Author Topic: Understanding three phase line EMI filters  (Read 3285 times)

0 Members and 1 Guest are viewing this topic.

Offline XaviPachecoTopic starter

  • Regular Contributor
  • *
  • Posts: 243
  • Country: do
Understanding three phase line EMI filters
« on: June 14, 2019, 04:07:36 am »
Hi,

I'm reviewing the schematic attached (EMI1) which I found inside an application note on a power supply design. I want to understand how to select the proper filter design for the three phase line. In the schematic, they have used a resistor and inductor in parallel with a set of capacitors in star connection. What kind of filter is this? Why do they put a resistor in series with the varistor? In other designs, I have seen just an inductor in series with each line like shown in the attachment Power1 where they are using individuals 680uH inductors at each line. How do I decide the proper filter?
 

Offline MosherIV

  • Super Contributor
  • ***
  • Posts: 1530
  • Country: gb
Re: Understanding three phase line EMI filters
« Reply #1 on: June 14, 2019, 06:53:01 am »
Hi. I think the resistors across the inductors are to prevent a spark when you unplug the equipment, gives a path for back emf from the filter inductors. They should have no (little) effect when in circuit, too hugh resistance.
I think the 10 \$\Omega\$ in line with the MOVs is to stop them blowing up when there is a massive pulse. The extra resistance should reduce the energy enough the the MOV will not physically split apart.

That is my guess  ;)
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Understanding three phase line EMI filters
« Reply #2 on: June 14, 2019, 09:15:48 am »
Resistance in series with a MOV defeats its purpose.  I would hope those aren't actually 10 ohm resistors, but placeholders for something like a thermal fuse, which disconnects the MOV if it fails (overheats due to excessive leakage current).

The 10ks are probably just for tweaking the response.  A CMC has a parallel resonant peak, below which it is inductive, and above which it is capacitive.  Resistance in parallel reduces this peak, but also broadens it, and maybe that's useful in combination with whatever capacitances or other parasitics are in that particular circuit.

Large value resistors don't do anything for ESD or surge.

Note that the inductors are coupled, so that mains current is balanced between lines and the core can be a high-mu material.  Individual inductors don't have this advantage, and must be rated for saturation > full peak load current, otherwise they stop being effective and some noise gets let through.

The neutral rectification is bizarre, or at least very unusual.  I can think of reasons to do it, but it shouldn't be done for most three phase equipment.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Understanding three phase line EMI filters
« Reply #3 on: June 14, 2019, 11:54:13 am »
Where the heck did you find that first circuit, OP? It makes no sense whatsoever. The second one appears to be for very low power, so likely for control circuitry or the like.

Regardless, it is common to use a 3-winding common mode choke on 3-phase power followed by 3 capacitors tied to either a virtual, or actual, neutral point. The insertion of additional inductance in series with the neutral makes no sense unless, perhaps, there is a serious problem with 3rd harmonic currents.

EDIT - and paralleling two bridge rectifiers with one supplied by line-to-line and the other supplied by line-to-neutral voltage also makes no sense.
 

Offline SparkyFX

  • Frequent Contributor
  • **
  • Posts: 676
  • Country: de
Re: Understanding three phase line EMI filters
« Reply #4 on: June 14, 2019, 01:12:58 pm »
The rectifiers do not make much sense, one will have sqrt(3)*Phase Voltage (for 120° 3~ sine wave), the other will have phase voltage. after redrawing it, it looks like the typical 3 phase rectifier, but with a non obvious reason for the connection of the neutral on the second half bridge.

Not all inductors must be EMI filters, they could also serve as inrush current limiters as component protection.
« Last Edit: June 14, 2019, 01:34:50 pm by SparkyFX »
Support your local planet.
 

Offline XaviPachecoTopic starter

  • Regular Contributor
  • *
  • Posts: 243
  • Country: do
Re: Understanding three phase line EMI filters
« Reply #5 on: June 14, 2019, 01:20:51 pm »
This is where I took the first circuit from: 12 W three-phase power supply for smart meters and home appliances

I haven't studied it in depth. Maybe I misunderstood something.
 

Offline max_torque

  • Super Contributor
  • ***
  • Posts: 1327
  • Country: gb
    • bitdynamics
Re: Understanding three phase line EMI filters
« Reply #6 on: June 14, 2019, 01:42:03 pm »
Take a look at the datasheet for commercial 3 phase filters, like those produced by Schaffner:





https://www.mouser.co.uk/ProductDetail/Schaffner/FN3310-250-99-R5



 

Offline duak

  • Super Contributor
  • ***
  • Posts: 1048
  • Country: ca
Re: Understanding three phase line EMI filters
« Reply #7 on: June 14, 2019, 06:14:52 pm »
I think the fourth rectifier diode pair is connected to neutral to allow for any combination of input connections to get power.  eg., for the degenerate case of one line and neutral.  Makes sense if this for a smart power meter - The PoCo will want to make sure you are charged even if two phases are out.


 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Understanding three phase line EMI filters
« Reply #8 on: June 14, 2019, 07:37:26 pm »
Yeah, you wouldn't want to do that for a substantial load, but for a control, to get more reliable power, that would be okay.  Makes sense.

The problem is neutral currents, which ideally shouldn't be drawn.  It's usually a smaller conductor, to handle leftover unbalanced current, not full load.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline jbb

  • Super Contributor
  • ***
  • Posts: 1265
  • Country: nz
Re: Understanding three phase line EMI filters
« Reply #9 on: June 14, 2019, 09:12:53 pm »
The 3 phase + neutral connection also gives somewhat higher output voltage, which could be useful in some applications.
The downside is that it generates neutral currents and 3rd, 9th etc. harmonic (aka ‘triplen’ harmonics) currents.
A standard 3 phase diode rectifier generates no 3rd, 9th etc harmonics.
 

Offline langwadt

  • Super Contributor
  • ***
  • Posts: 4857
  • Country: dk
Re: Understanding three phase line EMI filters
« Reply #10 on: June 14, 2019, 09:31:42 pm »
Yeah, you wouldn't want to do that for a substantial load, but for a control, to get more reliable power, that would be okay.  Makes sense.

The problem is neutral currents, which ideally shouldn't be drawn.  It's usually a smaller conductor, to handle leftover unbalanced current, not full load.

Tim

? like everyone else around here I have 3 three-phase power, when I use a single phase load like a kettle only one phase and neutral is used so the phases and neutral must be roughly the same size
 

Offline duak

  • Super Contributor
  • ***
  • Posts: 1048
  • Country: ca
Re: Understanding three phase line EMI filters
« Reply #11 on: June 14, 2019, 10:42:21 pm »
As long as there are 2 or 3 phases present, there won't be any neutral current.  I see the rectifier diodes switching to deliver the highest peak dfferential voltage at any instant.  If the load was a bipolar supply with the common center tied to neutral then there could be.  But here, the load shouldn't be referenced to neutral.  I looked at the app note, and it looks like the author used the ground symbol for circuit common but it doesn't mean it should be connected to neutral - that could be exciting!

I've worked with a few 3 phase systems, but they've always had a wye transformer source with the center connection to neutral.  There are apparently other PoCo connections, but I haven't seen them. eg., corner grounded delta (2nd image) and high-leg delta (3rd image).  I think the circuit above will work fine with these configurations.

The only configuration that would cause a problem would be an ungrounded source transformer.  Conceivably, there could be higher phase to neutral voltages than phase to phase voltage.  It would take a particular failure or wiring mistake but it's possible.

Here's a triplen current story:  I traced some down in a delta-wye step-down transformer operated as a step-up.   If memory serves, the line fuses were 150 A and would occasionally open under load.  There was some 50-60 A in the transformer's neutral, even when it was unloaded.  It was pretty damn hot too.  I disconnected the neutral and all became as it should be.  That poor transformer was probably conducting a good bunch of the triplen currents generated in the building.
« Last Edit: June 15, 2019, 04:33:46 pm by duak »
 

Online coppercone2

  • Super Contributor
  • ***
  • Posts: 11335
  • Country: us
  • $
Re: Understanding three phase line EMI filters
« Reply #12 on: June 15, 2019, 05:06:07 am »
for understanding the mysterious inductor resistor parallel combo

http://ecee.colorado.edu/~rwe/papers/APEC99.pdf

The first page explains quite a bit
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf