MLCC Dielectric Codes: Walsin’s Four Smallest Are 51 Part Numbers of 24,026

We counted the MLCC dielectric codes in Walsin’s export: ten distinct values across 24,026 rows. NPO holds 15,337 and X7R 7,544. The four smallest — Y5V, X7T, X7S and X8R — hold 51 part numbers between them, and 2 of those 51 are the whole Y5V range.

---

Venatech Limited, an authorised Walsin distributor in Hong Kong, holds the full parametric export published by Walsin Technology (Walsin) and 28 Walsin datasheets covering 451 pages. This post is measured from 00_ALL_MLCC_parts_combined.xlsx, 24,026 rows, checked against five of the approval sheets on our shelf. Every part number is real, and every count is of part numbers rather than pieces.

How many MLCC dielectric codes does Walsin’s catalogue actually carry?

Ten, and the distribution is extremely lopsided. We parsed every row of 00_ALL_MLCC_parts_combined.xlsx. Two codes cover 95.3% of the catalogue. Four codes cover 51 part numbers — 0.2% — and one of those four, the 390 entry, is not a dielectric at all.

Dielectric code	Part numbers	Share of 24,026	Where it lives
NPO	15,337	63.8%	Every series; the Class I workhorse
X7R	7,544	31.4%	General Purpose, Middle and High Voltage, SH, MT, ST, MG
X5R	751	3.1%	General Purpose, SH, TT, MT
X8G	209	0.9%	RF microwave, HT high temperature, and 3 automotive rows
X6S	133	0.6%	General Purpose
X8R	26	0.11%	HT high temperature only
X7S	21	0.09%	General Purpose only
Y5V	2	0.008%	General Purpose only
X7T	2	0.008%	SH soft termination only
390	1	—	A capacitance code in the wrong column

We have not seen this counted anywhere. The practical consequence is that four of Walsin’s ten published dielectric codes are not a shelf you can select from. They are four specific answers to four specific problems, and three of them resolve to a single case size.

Why are there only two Y5V part numbers, and why would you want one?

Because the only thing Y5V buys you here is height. The two parts are 1206F474M500 and 1206F474Z500 — both 1206, both 470 nF, both 50 Vdc, differing only in tolerance. We read their dimensions: both are 0.80 ±0.10 mm thick. We then pulled every X7R at the same 1206, 470 nF, 50 Vdc position: every one of them is 1.60 mm nominal. The Y5V pair is exactly half the height.

Part number	Dielectric	Capacitance	Tolerance	Rated voltage	Thickness	Match / mismatch vs 1206 470 nF 50 V, 1.0 mm max height
1206F474M500	Y5V	470 nF	±20%	50 Vdc	0.80 ±0.10 mm	Match on all five; the only Walsin 1206 470 nF 50 V part under 1 mm
1206F474Z500	Y5V	470 nF	-20+80%	50 Vdc	0.80 ±0.10 mm	Match on height and voltage; mismatch — asymmetric tolerance the sheet’s ordering key does not list
1206B474M500	X7R	470 nF	±20%	50 Vdc	1.60 mm nominal	Match on value, voltage, tolerance; mismatch — twice the height, will not fit
1206B474K500	X7R	470 nF	±10%	50 Vdc	1.60 mm nominal	Match on value and voltage, tighter tolerance; mismatch — twice the height
SH31B474K500	X7R	470 nF	±10%	50 Vdc	1.60 ±0.50 mm	Mismatch — soft termination, 1.60 mm, widest thickness tolerance of the group
MT31B474K500	X7R	470 nF	±10%	50 Vdc	1.60 +0.30/-0.10 mm	Mismatch — automotive AEC-Q200 compliant series, 1.60 mm

We will not tell you a part is interchangeable. We will show you where it differs — and here the difference is 0.8 mm of z-height against a dielectric that drifts far harder over temperature and bias.

Two details deserve a note. The General Purpose approval sheet we hold, 01_General_Purpose.pdf (ASC_General Purpose Series, Sep. 2026), names five dielectrics on its cover — NPO, X7R, X6S, X7S and X5R — and gives each a column in its general electrical data table. Y5V has no column. It appears in the ordering key as F=Y5V and in two test tables, and nowhere else. And the Z tolerance code on 1206F474Z500 is not in that ordering key either, which lists only A, B, C, D, F, G, J, K and M. We counted the Z code across the whole export: 15 rows use it, spanning X7R, X5R and Y5V.

Is SH55T225K251 a real part number, or is 55 a bad size code?

It is real, and we checked it specifically because 55 is not one of the size strings the export prints. The export’s Size column reads 2220 for both X7T rows, while the part number’s size field reads 55. We opened 19_Soft_Term_SH.pdf (ASC_Soft termination_(SH), Sep. 2026) and the ordering key prints 55=2220 (5750) in black and white. It also appears as 55=2220 (5750) in 16_Safety_X1Y2_General_S2.pdf. We then counted every prefix-style part number carrying 55 in that position: 70 rows, all of them 2220. Not a defect.

The same page resolves the whole X7T line. The SH sheet’s general electrical data table gives X7T one capacitance value — 2.2 µF — two tolerance codes, K (±10%) and M (±20%), an operating temperature of -55 to +125 °C and a capacitance characteristic of +22% to -33%. Two tolerances at one capacitance is two part numbers, and the export holds exactly two: SH55T225K251 and SH55T225M251, both 2220, 2200 nF, 250 Vdc. The datasheet predicted the count. That is the only dielectric in the catalogue where we can say that.

Note what X7T is for, and what it is not. A dielectric whose capacitance may fall by a third across its rated temperature range is a bulk and smoothing part, not a timing or filter part. The figure is printed in a Walsin document we hold, so you do not have to take our word for it.

Why do 26 X8R parts sit in a series whose datasheet never offers X8R?

This is the finding that should change what you ask for. All 26 X8R part numbers are in the HT high temperature series — HT18R…, HT21R… and HT31R… at 0603, 0805 and 1206, all 50 Vdc, from 100 pF to 100 nF. We read all 11 pages of 15_High_Temperature_HT_150C.pdf (WTC_MLCC_High_Temperature_HT, Apr. 2026). Its cover says X8G Dielectrics. Its ordering key offers exactly one dielectric code, G=X8G. Its general electrical data table has one column, X8G.

X8R appears once in those 11 pages, in a constructions table that contrasts the two ceramic systems: X8G is calcium-zirconate based, X8R is barium-titanate based. That is the entire published content on X8R in the sheet that covers the series those 26 parts belong to.

So for HT21R223K500 or HT31R104K500 we hold no capacitance range table, no tolerance set, no rated-voltage list, no operating temperature and no capacitance characteristic. The ordering key does not even define the R that sits in their dielectric position. We have not seen this gap written down anywhere, and we are stating it rather than letting the export imply coverage we do not have.

X8R also appears in no automotive series at all. The only automotive high-temperature Class I rows are three X8G parts, and all three are the same 0402 18 pF 50 Vdc position at three tolerances — MT15G180F500, MT15G180G500 and MT15G180J500. If your design wants an automotive-grade X8R, Walsin’s export does not offer one.

Where does X7S stop, and does the export agree with its own part numbers?

Not everywhere. We read all 21 X7S rows. They span 0201 to 1210 and 6.3 Vdc to 100 Vdc, from 100 nF up to 100 µF at 1210A107M6R3. The General Purpose sheet’s X7S column claims 0201 to 1812, 4 V to 100 V, 0.1 µF to 100 µF, ±22% and -55 to +125 °C. We checked: 1812 appears on zero X7S rows and 4 Vdc appears on zero X7S rows. The sheet’s envelope is wider than the catalogue’s.

One of the 21 also disagrees with itself. 0805A474K101 carries a capacitance code of 474, which the published ordering key decodes as 47 followed by four zeros in picofarads — 470 nF. Its Capacitance column reads 1000 nF. The row directly beside it, 0805A474M101, carries the same 474 code and reads 470 nF. We ran the same decode across all 24,026 rows: 27 rows disagree with the capacitance code in their own part number, and 0805A474K101 is the one that falls inside these four dielectrics. If you filter by the Capacitance column, you will get the wrong part; if you filter by the code, you will get the right one.

Does the Dielectric column always contain a dielectric?

No. Exactly one of the 24,026 rows holds a capacitance code there instead. MT18N391J101 is a 0603 390 pF 100 Vdc automotive part whose Dielectric column reads 390, and whose L, W and T fields are empty. Its part number says N, which the MT ordering key decodes as NPO, so the dielectric is recoverable — from the code, not from the column.

That single row is why the dielectric count reads ten rather than nine, and it is a worked example of the rule we apply to every cross: the part number is the specification, and the parametric column is a convenience that can be wrong. We read both, and when they disagree we say so instead of picking the one that reads better.

Can you buy any of these 51 parts out of Hong Kong stock?

One of them. We checked all 51 against the ready-stock snapshot as at 23 September 2026, which lists 1,125 catalogue codes. 1206F474Z500 is present — the Y5V part with the asymmetric tolerance code the ordering key does not list. Y5V 1206F474M500 is not. None of the 2 X7T, none of the 21 X7S and none of the 26 X8R parts is in the snapshot.

That is what a 0.2% dielectric looks like on a delivery schedule, and it is the honest reason to decide early. The trade press reported MLCC lead times in the 26 to 32 week range through 2026 as AI server and electric vehicle build-outs drew on the same capacity, so a part with 26 siblings in the whole catalogue is not the line item to leave to the last review. Our credential for saying this is checkable rather than asserted: we are an authorised Walsin Technology distributor in Hong Kong, confirm it with Walsin, and UL file E182369 covering Walsin’s safety capacitor line is searchable in UL’s own public database in about thirty seconds.

The short version

Walsin publishes ten MLCC dielectric codes across 24,026 part numbers, and four of them hold 51 parts between them. Y5V is two 1206 470 nF 50 V parts whose only advantage is 0.80 mm of height against X7R’s 1.60 mm. X7T is two 2220 2.2 µF 250 V soft-termination parts, and the SH approval sheet predicts exactly that count while printing a +22% to -33% capacitance characteristic. X7S is 21 parts, one of which disagrees with its own capacitance code. X8R is 26 parts in a series whose approval sheet documents only X8G and never defines the R code at all. The suspected bad size code 55 turned out to be printed in two Walsin sheets as 2220. One of the 51 was in Hong Kong on 23 September 2026. Qualification stays with you.

Send us the dielectric codes off your BOM and we will tell you which of them Walsin answers with 20,000 part numbers and which with two — and for the ones with two, exactly which specification we can source and which we cannot. If a selector offered you X8R and implied it came with the 150 °C figure, tell us which selector: that is the claim we most want to see a document for.

For the Class I and Class II basics, see NPO vs X7R vs X5R MLCC dielectric. For the code positions themselves, see tolerance, TCR and dielectric codes decoded. For the series behind the X8R rows, see high temperature MLCC to 150C, the HT series. Cross a whole list in one pass with the free MLCC cross-reference tool — paste up to 500 part numbers or drop your Excel or CSV, matched against Walsin’s published catalogue, with every difference from your original stated rather than hidden, and filters for soft termination, high temperature, safety certified, high Q / low ESR and low profile. Then check ready stock, a dated snapshot as at 23 September 2026 rather than a live feed. The range sits on the Walsin MLCC page.

---

Frequently asked questions

---