New Coilcraft AGM2222 inductors for output filters for Class D amps?


2009-06-24 12:04 pm
Here is the press release:
"Coilcraft’s new AGM2222 Series power inductors offer current ratings up to 110 Amps and very low DCR in a package measuring just 22 X 22 X 23 mm, a 73% volume reduction from previous-generation products. This significant improvement in power density makes the AGM2222 ideal for high-current applications including automotive bi-directional 12 V – 48 V DC-DC converters.
The AGM2222 Series is available in 13 inductance values from 1.9 to 10 µH. It is qualified to AEC-Q200 Grade 1 (-40°C to 125°C ambient), making it ideal for automotive and other high-temperature applications. Soft saturation is stable over temperature and enables high peak current handling. Magnetic shielding reduces EMI at high switching frequencies."

Here are the specifications: AGM2222 High Current Power Inductors | Coilcraft
Here are Inductance vs Current graphics: Typical Current vs Frequency - AGM2222 Shielded Power Inductors
Here are Inductance vs Frequency graphics: Typical Inductance vs Frequency - AGM2222 Shielded Power Inductors

I asked Coilcraft for the voltage ratings of these inductors and they responded, that all are tested at 100Vrms for 1 second, when they are manufactured.
So, I think these inductors will be very good for TPA3255 and similar power amplifiers, with single power supply under 80V, or for dual power supply up to +/-40V.

What do you think about them?
Will they be good for Class D amplifiers?


2009-06-24 12:04 pm
I've received 3 free samples of AGM2222 with 10uH inductance:

When I have time, I will try to compare AGM2222 to the SER2918H, but it is 22uH...
Both look very massive!
And both are heavy :)


Disabled Account
2010-01-17 9:20 pm
All measurements I've seen proved gapped ferrite being the most linear core material. Any kind of iron powder including carbonyl is inferior.
Btw coilcraft recommends these inductors for dc-dc converter claiming compact size and low loss - but no recommandation for class-D here.
Linearity is not related to low loss - I found lowest losses with carbonyl cores.
Last edited: