A novel way to measure magnetic fields, and DC current without a shunt?

field,

ferrite core of one of those,

Tell that to the poor sods that had to adjust pre-digital mobile radio communication equipment. I think they'll beat you to death using tooth picks :-)

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

gnetic field,

ferrite core of one of those,

e precise.

Hall sensors.

e as a simulated magnetic field from

n I expected!

=A0 =A0 =A0 =A0 =A0 =A0 =A0--------------------- +5V from USB

=A0 =A0 =A0 =A0 =A0 =A0 | =A0 =A0 =A0 =A0 =A0|

=A0 =A0 =A0 =A0 =A0 =A0 | =A0 =A0 =A0 =A0 =A0|

=A0 =A0 =A0 =A0|---- =A0d =A0 =A0 =A0 =A0|

=A0 =A0 =A0 =A0|

=A0 =A0 =A0 |---- =A0s =A0 =A0 =A0 =3D=3D=3D

80p =A0 =A0 =A0 =A0| =A0 =A0 =A0 =A0 ---

__ | =A0 =A0 =A0 =A0 =A0| 47 uF

=A0 =A0 =A0| =A0 =A0 =A0 =A0 =A0|

=A0 =A0[ ]1k5 =A0 =A0 =A0|

=A0 =A0 =A0 =A0 =A0 =A0| =A0 =A0 =A0 =A0 =A0|

=A0 =A0 =A0 =A0/// =A0 =A0 =A0 =A0///

st the inductor!

ture effects get cancelled,

you can do all sorts

se...

your sqrt LC if you like),

o play with, as they are :-)

0... 9...

pg

jpg

lemented in the PIC.

t

de quoted text -

Yup all that, And if you are measuring a current then you have to be very careful about the other magnetic items near your coil.... If someone brings a piece of iron near your magnetic field sensing coil that could cause an additional error.

But it sounds like fun.

George H.

periods...

meter with RS232 output.

A flux gate is more accurate, and doesn't take a lot more work. Flipping it around helps get rid of remanence.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

etic field,

I think I just thought of a very different way to do this.

An inductor with a field biasing its core will show a nonlinear effect that is asymmetrical. If the inductor is fed with the sum of two square waves at different frequencies, it will make sum and difference frequencies on its voltage.

Drive: F1 =3D F/3 F2 =3D F/5

Output: Fsum =3D F/3 + F/5 =3D 8*F/15 Fdif =3D F/3 - F/5 =3D 2*F/15

On a sunny day (Mon, 15 Feb 2010 13:48:16 GMT) it happened snipped-for-privacy@puntnl.niks (Nico Coesel) wrote in :

magnetic field,

ferrite core of one of those,

I had a commercial one myself, and only on SSB was the (tube) drift annoying. Making a good VCO is a art. There are zilions of tuned circuits still in use at the frequency I am using here. Every early PAL receiver had a h/2 (7.8kHz) tuned circuit. Never ever was a failure due to 'frequency drift'. Zillions of MW radios work just fine. AFC only became important at much higher frequencies like FM.

field,

ferrite core of one of those,

precise.

sensors.

simulated magnetic field from

expected!

from USB

inductor!

effects get cancelled,

can do all sorts

sqrt LC if you like),

with, as they are :-)

implemented in the PIC.

Two problems with using the oscillator as the null detector in a current-feedback loop:

1. The impedance of the external and feedback circuits will both affect the oscillation frequency.
2. The sign of the current error is lost, so closing a loop is tricky.

The classic DCCT uses second-harmonic distortion to detect the null between the unknown current and the feedback current. That preserves the sign of the error. They also use two cores and a clever threading scheme to keep the AC impedance of the "unknown" circuit from affecting the local control loops.

John

inductor.

periods...

frequency meter with RS232 output.

My flux gate controller chip, posted about earlier, did "flip" and auto-zero between EVERY measurement phase. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

field,

Look at the Almost All Digital Electronics (AADE) LCR meter. It has one oscillator and pushbutton switches to substitute an unknown L or C for the known L and C in its SINGLE oscillator. A uP counts the frequency differential and calculate the unknown.

You could do something similar with two identical inductors and a relay contact to switch between the "known" unbiased inductor and the "unknown" biased inductor. You coould also have a calibration routine to cancel out any small differential between the two inductors.

--
Joe Leikhim K4SAT
"The RFI-EMI-GUY"©

"Use only Genuine Interocitor Parts" Tom Servo  ;-P

Correct, but that 'nonlinear' effect is highly amplitude-dependent. So your 'sum of two square waves' has to have a drive impedance that makes the core's impedance irrelevant - that might be hard to do. It's also hard to age/temperature predict the effect (magnetic parts do things like vibrate, which can do troublesome things).

Am 15.02.2010 22:32, schrieb Jim Thompson: ...

I couldn't find any newer flux gate magnetometer from Honeywell but many AMR type chips. Those had the Set/Reset-Coils.

1A@3V sounds like a lot of power, but the duration of the pulses is ~1µs and you need two pulses vor every measurment.

I used a simple push-pull and a 1µF capacitor before Honeywell introduced the HMC5843 with integrated analog circuitry.

Falk

field,

ferrite core of one of those,

precise.

sensors.

simulated magnetic field from

expected!

from USB

inductor!

effects get cancelled,

can do all sorts

sqrt LC if you like),

with, as they are :-)

implemented in the PIC.

Ahh...what i suggested..feedback to null out core drive.

It seems that making the current a constant value would be the easy part. Making it not react to the temperature shifts would be the harder part. For that it may be best to provide another winding that brings the field seen by the core down to zero. The servo to hold the field at zero, would give an output that didn't depend on the "gain" of the system. This would also make the current control issue less important.