Author Topic: Inductor current rating and temperature  (Read 2578 times)

0 Members and 1 Guest are viewing this topic.

Offline smbakerTopic starter

  • Regular Contributor
  • *
  • Posts: 211
  • Country: us
    • Scott's Electronics & Sandrail & Old BBS Game Blog
Inductor current rating and temperature
« on: March 30, 2019, 01:34:38 am »
I have some switching power supply projects, and I chose this inductor, PA4309.104NLT, rated 3.5A. The issue I have is that I'm measuring it's temperature at around 130C with 1.4A going through it, and 190C with 2.2A going through it. This greatly exceeds the 125C rating of the inductor. Plus, it smells bad.

Datasheet says "The rated current as listed is the heating current that the part can withstand with acceptable temperature rise in typical applications".

So what am I missing here? It's not like one typically heatsinks an inductor, does one?

Scott
 

Offline Circlotron

  • Super Contributor
  • ***
  • Posts: 3362
  • Country: au
Re: Inductor current rating and temperature
« Reply #1 on: March 30, 2019, 01:55:02 am »
Is the rating with forced air blowing on it? Also, some inductors can change their magnetic properties when they are too hot, but come good again when they cool down.
 

Offline drussell

  • Super Contributor
  • ***
  • Posts: 1855
  • Country: ca
  • Hardcore Geek
Re: Inductor current rating and temperature
« Reply #2 on: March 30, 2019, 02:17:45 am »
How is the inductor being used in the circuit?
 

Offline nukie

  • Frequent Contributor
  • **
  • Posts: 802
  • Country: au
Re: Inductor current rating and temperature
« Reply #3 on: March 30, 2019, 02:44:55 am »
Not some but all magnetics will lose performance so it's best to go up a notch with the saturation spec. Keep heat away from magnetics.

I am pretty sure inductors are not suppose to smell bad. What is the peak current at the inductor?
« Last Edit: March 30, 2019, 04:17:12 am by nukie »
 

Offline smbakerTopic starter

  • Regular Contributor
  • *
  • Posts: 211
  • Country: us
    • Scott's Electronics & Sandrail & Old BBS Game Blog
Re: Inductor current rating and temperature
« Reply #4 on: March 30, 2019, 03:42:50 am »
How is the inductor being used in the circuit?

It's the main inductor in an LM2576 buck converter. LM2576 datasheet says to oversize by 20-30%, hence my choice of a 3.5A inductor. Perhaps I should have choose a greater value of inductance. I'm usually building 5V boost converters where the 100uH is appropriate, this time I built a 12V where 150uH or 220uH may have been more appropriate.

I can derate the circuit down to an amp or so and be within a more reasonable temperate. I'm just curious more than anything.

Scott
 

Offline AndyC_772

  • Super Contributor
  • ***
  • Posts: 4315
  • Country: gb
  • Professional design engineer
    • Cawte Engineering | Reliable Electronics
Re: Inductor current rating and temperature
« Reply #5 on: March 30, 2019, 06:42:43 am »
What's the efficiency of the power supply? How does it vary with load?

It sounds as though the inductor might be saturating, which means it'll behave more like a resistor, and losses will go up dramatically.

What do you see if you scope the switching waveform? Does it look clean?

Offline Jay_Diddy_B

  • Super Contributor
  • ***
  • Posts: 2766
  • Country: ca
Re: Inductor current rating and temperature
« Reply #6 on: March 30, 2019, 07:28:01 am »
Hi,

I think that the Pulse PA4309.104NLT is wildly optimistically rated for 3.5A.

Consider this:



The DC Resistance is given as 405m \$\Omega\$ at 25C.
adjusting for a typical operating temp of 100C and calculating power dissipation from DCR alone (Ignoring core losses)



And then dissipating that much power in this size device



Further If I search Digikey asking for 100uH 3 to 4A and in stock I get (partial result shown):



The pulse parts stand out, other vendors are 40 to 80m \$\Omega\$

Draw your own conclusion  :D

Have you tried using one of the, much bigger, inductors recommended on the TI datasheet?

Other consideration, the LM2576 is first generation simple switcher. You would benefit from a later device with a higher switching frequency.

Regards,

Jay_Diddy_B

« Last Edit: March 30, 2019, 07:37:56 am by Jay_Diddy_B »
 

Offline smbakerTopic starter

  • Regular Contributor
  • *
  • Posts: 211
  • Country: us
    • Scott's Electronics & Sandrail & Old BBS Game Blog
Re: Inductor current rating and temperature
« Reply #7 on: March 30, 2019, 03:30:07 pm »
Thanks Jay,

I do have some LM2596 on order, which is a 150 KHz regulator, and I have some XL4005 on order, which is a 300 KHz regulator. The LM2596 should be a drop-in replacement. The XL4005 has an annoyingly different pinout and is used in some cheap eBay switching, so I figured I'd give it a shot too.

I can't go with a physically larger inductor as board space does not allow.

Scott
 

Offline emu_herder

  • Newbie
  • Posts: 4
  • Country: us
Re: Inductor current rating and temperature
« Reply #8 on: March 30, 2019, 04:25:04 pm »
I'd stay away from those unshielded drum cores in general, they are all kinds of trouble.

In that footprint I'd guess that you'd need to be around 22uH to get up to 3.5A. Look at Bourns SRR1208-220ML for instance, that's a shielded 0.5" 22uH inductor with a 3.5A rating. At 150kHz you may need more like 33-47uH though. If you can handle being limited to 3A I'd try the 33uH from that series at 150kHz.
 

Offline smbakerTopic starter

  • Regular Contributor
  • *
  • Posts: 211
  • Country: us
    • Scott's Electronics & Sandrail & Old BBS Game Blog
Re: Inductor current rating and temperature
« Reply #9 on: April 05, 2019, 12:59:53 am »
I received my LM2596 150KHz switching regulators today! Pictures of the regulator and the switching waveform are attached!

I saved so much money buying these on eBay rather than digikey.


« Last Edit: April 05, 2019, 01:08:31 am by smbaker »
 

Offline jbb

  • Super Contributor
  • ***
  • Posts: 1265
  • Country: nz
Re: Inductor current rating and temperature
« Reply #10 on: April 05, 2019, 01:18:26 am »
Note inductors often have 2 current ratings: the saturation current (above which available inductance drops off rapidly) and thermal current. Saturation currents are often higher and thus may be used for marketing...
 

Offline Jay_Diddy_B

  • Super Contributor
  • ***
  • Posts: 2766
  • Country: ca
Re: Inductor current rating and temperature
« Reply #11 on: April 05, 2019, 02:22:25 am »
Note inductors often have 2 current ratings: the saturation current (above which available inductance drops off rapidly) and thermal current. Saturation currents are often higher and thus may be used for marketing...

I agree.

In the case of the Pulse datasheet, it is defined in note 3:



Jay_Diddy_B
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 9327
  • Country: fi
Re: Inductor current rating and temperature
« Reply #12 on: April 05, 2019, 05:49:10 pm »
The datasheet only lists ratings based on DC copper loss, which is only relevant when passing a DC current through the inductor.

In a switching converter, there is an additional AC loss, caused by copper AC loss (skin effect) plus core loss. Sometimes the AC loss is negligible; sometimes it's dominant! It's highly dependent on the material properties of the inductor core. Some core materials are completely unsuitable for switch mode converters. Some are so freaking good that you can ignore the AC loss completely and simply work with the datasheet DC current rating.

But, with no data about the AC loss in the datasheet, your only way is to measure, which is basically what you have done (based on temperature measurement), and found the inductor unsuitable.

Nowadays, many inductor manufacturers do provide AC loss analysis software (often web based), based on their measurements and modeling. If available, use such software to find the total loss.
« Last Edit: April 05, 2019, 05:51:15 pm by Siwastaja »
 
The following users thanked this post: Someone

Offline smbakerTopic starter

  • Regular Contributor
  • *
  • Posts: 211
  • Country: us
    • Scott's Electronics & Sandrail & Old BBS Game Blog
Re: Inductor current rating and temperature
« Reply #13 on: April 06, 2019, 06:54:56 pm »
I received my LM2596 150KHz switching regulators today! Pictures of the regulator and the switching waveform are attached!

I saved so much money buying these on eBay rather than digikey.

I'm surprised nobody commented on (what I believe to be) fake LM2596. Maybe it flew under the radar. ;) I've received fake vintage ICs before, most notably the SP0256A-AL2, but this is the first time I've received a fake on something as mundane as a voltage regulator. The IC is clearly marked LM2596, but the waveform appears to be 50KHz, not 150KHz. Can someone confirm my logic that this has to be a fake?
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf