Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

[SOLVED] Choosing the right inductor for smps

Status
Not open for further replies.

Salvu

Junior Member level 1
Joined
Jan 21, 2017
Messages
17
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
189
Dear all, I have been experimenting with SMPS for quite a while now but I have found that when choosing output inductors a few problems occurred especially regarding voltage drop and the output inductor getting hot.

I have been using Excellent IT during my design for half bridge topology and this states the value of the output inductor according to the minimum load and also its value if the smps is being regulated or not. This will change to a higher value if the smps is regulated
I also learnt that the output inductor has to be AIR GAPPED in its construction of the material used.

My experience so far is that I noted considerable voltage drop and one has to allow quite some headroom in the output voltage to compensate for this being in the region of 25% Choosing the right core and its size comes somewhat difficult at times as some compromises has to be dealt with regarding minimum load and maximum voltage drop allowed at full load, and also the size of the core as these tend to get rather hot. When they do it may seem that they will tend to change their value and I don't really know if they tend to saturate or not.

There are a lot of different ferrite cores available on line but choosing the right one comes to get me scratching my head and could not make up which is the right one

My Questions.

1) How can I determine the right core physical size for the output inductor according to the output current?

2) Having a split output say 50-0-50v is it best to have a single ring core (common mode choke) or a separate core for each leg?

3) Why these core get so hot when loaded at high current say 8-10 amps?

4) Is it best to wind these core with Litz wire or solid enamel copper wire?

5) Any suggestions regarding color code and frequency of operation SMPS normally are working in the region of between 20 and 150Khz more or less?

6) Can anyone suggest where can I get hold of some kind of chart or list specifying material and what is it suitable for etc?

Regards, Salvu
 

Hi,

SMPS development documentation should contain information to choose the right core material and core size.
SMPS IC manufacturers and core manufacturers should offer those documentations, too.
A simple internet search gives a lots of hits "how to select inductors fir SMPS"

Reading a document like:
Answers almost all of your questions.

Klaus
 
  • Like
Reactions: Salvu

    Salvu

    Points: 2
    Helpful Answer Positive Rating
Thanks Klaus, The information was quite a help. I made need more info to make use of my cores in my junk box. Example color code, What size in relation with current handling etc.
 

Hi,

The core size is not related to current ... directely.
It more depends on output power and SMPS type (working principle).

Klaus

Added: ...and core size decreases with frequency
 
  • Like
Reactions: Salvu

    Salvu

    Points: 2
    Helpful Answer Positive Rating
  • Like
Reactions: Salvu

    Salvu

    Points: 2
    Helpful Answer Positive Rating
- - - Updated - - -

Thanks KlausST, so this will bring out the fact that according to the frequency and total power traveling through the core will bring out the physical size of the core. The more output inductance involved the more voltage drop and less ripple. On a previous tread:- I had some problems with the right value of inductance needed for my 1000 watt smps. For the time being I managed to solve the problem but I still have the heating issue. The frequency of operation is 63Khz the output inductance is 15uH on each leg and the yellow ring core that I have is 30mm dia and 10mm high. The output inductor is wound in common mode fashion. The maximum regulated output of 60v-0-60v was achieved and peaked 1200 watts. Do you think I should use two cores on top of each other thus they will have a larger cross sectional area? I will wind the core again in case.

Regards, Salvu

- - - Updated - - -

Thanks Smijesh, The link was very helpful in selecting the right inductor
 

Hi,

I'm not sure if we talk about the same...
Maybe it's time for you to post a schematic we can discuss.

Klaus
 

Thanks once again for your patience Klaus. I am uploading the schematic for you to take a look (Here I am talking of L2/3 which is a bifilar choke on the main output) the wire gauge is 1.5mm enameled copper wire. Turns are around 15 for each leg as I don't remember exactly but the choke was measured with an inductance meter and is in fact 15uH each side.

View attachment 1000W smps V2 schematic.pdf
 
Last edited:

Hi,

Now this is somehow different than I assumed

To be honest, I've never seen a coil in this configuration, nor I understand it's function.
Is it your idea or do you have a reference circuit/document that you can show us.

For me this inductance is about useless...but as said: I don't understand the circuit, yet.

Klaus
 

The Idea was taken from these two schematics (attached) in the second one however I think the value is mistaken for 32uH rather than 320uH.

Does not nearly every smps has an output inductor to minimize ripple? Without a minimum inductance (5uH) it was difficult to load the smps even when it was unregulated. It was tripping the overload protection even at start up when the bypass relay on the soft start resistor came on. As you can see I got quite some capacitance on the output being 8000uf in all. The smps was seeing a dead short exceeding 13 amps which was my threshold limit for current. So what is your opinion about the output inductor, do you really think its not needed at all? In that case I have to alter the circuit and controlling the soft start pin on the SG3525, and with the current trafo I have to create an over current fold back instead of the short circuit protection. I guess this will eliminate the surge current every time I will load the smps and also when initially charging the secondary caps. I don't really know how this will behave as I did not test this configuration in the event of a short circuit.
Regards, Salvu

View attachment Esquema Revisado_05.pdf View attachment smps_110_220vac_134.pdf output inductor.jpg SMPS 1000w.jpg
 

Hi,

if there is a coil for positive voltage and an independent coil for negative voltage: then everything is working fine.

But in your circuit both coils seem to be coupled and in your description you speak about "common mode": This won´t work.
Because the current in one path is in opposite to the current in the other path. --> No magnetic field will be generated. No inductance. No voltage drop. No smoothing.

Klaus
 
  • Like
Reactions: Salvu

    Salvu

    Points: 2
    Helpful Answer Positive Rating
Hello KlauST, Sorry if I was a bit misleading but I wanted to make myself understandable. The inductor you see on the previous post has 4 ends 2 for each rail. when I got it hooked in the wrong phase the voltage drop was enormous and nearly I could draw no current at all. I can say that inductance is playing its part and also is the cause why the inductor gets hot. Above everything else how can I cure the problem with the core heating up on heavy load and power. Any suggestions from your end please?

Thanks and regards Salvu
 

Hi,

--> choose two independent coils. Exactely like in both the schematics of post#10.

Klaus
 

Thanks I will do 1 separate inductor for each leg.

Regards, Salvu
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top