what's planar transformer

planar transformer core -patent

could you give us a link or detailed explanation abt what is planar transformer

dose it built on PCB?
dose it need amulty layer PCB?
dose it needs a core ?

planar core transformer

Yes, you have to use a core the PCB has the windings,

check with w*w.ferroxcube.com, w*w.mag-inc.com for more info.

Tornado

rf types

At RF and microwave frequencies you can do this without the core. The coupling is less, but it works. With multilayer boards you can put more turns in a given surface area and increase the coupling.

thanks all

flatulent

my problem is to sens 37A square wave 45% duty 2 pulses/cycle

could i sens this current by making a PCB spiral of 25 turn as close as possible to the 37A trace
and connecting a burden resistor on this spiral as an I/V converter
dose this work?

directional coupler

With this power level (37 A) and your only wanting to detect a pulse, I think that an ordinary microwave parallel line directional coupler will work. Lets do some calculations. For 2 us, you have a frequency of about 500 kHz or a wavelength of 600 m. The sensitivity of a directional coupler is the sine of the line length divided by the wave length. lets try 10 cm to get sine (0.1/600) = sine (1.6e-4 radians) = 1.6e-4 coupling efficiency. this is multiplied by the coupling at 100% efficiency. [edited part. There is a factor of 4 improvement because of the quarter wave length being full coupling and a factor of 2 for FR4 which will make my calcuations around 1e-3 of efficiency which should allow reasonable signals with the 37 A main path signal.]

I would suspect that you could detect the pulse on the output of the coupler. Try an experiment with a piece of wire parallel to the main current wire and an oscilloscope to see if you can get a detectable current.

Another trick is to measure the voltage drop across a length of trace that the 37 A flows through.

Why don't you sense resistive with the resistance of your PCB trace and a difference amplifier??

In principle the spiral close to the 37A track will work. But your signal is strongly dependant to any ferromagnetic material that comes close to the spiral (increasing/decreasing your coupling between track and spiral) and certainly to any other transients or radio frequency signals.

A closed magnetic core with primary and secondary winding has a defined coupling that allows measuring a primary current accurately. This is a better choice.

But the choice depends on your application.

so tell us what you want to measure (frequency range, accuracy).

my application is to sens push pull converter current
this current is 37A peak with frequency 100Khz/90% duty

i don't use resistive sensing coz of power losses ( to get 1 volt representing this current i need 0.025 ohm/30 watt) too much losses

i though about using 0.001Ohm + a difference amplifier but i think nosie will be high comparable to detected signal (37mvolt)

also i need very cheap solution as 0.025 Ohm = 5$

do u have any solution for noise problem in case of .001 PCB resistor+amplifier

thanks 4 you all

emi problems

You are in a high EMI environment. How about putting an extra one turn winding on your transformer? This can be used to sense voltage across a large value resistor and thus not waste power.

This winding would be in parallel effectively with your load and would not waste power because of the load resistance, sensor resistance ratio could be made very small.

If your design can afford more costly components, you could put (glue) a Hall effect current sensor over the power trace and measure the current waveform that way.

sorry

what this single turn do?????????????????

dose it act as a current transformer ?????????

and what will be the gain of this turn V/A??



i'm thinking about sensing the secondary current instead of primary is that work??????

thanks

single turn

The single turn should give you an isolated measure of the voltage waveform across the load and drive.

It will not give you a measure of the current. I was not sure what you wanted to measure.

Many years ago there were FETs with second sources which had currents scaled down from the total. These were used for current measurement.

Just use a current probe (a single primary winding over a toroidal core bobin) as anyone else does it. Just make sure the core volume is sufficient not to saturate the ferite domains.