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.

Flyback power supply

Status
Not open for further replies.

pavandanyasi

Newbie level 6
Joined
Dec 9, 2017
Messages
14
Helped
1
Reputation
2
Reaction score
1
Trophy points
3
Activity points
127
Hai ,
I designed flyback power supply 65v/3amps using Topswitch262EN . I got the output 65v stable . when i put the load 1amp it is ok but when i increase the load 2amps the output is going off that is thermal shutdown. i got the drain to source wave form .i am not understanding the wave form is it ok or not please tell me that drain to source wave form are ok or not and also tell me ic is not supporting 3amps . i will attach the waveforms and implemented board.
 

Attachments

  • IMG-20171220-WA0002.jpg
    IMG-20171220-WA0002.jpg
    105.8 KB · Views: 69
  • IMG-20171220-WA0003.jpg
    IMG-20171220-WA0003.jpg
    113.8 KB · Views: 73
  • IMG-20171220-WA0004.jpg
    IMG-20171220-WA0004.jpg
    122.3 KB · Views: 69
  • 20171211_164731.jpg
    20171211_164731.jpg
    523.3 KB · Views: 83

Hi,

The waveform may be OK. It is the typical ringing when the current in the inductor becomes zero and thus the diode releases.

Did you read through other threads with SMPS problems?
Then you will already know the most likely errors:
* bad PCP layout --> stricktly follow the datasheet recommendations. Often there are reference designs.
* wrong part selection. Mainly the inductor. --> keep on the datasheet informations.

Added: I've just realized that you build the circuit using a stripboard.
--> With a stripboard it is impossible to comply with the PCB layout recommendations of the datasheet.
--> create a good PCB layout


Klaus
Thanks for your replay .tell me how to reduce the ringing .and what are the causes for getting the ringing. and one more thing is output voltage is not stable for 2amps load .and up to 1amps output is stable. for 2amps load topswitch ic is getting hot and output voltage goes to drop.
 

(Sorry for the confused post order)

Hi,

The waveform may be OK. It is the typical ringing when the current in the inductor becomes zero and thus the diode releases.

Did you read through other threads with SMPS problems?
Then you will already know the most likely errors:
* bad PCP layout --> stricktly follow the datasheet recommendations. Often there are reference designs.
* wrong part selection. Mainly the inductor. --> keep on the datasheet informations.

Added: I've just realized that you build the circuit using a stripboard.
--> With a stripboard it is impossible to comply with the PCB layout recommendations of the datasheet.
--> create a good PCB layout


Klaus

- - - Updated - - -

Hi,

The ringing is no problem. --> No need to reduce it.

The problem most likely is your PCB layout. It is not suitable for power switching circuits.
--> Create a suitable PCB layout and you are done.

It makes no sense to discuss about your stripboard design. It's a waste of time.
(For sure there may be working flyback circuits build on a stripboard ... but it is not reliable...it's more out of luck if they work)

Klaus
 

(Sorry for the confused post order)

Hi,

The waveform may be OK. It is the typical ringing when the current in the inductor becomes zero and thus the diode releases.

Did you read through other threads with SMPS problems?
Then you will already know the most likely errors:
* bad PCP layout --> stricktly follow the datasheet recommendations. Often there are reference designs.
* wrong part selection. Mainly the inductor. --> keep on the datasheet informations.

Added: I've just realized that you build the circuit using a stripboard.
--> With a stripboard it is impossible to comply with the PCB layout recommendations of the datasheet.
--> create a good PCB layout


Klaus

- - - Updated - - -

Hi,

The ringing is no problem. --> No need to reduce it.

The problem most likely is your PCB layout. It is not suitable for power switching circuits.
--> Create a suitable PCB layout and you are done.

It makes no sense to discuss about your stripboard design. It's a waste of time.
(For sure there may be working flyback circuits build on a stripboard ... but it is not reliable...it's more out of luck if they work)

Klaus
ok thank you.. i will do in PCB.
 

here is how to layout an smps on a pcb

- - - Updated - - -

basically, dont run high di/dt power switching currents through lengths of control gnd, and keep power switching current loops in as narrow area loops as possible.
 

Attachments

  • Basics of SMPS Layout _4.doc
    645.5 KB · Views: 48

here is how to layout an smps on a pcb

- - - Updated - - -

basically, dont run high di/dt power switching currents through lengths of control gnd, and keep power switching current loops in as narrow area loops as possible.


Thanku for your replay
mosfet drain to source am getting ringing . i used snubber circuit and clamp diode
also but not reducing ringing. leakage inducatance is 10mh. primary inductance is 139uh. please tell me how to reduce ringing . i will attaced the waveform pls find.
 

Attachments

  • IMG-20171220-WA0002.jpg
    IMG-20171220-WA0002.jpg
    105.8 KB · Views: 70
  • IMG-20171220-WA0004.jpg
    IMG-20171220-WA0004.jpg
    122.3 KB · Views: 74

Hi,

I wonder why you still are worried about the ringing.
Does the ringing cause a problem?

Klaus
 

Hi,

I wonder why you still are worried about the ringing.
Does the ringing cause a problem?

Klaus

Yes my sir has told me to reduce the ringing and he compared another flyback converter in that there is no ringing coming . and one more thing is i connected a resistor 250ohms in series with transformer primary pin to line dc positive line and reduced the ringing but resistor is getting heat and when i connected load output voltage drops . am not understanding what is the problem .
 

Attachments

  • 20171223_132201.jpg
    20171223_132201.jpg
    511 KB · Views: 74

..you show typical wave of drain of fet of flyback which is in dcm.
As Klaus hints you should not bother about it.

Do you mean 10uH of leakage.?
Did you interleave wind?
Did you follow PI EXPert guidance on powerint.com?

If you give me your LP,LS,VIN,VOUT,POUT,CORE etc, then i will try and do it for you.

or just give me vin,vout,pout and pwm controller wanted....and F(sw) if poss

How is your primary clamp?

- - - Updated - - -

Is your "sir" taking the micky?
If you do flyback CCM then yes you dont see all that ringing.

- - - Updated - - -

“ripping the micky” is (correct me if I am wrong) an English thing…typical in Brit engineering companies….it used to be good natured fun, but now its more of a way of senior Brit engineers to get rid of young brit noob engineers who are now seen as job competitors in todays Britain, where the industry is dying off rapidly.
(For example, at one Brit place the Brit seniors secretly bent a PCB till the circuit stopped working, and then it only actually worked when it was bent in a certain way (I guess so that the snapped connections reconnected)….they showed this to a noob Brit engineer and told him that he should bend PCBs to get them working, --and demo’d it with their “doctored” PCB…later the Noob got a heck of a rollocking when the gaffer caught him bending other PCBs whilst trying to get them working.}

- - - Updated - - -

The frequency of the main ringing you show will be about freq of L(pri) and the capacitance at the fet drain.
 

The amount of energy and ringing at turn off is closely related to the leakage inductance or lack of coupling of primary to secondary - many design engineers have spent many hours re-designing their transformers to increase the coupling to lower the leakage to reduce ringing and the heat in snubbers.

The ringing after the conduction period of the output diode is purely due to the magnetising inductance of the transformers two windings and the total capacitance seen by each of these windings ( e.g. the series capacitance of the main mosfet when off, the series capacitance of the output diode when off) the energy in this ring is low-ish - normally it is ignored by design engineers as it does no harm and only occurs at light loads - so as long as it does not cause EMC issues you can live with it.

- - - Updated - - -

you are adding a fair bit of leakage inductance to both sides simply with the long wires on each side - shorten these and reduce the loop area of these power circuits if you can, also put the pri side snubber close to the FET, the FET close to the Tx and a decoupling cap 630V 100nF close to the TX (from +HVDC to gnd). Similarly improve the output wiring - good luck ...
 

To clarify, your waveforms are "ringing" twice per period. First is during turn-off, when Vds rises. This ringing is due to leakage inductance, and should be avoided by minimizing leakage, or using snubbers. Second time is when the secondary current reaches zero and Vds drops back towards Vs. This ringing always happens when operating in discontinuous mode (DCM), and there's usually no reason to worry about it.
flyback_ringing.jpg
 

yes, and just to add a fourth dimension, some flybacks actually use a slow-ish RCD clamp diode so that the energy dissipated in the RCD clamp is less....this means that you actually intentionally allow the "leakage ring" to resonate more than usual.....it does this more easily because the slow diode doesnt 'snap it off'.......of course, when the "leakage ring" current is coming 'backwards' through the slow clamp diode, then it is actually shoving energy into the output, which is advantageous.

