Truncated match.
PICList
Thread
'Oscillator dependent on inductance'
1997\04\05@162911
by
71177.202
Dear PICsters:
Any thoughts on an oscillator whose frequency is directly (i.e.,
linearly) dependent on the value of an inductor? I think that this
would be a nice way to measure inductance with a PIC.
Thanks!
Alan Cohen
1997\04\06@205926
by
Steve Hardy
> From: Alan Cohen <spam_OUT71177.202TakeThisOuTcompuserve.com>
>
> Any thoughts on an oscillator whose frequency is directly (i.e.,
> linearly) dependent on the value of an inductor? I think that this
> would be a nice way to measure inductance with a PIC.
I don't think there is any 'natural' way to do this. No, if you want
to measure inductance by using the inductor as part of a tuned circuit
then you should be measuring period not frequency. Period (1/f) does
indeed relate linearly to inductance if you are using an LR type
circuit. On the other hand, period is proportional to the square
root of inductance with an LC circuit.
Regards,
SJH
Canberra, Australia
1997\04\06@225412
by
Harold Hallikainen

I think there are better ways to measure an inductance than to
stick it in an oscillator and measure the frequency.
You COULD use the technique used in a lot of digital multimeters
to measure capacitance. They typically have a sine wave oscillator
(typically a Wein bridge, but with the PIC you could generate the sine
wave with PWM). The sine wave oscillator outputs a known voltage. This
is run through the unknown capacitor into a current to voltage converter
(inverting input of an op amp with a feedback resistor around it). They
then measure the voltage out of this op amp, which corresponds to the
current through the capacitor. Since the current through the capacitor
at a specified voltage and frequency is proportional to the capacity,
they can directly display the capacity (no calculations required).
To measure the inductor, you could use the same circuitry but
would need to do a division, since the current would be inversely
proportional to the inductance.
Another approach would be to measure the time constant (T=L/R).
You could also just drive the inductor with a voltage step and wait for
the current to get to some specified value, since V=L di/dt, a fixed
voltage should cause a linear current ramp. The trick is to measure the
current without getting a voltage drop. The best way would, again, be an
op amp current to voltage converter. Might also be able to use something
like a Motorola SenseFET.
Finally, another approach would be to drive the inductor with the
sine wave, then using fast A/D (if the PIC is fast enough), measure the
amplitude and phase of the current. From this you can calculate an
equivalent R+jX. With the specified frequency, you can determine the
inductance and Q, noting that inductance is a constant for the component
and Q varies with frequency.
The same circuit could be used to measure any component,
displaying R+jX at the specified frequency. With appropriate
calculations, you could display capacitance, inductance and resistance.
If the A/D will not keep up, you can measure the amplitude using
a synchronous rectifier (avoiding knee voltage of typical diodes). Use a
comparator to detect zero crossings. This can then drive a switch like a
4066 to close on positive half cycles. Run the resulting signal through
a low pass filter and measure the resulting amplitude.
Phase can be measured by taking that comparator output and
driving one input of an exclusive or gate. Drive the other input of the
gate with a square wave equivalent of the sine wave drive. Run the
output of the exclusive or through a low pass filter and measure the
resulting DC. It will vary linearly with phase.
The sign of the phase can be determined by taking the output of
the above comparator and driving either the clock or D input of a D flip
flop. Drive the other input with a square wave equivalent of the sine
wave drive. The output of the D will be in one state for positive
(leading) phase and the other state for negative (lagging) phase.
Harold
1997\04\08@100643
by
Octavio Nogueira

> De: Alan Cohen <.....71177.202KILLspam@spam@COMPUSERVE.COM>
> Para: PICLISTKILLspamMITVMA.MIT.EDU
> Assunto: Oscillator dependent on inductance
> Data: Sabado, Abril 05, 1997 06:16
>
> Dear PICsters:
>
> Any thoughts on an oscillator whose frequency is directly (i.e.,
> linearly) dependent on the value of an inductor? I think that this
> would be a nice way to measure inductance with a PIC.
>
> Thanks!
>
> Alan Cohen
There is an article in Eletronics Now magazine in June 1996 issue
that has what you want, it's even connected to PIC.
Regards,
Octavio
========================================================
Octavio Nogueira
email: .....nogueiraKILLspam.....mandic.com.br
homepage: www.geocities.com/SiliconValley/Pines/6902/index.html
voice/fax: +55 11 2406474
"ProPic" The first Production PIC Programmer running in
Windows and under US$ 20.00.
========================================================
More... (looser matching)
 Last day of these posts
 In 1997
, 1998 only
 Today
 New search...