Given the 3-Pin inductor (small auto-transformer) seen in the image below:

How is it that the inductance between pins 1-2 is 260mH, between 2-3 is 5mH, and yet between 1-3 is 335mH?
One would think that the inductance of pins 1-3, should be 265mH, as both coils are in series. Notice that the windings resistance of 1-3 is in fact 810 (10 +800), as if the entire winding is one big series.

An explanation would be extremely helpful.
-Thank you

 

I agree. There appears to be an error in the data sheet.
Just curious but what is the part number for the coil?

 

Yes, it looks a bit mysterious.

However, it may be something with the square of the number of turns, when it is on the same core.

 

It's because of the frequency reference.

 

Given the 3-Pin inductor (small auto-transformer) seen in the image below:

How is it that the inductance between pins 1-2 is 260mH, between 2-3 is 5mH, and yet between 1-3 is 335mH?
One would think that the inductance of pins 1-3, should be 265mH, as both coils are in series. Notice that the windings resistance of 1-3 is in fact 810 (10 +800), as if the entire winding is one big series.

An explanation would be extremely helpful.
-Thank you

AFAICR: Mutual inductance is the key, if you wire 2x 47uH RF chokes in series; you get 94uH - if you put the same total number of turns on the same bobbin; you get more inductance than the 2 individual inductors.

 

I agree. There appears to be an error in the data sheet.
Just curious but what is the part number for the coil?

This is a custom component spec which was sent to me, after I sent that company a similar component I found and asked them to determine its values. For what its worth the part number is LCHB0912-3W-344M.

AFAICR: Mutual inductance is the key, if you wire 2x 47uH RF chokes in series; you get 94uH - if you put the same total number of turns on the same bobbin; you get more inductance than the 2 individual inductors.

It is unclear to me how mutual inductance could cause an increase in the total inductance of a coil. Can you possibly provide a reference describing this relationship?

 

Since the two inductors are coupled (i.e., on the same form),

Ltotal = (√L1 + √L2)^2.

√L1 = √260 = 16.12
√L2 = √5 = 2.24
16.12 + 2.24 = 18.36
18.36^2 = 337 ≈ 335

 

Since the two inductors are coupled (i.e., on the same form),

Ltotal = (√L1 + √L2)^2.

√L1 = √260 = 16.12
√L2 = √5 = 2.24
16.12 + 2.24 = 18.36
18.36^2 = 337 ≈ 335

Nicely done - mathematics isn't my familiar territory.

 

Since the two inductors are coupled (i.e., on the same form),

Ltotal = (√L1 + √L2)^2.

√L1 = √260 = 16.12
√L2 = √5 = 2.24
16.12 + 2.24 = 18.36
18.36^2 = 337 ≈ 335

Thanks, we learned something today.

 

Thanks, we learned something today.

So did I.

I had to figure it out on the fly, from the fact that inductance is proportional to the square of the number of turns. A certain amount of head-scratching was involved...

 

Nicely done - mathematics isn't my familiar territory.

You really need to do some math or designing circuits turns out like a monkey with a dart board.

 

You really need to do some math or designing circuits turns out like a monkey with a dart board.

Very true.

But looking back on 40+ years of designing, mostly analog, it kinda surprises me how little advanced math I actually used. I'd guesstimate that about 5% of the problems I had to work involved very basic calculus, 5% used geometry and/or trig, and the remaining 90% involved nothing more than first-year algebra.

But ALL of the work involved math of some flavor or another. Designing without math would be pure guesswork.

 

it kinda surprises me how little advanced math I actually used.

I feel a bit of shame to be repeating myself but, I've faked my way through life on nothing more than good algebra and a wee bit o' trig.
The Calculus light bulb never came on for me, but the algebra light bulb did! That's enough for a technician and almost enough for a designer. Some of the guys on here can embarrass me quite easily but I make up for it by knowing my limitations.

 

I feel a bit of shame to be repeating myself but, I've faked my way through life on nothing more than good algebra and a wee bit o' trig.

In my case, it's the above plus a bit of cunning and guile.

 

Very true.

But looking back on 40+ years of designing, mostly analog, it kinda surprises me how little advanced math I actually used. I'd guesstimate that about 5% of the problems I had to work involved very basic calculus, 5% used geometry and/or trig, and the remaining 90% involved nothing more than first-year algebra.

But ALL of the work involved math of some flavor or another. Designing without math would be pure guesswork.

Some people overdo the maths and neglect the science of controlling the electron.

I avoid maths whenever I can get away with doing so - that means more cobwebs to clear when I can't, but I usually get the job done.

The Wireless World magazine was pretty much a showcase of how far the author could fill a page with incomprehensible equations - but they did occasionally publish something useful.