Author Topic: DC to DC  (Read 1143 times)

0 Members and 1 Guest are viewing this topic.

Offline alizare368

  • Contributor
  • Posts: 43
  • Country: de
DC to DC
« on: April 17, 2019, 10:33:16 pm »
Hi
I design a dc to dc board ,that has an inductor of 47Uh , 5A ,when I use a Ring ferrite core, it doesn't get too hot .for simplicity, I used the design  EE16 core , but it gets very hot.

what is the problem?
Do you have any Idea?

Attached photo.
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 12717
  • Country: gb
  • Hero999
Re: DC to DC
« Reply #1 on: April 18, 2019, 01:36:15 am »
The wire appears to be on the thin side.

Please post a schematic.
 

Offline Buriedcode

  • Super Contributor
  • ***
  • Posts: 1156
  • Country: gb
Re: DC to DC
« Reply #2 on: April 18, 2019, 01:58:54 am »
Is the core gapped? What is the core material? 
 

Offline alizare368

  • Contributor
  • Posts: 43
  • Country: de
Re: DC to DC
« Reply #3 on: April 18, 2019, 11:06:13 pm »
hi
thank you for your help.
the core doesn't gap and the core material is ferrite .

 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 7118
  • Country: nz
Re: DC to DC
« Reply #4 on: April 18, 2019, 11:13:37 pm »
There are many different types of ferrite cores. Some will need a different frequency. Some will not work at all.

Your E inductor core is much smaller than your ring inductor core.
It maybe that the E core is too small to handle that much current.

Ring cores are very efficient shape, to get the same performance from an E core it needs to be bigger.
« Last Edit: May 02, 2019, 12:09:21 am by Psi »
Greek letter 'Psi' (not Pounds per Square Inch)
 
The following users thanked this post: alizare368

Offline Buriedcode

  • Super Contributor
  • ***
  • Posts: 1156
  • Country: gb
Re: DC to DC
« Reply #5 on: April 18, 2019, 11:27:25 pm »
Did you calculate how many windings are required for the desired inductance? Or Did you measure it?  If so, how did you measure it?

Ferrites have several parameters that are important when designing even a single inductor.  Size is just one of them.
 
The following users thanked this post: alizare368

Offline treez

  • Super Contributor
  • ***
  • Posts: 1046
  • Country: gb
Re: DC to DC
« Reply #6 on: April 19, 2019, 09:21:25 pm »
you calculate the saturation current of your ferrite with equation i(sat) = B(sat).A.N/L

A = core area
N = Number of turns
L = Inductance

You can also derive i(sat) by  Ampere’s Law…
I(sat) = [B(sat).dl ] / (uo.ur.N)
Eg for the torroid, if you don’t know it’s Area of cross section


By the way, if you wish,  here is a link to my free SMPS course
This is the link from my google drive

https://drive.google.com/open?id=1cmk81KxXA7-EliHHWr8ZimpEp7HCMIcY

Please tell me if it doesnt work.


..send me your full spec, and ill say what is your peak current etc.
eg tell your f(sw), vin, vout, schem, bom, etc etc
« Last Edit: April 19, 2019, 10:01:45 pm by treez »
 
The following users thanked this post: doktor pyta, 2N3055

Offline alizare368

  • Contributor
  • Posts: 43
  • Country: de
Re: DC to DC
« Reply #7 on: April 23, 2019, 01:07:18 am »
Hi
I'm working on this schematic,
and i set, the output is 12V . and my load  consumption is 3A.
When using the EE16 core, XL4016 is getting too hot.
 

Offline Buriedcode

  • Super Contributor
  • ***
  • Posts: 1156
  • Country: gb
Re: DC to DC
« Reply #8 on: April 23, 2019, 02:14:13 am »
Hi
I'm working on this schematic,
and i set, the output is 12V . and my load  consumption is 3A.
When using the EE16 core, XL4016 is getting too hot.

