therefore has one honest answer in this catalogue — 100 Vdc, which is 3,110 rows, or 12.9% of the shelf.

---

Venatech Limited, an authorised Walsin distributor in Hong Kong, holds all 203 of Walsin Technology (Walsin) parametric export files and 28 Walsin datasheets. This post is measured from 00_ALL_MLCC_parts_combined.xlsx and the 1,125-line ready-stock snapshot dated 23 September 2026. Every part number is real, every count is of part numbers rather than pieces, and two of them are there because the data is wrong.

Why does a 48 V bus have no natural MLCC voltage class?

Because the class that would serve it does not exist in this catalogue. We counted the Rated Voltage column across all 24,026 rows and found 25 distinct values: 4, 6.3, 10, 16, 25, 35, 50, 100, 200, 250, 400, 450, 500, 630, 1,000, 1,250, 1,300, 1,500, 2,000, 2,500, 3,000, 3,500, 4,000, 5,000 and 6,000 Vdc. There are zero rows at 63 Vdc and zero at 80 Vdc. The 35 Vdc class exists but holds only 16 rows.

A 48 V rail against a 50 Vdc part sits at 96% of rated voltage before any transient, and X7R and X5R dielectrics lose effective capacitance as DC bias rises, so the nominal figure on the line is not the capacitance you get. That is why 48 V designs normally reach for a 63 V or 80 V part. In Walsin’s catalogue that reach lands on nothing, and the next rung is 100 Vdc — a 2.08× derating ratio on a 48 V bus, which is generous on margin and expensive on case size.

Rated voltage class	Rows	Share of 24,026	Useful for a 48 V bus?
35 Vdc	16	0.07%	No — below the rail
50 Vdc	6,582	27.4%	96% of rating; no transient headroom
63 Vdc	0	0%	The class a 48 V design wants — absent
80 Vdc	0	0%	The second choice — also absent
100 Vdc	3,110	12.9%	Yes, at 2.08× derating
200 Vdc	1,066	4.4%	Yes, with a further case-size penalty

We have not seen this gap stated anywhere. It is not a shortage and not a lead-time problem; it is a catalogue structure that quietly forces a 48 V design onto a 100 V part.

How deep is Walsin’s 100 Vdc shelf?

Three thousand one hundred and ten rows, concentrated in the middle case sizes. We grouped the 100 Vdc rows by size: 0603 holds 879, 0805 holds 834, 1206 holds 486, 0402 holds 447, 1210 holds 175, 1812 holds 140 and 0201 holds 101. The remainder — 48 rows — sit at 2220, 2225, 1825, 1111 and 0505.

By dielectric the shelf is 2,164 NPO, 873 X7R, 67 X8G and a handful of X7S, X6S and X5R rows. That split matters because NPO at 100 V is almost entirely low-capacitance RF and timing work, while the bypass and bulk positions on a 48 V rail need X7R. So the usable 100 V bypass shelf is not 3,110 parts. It is the 873 X7R rows, and they are not evenly spread.

By series the 100 Vdc rows are 1,447 General Purpose, 434 Microwave RF, 376 Automotive MT, 360 High Q and Low ESR HH, 184 Soft Termination SH and 219 across the remaining six series. The soft-termination count matters on an automotive 48 V bus, because board flex is the failure mode soft termination addresses.

How much capacitance can you get at 100 V in a small case?

Far less than at 50 V, and the cliff is at 0402. We parsed the Capacitance field on every 100 Vdc row and the ceiling at 0402 is 4.7 nF, on 0402B472K101. Not one 0402 part at 100 Vdc reaches 100 nF. At 0603 exactly eight rows reach 100 nF and none exceeds it. At 0805, 27 rows reach 100 nF and three reach 1 µF.

Case size	Highest capacitance at 100 Vdc	Rows at 100 nF or above	Highest at 50 Vdc	Rows at 1 µF or above, 50 Vdc
0201	200 pF	0	10 nF	0
0402	4.7 nF	0	680 nF	0
0603	100 nF	8	2.2 µF	10
0805	1 µF	27	10 µF	23
1206	2.2 µF	75	10 µF	33
1210	10 µF	63	10 µF	45
1812	2.2 µF	55	10 µF	10
2220	22 µF	30	22 µF	13

Read the 0402 row twice. A 48 V design that wants a 100 nF bypass next to a controller pin cannot have it at 0402 from this catalogue, and at 0603 it has eight part numbers. Doubling the voltage class costs roughly two case sizes of capacitance density, before any DC bias derating.

Is there an AEC-Q200 option at 100 V?

Yes, and it is thin. The MT series — the one Walsin’s datasheets describe as qualified to AEC-Q200 — holds 376 rows at 100 Vdc, spread 146 at 0603, 108 at 0805, 79 at 1206, 33 at 0402 and 10 at 1210. We counted how many of those reach 100 nF: 17. The smallest is MT18B104K101, an 0603, and the largest capacitance in the whole MT 100 V set is 4.7 µF on MT32B475K101, a 1210.

