I am designing an isolated DC-DC converter with an input of 12V and an output of 5V@12A, at the moment, a flyback seems like the best option. I have read several books regarding the subject and I am proceeding to the implementation; however, I haven't been able to find a transformer that meets my requirements (I have searched on Mouser, Digikey and Newark to no avail). I suppose I've been searching incorrectly, since most of the transformers I have found can only manage up to 2A. I currently have no limitation regarding frequency or turns, but I would like to find the perfect match. So, where can I get the transformer I need? (Do have in mind that I require a website that can ship to Mexico)
-
2SMPS transformers are unfortunately not really available "off-the-shelf". This is an ongoing pain-in-the-ass of people like me who tend to make single/small-volume custom things. – Connor Wolf Jun 06 '13 at 20:17
-
1It's a shame, really... What's the cause of this issue? Do designers prefer custom-made magnetics? – CristianGuerrero Jun 06 '13 at 22:46
-
2@CristianGuerrero This would make an excellent separate question here on EE.SE. The causes are more practical/economical than physical. – Nick Alexeev Jun 07 '13 at 18:34
3 Answers
Sourcing a transformer a major challenge/headache during the design of an isolated power supply. The more power is required, the harder it is to find an off-the-shelf magnetic. Majority of the power supply transformers are batch-built.
Usually magnetics manufacturers' web sites have better search/filtering tools than distributors like DigiKey and Mouser. Here are my go-to places for shopping for off-the-shelf magnetics:
- Pulse Electronics
- Coilcraft
- Würth
- the supplier listed in the BOM of the reference design
Lead times may be long too.

- 37,739
- 17
- 97
- 230
-
It seems like the best option; however, I do have one more question (I hope you could answer it): I do not have much experience regarding transformers, it seems take there are several types (e.g. planar, toroidal, wound) for the same application but I haven't been able to know which is the best option for my application. Which one should I use? – CristianGuerrero Jun 06 '13 at 22:43
-
1@CristianGuerrero Again, this would make an excellent separate question on EE.SE. – Nick Alexeev Jun 07 '13 at 18:39
-
@CristianGuerrero Either of the construction methods could work for you. Planar xfrmrs are neat, especially when you don't need UL-rated isolation. But, to test the transformer design, you should use a wound xfrmr. Pot cores and other cores that have bobbins are handy, because they are easy to wind by hand. – Nick Alexeev Jun 07 '13 at 18:42
Another option is to wind your own, if you only need one or a handful.

- 5,650
- 26
- 36
-
1Since I am building a prototype, it is feasible, however, I have no technical knowledge regarding transformer design; I tried getting some literature about this, but most is excessively theoretical. – CristianGuerrero Jun 06 '13 at 22:35
-
Do you at least understand what the turns ratio should be? If you do then next is to understand tx losses and next is to be able to compensate for the losses. – Andy aka Jun 06 '13 at 23:03
Would it make it easier if you designed a flyback for an output of 12V dc and incorporated a buck convertor following producing the 5V. With a 12V DC output, the primary and secondary turns are more evenly matched and easier to source.
In fact, if cost permits virtually any output voltage up to 30V dc would suit as an input voltage to a hell of a lot of post-transformer buck regulators.

- 434,556
- 28
- 351
- 777
-
My input is a battery bank of 12V and I need an output of 5V, I think the design would complicate greatly with this approach – CristianGuerrero Jun 06 '13 at 22:32
-
Complication usually equals cost or maintainability. On the other hand, complication can mean finding an isolation transformer that fits the bill and giving you a product. It can also mean better performance. What version of complication do you mean? – Andy aka Jun 06 '13 at 23:02