We understand the application, but we are asking about the parameters of the inductor you have made.  Is the core gapped?  How many turns? Did you measure the inductance or just put some windings on a random core and assume it would be the correct value?
The best place to start would be to test the inductor you have made. It sounds like it is saturating, or that its inductance is too low.  Did you measure its inductance?
 
The following users thanked this post: alizare368

Online soldar

  • Super Contributor
  • ***
  • Posts: 1976
  • Country: es
Re: DC to DC
« Reply #9 on: April 23, 2019, 06:54:55 am »
I would not be building my own inductors unless I knew very well what I was doing and had the tools to measure all parameters.

A saturated inductor is very much like no inductor at all.

Unless you are knowledgeable about making inductors you will save much aggravation by just buying them.
All my posts are made with 100% recycled electrons and bare traces of grey matter.
 

Offline TimNJ

  • Frequent Contributor
  • **
  • Posts: 758
  • Country: us
Re: DC to DC
« Reply #10 on: April 23, 2019, 09:14:20 am »
In my experience, designing with un-gapped powder-core toroids is pretty straightforward. (Not always, of course.) The Here's a datasheet for a Chang Sung toroid:

http://eu0707.cafe24.com/_eng/download/prod/OD127.pdf

The effective permeability of the core drops with DC bias, and as a result, the AL value drops too. (You can see this in the datasheet plot.) The AL value, or "inductance factor" is the relationship between number of turns and inductance.

So, if you know your max nominal current through the inductor, you can determine if your "inductor" will really be acting like an inductor at all.

Gapped ferrite cores are a bit more tricky due to the geometry and addition of the gap. You can take a look at the attached images for a procedure I have used with success. It looks a bit obtuse, but it's not too bad. I usually make an Excel spreadsheet so I can play with the parameters and see how things change.
 

Offline TimNJ

  • Frequent Contributor
  • **
  • Posts: 758
  • Country: us
Re: DC to DC
« Reply #11 on: April 24, 2019, 01:48:31 am »
For anyone's reference, here's an Excel design spreadsheet I made using this method.

Note 1: This spreadsheet does not include Step 1 "Determine the Core Size", as the core size was already determined by the product footprint. But, you probably don't want to do this unless you have an "intuitive feel" for this.
Note 2: This spreadsheet is for a boost converter. You will have to use different formulas for determining proper inductance and ripple current for other topologies.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 8657
  • Country: us
Re: DC to DC
« Reply #12 on: April 24, 2019, 02:26:24 am »
That toroid looks like powdered iron, which IIRC behaves like a gapped core. As has been mentioned already, two identical looking ferrite cores can have very different properties and these properties are important.
 

Offline TimNJ

  • Frequent Contributor
  • **
  • Posts: 758
  • Country: us
Re: DC to DC
« Reply #13 on: April 24, 2019, 07:00:26 am »
Powdered iron cores are sometimes called "distributed gap" cores. Instead of a single gap, as used with ferrite cores, you can think of a powdered iron core has having thousands (millions? billions?) of little gaps between each iron powder particle in the core. Inductors wound on powdered iron cores will have a much "softer" saturation curve. That is, there isn't a single point at which the inductance drops off sharply. It's gradual.

There are also many types of powdered cores. The cheapest is simply "powdered iron", but there are other types like MPP, Hi-flux, and Sendust. All have their tradeoffs, of course.
 

Offline aju11

  • Contributor
  • Posts: 23
  • Country: in
Re: DC to DC
« Reply #14 on: April 24, 2019, 07:42:04 pm »
Saturation
 

Offline treez

  • Super Contributor
  • ***
  • Posts: 1046
  • Country: gb
Re: DC to DC
« Reply #15 on: April 27, 2019, 11:01:09 pm »
here is an ltspice sim (ltspice is free) of your power supply
I presumed your vin was 24v
and your f(sw) was 100khz

Please tell me if it isnt.
Also attached the quik excel design file.

Its always good to do a sim of a power supply so you can get into it before making it.
Check your inductor peak current calculation etc.
« Last Edit: April 27, 2019, 11:04:57 pm by treez »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf