Inductor SRF vs inductance should be a one-way street: more inductance, lower self-resonant frequency. We sorted every Walsin inductor ladder and counted 144 steps out of 1,569 where the published SRF(min) goes up instead. All 144 sit in five series. The other seven series break it on 0 of 489 steps.
Venatech Limited, an authorised Walsin distributor in Hong Kong, holds the 5,410-row inductor export published by Walsin Technology (Walsin), and this post is written from its Series, Size, Inductance, Unit and SRF(min) columns. We excluded the QA test row Test123999. Every part number below is real. We hold no Walsin inductor datasheet, so every figure here is an export figure, and the one unit question is marked in the note below.
Should SRF fall every time inductance rises in a chip inductor series?
Within one series at one case size, yes, as a first-order rule: more turns or more winding length adds parasitic capacitance, and the resonance moves down. We took the lowest SRF(min) published at each inductance value, sorted by inductance, and asked whether the next step up is lower or equal. In 1,569 steps across 39 series-and-size groups, it is higher 144 times, or 9.2 percent.
Two rules of the method matter. We only used groups with at least five distinct inductance values, so a two-part series cannot create a trend. We took the lowest SRF per inductance value, which is the conservative reading of a column called SRF(min). Nothing in that rule is generous to the data, and 144 steps still rise.
How many inductance steps in the Walsin export break that rule?
144 of 1,569, and they are not spread evenly. We parsed all 4,400 rows that have a positive SRF and an inductance in nanohenries or microhenries. Five wirewound and multilayer families carry every break: WLCW 43 of 273 steps, WCA 33 of 258, WCI 33 of 258, WQCW 29 of 239 and WLFW/WFI 6 of 52. Together that is 144 of 1,080 steps, 13.3 percent.
| Series | Steps tested | Steps where SRF rises | Share | Match / mismatch against a monotonic SRF ladder | Your incumbent part (fill in) |
|---|---|---|---|---|---|
| WLCW | 273 | 43 | 15.8% | Mismatch | |
| WCA | 258 | 33 | 12.8% | Mismatch | |
| WCI | 258 | 33 | 12.8% | Mismatch | |
| WQCW | 239 | 29 | 12.1% | Mismatch | |
| WLFW/WFI | 52 | 6 | 11.5% | Mismatch | |
| WLCM | 143 | 0 | 0% | Match | |
| MCI | 138 | 0 | 0% | Match | |
| MCI-TM | 67 | 0 | 0% | Match | |
| MCI Automotive | 67 | 0 | 0% | Match | |
| WLFI | 42 | 0 | 0% | Match | |
| WQCM | 25 | 0 | 0% | Match | |
| AIP | 7 | 0 | 0% | Match |
The four-series overlap is real. WLCW, WCA, WCI and WQCW repeat the same inductance ladders, which we covered in 3,146 chip inductors publishing identical data. Counted once per case size and inductance pair, the 144 rises collapse to 62 distinct steps. We have not seen this counted anywhere. Quote 62 if you want the distinct physical steps and 144 if you want the number of rows a script will trip over.
Which series never break it?
Seven, and they share a profile. We counted 489 steps in WLCM, MCI, MCI-TM, MCI Automotive, WLFI, WQCM and AIP, and SRF falls or holds on every one of them. These are the 0201 and 0402 multilayer families plus a few small power lines, where each size holds one construction rather than several.
At 0201 the effect is stark. We opened the 73 inductance values published at that case size, from WLCM0603Z0B1N0TB at 1 nH to the top of the ladder, and the lowest SRF per value falls from 10,000 to 900 without a single upward step across 191 transitions in the 0201 groups. That is the behaviour the textbook describes.
| Case size | Transitions tested | SRF rises | Match / mismatch against a monotonic ladder |
|---|---|---|---|
| 0201 | 191 | 0 | Match |
| 0402 | 343 | 29 | Mismatch |
| 0603 | 381 | 48 | Mismatch |
| 0805 | 259 | 44 | Mismatch |
| 1008 | 214 | 7 | Mismatch |
| 1210 | 136 | 4 | Mismatch |
| 1812 | 38 | 12 | Mismatch |
The 0603 and 0805 rows hold 92 of the 144 rises. Those are the sizes where one series offers more than one style of part under the same size label.
How large are the breaks, and where do they sit?
Mostly small, with a few that matter. Of the 144 rises, 131 are below 1.5 times the previous step’s SRF, 9 are between 1.5 and 2 times and 4 reach 2 times or more. The largest is 3.43 times.
| From (part, inductance, SRF) | To (part, inductance, SRF) | Ratio | Match / mismatch |
|---|---|---|---|
WLCW2012Z0J2N2PB, 2.2 nH, 3.0 |
WLCW2012HQJ2N5PB, 2.5 nH, 10.3 |
3.43x | Mismatch — SRF rises, and the part prefix style changes from Z0 to HQ |
WLCW2012Z0J2N9PB, 2.9 nH, 4.7 |
WLCW2012Z0J3N0PB, 3 nH, 7.9 |
1.68x | Mismatch — same style, SRF still rises |
WLCW1608Z0J3N6PB, 3.6 nH, 5.9 |
WLCW1608Z0J3N9PB, 3.9 nH, 6.9 |
1.17x | Mismatch — same style, SRF rises |
WLCW4532Z0K82NPB, 82 nH, 800 |
WLCW4532Z0JR10PB, 100 nH, 850 |
1.06x | Mismatch — small, same style, tolerance changes from 10 to 5 percent |
WLCM0603Z0B1N0TB, 1 nH, 10,000 |
next lower in the 0201 ladder | falls | Match |
The first row is the useful one. The 2.5 nH part is a different style from the 2.2 nH part, with a Q of 80 and a rated current of 1.6 A against 0.6 A, so the ladder is two ladders spliced on inductance value. Not every break is that explainable: the 2.9 nH to 3 nH step stays inside style Z0.
