I agree that a 259B does not make a good general purpose instrument for
measuring inductor values over the HF range. In fact, there are very few
instruments which are. One of the few I know of is the HP 4191A vector
impedance meter, but it's not likely to be found in many amateur
workshops. Making good impedance measurements at HF is often very much
more difficult than most people realize.

The 259B is, I maintain, a very good instrument for identifying core
materials and for use in the design of inductors, transformers, and
other magnetic components. I've used mine many times for the purpose and
gotten the results I expected. That was, I thought, the subject of this
thread, but it appears to have drifted elsewhere.

Roy Lewallen, W7EL

Owen Duffy wrote:

Roy, with respect, you are describing work-arounds for the inability of
the '259B to make useful measurements on inductors over about 200 ohms
reactance.

Sure, the instrument can be used to characterise a core, and that
information extrapolated to design an inductor with higher reactance, but
you cannot measure the larger inductance directly, or discover reliably,
the properties of the larger inductor like self resonance effects, or
loss.

I won't address defensive posts by others who seem to have chosen to
ignore my qualification "When talking about ferrite or powdered iron
cored inductors of reactance over 200 ohms".

An example of the traps: a chap recently confirmed to me that indeed mu
is frequency sensitive as demonstrated by the '259B measurement of the
inductance of an inductor over HF which showed inductance was highly
frequency dependent. The problem was that the inductive reactance was
over 250 ohms at most frequencies of measurement, and the '259B
calculates inductance without warning that the value is unreliable
because of the magnitude of reactance on which the inductance is
calculated. Try measuring a 30uH coil's inductance over 2-30MHz using a
'259B and you will see what I mean.

Owen

Roy Lewallen wrote:
I agree that a 259B does not make a good general purpose instrument for
measuring inductor values over the HF range. In fact, there are very few
instruments which are. One of the few I know of is the HP 4191A vector
impedance meter, but it's not likely to be found in many amateur
workshops. Making good impedance measurements at HF is often very much
more difficult than most people realize.


Look at the L/C meter from AADE (Almost All Digital Electronics)
www.aade.com which a 1 hour kit for $99. It is quite accurate and uses
a variable frequency oscillator so that smaller values of inductance and
capacitance are measures at higher frequencies. I own an 8753 HP
Network Analyzer and still use the L/C meter most of the time since it
is so convenient. It easily can tell the differences between
73/43/31/61 materials. I usually wrap 2 turns thru the torroid to do the
measurement.

73,
Larry, W0QE

Larry Benko wrote:

Look at the L/C meter from AADE (Almost All Digital Electronics)
www.aade.com which a 1 hour kit for $99. It is quite accurate and uses
a variable frequency oscillator so that smaller values of inductance and
capacitance are measures at higher frequencies. I own an 8753 HP
Network Analyzer and still use the L/C meter most of the time since it
is so convenient. It easily can tell the differences between
73/43/31/61 materials. I usually wrap 2 turns thru the torroid to do the
measurement.

73,
Larry, W0QE

That looks like a handy gadget. It wouldn't be very good for working
with the kinds of ferrites often used for baluns, wideband transformers,
or EMI suppression because of the very low Q encountered. Many of the
ferrites I use have a Q of one or less over at least part of the
frequency range of operation. And of course the L/C meter wouldn't be
useful in assessing the loss or Q of inductors at all. One of the unique
things about the HP unit is that it's able to resolve very small R in
the presence of large X or vice-versa, which is very difficult to do.
But then it costs a couple of orders of magnitude more.

A network analyzer is poorly suited for making measurements of
impedances which are a great deal higher or lower than 50 ohms.

The problem with any of these instruments is that they'll all happily
give you an answer, often to several decimal digits. But as Owen pointed
out for the 259B, the answer can be anywhere from good to total garbage.
The only way to know which is to understand the limitations of your test
equipment and something about the nature of the component you're trying
to measure. There just isn't any instrument which will do the thinking
for you.

Roy Lewallen, W7EL

Roy Lewallen wrote:

That looks like a handy gadget. It wouldn't be very good for working
with the kinds of ferrites often used for baluns, wideband transformers,
or EMI suppression because of the very low Q encountered. Many of the
ferrites I use have a Q of one or less over at least part of the
frequency range of operation. And of course the L/C meter wouldn't be
useful in assessing the loss or Q of inductors at all. One of the unique
things about the HP unit is that it's able to resolve very small R in
the presence of large X or vice-versa, which is very difficult to do.
But then it costs a couple of orders of magnitude more.

A network analyzer is poorly suited for making measurements of
impedances which are a great deal higher or lower than 50 ohms.

The problem with any of these instruments is that they'll all happily
give you an answer, often to several decimal digits. But as Owen pointed
out for the 259B, the answer can be anywhere from good to total garbage.
The only way to know which is to understand the limitations of your test
equipment and something about the nature of the component you're trying
to measure. There just isn't any instrument which will do the thinking
for you.

Roy Lewallen, W7EL

That is what I thought also Roy but the little L/C meter seems to work
pretty well with low Q inductors. It is not measuring impedance but
measuring a frequency of an inductor and known capacitor in an
oscillator. The Q does change the oscillating frequency but not that
much. As I said before it is super easy to tell the difference between
77/31/43/61 type ferrites. Of course the permeability difference
between those types is a factor of 2 or more so ultra accuracy is not
important. The original question was how to determine what type an
unknown core was and not what impedance it represented at a certain
frequency.

