Inductance calculation

[boylesg]

I believe this circuit is a Wein Bridge Oscillator.

If I was to use this to calculate the inductance of a coil as per here: **broken link removed**, then what would I connect the other end of the coil to?

 

[BradtheRad]

I made a simulation. It oscillates when I add a resistor at the end of the coil. The resistor's other end goes to ground.

A value above 1200 ohms should work okay. Peak current draw is 10mA. That is well within the output spec for a 741.

Frequency is 1580 Hz.

Once it is oscillating, I can detach the resistor, and the oscillations continue.

If you want to use this circuit to measure inductance at 1580 Hz, it will only work for henry values above 50mH or so. Below that value, the 741 does not deliver sufficient current to bring out variations in the inductive time constant. What this means it that you'll get 12VAC output when you attach any coil of too low a henry value.

I believe that if you raise the frequency, that will allow you to measure henry values down in the mH or uH range.

 

[FvM]

I believe that if you raise the frequency, that will allow you to measure henry values down in the mH or uH range.

Surely, but not with a 741 based oscillator.

Supposed, the device under test is a power inductor and shall be measured somehow within it's intended operation frequency range, the bridge impedance will be still in an ohm rather than kohm range and thus require a generator with more output current than a simple OP can provide.

The measurement method in the link is by the way discarding half of the available information (series resistance respectively loss angle) by only using two of the three circuit voltages in the calculation.

 

[boylesg]

I made a simulation. It oscillates when I add a resistor at the end of the coil. The resistor's other end goes to ground.

A value above 1200 ohms should work okay. Peak current draw is 10mA. That is well within the output spec for a 741.

Frequency is 1580 Hz.

Once it is oscillating, I can detach the resistor, and the oscillations continue.

If you want to use this circuit to measure inductance at 1580 Hz, it will only work for henry values above 50mH or so. Below that value, the 741 does not deliver sufficient current to bring out variations in the inductive time constant. What this means it that you'll get 12VAC output when you attach any coil of too low a henry value.

I believe that if you raise the frequency, that will allow you to measure henry values down in the mH or uH range.

Thanks for that Brad.

 

[BradtheRad]

Surely, but not with a 741 based oscillator.

Supposed, the device under test is a power inductor and shall be measured somehow within it's intended operation frequency range, the bridge impedance will be still in an ohm rather than kohm range and thus require a generator with more output current than a simple OP can provide.

Yes, a 741 can provide 20 mA. Its output impedance is on the order of 50 ohms. It is severely loaded by a 10 uH coil at 1591 Hz due to its reactance being 1/10 ohm.

However if the frequency were 1.59 Mhz, then the 10uH coil would have a 100 ohm impedance. This gets into a range which will yield usable readings.

The measurement method in the link is by the way discarding half of the available information (series resistance respectively loss angle) by only using two of the three circuit voltages in the calculation.

You're right.

I have a hunch as to why the fellow only measures across the resistor.

It's because when we measure volts across the coil, there is more going on than just current through the coil. We need to wonder whether coil emf is adding or subtracting from the volt reading. We can read high volts, yet little current is going through.

By taking a volt reading across the resistor, we know it is in direct proportion to current.

It's one of those things about coils that makes them interesting.

 

[FvM]

However if the frequency were 1.59 Mhz, then the 10uH coil would have a 100 ohm impedance. This gets into a range which will yield usable readings.

Yes, but not with a 741, which exposes less than 1 MHz gain-bandwidth-product and about 20 kHz "power bandwidth" (maximum frequency of indistorted full level output). But of course the problem can be solved by amplifier circuits with larger bandwidth and/or higher output current.