Walsin MLCC Series Index vs the Approval Sheets: 36 of 92 Testable Cells Disagree and 2 Rows Are Shifted One Column

A one-line-per-series MLCC index disagrees with Walsin’s own approval sheets in 36 of 92 testable cells. We compared sizes, dielectrics, capacitance range and voltage range for all 24 series sheets: 20 index rows carry at least one wrong cell, and 2 rows are shifted one column. Read the sheet, not the summary line.

Venatech Limited, an authorised Walsin distributor in Hong Kong, keeps a series index beside its 26 MLCC PDFs from Walsin Technology (Walsin): INDEX_MLCC_series.csv, one row per sheet with sizes, dielectric, capacitance range and voltage range. It is the table an engineer or a buyer reaches for first. This post is written from that index, section 6 “General electrical data” of each of the 24 series sheets, and the MLCC export 00_ALL_MLCC_parts_combined.xlsx. Every part number below is real. The question is simple: can you select a series from the summary line?

How many cells of a Walsin MLCC series index disagree with the approval sheets?

36 of 92. We opened section 6 of all 24 series sheets and wrote down the union across its dielectric columns: first and last case size, the dielectric set, the lowest and highest capacitance and the lowest and highest rated voltage. That is 96 cells. Two cannot be tested because the index cell is blank, and two because the sheet prints only a maximum, which leaves 92. 56 agree, 36 do not, and 20 of the 24 index rows carry at least one wrong cell.

Four rows have no disagreement: RH, HH, MG and the large-size sheet. Five rows carry three or more wrong cells. We have not seen this counted anywhere, and the method is repeatable: take the first and last value in each column of section 6 and compare it with the index cell.

The VT row is the sharpest case. The VT sheet prints four case sizes, 0603 to 1210, a 10 nF to 10 uF range and six voltage codes from 10 V to 100 V. The index describes one size, one capacitance and one voltage. VT also has no rows in the export, so nothing else in our data contradicts the index, and nothing confirms it.

Which two index rows are shifted by one column?

Rows 02 and 03, the ultra-small 01005 sheet and the large-size sheet. We parsed the file with a CSV reader rather than by eye. Both rows have an empty size cell, so the dielectric list sits in the sizes column, the capacitance range sits in the dielectric column, the voltage sits in the capacitance column, and the voltage column is empty. The other 24 rows have nine populated cells in the right places.

A voltage filter returns 22 of 24 series, and a filter on NP0 in the dielectric column misses both rows, because that column holds “0.2pF-0.1uF” and “10pF-47uF”. The ultra-small sheet prints case size 01R5 and 4 V to 50 V, where the index says 6.3-50 V. After un-shifting, row 02 still disagrees on voltage, because the X5R column starts at 4 V. Row 03 agrees on its three testable cells.

Which index rows are copies of each other, and do their sheets agree?

Three pairs are identical in all four cells, and in each pair the two sheets differ on three of four fields. We compared every pair of rows cell by cell. Rows 04 and 05, 16 and 17, and 23 and 24 are copies: Middle and High Voltage and Surface Coating, S2 and S2-H, Mega Cap M and Mega Cap V. A copied row means one of the two is wrong by construction.

The Mega Cap V sheet describes a vertical-stacked 2220 part at 100 V, while its index row claims 1210 to 2225 at 50 V to 630 V. Neither Mega Cap series has export rows, so the cross-reference tool cannot offer them and we do not guess.

When the index and the sheet disagree, which does the export side with?

The sheet, in 9 of 12 testable cells. The export carries rows for 16 of the 24 series, and on size, dielectric and voltage we could ask it to referee 12 cells where the index and the sheet differ. We measured the lowest and highest voltage, the dielectric set and the size span in the export for each series. It sides with the sheet 9 times and with the index 3 times.

The three exceptions matter more than the nine agreements. The HT sheet prints X8G alone while the export holds 26 X8R parts in the series, and the SH and FT sheets print voltages the export never reaches. We cannot say whether the sheet or the export is wrong in those cells; the question is on our request list.

How many export rows fall outside the index and outside the sheet?

265 outside the index, 55 outside the sheet. We checked every export row of the 15 series that have both rows and a rated-voltage envelope, 23,853 rows. The S2 series is left out because its export voltage is the impulse voltage. A row counts as outside if its case size, dielectric or rated voltage lies beyond the printed range.

One of the 265 is the known MT row that carries 390 in the dielectric column; the rest are not defects. A filter built on the index would exclude ST43N101K302, a 3000 V soft-termination NP0 part, before it started.

What did agree between the index and the 26 PDFs?

Everything mechanical. We ran pdfinfo on all 26 files and compared. The page count and the file size in kilobytes agree on 26 of 26 rows, the 26 download links are the same text in the index and in the folder’s manifest, and the 431 pages sum, with the two resistor sheets, to the 451 pages we hold.

The index is accurate on the columns a script can read from a file and inaccurate on the columns that need a person to read section 6. That is a property of every summary table, including ours, and it is why we send the sheet and not the summary. We hold 28 Walsin datasheets, 451 pages, and the sheet covering your part number will be in your inbox on request. The structure of the sheets themselves is covered in how to read a Walsin MLCC part number, and the range edges of sheets 13 to 18 against the export in 40 printed range edges checked.

Use the free cross-reference tool to see which Walsin parts exist for a requirement, then check the sheet. What is in Hong Kong is on ready stock: 1,125 part numbers as at 23 September 2026, a dated snapshot and not a live feed. The capacitor range sits on the MLCC page.

The short version

We compared a series index against the 24 Walsin series sheets, field by field: 36 of 92 testable cells disagree, 20 of 24 rows carry at least one, and rows 02 and 03 are shifted one column. Three pairs of rows are copies although their sheets differ on three of four fields. The export sides with the sheet in 9 of 12 cells and the index in 3. Pages, sizes and links match on 26 of 26. We will not tell you a series suits your design from a summary. We will send the sheet. Confirm against both datasheets and qualify in your own circuit.

Send us the part number you are replacing and the series you picked from a summary table, and we will tell you which edge of its sheet your requirement sits on. And a question back: how many of your approved-vendor-list lines were chosen from a summary line rather than from section 6?

Frequently asked questions