# Inductance in parallel

Discussion in 'Electronic Design' started by terry, Dec 4, 2006.

1. ### terryGuest

Very basic question. (I hope)

I have a tuned circuit that covers a range of approx 250 to 600
kilohertz. I desire to increase its frequency range to at least double
(i.e. around 500 to 1200; or even to cover, say 750 to 1600.
Hopefully in a manner that does not require removing/changing/modifying
the existing inductance coil. Just paralleling it with something.

Thinking is based upon.

If capacitors are ) In parallel their joint capacity increase. b) In
series the joint capacitance is decreased. A simple example would be
two 500 mmf caps in series would nominally (stray capacitance aside)
give 250 mmf. Also increase of capacitance in say, a tuned circuit.
normally 'decreases' frequency.

If inductance are in series the joint inductance increases. Assume no
mutual coupling?
If in parallel the inductance decreases .........................
yes???

So, for an example if one inductance is paralleled by an identical one
the total effective inductance would be halved? etc.

Using the half remembered formula f = 1/2pi root of L xC it would
appear that inductance would have to decreased by a ratio based on the
the square of the inductance? In other words one quarter the inductance
(thus reducing the product of LxC) would double the frequency?

Anybody care to make this complicated please; or what is missing?

I also half remember inductance being affected by the number of turns
squared, modified by the permeability of the core material etc.

But it WAS nearly 55 years ago! So take it easy. Been rather busy
working and living for the last 53 years.

2. ### Richard HenryGuest

I'm jst assuming that you have a computer with Internat access right in
front of you, so...

Googling "parallel inductance" delivers "about 1, 030,000" hits, at
least some of which are good.

3. ### Andrew HolmeGuest

Correct

It sounds like you're trying to make a long wave coil cover medium wave.

A simpler way to get the lower inductance might be to split the long wave
coil in half i.e. tap it half way down; but you could add parallel
inductance instead. The only problem with this is that inductors are lossy
and it will lower the Q. You need to use a good quality parallel coil, and
screen it, or, if your other coil is a ferrite rod antenna, mount it in such
a way that it reinforces the signal in the main coil.

The short answer is yes, paralleling two equal inductors halves the
inductance.

Just be careful that any magnetic fields from the two inductors dont
interact, else the resultant inductance may be considerably different,
and dependent on the physical arrangement.

--
Regards,

Design Engineer J & K Micro Systems
Microcomputer solutions for industrial control

5. ### HKJGuest

http://www.miscel.dk/MiscEl/miscelFo.html

This is corrent.

If your work without cores (i.e. air coils) it is more difficult:
http://www.miscel.dk/MiscEl/miscelAirCoil.html

6. ### Tom BruhnsGuest

Generally, yes, as others have posted, inductances in parallel combine
just as resistances: L(net) = 1/(1/L1 + 1/L2 + ... + 1/Ln) As you
noted, the inductances must not share magnetic field. If they do,
things get more complicated, but you can calculate the effects, or just
measure them. In fact, it's under the condition that the turns
completely share the magnetic field that the inductance goes as the
square of the number of turns. That is far from the case with typical
solenoid RF coils. For example, a 1" diameter coil of 25 turns, 1"
long, will have an inductance of about 11.2uH. A 2" long coil of 50
turns (two of the original end-to-end) will have an inductance of about
26.5uH, a far cry from four times 11.2. On the other hand, if you put
50 turns where the original 25 were--50 turns in one inch length--you
do get very close to four times the original inductance.

One thing that others have not pointed out, that I've seen so far
anyway, is that large coils can have self-resonances not far above
their operating range. You will need to be careful that those
resonances don't mess with operation of your circuit.

If there is any way you could switch the original inductance out and
switch a good single one in its place, instead of paralleling one, I
think you stand a better chance of being happy with the performance.
There are also other issues like LC ratio (operating impedance level)
that probably won't be optimum either, and limiting the "detractors" as
much as you can would be a good idea.

Remember that the resonance is 2*pi*f = 1/sqrt(LC). If the capacitance
range stays the same, you need to make the inductance 1/4 as large as
it is now to double the frequency. To triple the frequency, you'll
need to make the inductance 1/9 as large is it is now. That would mean
paralleling the original with 1/3rd its inductance, or 1/8th its
inductance, respectively.

Cheers,
Tom

7. ### terryGuest

Tom Bruhns wrote: in part ..............
..
Thanks to all for all the adviceand the caveats also. As somebody
asked; yes I am thinking of gettiing a low frequency recevier to 'move
up' into the broadcast band without tearing the guts out of it.
Possibly by just tagging some extra inductance in parallel with three
existing and themselves fully shielded range coils (it's a TRF).
At least I could remember some of the basic formulae without even
opening a text book!
reducing inductance to one quarter of a single inductor and then along
with various stray capacitances wondering about interstage coupling via
the existing arrangment and maybe ending with very low Qs and loss of
gain? sounded somewhat finicky.
Thankyou. Terry.

8. ### lemonjuiceGuest

Receivers use variable capacitors to tune LC circuits though some like
the WW II ones used tuning inductors which had powdered iron or an
iron core in the coil to vary the inductance. The tuning sections not
only make things work better but as you can easily tell seperate
signals on different wavelengths.

From the formula you wrote above you can see
that according to your frequency needs

C= 1/(39.5 * F^2*L)

L= 1/(39.5 * F^2*C)

Depending on what you want to change.
Parallel-series inductor has been thoroughly explained by the other
guys.
You want to know whether to add in series or parallel capacitors ?