Common-mode choke values convention?

Thread Starter

seanstevens

Joined Sep 22, 2009
144
Hello,

Hopefully a quick question. When I read the datasheet for (at least one manufacturer) the indicated inductance is actually half the measured inductance across one winding. For example, PLA10AN7720R7D2 is marked as 7.7mH, When I physically measure the inductance of a single winding it's showing as just over 14mH.
The same with PLA10AN3630R3D2, marked as 36.0mH, measuring it shows around 68mH.
Is this the convention for common-mode chokes? I only have the above two types, so I cannot measure others that I read the datasheets for.
I just wanted to check if they all show half of the value of individual winding, it just doesn't make any sense to me.

PLA10AN3630R3D2B

Thanks,
 

peterdeco

Joined Oct 8, 2019
484
Are you sure you're measuring a single winding? According to the data sheet the PLA10AN3630R3D2 is a dual 36mH inductor. Maybe you're measuring the two in series?
 

Thread Starter

seanstevens

Joined Sep 22, 2009
144
Yes. I am measuring one probe on one pin, and the other on any other that reads anything. As you can see from the datasheet, unless you short a couple of pins, it is impossible to measure anything else than just one winding. Also, if I measure for resistance, I can only measure around 1 Ohm between the same pins that I am measuring the inductance, any other is just an open circuit, very odd.
 

crutschow

Joined Mar 14, 2008
30,094
I just wanted to check if they all show half of the value of individual winding,
Yes.
Common-mode chokes normally are connected with the windings in parallel, one for the hot lead and one for the common.
That makes the common-mode inductance 1/2 of a single winding, since two equal inductors in parallel have half the inductance, just like two resistors in parallel.
 

Thread Starter

seanstevens

Joined Sep 22, 2009
144
Yes.
Common-mode chokes normally are connected with the windings in parallel, one for the hot lead and one for the common.
That makes the common-mode inductance 1/2 of a single winding, since two equal inductors in parallel have half the inductance, just like two resistors in parallel.
But they are not physically in parallel to drop to half value, are they? If they were physically in parallel then you would have a short between Live & Neutral. As shown in their datasheet they are physically positioned in parallel but the terminals/pins are not.
But do correct me if I am wrong.
 

crutschow

Joined Mar 14, 2008
30,094
But they are not physically in parallel to drop to half value, are they? If they were physically in parallel then you would have a short between Live & Neutral.
True.
Do you understand what common-mode is?
Common-mode is the same signal appearing on both inputs, thus the common-mode inductance is measured with the two shorted together.
The choke has no significant inductance to normal-mode signals.
 

Thread Starter

seanstevens

Joined Sep 22, 2009
144
True.
Do you understand what common-mode is?
Common-mode is the same signal appearing on both inputs, thus the common-mode inductance is measured with the two shorted together.
The choke has no significant inductance to normal-mode signals.
Your first sentence is 'True' to my comment that Live & Neutral would short.
Yes, I think I know what common mode is.
Please tell me in the attached diagrams where the inductors are physically connected together when you feed live in one coil and neutral in the other so that you can get rid of common noise or EMC noise present on both live and neutral or any other pair that carry the same common noise.
 

Attachments

crutschow

Joined Mar 14, 2008
30,094
Please tell me in the attached diagrams where the inductors are physically connected together when you feed live in one coil and neutral in the other so that you can get rid of common noise or EMC noise present on both live and neutral or any other pair that carry the same common noise.
Of course they are not connected together in the actual circuit.
They are tied together to apply equal signals to both inputs for testing purposes only.
 
Last edited:

crutschow

Joined Mar 14, 2008
30,094
But not if they are wound on the same core. It would be just like winding them bifilar.
Good question.
I believe the answer is that a bifilar winding is basically one winding with a given number of turns, which is not the same as two separate windings on the same core with both having the same number of turns.

seanstevens, to help settle this question, can you measure the inductance with both inputs and outputs tied together?
 
Last edited:

Ian0

Joined Aug 7, 2020
5,486
Good question.
I believe the answer is that a bifilar winding is basically one winding with a given number of turns, which is not the same as two separate windings on the same core with both having the same number of turns.
And just to add to the confusion, the two windings of a common mode choke are deliberately not very well coupled to each other, so it may act as a single winding with two wires at low frequency and two separate windings at high frequency!
 

DickCappels

Joined Aug 21, 2008
8,804
Note: that the two inductors are magnetically coupled and in all cases I have looked into are actually on the same core. For that reason I agree with @crutschow 's statement in his post #4.
1652468203879.pngAt high frequencies the inductance between one side of the choke and the other appear as if the two inductors are in parallel at those frequencies. Consider the capacitors across the input and output of the choke at say, 1 MHz.
 

Thread Starter

seanstevens

Joined Sep 22, 2009
144
seanstevens, to help settle this question, can you measure the inductance with both inputs and outputs tied together?
Good idea curtschow. So I have the results and it is even getting weirder!
I used a 7720R7 choke, the same datasheet indicates its common mode inductance is 7.7mH with the resistance of the winding ~0.7 Ohm. As before, my measurement showed ~14.4mH, so as suggested, I paralleled another choke and measured the result of the two windings in parallel - guess what, it still reads ~14.5mH - no change at all yet the resistance of the two winding in parallel dropped to ~0.3 Ohms. Isn't that odd!

I think we need to get the definitions right here so as to avoid misunderstanding. As far as I know, common mode chokes are two identical windings on the same core with the dot notation indicating the direction of the insput/outpu/signal and the winding.

To work as a common mode, the two signals should be fed into the dots carrying the same noise on both winding inputs to the choke. As the magnetic couplings are in the same directions, the total inductance then looks like double the individual windings to the noise, therefore, presenting a large impedance at the designed frequency.

having said all that - I still don't see what is happening here! I have used them for years and I know they work and remove EMC/noise but never actually measured one, just went with the datasheet.
 

DickCappels

Joined Aug 21, 2008
8,804
@seanstevens , keep in mind that the instrument you use to measure inductance probably only shows you the inductance at one or two frequencies and may not indicate the effective inductance at frequencies of interest for the purpose of RFI/EMI compliance work (tens and hundreds of MHz and these days up).

In my limited experience in electromagnetic interference compliance, one really needs to actually test these filters in product you are trying to get to pass because, like ferrite beads, the written specifications are only indicators of how useful they might be. Because of the many factors involved, many of which are not readily quantifiable you don't know what the effect will be until you do the measurments.
 

Danko

Joined Nov 22, 2017
1,357
I used a 7720R7 choke, the same datasheet indicates its common mode inductance is 7.7mH with the resistance of the winding ~0.7 Ohm. As before, my measurement showed ~14.4mH, so as suggested, I paralleled another choke and measured the result of the two windings in parallel - guess what, it still reads ~14.5mH - no change at all yet the resistance of the two winding in parallel dropped to ~0.3 Ohms. Isn't that odd!
All is correct. When windings are connected in parallel, then it seems like one winding with the same number of turns, but more thicker wire.

1652698863434.png_______ 1652699698430.png
 

Ian0

Joined Aug 7, 2020
5,486
Yes. The reluctance has not changed, so the relationship between magnetomotive force and flux has not changed.
Two inductors in parallel which are not wound on the same core would have half the reluctance, (the two reluctances would be in parallel) so would need twice the magnetomotive force for the same flux, so half the inductance.
 

Thread Starter

seanstevens

Joined Sep 22, 2009
144
@seanstevens , keep in mind that the instrument you use to measure inductance probably only shows you the inductance at one or two frequencies and may not indicate the effective inductance at frequencies of interest for the purpose of RFI/EMI compliance work (tens and hundreds of MHz and these days up).

In my limited experience in electromagnetic interference compliance, one really needs to actually test these filters in product you are trying to get to pass because, like ferrite beads, the written specifications are only indicators of how useful they might be. Because of the many factors involved, many of which are not readily quantifiable you don't know what the effect will be until you do the measurments.
@DickCappels, yes I do realise that, you are right however, I am only using a £2000 LRC bridge to read the inductor value only.
As for using chokes or ferrite both of which I use, you are right too. I always start with the datasheets and then prototypes and tests as theoretical design and actual finished prototype may not follow the same line.

@Danko, what happens to the value of the two 10mHs when in parallel? I mean what happened to the parallel inductor values?

Anyhow, thank you all for your input. The mystery goes on but I shall leave it there.
 

Danko

Joined Nov 22, 2017
1,357
@Danko, what happens to the value of the two 10mHs when in parallel? I mean what happened to the parallel inductor values?
Anyhow, thank you all for your input. The mystery goes on but I shall leave it there.
Two coils with inductance L and K=0 (no common magnetic field) connected in parallel, have inductance=L/2.
Two coils with inductance L and K=1 (common core and magnetic field) connected in parallel,
have inductance=L if connected in the same phase
and have inductance=0 if connected in opposite phase.
------------------------------
Maybe you will interested in this:
Testing Common Mode Chokes
Worked Example of Suitable Tests
https://www.voltech.com/applications/test-examples/common-mode-chokes/
 

Thread Starter

seanstevens

Joined Sep 22, 2009
144
Two coils with inductance L and K=0 (no common magnetic field) connected in parallel, have inductance=L/2.
Two coils with inductance L and K=1 (common core and magnetic field) connected in parallel,
have inductance=L if connected in the same phase
and have inductance=0 if connected in opposite phase.
------------------------------
Maybe you will interested in this:
Testing Common Mode Chokes
Worked Example of Suitable Tests
https://www.voltech.com/applications/test-examples/common-mode-chokes/
Thank you Danko, understood.
 

MisterBill2

Joined Jan 23, 2018
12,396
The simple solution will be to contact the manufacturer and discover the conditions under which the inductance is specified.
With common mode chokes, at the frequencies that the inductance is intended to impede, the two sides are effectively in parallel especially since very often there are capacitors across the lines near that choke. Understanding a data sheet requires knowing the conditions under which the data is specified, which are not always obvious.
This information may even be present on the data sheet some place.
 
Top