AEC-Q200 MLCC Cross Reference: 2,081 of 15,606 Commercial Specifications Have an Automotive Version, and X5R Has Zero of 630

Our cross-reference tool shows the commercial part first and lists the AEC-Q200 version beneath it where one exists. We counted how often one exists. Across Walsin’s MLCC export, 2,081 of 15,606 distinct commercial specification signatures have an AEC-Q200 counterpart — 13.3%. For X5R the answer is zero of 630.

Venatech Limited, an authorised Walsin distributor in Hong Kong, publishes a free cross-reference tool with a grade filter. The filter’s promise is modest and honest: commercial shown first, AEC-Q200 version listed beneath where one exists. Nobody, including us until now, had measured how often that condition is met. This post is written from 00_ALL_MLCC_parts_combined.xlsx, Walsin Technology (Walsin)’s own MLCC export, 24,026 rows. Every part number below is real and checkable against it. We have not seen this counted anywhere.

How often does an AEC-Q200 version of a commercial MLCC actually exist?

For 2,081 of 15,606 distinct commercial specification signatures — 13.3%. We defined a signature as case size, dielectric, capacitance, rated voltage and tolerance, the five fields a selector filters on. We split the export into the 2,294 rows Walsin files under its two AEC-Q200 series, MT and ST, and the 20,944 rows that are neither those nor MG, then asked one question of every commercial signature: does an AEC-Q200 row carry the identical five fields?

Read the last two rows carefully, because they are the two things a buyer gets wrong. Relaxing tolerance does not rescue the yield — it moves it from 13.3% to 17.4%, which is a different denominator and the same disappointment. And MG adds almost nothing: only 23 commercial signatures gain a counterpart when you allow MG in, and MG is the series whose own ordering key states it carries no AEC-Q200 certification on all 788 rows. We cover that split in automotive MLCC, MT against MG.

AEC-Q200 is “compliant” and “qualified to” in Walsin’s own wording, never “certified”. We hold no qualification report for any part number, and you should ask for the report covering your exact part number before you design it in.

Which case sizes carry the automotive upgrade, and which carry none?

Six of the sixteen case sizes in the export carry a meaningful automotive shelf, and seven carry literally none. The best yield is 0603 at 20.9%; the worst non-zero is 1812 at one signature in 889. We grouped all 15,606 commercial signatures by the Size column and counted the matches in each group.

One row in the export is not a case size at all: a single Middle to High Voltage record has the series name and a tab character jammed into its Size field. We counted it separately and it matches nothing. It is a known defect in Walsin’s export, not a part.

The practical consequence is blunt. If your automotive programme is built on 1808 or 1812 capacitors — and high-voltage designs often are — the grade filter has nothing to offer you, and no amount of filtering will change that. See middle and high voltage MLCC for what that shelf actually holds.

Why does every X5R specification return nothing on the automotive filter?

Because Walsin’s AEC-Q200 MLCC series publish three dielectrics and X5R is not one of them. We grouped every commercial signature by dielectric. NPO returns 1,204 of 9,754 and X7R returns 877 of 4,840. Every other dielectric returns zero, and the largest of those is X5R at 0 of 630.

0402X104K100 is a real example: General Purpose, 0402, X5R, 100 nF, 10 Vdc, ±10%. Paste it and the grade filter will show it and show nothing beneath it, because the 2,294 AEC-Q200 rows in our extract carry only NPO, X7R and three X8G parts. The honest answer is not “here is the automotive version” but “there is no automotive version of this dielectric, and your choice is an X7R at a different capacitance-versus-voltage curve”. The dielectric trade is in NP0 against X7R against X5R.

Does the AEC-Q200 part keep the special characteristic of the part it replaced?

Usually not. This is the finding that should change how you read a grade-filter result. Of the 1,487 soft-termination SH signatures, 579 have an AEC-Q200 counterpart — but only 65 of those 579 are matched by ST, the automotive soft-termination series. The other 514 are matched by MT, which is not a soft-termination part. The five fields agree and the feature you selected the part for does not survive.

SH18B101K100 is a 0603 X7R 100 pF 10 Vdc ±10% soft-termination part. Its AEC-Q200 counterpart on all five fields is MT18B101K100, which is an MT part, not an ST part. If you chose SH for flex-crack margin on a board that bends, that margin is gone. Read soft termination SH, SG, ST and VT before accepting that swap.

If all five fields match, are the two parts physically interchangeable?

No, on 218 of the 2,081 matched signature pairs — 10.5%. We took every matched pair and compared the L, W and T columns, the dimensions the export publishes. One pair in ten disagrees on at least one of them. The five fields a selector shows you are not the fields that decide whether the part fits your land pattern and your stencil.

The fourth row is the one to stare at. Five fields identical, and a three-terminal feed-through capacitor against a two-terminal chip. We will not tell you a part is equivalent. We will show you where it differs, and in that row it differs in the pin count. 89 of the 2,294 AEC-Q200 rows publish no L value at all, so on those the dimensional comparison cannot be made from the export and has to come off the datasheet.

What do you do with the 13,525 specifications that have no automotive version?

You change a field, and you decide which one. 13,525 of 15,606 commercial signatures have no AEC-Q200 counterpart, so for seven specifications in eight the question is not “which automotive part matches this” but “which field am I willing to move”. Dropping tolerance from the comparison lifts the matched share from 13.3% to 17.4%. Moving dielectric from X5R to X7R opens a shelf that was empty. Moving case size up from 1812 to 1210 goes from one signature in 889 to 78 in 1,074.

1. Paste the list, do not check parts one at a time. Up to 500 part numbers, or drop your Excel or CSV, into the free cross-reference tool, then set the grade filter to automotive and read what comes back — including the lines that come back empty.
2. Tell us which field you can move. Tolerance, dielectric, voltage or case size. We will tell you what that buys in part-number terms, because we have the counts above and they are repeatable.
3. Check what is physically in Hong Kong on ready stock — 1,125 part numbers as at 23 September 2026, a dated snapshot and not a live feed.

The full capacitor line sits on the Walsin MLCC page. Our line runs from Hong Kong, Shenzhen, Singapore and Jiangmen, with Chengdu in preparation, and nowhere else. Unit 207, 2/F, Sun Ling Plaza, 30 On Kui Street, Fanling, N.T., Hong Kong. (852) 2628 4777.

The short version

The grade filter promises an AEC-Q200 version “where one exists”, and one exists for 2,081 of 15,606 commercial specification signatures — 13.3%. Seven case sizes have none at all. X5R has none at all: 0 of 630 signatures. Of the 579 soft-termination SH signatures with an automotive match, only 65 keep soft termination. 218 of the 2,081 matched pairs disagree on at least one published dimension. AEC-Q200 is compliant, never certified, and we hold no qualification report for any part number. Qualification stays with you — our job is to make sure the list you qualify against is the real one.

Send us the automotive lines off your bill of materials and we will return the grade-filter position on each one: matched, matched with a feature lost, matched with a dimensional difference, or no automotive part at this specification. And a question back: when your selector showed you an automotive upgrade last month, did it tell you which of those four it was?

Frequently asked questions