| Author |
LC oscillator to determine inductance
|
|
| gfretwell@aol.com 2007-05-06, 3:25 am |
| I have a wound coil about 3 feet long and an iron rod that goes
through it (in and out). I want to measure the travel. Will a LC
oscillator (Colpitts?) work to determine how much of the rod is in the
core by measuring the frequency? Is it linear?
| |
| Salmon Egg 2007-05-06, 1:25 pm |
| On 5/5/07 10:09 PM, in article 9goq33tss97ngmqt2t9n0skqhrolnep9j9@4ax.com,
"gfretwell@aol.com" <gfretwell@aol.com> wrote:
> I have a wound coil about 3 feet long and an iron rod that goes
> through it (in and out). I want to measure the travel. Will a LC
> oscillator (Colpitts?) work to determine how much of the rod is in the
> core by measuring the frequency? Is it linear?
What frequency are you talking about? Is the rod solid? Laminated? Powdered?
Just about anything you measure will will be difficult to correlate with
frequency other than by making an experimental table. Moreover, the
inductance will be nonlinear.
Bill
-- Fermez le Bush--about two years to go.
| |
|
| gfretwell@aol.com wrote:
> I have a wound coil about 3 feet long and an iron rod that goes
> through it (in and out). I want to measure the travel. Will a LC
> oscillator (Colpitts?) work to determine how much of the rod is in the
> core by measuring the frequency? Is it linear?
Interesting experiment. This may help:
http://myweb.cableone.net/adamsmed2/pto.htm
Ed
| |
| gfretwell@aol.com 2007-05-07, 3:25 am |
| On Sun, 06 May 2007 20:08:52 GMT, ehsjr <ehsjr@bellatlantic.net>
wrote:
>gfretwell@aol.com wrote:
>
>Interesting experiment. This may help:
>http://myweb.cableone.net/adamsmed2/pto.htm
>
>Ed
Thanks Ed, This is just an extension of my linear transformer idea., I
ran into "production" problems and I decided to reduce the windings by
50%. (the transformer worked but it didn't last)
It turns out frequecy measuring is easier than voltage measuring
anyway.
Current plan is a 3/8 or 1/2" hot dipped galvanized rod transiiting a
3' section of 3/4" PVC with a buttload of 30ga windings on it (some
good weather tolerant insulated wire this time)
Can I deal with the lack of linearity by loading up the windings
toward one end?
I guess I really slept through reactive circuits in AC basics ;-)
| |
| gfretwell@aol.com 2007-05-07, 3:25 am |
| On Sun, 06 May 2007 20:08:52 GMT, ehsjr <ehsjr@bellatlantic.net>
wrote:
>gfretwell@aol.com wrote:
>
>Interesting experiment. This may help:
>http://myweb.cableone.net/adamsmed2/pto.htm
>
>Ed
BTW the "brass" is not lost on me but I can live with steel if it
gives me a better delta. This is going to be over salt water. Brass is
certyainly attractive if expensive.
| |
|
| gfretwell@aol.com wrote:
> On Sun, 06 May 2007 20:08:52 GMT, ehsjr <ehsjr@bellatlantic.net>
> wrote:
>
>
>
>
>
> Thanks Ed, This is just an extension of my linear transformer idea., I
> ran into "production" problems and I decided to reduce the windings by
> 50%. (the transformer worked but it didn't last)
> It turns out frequecy measuring is easier than voltage measuring
> anyway.
>
> Current plan is a 3/8 or 1/2" hot dipped galvanized rod transiiting a
> 3' section of 3/4" PVC with a buttload of 30ga windings on it (some
> good weather tolerant insulated wire this time)
> Can I deal with the lack of linearity by loading up the windings
> toward one end?
You can experiment with that, but I'm not sure what would be
most practical. That's a big coil - seems like it would be
awkward to wind/rewind/rewind/rewind until you get it as
good as you can. It might be easier to use a lookup table with
a linearly wound coil.
How did you deal with the non-linearity with the coil
that worked?
Ed
> I guess I really slept through reactive circuits in AC basics ;-)
>
| |
|
|
gfretwell@aol.com wrote:
>
> Thanks Ed, This is just an extension of my linear transformer idea., I
> ran into "production" problems and I decided to reduce the windings by
> 50%. (the transformer worked but it didn't last)
> It turns out frequecy measuring is easier than voltage measuring
> anyway.
Um, maybe! "Linear" transformers are the standard way to do this!