I just grabbed 3 FairRite 59XX003801 torroids (2.4" OD x 1.4" ID x .5")
in 3 materials and measured a single turn with the L/C meter.

Material 61 (u=125) calc. inductance = .173uH, measured = .177uH
Material 43 (u=800) calc. inductance = 1.07uH, measured = .930uH
Material 75 (u=5000) calc. inductance = 6.85uH, measured = 7.39uH

The calculated inductances came from the FairRite catalog. I would say
the "low Q" inductors measured pretty well and the materilas were very
easy to distinguish.

73,
Larry, W0QE

Larry Benko wrote:

That is what I thought also Roy but the little L/C meter seems to work
pretty well with low Q inductors. It is not measuring impedance but
measuring a frequency of an inductor and known capacitor in an
oscillator. The Q does change the oscillating frequency but not that
much. As I said before it is super easy to tell the difference between
77/31/43/61 type ferrites. Of course the permeability difference
between those types is a factor of 2 or more so ultra accuracy is not
important. The original question was how to determine what type an
unknown core was and not what impedance it represented at a certain
frequency.

I just grabbed 3 FairRite 59XX003801 torroids (2.4" OD x 1.4" ID x .5")
in 3 materials and measured a single turn with the L/C meter.

Material 61 (u=125) calc. inductance = .173uH, measured = .177uH
Material 43 (u=800) calc. inductance = 1.07uH, measured = .930uH
Material 75 (u=5000) calc. inductance = 6.85uH, measured = 7.39uH

The calculated inductances came from the FairRite catalog. I would say
the "low Q" inductors measured pretty well and the materilas were very
easy to distinguish.

Yep, that's perfectly adequate to distinguish the types. You must have
made the measurements at a pretty low frequency. Type 75 has a Q of 1 at
about a MHz, and 43 at a few MHz, and they drop at higher frequencies.
Will the circuit oscillate with Q that low?

Roy Lewallen, W7EL

Roy Lewallen wrote:


Yep, that's perfectly adequate to distinguish the types. You must have
made the measurements at a pretty low frequency. Type 75 has a Q of 1 at
about a MHz, and 43 at a few MHz, and they drop at higher frequencies.
Will the circuit oscillate with Q that low?

Roy Lewallen, W7EL

Referring to http://www.aade.com/lcm2binst/HP.html the measurement freq.
was about 750KHz for the type 61 and 43 materials and about 70KHz for
the 75 material based on the inductances that were displayed.

73, Larry W0QE

Larry Benko wrote:
Roy Lewallen wrote:


Yep, that's perfectly adequate to distinguish the types. You must have
made the measurements at a pretty low frequency. Type 75 has a Q of 1
at about a MHz, and 43 at a few MHz, and they drop at higher
frequencies. Will the circuit oscillate with Q that low?

Roy Lewallen, W7EL


Referring to http://www.aade.com/lcm2binst/HP.html the measurement freq.
was about 750KHz for the type 61 and 43 materials and about 70KHz for
the 75 material based on the inductances that were displayed.

73, Larry W0QE

Darn typo. That should be 750KHz for the type 61 and 43 and 700KHz for
the type 75. Sorry.

73,Larry W0QE

Roy Lewallen wrote:

The 259B is, I maintain, a very good instrument for identifying core
materials and for use in the design of inductors, transformers, and
other magnetic components. I've used mine many times for the purpose
and gotten the results I expected. That was, I thought, the subject of
this thread, but it appears to have drifted elsewhere.


It has certainly drifted away from Australia, and towards the USA.

In the USA, it's a good bet that an unknown ferrite core will be made by
Fair-Rite, and probably one of the more common materials; or else it's
probably a dust-iron core from Micrometals. With help from the catalogs,
and a few known cores for reference, even quite limited test equipment
will have a good chance of identifying the specific product.

But that may not be true in the rest of the world. You may have a core
that is marketed in your home country but imported from another, but
having been manufactured in a third country using a process licensed
from... well, who knows any more? There are no world-standard sizes, and
no direct equivalents between magnetic materials from different
manufacturers. The best you can hope for is to identify the material as
being "somewhat like" a known Fair-Rite mix.

With such uncertainties about the material itself, you can afford far
fewer uncertainties about the measurement. If you don't have advanced
test equipment (or an advanced understanding of the limitations of
simpler equipment) then it may be better to forget about characterizing
the magnetic material. If you want to know if it will work in an HF
balun, it may be much easier to *make* a balun.

On the other hand, if you really want to chase down the problem of
identifying and characterizing unknown cores from anywhere around the
world, the following links may help.

http://users.catchnet.com.au/~rjandusimports/
No coincidence that this site is .au - they know about this problem in
Australia.

There are very useful international cross-reference pages at:
http://users.catchnet.com.au/~rjandu.../xref_mat.html
http://users.catchnet.com.au/~rjandu...ze_toroid.html
Do remember that these are not exact equivalents, only the closest
available. Also note the huge gaps in the tables, where nothing even
comes close.

Another very useful resource is DL5SWB's Mini Ring Core Calculator:
http://www.dl5swb.de/html/mini_ring_core_calculator.htm
This software contains extensive libraries of cores from international
manufacturers, including dimensions and paint colours as well as
magnetic properties. If you know the identity of the core, it will
calculate the inductance from the number of turns. If you have an
unknown core, and can make some meaningful measurements, there are
separate functions to help identify it.


--

73 from Ian GM3SEK 'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek