How to measure leakage inductance

[percy]

How can I find out the leakage inductance of any transformer ?
I know about the method of shorting the secondary and meaasuring the primary inductance, but is that all ? I tried that and I get about 55mh for a small power transformer (12-0-12 48VA). Is that real ? or is there more to it ?

[lndm]

http://www.diyaudio.com/forums/showt...219#post871219

[AndrewT]

Hi Percy,
are you any further forward with the shorting of primary or secondary?

Have you tried the alternative to find how much different the two measurements are?

I would guess that the smaller the transformer (more turns) the higher the inductance. Also the acceptable DC will also be lower for a small transformer, due to saturation of the very small area of iron.

[percy]

Hi Andrew,
last night I measured everything and here are the results. Below are reactances of the said windings at 60hz. Calculated inductance for that reactance and frequency is in parantheses ().

Primary:-
Open Secondary = 350ohm (.93H) *see note below
Short Secondary = 28ohm (74.3mh)

Secondary:-
Open Primary = 31ohm (82.3mh) **see note below
Short Primary = 1.7ohm (4.5mh)

*With open pri, the sec showed a distinct resonance peak at 9.5Khz.
**With open sec, the pri did not show a clear resonance like the sec but the reactance started dropping after about 11-12Khz.

Now how do I interpret these results ? How do I figure out the leakage inductance ? What does the resonance tell me ?

C'mon guys..help me out here..

[AndrewT]

Hi,

I hope someone comes to our aid.

You have to choose between 74mH and 4m5H.

I was hoping by asking you to measure them that the answers might have turned out the same. Some dreamworld I'm in.

Thanks for taking the trouble.

[subwo1]

I am going to go with the idea it is the primary inductance with the secondary shorted. One reason for my view is that the drive is on the primary. A good explanation I read is that it is an inductor in series with the primary which is not coupled with the secondary.

One complication with trying to factor in secondary leakage inductance into the figure is in the case in multiple secondaries. I'd say that the primary's figure is the one to use.

[mzzj]

Originally posted by percy
Hi Andrew,
last night I measured everything and here are the results. Below are reactances of the said windings at 60hz. Calculated inductance for that reactance and frequency is in parantheses ().

Primary:-
Open Secondary = 350ohm (.93H) *see note below
Short Secondary = 28ohm (74.3mh)

Secondary:-
Open Primary = 31ohm (82.3mh) **see note below
Short Primary = 1.7ohm (4.5mh)

*With open pri, the sec showed a distinct resonance peak at 9.5Khz.
**With open sec, the pri did not show a clear resonance like the sec but the reactance started dropping after about 11-12Khz.

Now how do I interpret these results ? How do I figure out the leakage inductance ? What does the resonance tell me ?

C'mon guys..help me out here..

What are the DC resistances of pri and sec?

[Eva]

You should substract DC winding resistances from the measured reactance values before using them (both the DCR from the winding being measured and the "reflected" DCR from the winding being shorted, that has to be adjusted multiplying it by by turns1^2/turns2^2).

The leakage inductance is actually a single magnitude, but it's not seen in the same way from the primary and from the secondary due to different turn counts. However, the relationship between both values must be also turns1^/turns2^2, otherwise something has gone wrong in the measurement.

[percy]

Pri DCR is 9.78 ohm and Sec DCR is 0.78 ohm.
as for the voltage/turns ratio - the ACV measured at the primary was 118.5V and at the secondary was 28.38V. So its 4.175:1 .

(both the DCR from the winding being measured and the "reflected" DCR from the winding being shorted, that has to be adjusted multiplying it by by turns1^2/turns2^2).

could you explain a little more. do you mean subtract 10.56 (9.78+0.78) from all reactance readings ? and then adjust so that the impedance ratio is close to the square of the voltage/turns ratio ?

[Eva]

You have to reflect the DC resistance of the secondary winding back to the primary and vice-versa.

So 0.78 ohms in the secondary are seen as 0.78*4.175^2=13.6 additional ohms in the primary. Also, 9.78 ohms in the primary are seen as 9.78/(4.175^2)=0.56 additional ohms in the secondary. The reflected value is usually very close to the value from the own winding, otherwise there may be some error in the measurements.

Thus, when measuring from the primary with the secondary shorted, you have to substract the primary DC resistance plus 13.6 ohms from the primary measured impedance in order to get the actual inductive reactance.

Also, when the primary is shorted and the measurement is done from the secondary, you have to substract the secondary DC resistance plus 0.56 ohms from the measured impedance in order to get the actual inductive reactance.

Furthermore, you should obtain a primary inductance value nearly 4.175^2 times bigger than the secondary inductance, otherwise something is wrong with the measurements.

Remember to substract from your measurements the resistance value that the multimeter produces when you touch directly both tips, otherwise this typical 0.3 or 0.4 ohm offset error will ruin your experiment (as secondary DCR is in that range).