Printed coil design, Help

[atakan_1907]

Hi all,
Im new to rf electronics and want to design an on-pcb coil to use in a tda7000 receiver circuit. I calculated the desired inductance as 250nH.
Altought im searching and reading about it for a while, couldnt make it. I tried asitic, agilent ads etc but ... no!
How can i calculate it and draw on proteus or even ms-paint =)
Could someone guide me to understand the principle and design my own?
Please dont just give me links, explain. I dont even know the thickness of copper on a pcb =)

 

[RCinFLA]

250 nH is a pretty large PCB inductor. You would be better off using an airwound coil.

 

[klystron]

This calculator computes the inductance of a microstrip etched on a PCB, which according to the ARRL handbook is:

L= 0.00508*b*(ln(2*b/(w+h))+.5+0.2235*(w+h)/b)

where:

w is width of the strip in inches,
b is the length in inches,
h is the distance between the strip and the ground plane, and
L is inductance in uH.

 

[atakan_1907]

How big, 4 cm2 or larger?
Is that formula for just a strip? Sth like this? ========================

 

[klystron]

Yes, it is for a strip. For other configurations see Inductance Calculator

 

[atakan_1907]

Sorry, i didnt get a good answer
Could someone explain how to calculate and draw an etched coil?
I saw the inductance calculator before but its not clear enough.

 

[kalbun]

I can suggest you this link:

Stanford Microwave Integrated Circuits Laboratory -- Stanford University

Here you can find not only a quite good calculator, but also an academic article explaining how to do.
I used the article to develop a software that, in facts, given the desired inductance and the physical constraints, gives you the best match.

But please be careful with printed coils. I did some experiments and at the end replaced PCB coils with good quality, traditional inductors due to parasitic capacitances and other side effects (I was working at 500 MHz).

 

[FvM]

You have to find out the structure width that can be reproduced in your PCB technology, e.g. 150 um. A square spiral will be most easy to draw.

Although I'm sure, a printed inductor will work for FM radio frequencies, it will considerably larger than a wire wound SMD inductor.

 

[neazoi]

Or be a little bit moe practical. By trial and error... **broken link removed**

 

[volker_muehlhaus]

The equations are well known, so trial and error is not needed.

Simple Accurate Expressions for Planar Spiral Inductances

 

[biff44]

That large of an inductance is going to require multiple turns. A single loop will not have the inductance. Since you will need to jumper over the coil to get to the middle (start) with a multilayer trace, it is best to simulate such a thing in an EM simulator to model the full effects. Anything else will be a crude approximation.

 

[volker_muehlhaus]

@biff44:

The equations that I linked to, and the equations used in the online calculator in post #7, are designed for multi-turn inductors.
I am working on inductor analysis with EM simulators, but use these equations as a starting point. click!
From my experience, the equations are often accurate to within 3% or so, if we compare DC inductance to EM analysis results.

I agree that for the stacked multi-layer inductors in post #9, you would better use an EM simulator because the coupling between the segments on different layers is not know exactly.