Add the automotive soft-termination ST series and you gain 35 more rows at 100 Vdc. Walsin’s datasheets say AEC-Q200 compliant, not certified, and Venatech makes no qualification claim of its own — ask for the qualification report covering the exact part number. We hold 26 MLCC datasheets covering all 16 export series, including 11_Automotive_MT.pdf and 12_Soft_Term_Automotive_ST.pdf, and we will send the one covering your part number rather than describe it.

What does one 100 nF 100 V position look like across Walsin’s series?

Eight real parts against one requirement — 100 nF, 100 Vdc, 0603 or near. We pulled every 0603 row at 100 nF and 100 Vdc and there are exactly eight, across four series, with thickness bands that differ by a factor of three. We will not tell you a part is interchangeable. We will show you where it differs.

Part number	Series	Size	Capacitance	Tolerance	Thickness (mm)	Match / mismatch vs 100 nF 100 V 0603
0603B104K101	General Purpose	0603	100 nF	±10%	0.80+0.15/−0.10	Match on all five fields
0603B104J101	General Purpose	0603	100 nF	±5%	0.80+0.15/−0.10	Match; mismatch — tighter ±5% tolerance
MT18B104K101	Automotive MT, AEC-Q200 compliant	0603	100 nF	±10%	0.80+0.15/−0.10	Match; mismatch — automotive series, request the qualification report
ST18B104K101	Automotive ST, soft termination	0603	100 nF	±10%	0.80±0.30	Match; mismatch — soft termination, double the thickness band
SH18B104K101	Soft Termination SH, Ag-poly	0603	100 nF	±10%	0.80±0.30	Match; mismatch — Ag-poly soft termination, double the thickness band
MT21B104K101	Automotive MT, AEC-Q200 compliant	0805	100 nF	±10%	—	Mismatch — 0805 footprint, 2.25× the board area
0603B105K500	General Purpose	0603	1 µF	±10%	—	Mismatch — ten times the capacitance but only 50 Vdc, unusable on a 48 V rail
MT31B105K101	Automotive MT, AEC-Q200 compliant	1206	1 µF	±10%	—	Mismatch — 1 µF at 100 V exists, at 1206 and AEC-Q200 grade

Five of the eight match on capacitance, case size and voltage class and still differ in a field that decides the build: tolerance, termination chemistry, or a thickness band three times wider than the general-purpose part. Confirm against both datasheets and qualify in your own circuit — that step stays with you.

Is the 100 V class in stock in Hong Kong?

Thinly, and we would rather you heard it from us. We matched every catalogue code in the 23 September 2026 snapshot against the MLCC export: 764 of the 1,125 codes resolve to an MLCC row, and of those 371 are 50 Vdc while 56 are 100 Vdc. The next classes down are 25 Vdc on 61 codes and 250 Vdc on 50. So the class a 48 V design is forced onto carries about a seventh of the stocked 50 V depth.

Stock is a snapshot and never a permanent quantity. The planning rule is what matters: on a 48 V rail, settle the 100 V positions early, because the shelf behind them is shallower than the one your 12 V and 24 V rails have been drawing on.

Does the part number agree with the capacitance column at 100 V?

Usually, and we found where it does not. Walsin’s MLCC part numbers carry a three-digit capacitance code, so the code and the column can be checked against each other. We decoded all 16,661 part numbers whose format allows it: 16,641 agree, and twenty disagree — a defect rate of 0.12%, low and not zero.

Two of the twenty sit on the 100 Vdc shelf. 0805A474K101 carries the code 474, which decodes to 470 nF, while its Capacitance column reads 1000 nF. MT32N333J101 carries 333, decoding to 33 nF against a column reading 3300 pF — a factor of ten, in an AEC-Q200 series. Seven of the twenty are MT parts. We have not seen this cross-check published, and it is the cheapest sanity test available on a pasted BOM: if the code and the column disagree, stop and ask for the datasheet.

What should a buyer in the United States or Germany do about a 48 V rail?

Separate the positions that genuinely need 100 V from the ones that inherited it, then cross the rest in one pass. A 100 nF bypass behind a local 5 V regulator does not need the bus rating; the input bulk and the hot-swap path do. On this catalogue that distinction is worth two case sizes per position.

1. List the 48 V positions by what they actually see. Anything downstream of a point-of-load converter is a 5 V or 12 V part. Only the parts across the bus itself need the 100 V class, and on 0402 that class effectively does not exist above 4.7 nF.
2. Cross the whole BOM at once. Paste up to 500 part numbers or drop your Excel or CSV into the free cross-reference tool. It is matched against Walsin’s published catalogue and returns candidates field by field, with every difference from your original stated rather than hidden. Filter by commercial or automotive grade, or for soft termination, high temperature, safety certified, high Q / low ESR or low profile.
3. Check what is physically in Hong Kong on ready stock before you commit. That page is a dated snapshot as at 23 September 2026, not a live feed, and 56 of its matched MLCC codes are 100 Vdc.

