The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Inductor with small inductance and low-resonance frequency
https://designers-guide.org/forum/YaBB.pl?num=1397125673

Message started by rf_man on Apr 10^th, 2014, 3:27am

Title: Inductor with small inductance and low-resonance frequency
Post by rf_man on Apr 10^th, 2014, 3:27am

Hi Gurus,

A high shunt impedance is needed at low-frequency (~500MHz), 30nH inductance is approx. ok. However, this inductor is huge on silicon(SiGe Bicmos technology), its fres is about 2GHz. Wanna to use an inductor which has fres at ~500MHz, then the impedance at fres=500MHz is high enough. But I think the inductor with fres=500MHz is not possible to be made smaller than the original one: L=30nH, fres=2GHz. Because, fres=1/sqrt(LC), for lower fres, L needs to be large, maybe change to lower metal layers can help increase C, but fres will not be reduced too much.
Am I right? or there are other methods to make a relatively small inductor but with very low fres?

Your help is highly appreciated.

An rf-design beginner.

Title: Re: Inductor with small inductance and low-resonance frequency
Post by aaron_do on Apr 10^th, 2014, 5:54pm

Hi,

its easy to lower the resonant frequency. Just add a parallel cap to form an LC tank. But you seem to be disregarding the inductor's Q factor. At resonance, a smaller inductor will appear as a smaller shunt resistance which is proportional to L (Q.ωL). i.e a bigger inductor will get you better performance.

That said, at 500 MHz, you may not want to use an inductor at all. A resistor or a transistor might do a better job. As you pointed out, 30 nH inductors are not very practical on chip.

regards,
Aaron

Title: Re: Inductor with small inductance and low-resonance frequency
Post by rf_man on Apr 11^th, 2014, 2:00am

Hi Aaron,

Thank you for your reply.

The 30nH inductor is actually used as an inductive shunt for ESD protection. Adding a parallel cap will this cause possible oscillation problem?
And using inductor at resonance is a very narrow band solution for RF performance, right?

Kind regards
rf_man

Title: Re: Inductor with small inductance and low-resonance frequency
Post by aaron_do on Apr 11^th, 2014, 5:49pm

I see. 30 nH isn't very practical on-chip. It depends on your design, but roughly speaking, at 500-MHz you would need inductors of that size. How narrow the bandwidth would be depends on your Q. I don't know anything about your design so...

I assume you've considered protection diodes? You could have a protection diode on-chip just to get a basic level of protection, and then use an off-chip inductor...

Aaron

Title: Re: Inductor with small inductance and low-resonance frequency
Post by rf_man on Apr 14^th, 2014, 12:19am

yes, I considered the ESD diodes. Due to the parasitic diodes to substrate, the stacked diodes solution does not work.

Could you please indicate whether there is risk of oscillation due to adding parallel C to form an LC-tank?

Many thanks.

rf_man

Title: Re: Inductor with small inductance and low-resonance frequency
Post by aaron_do on Apr 14^th, 2014, 2:05am

Hi,

yes there is definitely a risk of oscillation. There would be a risk of oscillation even without adding the C. You always need to check for instability before fabricating your circuit, and addition of inductors will increase the risk.

What kind of circuit is this? No problem if you can't share...

Aaron

Title: Re: Inductor with small inductance and low-resonance frequency
Post by rf_man on Apr 15^th, 2014, 5:11am

Hi Aaron,

It's a front-end IC.

Suppose that use a relatively small inductor together with a parallel C to tune the resonance frequency to 500MHz. Then the risk of oscillation is that it may oscillate at 500MHz? If just use the 30nH inductor without adding parallel C, then it might oscillate at a higher frequency (determined by 30nH and the parasitic C to the ground)?

Now I am confused with the self-resonance frequency, e.g. an inductor 's fres=2GHz, then it means that it will always oscillate at 2GHz? Below 2GHz, it functions as an inductor; above 2GHz, it functions as an capacitor. But many inductors are used in a circuit which operates far above the fres of the inductor without oscillation at fres of the inductor, right? Please help me to understand this.

Kind regards,

Title: Re: Inductor with small inductance and low-resonance frequency
Post by loose-electron on Apr 15^th, 2014, 11:59am

could you provide a schematic?

Title: Re: Inductor with small inductance and low-resonance frequency
Post by aaron_do on Apr 15^th, 2014, 10:22pm

Hi,

apart from the above request...

resonance does not imply oscillation. First, when we say oscillate, we mean oscillate forever or until the oscillation is disrupted. It you are talking about an oscillation that gradually dies down, then we usually call it "ringing". In order for a resonator to oscillate, the same amount of energy that is dissipated in the parasitic resistance must be added back, usually by positive feedback (unwanted in your case). In fact, there is always a risk of oscillation even without an inductor. But I would say that inductors increase the risk.

have I answered your question?

Aaron

Title: Re: Inductor with small inductance and low-resonance frequency
Post by rf_man on Apr 18^th, 2014, 12:01am

Hi Aaron,

sorry that I can't provide the schematic.

hank you a lot for your help and patience.

Kind regards,

rf_man