This post talks about this.....
https://www.edaboard.com/showthread.php?t=363493

- - - Updated - - -

By the way, using a slow diode in the RCD snubber is not really recomended....even though i worked at a place where they were doing it and were selling a 70W offline Flyback SMPS throughout europe with huge volumes....having said that, this product was a cosmetic product, and wasnt going to be used on full power
all the time. The basic schem is as follows......it was a Chinese design, and i noted in their SMPS test report that the one component that they provided no thermal data on was the 51R RCD clamp resistor, and that suggests to me that maybe it was running too hot........i certainly burned my finger touching it.
 

Attachments

  • Schematic_Flyback slow diode clamp.pdf
    32.5 KB · Views: 88

@ treez: Most modern designs (designers) use resonant flyback to reduce losses in the fet, snubber diode and output diode and to reduce RFI - once you start building them it is hard to go back to a standard flyback even for very low power...

- - - Updated - - -

@ pavandanyasi a good sized snubber on the sec side diode will help reduce overshoot on the fet drain at turn off also, say 2.2nF 200V cap and 22 ohm, 3W non inductive R.

- - - Updated - - -

just noticed: there appears to be no heat-sink on your PI device? to keep it cool...? please explain, I assumed initially it was on the bottom of the wiring board... :eek:/

- - - Updated - - -

Also your main output diode looks to be a slow type? it needs to have Trr < 50nS, similarly for the snubber diode on the pri side, e.g. BYV26C or better - all these changes will help as the fet will then have less of a current spike at turn on. I assume you are using a TOP-271 or similar?
 

I designedpaint1.png fly back power supply 60v/3amps . circuit should be work cv and cc mode.from power integration reference circuit . actually the reference circuit is 23v/3amps . i made some changes in the circuit for 60v. and i got the constant 60v output iit is working fine. but, in the reference circuit opamp lm358 section is designed for 23v/3amps and now i want to set the 23v to 60v in opamp circuit section.please tell what changes to do in the circuit . i know haveto change resistor divider but am confusing. please help me exatly what resistors have to change . i attached the reference circuit diagram
 

Need to generate an auxiliary supply voltage in a 15 - 25 V range. You can copy some ideas from the high voltage output reference designs like der580.


Please explain me how the lm358 circuit will work in the above circuit diagram. am not understanding
 

Need to generate an auxiliary supply voltage in a 15 - 25 V range. You can copy some ideas from the high voltage output reference designs like der580.

- - - Updated - - -

You have voltage and current PI controllers, their outputs are ANDed by a diode network.

The problem with 60 V output is the limited LM358 supply voltage range.
 

the LM358 needs a zener on the Vcc to limit the supply volts and an R to the 60V o/p

- - - Updated - - -

the opto needs to run off a lower Vcc too
 

just do it for the output voltage regulation first, then add in the error amp (lm358) for the current regulation....you will then disentagle them in your mind and it will be obvious to you.
There is actually an IC which has the two opamps and reference on it which saves you needing all that circuitry.........people here will know its name as i have forgot.........it is literally made for the application which you show here...ie, voltage and current regulation.............as you know, its usual for the v regulation to work first, and the current reg only kicks in if the circuit is overloaded.

- - - Updated - - -

Also, if you dont understand that circuit that you show, then you can use here the ltspice simulation of it, and experiment with changing the load and running it...this will definetely show you how it works. If you do not understand after then please get back and i will make it clear

- - - Updated - - -

Give me a shout if you want an ltspice quik start guide.
But basically just change file to .asc type and open it, then hit running man icon.
 

Attachments

  • Flyback_60V current limited.txt
    8 KB · Views: 51

please send me the ltspice simulation software. the above circuit is implemented by power integration please check der566

- - - Updated - - -

the above circuit is implemented by power integration pls check der566 file and pls explain me how the lm358 section works and i want to run the circuit in 60v/3amps . and pls tell me what changes have to do .its so important to me . pls help
 
  • 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