> Current plan is a 3/8 or 1/2" hot dipped galvanized rod transiiting a
> 3' section of 3/4" PVC with a buttload of 30ga windings on it (some
> good weather tolerant insulated wire this time)
> Can I deal with the lack of linearity by loading up the windings
> toward one end?
A coil can be made to determine the frequency an oscillator of some
type.
But there are a number of problems here. The first question is do you
HAVE to use an iron rod because of the rest of the application. A
solid iron rod not only has frequency limitations (which means you'll
probably have to use beaucoup turns of fine wire to get frequency
down, but also has conductivity.
The conductivity of the rod causes the "metal detector" effect. With a
standard oscillator type metal detector, the presence of conductors
allows eddy currents which send the frequency higher by reducing the
inductance of the coil (see brass screw example in the link someone
posted) The zinc coating can help do this as well. The ferromagnetic
properties of the rod will increase the inductance as it goes into the
coil lowering the oscillator frequency. The two effects tend to
fight each other. Best would be to use a ferrite rod in the coil
which solves the problem. But doesn't solve the problem of where to
get said rod! Custom ferrites are not so easy to find!
The final result won't be particularly linear but this is the 21st
century and that's what microprocessors are for. And you didn't raise
the question of accuracy. This kind of sensor will have a lot of
accuracy drift as well.
Good luck!
Benj
| |
| gfretwell@aol.com 2007-05-08, 1:25 pm |
| On Tue, 08 May 2007 04:21:57 GMT, ehsjr <ehsjr@bellatlantic.net>
wrote:
>You can experiment with that, but I'm not sure what would be
>most practical. That's a big coil - seems like it would be
>awkward to wind/rewind/rewind/rewind until you get it as
>good as you can. It might be easier to use a lookup table with
>a linearly wound coil.
>
>How did you deal with the non-linearity with the coil
>that worked?
I was still playing with the PC interface when the coil failed. A
little salt air destroyed magnet wire in a heartbeat.
This time I am going with waterproof insulation and potting the whole
thing. Absolute accuracy is not as important as repeatability. I was
just curious what linearity problems I will have with a core that
moves in an inductor. Is this log scale?
I may end up making more than one. Fortunately I have a huge spool of
the wire.
| |
| gfretwell@aol.com 2007-05-08, 1:25 pm |
| On 8 May 2007 00:16:32 -0700, Benj <bjacoby@iwaynet.net> wrote:
>
>gfretwell@aol.com wrote:
>
>Um, maybe! "Linear" transformers are the standard way to do this!
>
>
>A coil can be made to determine the frequency an oscillator of some
>type.
>
>But there are a number of problems here. The first question is do you
>HAVE to use an iron rod because of the rest of the application. A
>solid iron rod not only has frequency limitations (which means you'll
>probably have to use beaucoup turns of fine wire to get frequency
>down, but also has conductivity.
>
>The conductivity of the rod causes the "metal detector" effect. With a
>standard oscillator type metal detector, the presence of conductors
>allows eddy currents which send the frequency higher by reducing the
>inductance of the coil (see brass screw example in the link someone
>posted) The zinc coating can help do this as well. The ferromagnetic
>properties of the rod will increase the inductance as it goes into the
>coil lowering the oscillator frequency. The two effects tend to
>fight each other. Best would be to use a ferrite rod in the coil
>which solves the problem. But doesn't solve the problem of where to
>get said rod! Custom ferrites are not so easy to find!
>
>The final result won't be particularly linear but this is the 21st
>century and that's what microprocessors are for. And you didn't raise
>the question of accuracy. This kind of sensor will have a lot of
>accuracy drift as well.
>
>Good luck!
>
>Benj
Thanks Ben. Brass indeed may be the way to go.
| |
|
| gfretwell@aol.com wrote:
> On Tue, 08 May 2007 04:21:57 GMT, ehsjr <ehsjr@bellatlantic.net>
> wrote:
>
>
>
>
>
> I was still playing with the PC interface when the coil failed. A
> little salt air destroyed magnet wire in a heartbeat.
> This time I am going with waterproof insulation and potting the whole
> thing. Absolute accuracy is not as important as repeatability. I was
> just curious what linearity problems I will have with a core that
> moves in an inductor. Is this log scale?
> I may end up making more than one. Fortunately I have a huge spool of
> the wire.
I don't know, but I would think is it is a whole bunch more
linear than logarithmic. On small coils it is. If you don't
need absolute accuracy, my bet is a regular winding will be
fine.
Ed
|
|
|
|