Hello all. New member, first post. I am helping a friend make a step-up transformer. The plan is to use a square spit ferrite core. I will be machining plastic spools to wind the coils on. Does it matter if the i.d. of the spools are round vs. square to match the cross section of the ferrite? This will be for a low amperage, low power application. Coils will be wound with 18 ga formvar coated magnet wire, 500 turns on the primary, 3000 turns on the secondary. I am a novice, mechanical engineer with limited EE knowledge, i.e. "V = I R, touch one wire at a time and wear tennis shoes." Please let me know if my questions are better suited for a different forum.

Thanks

 

Is this transformer for a mains frequency, 50Hz or 60Hz? Ferrite is usually for high frequency transformers like in a switch mode power supply. Do you have a link to what core you're using? They are kind of dependent on what is being done too, different materials for different purposes.

 

Correct, it is for a high frequency application. We were able to find round cross-section ferrites. It will make machining the coil spools easier.



Here is the circuit model:



The primary input will be between 5V to 9V at 5khz to 12khz

 

Did you look on the website of the supplier of the core? Many times the copanies will sell things that go together with other products or tell you where to look for them.

 

Good suggestion, thanks!

 

Can you put a number to "low amperage"?
I haven't calculated the volume that 500 turns of 18AWG will occupy (have you?), but to me that core looks a tad on the small side, by the time you take into account bobbin dimensions and insulation layers.

 

It will make machining the coil spools easier.

I missed it in your "circuit" schematic, but, is this for the one marked "toroidal core"? If so you do know/understand the core you showed in the same post isn't a toroidal don't you? They don't even use bobbins/spools, the wire gets wrapped directly on the core. And like Alec said 500 turns of 18AWG takes a lot of real estate.

 

Ok, I did some more design work. What is labeled as "toroidal core" is actually a split square core. I have modeled it to scale using the actual dimensions of the ferrite, 29n018 from alltronics.com If I machine the bobbins from 1.75" o.d. 6/6 nylon or HDPE with a length of 1.25" I can fit 500 turns of 20 ga on the primary (left) and 3000 turns of 28 ga on the secondary. I calc'd the dimensions with an online coil calculator. I will have to go back and verify if the calculator included the thickness of formvar insulation but there is room to stretch the bobbin lengths and tweak the gauge sizes if necessary. Not sure of the wire gauge is sufficient for wattages or heat generation. Initial specs call for 5-9V dc primary input at .5 amps. Do I need to be concerned with fixturing the bobbins so they don't rotate? glue, dielectric set screw, etc? The actual ferrites have an 1/8" perimeter groove. I'm assuming I will also need to clamp the ferrite halves together with tie-rods or such.

 

Initial specs call for 5-9V dc primary input

?? Transformers don't work on DC.

 

Yes, we realize that. The DC source will power a 555 timer circuit, and the conditioned output will supply the step-up transformer primary.

 

Also be aware that split cores like that have an air gap where the two halves come together, no matter how close they fit together. That gap has consequences in some circuits especially high frequency circuits.
http://www.encyclopedia-magnetica.com/doku.php/air_gap

 

Is there such a thing as a ferromagnetic paste one could apply to the faces?