Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Offline SMPS filter with diodes across filter inductors?

Status
Not open for further replies.
T

treez

Guest
Hello,
The below schematic shows an offline SMPS with diodes across the filter inductors. The diodes prevent big turn on voltages happening due to LC resonance at switch on (especially if switch on occurs at the main voltage peaks).

Why is this technique not more used?

(LTspice simulation also attached)
 

Attachments

  • PFC _BCM _LT3799.pdf
    22.3 KB · Views: 84
  • PFC _BCM _LT3799.txt
    10.8 KB · Views: 50

The way I've seen in other offline SMPS to avoid high peak turn-on currents, is to use a NTC thermistor; or in higher powered units a wirewound resistor which after a second or so is bypassed with a relay.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
thanks, but a relay or an NTC is lots more expensive than a simple diode....so whats wrong with the diode?
 

There is nothing wrong with the diode, it is just a design preference or cost/performance tradeoffs.

With a thermistor one can control the peak inrush current charging the bulk caps.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
thermistors don't work well enough when hot....also, thermistors tend to only be for higher power smps's. Also, thermistors are more expensive than simple diodes.
The way I show here is the cheapest, but nobody uses it..why?
 

-Design preferences/ bias
-Legacy design rules
-Copy/Paste design methodology
-The company has a 1,000,000 thermistors left over from other project
-This is the way I've always done it....
-etc., etc.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Why not use a simple inrush current limiting circuit instead of inductors, yes inductors can be used for filtering purposes but with an active inrush current limiting circuit you do not need passive limiting solutions. Additionally, you can adjust charging time by simply changing a 0805 package cap.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
sorry but filter inductors are needed to filter out the switching current......you cannot just always do that with resistors alone, its too dissipative
 

NOW I understand...you were concerned about the inductor ringing from a phase-control waveform!
Initially I had thought you were only talking about the power-on phase. Sorry!

Well, like everything else, one can have great ideas on paper and simulations, but actual testing may dictate your next steps. As the attached app note indicates, these components are for EMI purposes. So they may have tested the prototype and found that resistor damping helped them meet the EMI requirements for that circuit.

One thing that I have learned about "Reference Circuits" is that they provide a good platform for the engineer to "play" with a particular IC. But the circuit most likely is not optimized, may not fully apply to your requirements, or use certain components which may be difficult to obtain. As such, substitutions and changes are inevitable, and each engineer is free to add his own set of design guidelines.

So, if you can meet EMI, cost and size requirements with a diode solution, go for it.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
I suggest a different method to design a ruggedized off-mains switcher.

First provide it with surge immunity as required by IEC61000-4-5, e.g. by Varistors or TVS diodes. I assume the circuit won't have problems with input filter ringing after it.

The quoted "active inrush current protection" is intended to reduce periodic peak currents with phase angle control, the diodes won't help against it.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
If the dc bus caps are film type then nobody will care about periodic inrush currents into them...its just the overvoltage that needs guarding against, and the simple diode solution, does exactly that...so why does nobody do it?

ok the IEC stuff, but why not the simple diode to go with that?.....I admit the diode wont stop mains transients
 

Hello, this is relevant to this thread, so if i may i would like to put it here...

Hello,
Do you know what is the saturation current of the 250uH of “stray” inductance in the 47mH common mode choke of part number = B82734RZ132B030 ?

47mH Common mode choke datasheet: (B82734RZ132B030)
https://en.tdk.eu/inf/30/db/ind_2008/b82734r_w.pdf

We need to know because we are investigating the damaging effects of the LC ringing which might happen when the power supply is connected (accidentally) to a leading edge dimmer which is turning on at the mains peak in every mains half cycle.
If the stray inductance (actually the leakage inductance) has a low saturation current, then we neednt worry about it ringing with input capacitors, as it will have saturated and will not have its inductance).
Our power supply has a boost PFC, so there is no big electrolytic after the bridge rectifier with which to quench such unfortunate ringing.
 

to answer the original poster, this method is widely used but more often with bi-directional zeners back to back, or a bi-directional TVS, just using a diode limits the swing one way and hence limits the filtering action....
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top