High current LISN

[T3sl4co1l]

Currently doing a higher current project so figured it would be nice to have some kind of LISN on hand to monitor it:



Circuit is standard,



with a few variances for parts on hand.  Inductance is a little more than the more standard 5uH, but it comes down to that at rated current, so I figure a little extra won't hurt.  Anyway, not going for any particular standard here, just want some bandwidth.  Getting to 100kHz or below is more than fine here.

I labeled it for "100A"... wire is 8AWG, not really suitable for that, but it'll handle for a few minutes before the wire gets concerningly hot.  That's really just what the inductance was designed for... great excuse, I know.   I'm not using it quite that high for immediate purposes so it's fine anyway.

(Cores are MS-300060-2 if you were wondering.  Picked up a small pile of them a while ago for a different project, since been using them from time to time.)

Tim

[jonpaul]

Tesla, very fine work, for DC test not power mains 120/240V?

Micrometals are a favorite,

A fine HP or tektronix SA whose input front end mixer killed by LISN overload will ruin your day....
I would add Spectrum Analyzer protection in any LISN, as one can conceive overload of a 50 ohm mixer protection eg crowbar of DC feed, connection or disconnection, etc.

The energy coupled via the DC block cap, from a fault transients should be measured or simulated. Check out your SA input overload specs.

My practice is to add 20..30db wideband attenuators before the SA, remove only after connections and measurement is stable Use  wideband low  capacity transients protection.



Bon chance

Jon

[mag_therm]

Neat job.

On my DC-DC converter test recently, home brew SA interface was a
divider:  {2.2nF 1kV disk cer > 330 R series >76 R shunt}
 making about -12dB, 400 Ohm in  60 Ohm out

That into SMA with a -2 dB screwed in.

I thought that 2.2 nF error was not too high, being -j36 Ohm at 2 MHz
The LISN inductor (5 uH) is j63 Ohm @ 2 MHz  > j630 Ohm @ 20 MHz

Here is an easy T calculator:
https://www.allaboutcircuits.com/tools/t-pad-attenuator-calculator/

[T3sl4co1l]

Indeed, sudden connection/breaking of that many amperes could make quite a transient into 50 ohms.

I'll probably add at least a 3dB pad and clamp diode.  Thinking one of those quad clamp diodes plus 5V TVS in SOT-23-6, as a flying zener (i.e. pins 1 and 3 to signal, pins 4 and 6 to GND, 2 and 5 (the VCC/GND pins) floating).

Caps are all rated 400V or more, so mains is technically possible, but I don't intend to use it beyond LV.

Tim

[coppercone2]

I still need to straiten out the warped sheet metal box for my 12AWG air core LISN. The distortion is so bad I can almost bounce on the chassis top like a trampoline (welded out of angle irons and 10 gauge plate) if I was a little fatter.. and I also need to make the coupler. Tried to make a box ended up with a armored beetle

Also the problem was when you put the chassis together the 50uH became a 40uH so I need to add some windings to the 250uH inductors and remake the 50uH inductors, maybe add a section or change the distance between the coils.

next up is the 1000A bus bar LISN

and I like that the OP LISN has torroidal inductors. Its usually made with open coils

[T3sl4co1l]

I still need to straiten out the warped sheet metal box for my 12AWG air core LISN. The distortion is so bad I can almost bounce on the chassis top like a trampoline (welded out of angle irons and 10 gauge plate) if I was a little fatter.. and I also need to make the coupler. Tried to make a box ended up with a armored beetle

Ah yeah, gotta be careful with those welds... didn't tack up, or too many/too long seams at a time?

For my part, I stick to (thinner) sheet metal, and bolted members... but mainly because I don't have a workshop to speak of.

next up is the 1000A bus bar LISN

and I like that the OP LISN has torroidal inductors. Its usually made with open coils

Heh... I wonder what you'd even need 1kA for.  Do telecom "rectifiers" go that high?  Maybe...

Some power line facing equipment would (while also at much higher voltages(!!)), but I assume that's all by industry standards, or custom (as specified in contract).  Or other specialty industrial equipment, but they're largely not going to care about emissions, or would evaluate the whole system (say, an aluminum smelting rectifier, as a unit, or as the whole facility).

Also used toroids on another one,



The various values ballast a particular isolation transformer (240VCT output, so I can even use it like a NA outlet if need be), which I measured having leakage inductance as the balance of the big inductors.  Hence the single core for the longest winding, three cores for the other, and five for neutral (no inductance: normally grounded).  The row of three single cores are the 50uH load-facing inductors.  These were calculated to do, I think it was 15A RMS, 30A peak, at -30% inductance or something like that.  Good enough for light loads and active PFC, just avoid using very peaky or high power loads.  No biggie.

Tim

[uer166]

Picked up a small pile of them a while ago for a different project, since been using them from time to time.)

May I ask where? I'm not too interested in powdered iron, but would like to expand supplier options regardless.

[T3sl4co1l]

Lodestone Pacific.

Used to be a few others, want to say Elna used to have online ordering, or at least inventory? But I don't see that anymore? I forget. I guess you can still order the old fashioned way.  Lodestone is also very old fashioned as far as ordering.  Dexter https://dxtmagnetics.com/ has online ordering, and there's always Amidon, and the usual suppliers (Digi-Key, Mouser, etc.) who are stocking surprisingly a lot of ferrites and accessories these days, but not so much powder (which maybe was to your point ).

Can also take a gamble on Ali, which... probably isn't too bad, there's no shortage of materials made over there of course, and you can at least test these kind of parts without *too* much difficulty and see if they are to spec.  I don't really have any recommendations in that direction; I did once get some bobbins from Xuyi Ouge Electronic Co., Ltd, which came in just fine.

Tim

[TimNJ]

Old school website, but always reliable: https://www.cwsbytemark.com/index.php?main_page=index&cPath=206

There is also MHW, but their website is a total disaster (broken?): https://mhw-intl.com/products/magnetics/csc-powder-cores/

I cannot figure out, for the life of me, how to search their inventory or order anything (anymore)

[jonpaul]

Core suppliers

Powdered iron

Micrometals

Ferrite
Fair-Rite
TDK
Nippon Céramique

Small diy qty from epay, ham fleas, but Micrometals will sell to individuals.

Some LISN use aircore solenoids for no saturation.
We use the commercialy available Solar LISNs.

Bon chance

Jon

[coppercone2]

are you going to do the thing in the input where there is a circle of capacitors around the input to the chassis like a feed through/iris that I seen in some lisns even though the VNA seems to say that arrangement does nothing?

[T3sl4co1l]

Nah, the 22nF is good enough bypass. I'm not going to put an enclosure on it, it doesn't need extreme isolation.

Tim

[uski]

Hella cool! Where did you find those MS-300060-2 cores? I could not find anywhere (am in the US). Do you have a secret source? 

[edit: attachments removed]

[T3sl4co1l]

Hella cool! Where did you find those MS-300060-2 cores? I could not find anywhere (am in the US). Do you have a secret source? 

Read better?

May I ask where? I'm not too interested in powdered iron, but would like to expand supplier options regardless.

Lodestone Pacific.

BTW those images are self-hosted intentionally, I'd appreciate not re-hosting them, thanks.

Tim

[uski]

Lodestone Pacific.

Thanks!

BTW those images are self-hosted intentionally, I'd appreciate not re-hosting them, thanks.

Aww. Sorry - I was trying to help others in the future. I 100% respect your decision, so I have removed the attachments.

At least once a week, I see an interesting post on this forum posted a few years back, where the person decided to host the images elsewhere, and then closed their website for some reason, or used an image provider that stopped working, and then the post becomes useless so that's why I thought it was a good idea to host a copy here, for posterity - because that LISN is a beauty!