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.

[SOLVED] Transformer design for SMPS

Status
Not open for further replies.

ashugtiwari

Full Member level 4
Joined
Dec 12, 2008
Messages
218
Helped
45
Reputation
90
Reaction score
44
Trophy points
1,308
Activity points
3,310
Hello,

This is my first time making SMPS for my application, so parden my silly mistakes and questions if any.

I stay in India where 230V ac rms @ 50Hz is available. I took 210V as min and 250V as max input. I need to get 5V@2A as output and using TNY266 as offline switcher for SMPS regulation. I referred following 2 links for making my first design:
https://www.mouser.com/pdfdocs/2-8.pdf

https://www.fairchildsemi.com/application-notes/AN/AN-4137.pdf

I calculated all the values which I attached in excel sheet.
https://1drv.ms/x/s!AhoWru2BCG9EgoRuFUzRit8UTTyGbg

It seems I need to wind transformer myself. So, I calculated primary and secondary turns used the table given in below link:
https://www.powerstream.com/Wire_Size.htm

So, primary is 144 turns of AWG 27 wire based on switching frequency and current and secondary is 4 turns of AWG 15 based on current but the switching frequency of that AWG is very low compared to the switching frequency of TNY266 i.e. 132 kHz. Did I select the correct AWG of wire for primary and secondary? And as per Infineon-Mouser design guide, I have decided to use EE16/8/5 core, is this the correct selection?

Am I doing correct calculations? If not, what are the corrections? Any recommended readings on subject to help design power supply and transformer design?

Thanks

Ashutosh
 

Did you try and use their PI Expert software?....it does the transformer design for you.
Anyway, you should just check that core does not saturate with peak primary current.
Also, that core losses are not too high.
Also, you need to be sure you are not reflecting too much voltage to the primary, otherwise mosfet will blow.
Also, you need to make sure you do not reflect too much voltage to the secondary, otherwise diode will blow.
Also, you must have gap for flyback.
Also, you should have proper mains isolation, so use triple insulated wire or enamelled copper wire and margin tape to keep good clearance distance.
Remeber that if you have a too high NS/NP ratio, then your duty cycle will get too small, and your on time will go below the controllers minimum on time.
But if you have too low NS/NP ratio then your duty cycle will get too high, and you will be cramming the secondary conduction (the highest current one) into too small a time and secondary rms will be higher than wanted.
Also, the inductances must be high enough to keep you out of too deep discontinuous conduction mode...but DCM is a good mode...just dont go too deep into it else peak current is too high.
It is nice for you to do DCM...because you dont get noisy sec diode turn off.
 
I used the PI expert app to find all the transformer values. But, just a small doubt. Since, before this app I did manual calculations, the operating frequency of tinyswitch-3 is 132kHz and for this frequency the ideal wire given in wire gauge chart is 27 AWG which also has current carrying capability higher than what I have calculated for primary and for secondary I planned to use 3 of these wires for higher current at same frequency.
But, the PI expert gave primary wire as 33 AWG 1 magnet wire and 26 AWG 3 secondary wire which has higher and lower values of frequency but with lesser current capability, will it not affect the heat generation of wires or losses in wires?
Sorry, if this question sound stupid and app is giving correct output, I just have this doubt to be cleared for my understanding.
 

Flyback has DC in the coils, so skin effect is not so important.....also, if you only have a few turns inyour secondary, then even with skin effect, there will not be a big i^2R loss.....i think with SMPS it often causes confusion, because people take [Amps/square mm] values from mains transformers, which have huge numbers of turns...with flybacks, there are not that many turns, so not that much resistance, and not that much loss due to conduction....

Do you i^2.R calculation and see how it comes out......remember skin efect in flyback is not that significant unless you have wire that is more than three skin depths in diameter.
 
the PI expert gave primary wire as 33 AWG 1 magnet wire and 26 AWG 3 secondary wire which has higher and lower values of frequency but with lesser current capability.

Of course you can use lower gauge but that may increase the heating. It is not at all critical because your output power is only 10W.

If you use fatter wire (lower number of AWG) then the transformer window area may not be sufficient. Generally +/-1 deviation from the AWG value is not at all a problem.

But you should stick with the freq suggested by the software. Yes, it is a good practice to use 2-3-4-5 strands (depending on the current) of the thinner wire (nicely twisted) as the secondary coil. You will need only one AWG wire.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top