For why the nominal capacitance is not the working capacitance, see MLCC DC bias and voltage derating. For the automotive series split behind the MT and MG counts, see AEC-Q200 MLCC: MT vs MG series. The full range is on the Walsin MLCC range page.

The short version

A 48 V bus MLCC voltage rating has no middle option in Walsin’s catalogue. We counted 25 rated-voltage classes across 24,026 rows and found zero at 63 Vdc and zero at 80 Vdc, so a 48 V rail lands on 100 V at 2.08× derating. That shelf is 3,110 rows, only 873 of them X7R. At 0402 it stops at 4.7 nF with nothing at 100 nF; at 0603 exactly eight parts reach 100 nF and none goes higher. The AEC-Q200 MT series offers 376 rows at 100 V, 17 of them at 100 nF or above. In the 23 September 2026 snapshot, 56 matched codes are 100 Vdc against 371 at 50 Vdc. Qualification stays with you.

Send us the 48 V rail — the bus positions, the capacitance you need and whether AEC-Q200 is in scope — and we will come back with Walsin candidates field by field, differences included, plus which of them was in Hong Kong on 23 September 2026. Then tell us something: would you rather take a 100 V part two case sizes up, or run a 50 V part at 96% of rating and derate the capacitance? We have an opinion, and real boards keep disagreeing with it.

---

Frequently asked questions

<!– wp:rank-math/faq-block {"questions":[{"id":"faq-question-01","title":"Does Walsin make a 63 V or 80 V MLCC?","content":"Not in the parametric export we hold. We counted the rated voltage of all 24,026 MLCC rows and found 25 distinct classes with zero rows at 63 Vdc and zero at 80 Vdc. The ladder runs 35, 50, then 100 Vdc. This is a statement about our export file, not about Walsin's full product range, and the question is on our request list.","visible":true},{"id":"faq-question-02","title":"What voltage rating should a 48 V bus MLCC have?","content":"In this catalogue, 100 Vdc, because nothing sits between 50 V and 100 V. A 48 V rail against a 50 Vdc part is at 96% of rating before any transient, and X7R and X5R lose effective capacitance as DC bias rises. The 100 Vdc shelf holds 3,110 rows, of which 873 carry an X7R dielectric.","visible":true},{"id":"faq-question-03","title":"How much capacitance can I get at 100 V in an 0402?","content":"4.7 nF, on 0402B472K101. We parsed the Capacitance field on every 100 Vdc row: zero 0402 parts reach 100 nF at that voltage, while 0603 has exactly eight parts at 100 nF and none above it. The same 100 nF is ordinary at 50 Vdc in both case sizes.”,”visible”:true},{“id”:”faq-question-04″,”title”:”Is there an AEC-Q200 MLCC at 100 V from Walsin?”,”content”:”Yes. The MT series holds 376 rows at 100 Vdc, of which 17 reach 100 nF or more, the smallest being MT18B104K101 at 0603 and the largest MT32B475K101 at 4.7 µF in a 1210. Walsin says AEC-Q200 compliant, not certified. Request the qualification report for your exact part number.”,”visible”:true},{“id”:”faq-question-05″,”title”:”How many 100 V MLCC are in Venatech’s Hong Kong stock snapshot?”,”content”:”Fifty-six of the 764 snapshot codes that we could match to the MLCC export, against 371 at 50 Vdc. The snapshot is dated 23 September 2026 and covers 1,125 part numbers in total. Stock is never a permanent quantity, so treat that as a planning signal and ask us before committing a production slot.”,”visible”:true},{“id”:”faq-question-06″,”title”:”Can I check a Walsin MLCC part number against its capacitance?”,”content”:”Yes, and you should. The part number carries a three-digit capacitance code. We decoded all 16,661 MLCC part numbers whose format allows it: 16,641 agree with the Capacitance column and 20 do not. Two of the 20 are at 100 Vdc, including 0805A474K101, whose code decodes to 470 nF while the column reads 1000 nF.”,”visible”:true},{“id”:”faq-question-07″,”title”:”What is the fastest way to second-source a 100 V MLCC?”,”content”:”Separate the positions that truly see the bus from the ones downstream of a converter, then cross the whole list in one pass. The free cross-reference tool takes a pasted list of up to 500 part numbers or an Excel or CSV file, matches it against Walsin’s published catalogue and states every difference from your original. Then check the dated ready stock snapshot.”,”visible”:true}],”listStyle”:””,”titleWrapper”:”h3″,”titleCssClasses”:””,”contentCssClasses”:””,”sizeSlug”:””} –>

---