In this article we are going to understand in detail how we can create air gap in a ferrite core based transformer which we are using in smps circuits like flyback or boost or forward type.
Introduction
We must understand that ferrite cores have very high permeability and they saturate very fast if we do not put gap in it.
The air gap helps to store magnetic energy and prevent saturation. Without air gap the transformer may work for low power but it will not work properly in high power smps where energy is stored and released during switching.
So we must put proper air gap to make the transformer work in flyback or boost converter.
Why we need to put air gap
So in smps circuits like flyback or boost we are using the inductor or transformer to store energy when mosfet is on and then deliver that energy to output when mosfet is off. This energy is stored in the magnetic field of the core.
But ferrite core gets saturated very fast when magnetic flux becomes high. So then if we do not put gap in the core, it will saturate quickly and transformer will stop working and mosfet may burn also.
The air gap makes the core store more energy because it reduces the effective permeability and increases magnetic reluctance.
So now core can handle more flux without getting saturated. That is why in all energy storage type smps we always put air gap.
When to put air gap
We must remember that we do not put air gap in all types of transformers. We only put air gap in those designs where core is used for energy storage purpose. For example in flyback, boost, buck converter we need air gap because energy is stored in core during switch on time.
But in push pull, full bridge, half bridge forward type we may use very small air gap or no air gap because energy is transferred directly and not stored. So we must check what type of smps we are building and then decide to keep gap or not.
Types of core suitable for air gap
Mostly we use E-core or EE-core or EI-core or UU-core or ER-core or pot-core for making ferrite transformer. All these cores can be gapped easily by putting small spacer between two mating faces.
But toroid core is not suitable for gapping because it has circular shape and no flat mating surface. So we always choose flat core where we can easily insert paper or plastic or washer to make uniform air gap. Most common core is ee ferrite core.
How to add air gap practically
So now we will understand how we can actually add air gap on ferrite core. We take both pieces of ferrite core, like e and e half. Normally they touch each other flat without gap. But we need to separate them little to add gap.
Now we can take thin non magnetic spacer material like paper or plastic or fiber sheet or even thin tape. The most common way is to take thin cardboard or paper from matchbox or capacitor box. We cut small square piece and put it on the center leg of the core.
We do not put spacer on all legs. We only put spacer on center leg which carries magnetic flux. The two side legs are for mechanical support. Then we keep the second e core on top and press it. Now both cores are not touching directly and there is small air gap equal to thickness of paper.
Can we put air gap on side limbs of the E-core, instead of center limb?
So now we see that in most ferrite E-cores we have three limbs. The middle limb is the one which carries most of the magnetic flux.
The two outer limbs are there for completing the magnetic loop and mechanical stability. Normally we are supposed to put the air gap only on the center limb because that is where the magnetic flux flows and that is where the energy storage happens.
But in real practical work many times the bobbin is already fitted and wound, and the winding covers the full center leg.
Then we cannot insert paper or tape on center limb unless we remove full coil. So that time people think can we put the air gap on side limbs instead because that is open and flat.
The answer is yes we can do it but we must understand what will happen. When we put the spacer on outer limbs instead of center limb, then magnetic reluctance is still increased, so the air gap still appears in the flux path.
The overall effective permeability of the core is still reduced. So energy storage becomes possible and saturation is avoided.
But the problem is that now the flux is passing through center limb fully but gap is on side limbs. So fringing flux will increase near the edge of the gap. This will cause little more leakage inductance.
Also there can be small magnetic imbalance because gap is not placed where flux is densest. This may slightly reduce efficiency or increase emi, but not always.
Hence in low to medium power smps circuit we can safely put the gap on side limbs if we do not have option to gap the center limb. Many people do this trick to avoid re-winding the transformer.
It still works nicely and gives good result. Only thing is we must try to divide the gap equally between two outer limbs, so that flux stays balanced.
For example we can uniformly put 0.2 mm paper on both outer limbs, and leave the center untouched. This way the core can still store energy and prevent saturation.
But in very high power smps, or when we are designing precision transformer with low leakage, then it is always better to put gap on center limb only.
In such cases we can pre-gap the core before inserting the bobbin. Or we can buy factory-gapped cores where center limb is already machined.
How much gap to add
This depends on power level and transformer turns and core area. But normally for smps transformer we start with gap of 0.2 millimeter to 1 millimeter depending on the application. For example in flyback for 20 watt we can keep 0.3 mm gap.
For 100 watt we can keep 0.5 mm or more. If we make gap too big then magnetizing inductance becomes low and current increases.
If we make gap too small then saturation comes early. So we have to test and find the best value. We can also calculate the gap by formula but practically we try different gaps and test current and efficiency.
Where to put the gap
We must always put the gap only on the center leg of the core. If we put gap on all three legs then the magnetic path becomes unbalanced and leakage flux increases and performance becomes poor. Also core loss may increase.
So we take only the middle leg and put gap there. If we use both e cores then we can divide gap into two equal parts and put half on each center leg.
For example if we need 0.4 mm total gap then we put 0.2 mm spacer on each core. That helps to make magnetic path centered and reduces fringing effect.
How to keep the cores fixed after adding gap
After putting the spacer between center legs, we press both cores and align properly. Then we use plastic tape or insulation tape and wrap the core tightly. We can also use cable tie to fix the cores. In high power design we can use clamp or glue to hold the cores tightly.
We must make sure that the gap does not change due to vibration or pressure. So we should always fix the core tightly after gapping.
How air gap affects the winding
When we add air gap the leakage flux increases slightly so we must keep primary and secondary winding close to each other to avoid leakage inductance.
Also air gap reduces inductance value so we need more primary turns to reach same inductance. But the benefit is that now core can store more energy and not saturate easily. So we get better current handling and more stable operation.
How to know if gap is correct
We can test by applying dc current to primary winding and check waveform. If core is saturating then current will rise sharply. We can also measure inductance using lcr meter and calculate peak current.
If transformer is working well without overheating and waveform is clean and output is correct then our air gap is correct. Otherwise we can increase or decrease the spacer thickness and test again.
Conclusion
So here we saw how we can add air gap in ferrite core for smps transformer. We only need to cut small piece of paper or plastic and insert between center legs of the core.
This helps to store more energy and prevent core saturation. We always put gap on center leg only and fix the core tightly after that.
This method is very useful in flyback, boost, and buck converter design. Without air gap transformer will not work correctly in energy storing smps. So air gap is small thing but very important for good smps transformer.