A toasty, toasty choke / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

A toasty, toasty choke

1 2 next

Move Thread

LAN_403

Adam Munich

Tue Dec 29 2015, 05:54PM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

Hi all,

I designed a USB-powered magnetic field generator / inductive charger based on a royer oscillator with some modifications made to keep it stable. However, the inductor (orange) on the device gets hot enough to melt off its solder pads!

1451411637 2893 FT0 12380187 1024072150990540 992557104 N

As I understand it, the purpose of this inductor is to soften the supply rail such that when my mosfets turn on and dump energy into the LC tank, they don't have a dI/dt that's great enough to latch them on.

The inductor doesn't see any imaginary current, only real current. At 1A, this is below its DC rating.

However, the current is *pulsed* (not oscillating) at Fres * 2, in this case 1.6MHz

My theories are;

- The inductor is heating itself by running at 1.6MHz,
- The field coil is heating the inductor.

I'm leaning toward the former.

Has anyone encountered something similar? And if so, have you found a solution?

johnf

Tue Dec 29 2015, 05:58PM

Registered Member #230 Joined: Tue Feb 21 2006, 08:01PM
Location: Gracefield lower Hutt
Posts: 284

The heating is the losses in the ferrite that inductor uses.
It was probably designed for use around 100kHz.
This info is often left off the datasheet

Adam Munich

Tue Dec 29 2015, 06:02PM

Registered Member #2893 Joined: Tue Jun 01 2010, 09:25PM
Location: Cali-forn. i. a.
Posts: 2242

As I imagined, yeah.

I need a ~1A 39 to 100uH inductor with much less magnetizing reluctance.

Anyone know of a good inductor series?

DerAlbi

Tue Dec 29 2015, 10:42PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

The inductor doesn't see any imaginary current, only real current. At 1A, this is below its DC rating.

Can you explain what you mean by that (and where you would learn such thoughts)?

Anyway: you didnt tell the inductance nor the partname of the inductor.... Check out the datasheet yourself then. It should give you a self resonant frequency (SFR) of the inductor. Thats basically the highest frequency you can use the inductor.
1.6MHz has serious implications concerning the active wire-diameter due to skin-effect. The current rating is not a usefull rating here. There is also a second rating to think about: called Isat.
Also the ferrite does not support anywhere near that frequency (most likely).

Solution: use an inductor that is specified for your use. At such high frequency do you really need 10s of uH ?

Anyone know of a good inductor series?

You do know that for a usefully hint we would need to know your PCB layout ?
There is also no shame in using the parametric search of digikey to get a good overview. How did you find the components in the first place?

Electra

Wed Dec 30 2015, 01:32AM

Registered Member #816 Joined: Sun Jun 03 2007, 07:29PM
Location:
Posts: 156

How about DR127 Series ?

Test freq is yes 100khz but I believe it'll work at higher frequancys because your well below the V/us ratings allowed, if its running of USB I take it its 5v.
While some ferrite's are good to the low Mhz range It did not state the actual material to confirm this.

...

Wed Dec 30 2015, 02:25AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

I think that you misunderstand how the 'mazilli' circuit works, the inductor is doing a lot more than '...soften the supply rail such that when my mosfets turn on and dump energy into the LC tank, they don't have a dI/dt that's great enough to latch them on.'

It is not a normal royer (certainly the one that you have is not a royer, since there is no core to saturate), if anything is is a 'current fed resonant royer'. The inductor serves to convert the DC supply to a constant current source, and will have a significant voltage across it in operation (approximately Vin IIRC) so you need to have an inductor which can store enough energy in its core without saturating.

Sulaiman

Wed Dec 30 2015, 01:30PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

Assuming 800 kHz operating frequency (what FETs are you using ?)
skin-depth in copper means that c0.2 mm or less diameter wires should be used,
'Litz' wire would be very suitable.

Many ferrite cored inductors use manganese-zinc ferrite which is VERY lossy at 800 kHz
use a nickel-zinc ferrite core (e.g. a.m. radio ferrite rod) or air core

Align the d.c. constant-current inductor(s) to not pick up the magnetic field of the resonant inductor

you may also consider using 'Litz' wire for the resonant inductor to reduce heating there also.

DerAlbi

Wed Dec 30 2015, 01:58PM

Registered Member #2906 Joined: Sun Jun 06 2010, 02:20AM
Location: Dresden, Germany
Posts: 727

I know its a bit off-topic but how do you came to 0.2mm ?
At 800kHz the skin depth is 73um. At 1.6MHz its 51um.
I usually just double the skin depth (because its a radius) to determine the wire diameter... resulting in 0.1mm in that case. (which a 1A coil is surely not made of)
Could you tell me why you used factor 4 instead of factor 2? I mean.. if i was choosing the too small diameter all the time i would like to know

Sulaiman

Thu Dec 31 2015, 06:56AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

Wikipedia:
A convenient formula (attributed to F.E. Terman) for the diameter DW of a wire of circular cross-section whose resistance will increase by 10% at frequency f is:

D = 200mm / sqrt(F)

which comes out to 0.223 mm, I rounded down to a common gauge.

I have used this approximation many times (since before Wikipedia existed !) and it seems OK
of course the 'proximity effect' means measured ac resistance is usually higher.

ConKbot of Doom

Fri Jan 08 2016, 04:03PM

Registered Member #509 Joined: Sat Feb 10 2007, 07:02AM
Location:
Posts: 329

I find that coilcraft is good for magnetics, since they have a lot of information available on their website. Even a core loss calculator. For example, the RFB1010 series looks like it would take to this well.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob