Staggered 50uH and 250uH inductor design for LISN

[floobydust]

Why guess about LISN construction? The are precisely defined and full details are given in the CISPR documents about their construction. Same goes for the MIL and IEEE standards.
...

The discussion led to Jay Diddy B proving ferrite cores are suitable to 30MHz for high-current inductors, resulting in his LISN with physically smaller inductors than the CISPR standard-build with its massive air-core inductors.

You don't need EMC lab certification-grade accuracy. A dB or two error is perfectly usable for pre-scans or design work, and even that could be corrected for with a little extra work.
It's strange some commercial LISN's have corrected flat frequency response, and others are bumpy, just copying verbatim the CISPR schematic.

[CopperCone]

i think this is an interesting discussion in general and I don't see why someone is trying to shut us down

making the LISN better in terms of linearity is always an option too. No reason to bow to a standard or compliance laboratory

I am a fan of the mini LISN because the one I made is gigantic.

[dazz1]

Hi
One of the benefits of standards is that they enable consistent results wherever you might be.
One of the side effects of standards is that they stifle innovation. 

The standards should be written to account for the deficiencies of the standard design. If you make a LISN that is better than the standard, it may no longer comply with the standard.  If you are testing for compliance, the better but non-compliant LISN isn't much use.

Innovation and standards are both great.  Both have limitations and side effects.

Dazz

[slloyd]

according to CISPR16-1 figure27 the L1 inductor is wound in one go but has 9 taps off it for some resistors to cancel some high frequency noise.  when i recreate this coil in the air coil calculator i get 62.5uH.  i'm assuming inside the metal box it must be ~50uH. 

https://ibb.co/mJFOR7

  if you were building this inductor knowing your resistors were SMD you'd have to build this inductor in 9 separate coils similar to how the tekbox 5uH does it.  the part that confuses me if i follow the number of turns in the picture and change the calculator to look at just 4 turns, but change the length to get the same pitch, i end up with 3uH for that 4 turn inductor.  if i do the same thing for each separate coil, then sum them up i am no where near 62.5uH. 

https://ibb.co/jgYcm7

ps. sorry,  i'll have to figure out how to link pictures properly later.  gotta run!

[CopperCone]

Keep in mind the spacing between 5he coil section messes with inductance due to mutual inductance. I think i got within 5% measured on the calculator program you used with my finished inductor.

I forgot about the god damn box messing with stuff. I might have made a 40uH LISN if the box reduces the inductance as much as you say

[floobydust]

https://ibb.co/mJFOR7

Your wire diameter is huge at 6mm, check and try #12AWG, 2mm or something smaller

[CopperCone]

yea 6mm is kinda big thats like a truck suspension

[slloyd]

i was just verifying the math using the 6mm with 8mm pitch that CISPR 16 uses in their example.  in my own project i'll be using #12 or #14AWG.

btw, if you want to read CISPR16 to see where all these circuits originated from, its here:  https://law.resource.org/pub/in/bis/manifest.litd.9.html

in the reference they show in fig23 that the circuit is good to 100A so perhaps that's why they used 6mm wire

[slloyd]

in CISPR16, figure 23, is this R5 a type-O?  is it supposed to be R3?

[T3sl4co1l]

Heh, looks like it.

Tim

[floobydust]

Yes, typo.

Careful following those CISPR schematics verbatim,  missing are safety-discharge resistors for C1, C2.

[T3sl4co1l]

Alternately, run it from an isolation transformer, which discharges the caps ~instantly.  Which is a good idea because of the large capacitance to ground, too!

Tim

[slloyd]

if i can find an isolation transformer that is 1.5kW rated and doesn't cost a fortune.  otherwise, i'll just power it off a wall socket that is not GFI protected and use proper safety measures

[slloyd]

shown as X2 capacitors, probably should be Y2 ?

[slloyd]

Careful following those CISPR schematics verbatim,  missing are safety-discharge resistors for C1, C2.

Yes, i see that. 

[floobydust]

shown as X2 capacitors, probably should be Y2 ?

That's the as-built Compower 115A LISN,  X2 was actually used if you look at the pictures (for the 1uF parts) and not sure about the big motor starting caps (?) and series resistor module.
https://www.eevblog.com/forum/rf-microwave/50uh-and-250uh-inductor-design-for-lisn/msg1454320/#msg1454320

[dazz1]

Hi
Line to Neutral should be Cx
Neutral to Ground should be Cy
https://www.allaboutcircuits.com/technical-articles/safety-capacitor-class-x-and-class-y-capacitors/

[slloyd]

That's the as-built Compower 115A LISN,  X2 was actually used

agreed, that's what they used.  not sure why though, seems to me Y2 would be the appropriate choice.  perhaps this is why 115A has been discontinued? 

any idea of the part number on those big yellow "motor start" capacitors?

[T3sl4co1l]

X/Y ratings aren't very applicable here, partly because you can't get them in values large enough.

Instead, use conventional parts, rated for the AC voltage, and protect against surges with MOVs.

Another good reason for an isolation transformer, the added inductance (and potential for core saturation) helps attenuate surges, reducing peak current into MOVs and such.

Tim

[slloyd]

the unmeasured line should be 50ohm terminated

[slloyd]

looks like some newer X2 on the market with high enough uF values now available.  check out B32928A4825K by EPCOS. 

quantity 4, (i.e. not in parallel?) could be used here:

[slloyd]

based on CISPR and the info provided from the good members of eevblog.. i put together this schematic for a bench top LISN rated to 25A.  the top half of the schematic will be free wired and the bottom half will be on a PCB.  i still need to assign part numbers.  schematic is posted as a DRAFT for comment and is not meant for construction.

please feel free to comment 

next i'll be working on the inductor build

[slloyd]

speaking of building inductor.  CISPR 16 has this to say about L2, 250uH inductor:

The inductance L2 should have a Q-factor not less than 10 over the 9 KHz to 150 kHz frequency range. In practice, it is advantageous to use inductors coupled in series opposition in the live and neutral lines (common-core choke).

what exactly do you think they meant by that?  when i read this it sounds like they wanted L2 to be wrapped on a gigantic torroiod.  both L2 and L2*, in common mode configuration.  but generally that's not what i see in the pictures posted of commercial LISNs.  looks like two separate air core inductors each wound on a bobbin of any random size.

[CopperCone]

There was one LISN in this thread posted that has a CM EI core transformer for the 250uH bit.

I assume you get some kind of reasonable cm effect through the chassis.

T3slacoil I think the problem with MOV is that they are nonlinear. I think the error would be small though.
And the same goes for using a cored transformer, it can saturate and distort on inrush and stuff.

Most of my capacitors are X rated, a few are just film  caps from a plasma television, they are 'protected' by the 5 ohm resistor at least.

I think mine will just ride dirty with a input GDT lol.

I never saw an air coil torroid before, its hard to imagine for some reason since its just air not a flux focuser.

If you want  to try:
http://coil32.net/toroid-air-core-coil.html

Too late for me.

But,. that link is not really applicable, since for a decent CM response you want two half moon windings...

[T3sl4co1l]

T3slacoil I think the problem with MOV is that they are nonlinear. I think the error would be small though.
And the same goes for using a cored transformer, it can saturate and distort on inrush and stuff.

You don't put the MOV on the EUT side.

Measurements are not taken during inrush.

Tim