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.

Push Pull Converter transformer design for 120 watt power

Status
Not open for further replies.

adnan012

Advanced Member level 1
Joined
Oct 6, 2006
Messages
468
Helped
2
Reputation
4
Reaction score
2
Trophy points
1,298
Activity points
4,923
hi
i want to design push pull transformer with the following specs

Input voltage range 10-18VDC
Output voltage 120VDC
Output power 120 watts
Switching Frequency 100kHz

Transformer core EFD25/13.

I need design equations for transformer.

Regards
 

Design it as a Buck in the first place...a "virtual buck"......pick your Dmax and do it......from your Dmax and vout you will get your virtual vin....this is the Vin that the transformer will put at its secondary..................so thats why i called it a virtual buck.......

so then all you need to do is use vp/vs = np/ns

...and find what turns ratio you need to make your actual vin equal to your "virtual buck vin"

...because in effect, the push pull is just like a buck, but with a transformer there....so you use the transformer to give you the input voltage to your virtual buck that you would want, if it were really, a buck.

To design the transformer....
Any numbers of turns will do...
as long as the turns ratio is right.
As long as its physically manufacturable.
As long as the core doesnt saturate on the magnetising current.
As long as you can terminate the turns to the bobbin
As long as you dont get too much skin effect loss ..use litz
As long as you dont get too much copper loss.
as long as you dont get too much core loss....dont have too high Bmax
As long as you dont have way too much leakage inductance (try and have full layers of turns)
As long as you dont have too much proximity effect loss......dont use loads of layers for one coil...then on top of that, loads of layers of the other coil.....
As long as you have the required voltage isolation...use enammeled copper wire or triple insulated wire

- - - Updated - - -

So anyway.....you know vout.
You know your min vin.
Choose a Dmax....say 0.7 if you like

Then "vin" = vout/D.

...THATS the Vin to the "buck"...

Now what transformer turns ratio converts your actual min vin to your Buck "vin"?
...there you have it.

You may like to choose a lower D....say less than 0.5 so that you dont need slope compensation.

- - - Updated - - -

remember you have two primaries (split)......

- - - Updated - - -

you can also do like a current doubler type secondary with two output inductors...and one sec coil...but can also have two secondaries for either polarity.

- - - Updated - - -

Have a go and we can also talk of getting the primary1 coupling with secondary1 the same as the coupling between primary2 and secondary2....i am sure you already understand.
You could eg do interleaved and have pri1//sec1//pri2//sec2.

Or, since its only 120W...you could just do a 2 tran forward and have simple pri and sec...use a simple bootstrap high side driver for the high fet
 
Last edited by a moderator:
Simulation comparing various primary values in a transformer. It shows how your specs might be achieved (although your topology is no doubt different).

transformer 4 primary values compared 10V sq wa x16 stepup.png

A primary value in the tens of microHenries appears to yield satisfactory performance.

Notice that a mere 1/10 ohm parasitic resistance requires that the step-up ratio be greater than the amount predicted in theory. Conventional wisdom tells us to expect 10 or 15 percent loss of efficiency.
 

...because in effect, the push pull is just like a buck, but with a transformer there....so you use the transformer to give you the input voltage to your virtual buck that you would want, if it were really, a buck.
The converter topology should be clarified. The transformer itself can't act as a buck converter, there must be an output inductor performing energy storage. But that's probably what the OP intends.
 

The converter topology should be clarified. The transformer itself can't act as a buck converter, there must be an output inductor performing energy storage. But that's probably what the OP intends.
Thanks, yes , i agree entirely, sorry if i didnt make it clear above.
Yes i certainly didnt mean the transformer acts as a buck converter.

- - - Updated - - -

if the OP picks a duty cycle max..say 0.7...then take vin = 171v, vout = 120v at 120w....then design a CCM buck converter like that...ill show OP how OP can then very simply convert this into a pushpull of the same spec.

Note that I have used min vin = 171v when real vin min is 10v…….thats where the transformer will come into it….and where we will convert the miscreant buck into the swan-like pushpull

- - - Updated - - -

Here is a push pull that meats your spec in the free simulator from ltspice…….convert to .asc format, then hit run.
I didn’t put any leakage in the transformer, …so you will need to add that and add some means of snubbing it

- - - Updated - - -

I think one of the main things for the transformer, is getting the push pull design right first....and of course......you need to wind full layers of turns in general otherwise its too difficult to wind a layer on top of the one you've just wound....you can always cheat and use double or triple strands to fill the layer if need be........do you know how to use the core's "AL" value to find the inductance....also, use a ferrite thats fairly matched to your f(sw)

- - - Updated - - -

I dont think theres a formula for doing a transformer, since you are limited to offtheshelf cores generally...so pick one....wind it.......see if it meets the criteria above...if not, pick a different core........you can make an excel spreadsheet which makes it quick to do this.....often you pick a core too small, and it would be in saturation, so you just choose the next size up....or often, you find your turns simply dont fit on the bobbin..they overflow it..so again, you need to pick the next size up and re-do it.

- - - Updated - - -

all the maths you need to do the transformer is in this doc
https://massey276.wixsite.com/maths

- - - Updated - - -

Actually sorry I mentioned a current doubler rectifier above…..this could be do-able, but not worth it for your spec, as you have a high step up ratio and rel’y low output current.

- - - Updated - - -

Do you understand the difference between magnetising current and power current in a push pull?
This is key to winding the txfmr.
The attached sim shows an expression for the magnetising current so that you can see it in the pushpull ltspice simulation attached
 

Attachments

  • pushpull.pdf
    22 KB · Views: 151
  • pushpull.txt
    9.1 KB · Views: 60
  • pushpull with magnetising current show.txt
    9.3 KB · Views: 108
Last edited by a moderator:
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top