# Measuring transformer inductances without LCR meter

#### cupoftea

Hi,
I need to measure inductances on a switch mode transformer. I have no LCR meter. Txfmr has single Pri and split sec.
But i do have a scope, and a DC power source.
I also have a Frequency response analyser....
(Voltech TF2000)
(frequency is adjustable from 100Hz to 100Khz and its a sine output)

I only have ceramic caps between 100pF to 470nF available.

Measure L(pri)
For the primary, i put a 12nF ceramic cap in series with the primary, and then fed a 5V peak sine wave to this LC series circuit, from the TF2000. I put the scope across the inductor. I found that the voltage amplitude peaked at approx 31kHz. As such , i calculate L(pri) to be approx 2.2mH.

Measure L(sec)
The secondary is obviously much lower inductance. So to measure L(sec) i put whole sec oil in parallel with a 470nF ceramic capacitor. I then put a 150R in series with this LC. I then put the scope across the LC. I then instantaneously 'tapped' 5VDC across the R--L//C.
...The resulting oscillation had T = 32us. This gives total L(sec) of 55.2uH. So each sec half is (55.2/2)uH

Can you think of any better ways to measure these inductances with this kit?

#### dick_freebird

Pulse generator w/ 0 - 1V narrow pulse output
such that core is reset after every pulse
series 1-ohm burden resistor (or CT-1, CT-2)

Observe current wvfm on 'scope, for dI/dt slope
V=L*dI/dt so L=(1V)/(dI/dt)

I'd put probes on both ends of inductor just so
you know Zsrc doesn't enter into it.

You can stretch the pulse width to find saturation
knee as well, if this is an interest (good luck, with
LCR meter) and the pulse gen has that capability.
If not consider using a wet relay and a power

cupoftea

### cupoftea

Points: 2

#### Easy peasy

clip a high Q cap in parallel with the Tx pri wdg, say 100nF, 400V, attach a 2k2 R to one end, drive the 2k2 and the other end of the Tx/cap combo with the sig gen, gnd to Tx/cap side,

put a scope probe across the Tx wdg ( this is also the cap ) AND look at the output of the sig gen too ( 2 channels ) common gnd

vary the sig gen freq, when you see the volts rise across the Tx - try and get the max when the 2 signals are in phase - this is resonance - note the freq, Fo.

2. .pi. Fo = 1 / SQRT ( L.C ), so 1 / L = C * ( Fo.2.pi. ) ^2

If you now short the sec side you can measure the leakage of the Tx - all referred to the primary ( much higher freq ).

#### cupoftea

clip a high Q cap in parallel with the Tx pri wdg, say 100nF, 400V, attach a 2k2 R to one end, drive the 2k2 and the other end of the Tx/cap combo with the sig gen, gnd to Tx/cap side,

Thanks, Do you mean a square wave output sig gen?....though i dont have a sig gen....only have a sine source, as discussed above.

#### Easy peasy

sine wave is usual ....

#### KlausST

##### Super Moderator
Staff member
Hi,

I guess you have some mains transformer and a scope.

low voltage mains frequecy --> resistor --> transformer

U and I should be in a useful value. I should be low enough not to saturate the core. Check with the scope.

Now you are free to do simple ... to advanced measurements and calculations.

From simple ideal_R, ideal_L
over additional winding resistance = coil loss
and phase shift to core loss calculation

Klaus

#### Taihung

##### Member level 1
Funny !
The useful way is use an opam with LC oscillator. If you known C and use scope to measure frequency, you can calculate L.
It also is the cheapest way to DIY LC meter (when using microcontroller to measure frequency). Just few \$.

Brg!