Why does the 10 uH to 12 uH step rise in the WLFW and WFI ladder?
Because the step changes part-number family. In the WLFW/WFI 0805 ladder the lowest SRF at 10 uH is 21 on WLFW2012Z0J100PB and at 12 uH it is 37 on WFI2012FS120J: 1.76 times higher, with a different prefix and a Q of 0. That is one of six rises in the series.
We checked the 0.68 uH to 0.82 uH step as well. WLFW2012Z0JR68PB publishes 400 and WFI2012FSR82J publishes 600, again across the prefix change. Both WFI parts publish a Q of 0, a field we covered in Q factor, where 1,082 rows publish zero. A ladder that crosses a prefix is not one product line, whatever the series label says.
What should a Hong Kong designer do with an SRF figure that rises?
Treat the series label as a grouping, not as a promise of one construction. When we decoded the ladders, the rises clustered where the part number changed style or prefix, which is information the Series column does not carry. For how the families differ, see how to tell Walsin inductor series apart. For self-resonance itself, see the SRF explainer.
Three rules, each measured above:
- Do not interpolate SRF between two inductance values without checking both part numbers share a style code. 62 distinct steps break the rule.
- Pick the part, not the ladder. At 2.2 nH and 2.5 nH in the 0805 ladder the published SRF differs by 3.43 times.
- Choose 0201 families if you want a clean ladder. Zero of 191 steps rise there.
Cross a part you already use through the free cross-reference tool, and we will tell you which fields differ rather than which fields fit. The inductor range is on the inductor products page, and what is held in Hong Kong is on ready stock, 1,125 part numbers as at 23 September 2026. It is a dated snapshot and not a live feed.
The short version
We counted 1,569 inductance steps across 39 series-and-size groups in the Walsin export. SRF(min) rises on 144 of them, 9.2 percent, and all 144 sit in WLCW, WCA, WCI, WQCW and WLFW/WFI. Seven series break the rule on 0 of 489 steps, and the 0201 groups on 0 of 191. Collapsed across the repeated ladders, 62 distinct steps rise, the largest 3.43 times from WLCW2012Z0J2N2PB to WLCW2012HQJ2N5PB. We will not tell you a part matches on SRF. We will show you the number the file holds. Confirm against both datasheets and qualify in your own circuit.
Send us your inductor part number and the frequency you run it at, and we will show you which fields differ and which do not. And a question back: when your last SRF check passed, did it compare two parts of the same style?
Frequently asked questions
Does SRF always fall as inductance rises?
As a rule inside one series and case size, yes, but the Walsin export breaks it on 144 of 1,569 steps, 9.2 percent. All 144 sit in five series: WLCW, WCA, WCI, WQCW and WLFW/WFI. Seven other series break it on none of 489 steps, so the behaviour depends on the series rather than the physics.
Which Walsin inductor series have a clean SRF ladder?
WLCM, MCI, MCI-TM, MCI Automotive, WLFI, WQCM and AIP. We counted 489 inductance steps across them and SRF(min) falls or holds on all 489. At the 0201 case size the lowest SRF drops from 10,000 to 900 across 191 transitions with no rise at all.
How big are the SRF rises?
Mostly small. Of 144 rises, 131 are under 1.5 times, 9 are between 1.5 and 2 times, and 4 are 2 times or more. The largest is 3.43 times, from 3.0 on a 2.2 nanohenry part to 10.3 on a 2.5 nanohenry part in the WLCW 0805 ladder.
Why do WLCW, WCA, WCI and WQCW show the same breaks?
Because they repeat the same inductance ladders. Counted once per case size and inductance pair, the 144 rises collapse to 62 distinct steps. We cover the overlap in a separate post on 3,146 inductors that publish identical data. Quote 62 for physical steps and 144 for rows.
What unit is the SRF(min) column in?
The export states none. WLCW values for the same ladder sit about 1,000 times below the WCA, WCI and WQCW values, which suggests two conventions. Every comparison in this post stays inside one series, so the scale does not affect it. Confirm the unit with Walsin before comparing across series.
Is there a datasheet to check these SRF figures against?
We hold 28 Walsin datasheets, 451 pages, and none covers an inductor. Thin film, inductor, disc, RF, MOV and antenna lines have no datasheet in our file. The export is the only figure we can show you for SRF, and it should be confirmed against Walsin’s own site.
Can Venatech tell me which part in a ladder matches mine?
We can show you the fields that match and the fields that differ between your part and a Walsin part, through the free cross-reference tool. We state which fields match and which differ. Qualification stays with you, in your own circuit, against both datasheets.