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.

Output voltage drop in flyback SMPS

Status
Not open for further replies.

Mahdi.sh

Newbie level 4
Joined
Mar 5, 2016
Messages
6
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
45
Hello everyone!
Im newbie with SMSP design. I have problems with my flyback SMPS.
Goal is to get 5V with 7Amps in output. At no load everything is OK and average voltage is 5V fixed, and also working good with 5ohm load. But unfortunately when I put a 1.5ohm (which means almost 3amps) average voltage decreases to 3V (and I see a lot of oscillation there).:-(
I use UC3843 as PWM controller , a EI28 ferrite core. Working frequency is 80KHz. I've attached the schematic.
My case is single output (just 5V 7A).

Actually Im suspicious of my transformer winding. How much is it important? I mean how much bad winding can effect on result??
Primary inductance : 300uH
N1/N2 = 5




smps.png
 

T

treez

Guest
You have no capacitor from opto collector to ground, -there should be.
-You have no compensation components from cathode to reference of the tl431
-You have an output LC filter which has too high inductance...also, the most of the output capacitance should be upstream of the output inductor...also, the LC resonant frequency of the output filter is about 700Hz, and that's too close to what is likely the crossover frequency.

- - - Updated - - -

Your 180k, 1w supply resistor is not of sufficient power rating
-You have no capacitor on the vref pin, eg 100n to 1uf

- - - Updated - - -

-You need a ceramic cap say 100n to 1u on the vcc pin of uc3843

- - - Updated - - -

Are you now saying you only need the 5v output, and not the 15v output?

- - - Updated - - -

Stacked winding will be troublesome with so much current flowing in the 5V secondary, it means you will get cross-regulation problems, but it may work to a degree.
 

Mahdi.sh

Newbie level 4
Joined
Mar 5, 2016
Messages
6
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
45
Thank you very much treez!
You're right, Im admitting that there are a lot of shortcoming in my circuit.
But let me expect that it should work to some better levels.(Am I right?!!)
- yes I only put 5V output

I'll do as you told
I hope that it works
:)
 

T

treez

Guest
so you do not now want the 15v output? or you are just trying to get 5v output right first?
 

Mahdi.sh

Newbie level 4
Joined
Mar 5, 2016
Messages
6
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
45
No I dont need 15V output, and I didnt put it in my test circuit.
 

T

treez

Guest
Here is a 5Vout 7A current mode, ccm flyback.
It runs off the DC bus after the mains rectifier, and since you indicate 220VAC mains, I made it for 195vac to 265vac

The duty cycle at min vin has been made small so that there is a decent length of secondary conduction time to make irms sec small. But the duty cycle was not made so small that the reverse secondary diode voltage was too big, because I wanted you to be able to use a schottky on the secondary.

I made it a 1/11 = NS/NP ratio…..and I expect you will interleave the secondary between two primary halves in series.
The primary can be TIW.

tHE SIMULATION RUNS in the free download ltspice simulator from linear.com
Just call the .txt file to .asc, then open it, then hit the running man icon, if you wish to,.

Also, attached is the excel design file.
The LT124X range is pin for pin similar to the UC384X range
 

Attachments

  • Flyback _5V 7A.txt
    6.5 KB · Views: 15
  • Flyback _5V 7A.pdf
    17.7 KB · Views: 25
  • Flyback _5V 7A.zip
    5.2 KB · Views: 14

Mahdi.sh

Newbie level 4
Joined
Mar 5, 2016
Messages
6
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
45
Thank you a lot treez!
I saw simulation...
It works very good but isolation (HV from LV) is critical for me.

I think I've found my mistake! I used a slow opto, so for high loads UC couldn't follow changes.
Actually I had used pc817 instead tlp521.
Im going to utilize a fast opto. pray for mee! :wink:
 

T

treez

Guest
ok, but a slow opto should be ok as long as the other feedback compensation components take it into account.
Did you sandwich wind the trxfmr, as per AN-18 of power integrations?
 
T

treez

Guest
Feedback compensation components as in attached.
Also, the cap across the opto is one too.

- - - Updated - - -

The output cap and its esr is also a feedback loop parameter.
So if you are using a slow opto, then you can compensate for it with other things in your circuit.
By slow opto yuo are of course referring to the BC capacitance of the opto transistor.

- - - Updated - - -

Also attached is onsemi app not explaining it.
 

Attachments

  • Basso 287.pdf
    606 KB · Views: 15
  • TL431_OPTO ..ONSEMI ..TND381-D.pdf
    581.2 KB · Views: 16
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top