about Q factor of varactor diode at any freq

[amicloud]

Hello all,
I learned that M/A COM MA46 Series varactor diodes have very high Q at 50MHz(as given in datasheet), for example 4000(amazing). But what is the Q at 1000Mhz，1500MHz，2000Mhz,2500MHz? No value is given. So what is the formular to calculate it?
has anybody the experience of using the diodes? how much is the Q in reality?

Best regards,
Yun Liu

 

[chuckey]

I would say that the Q at 50 MHZ is a theoretical value (though it might be correct). If you assume that the loss resistance is in the connection in series with the capacitive element, though some will be the parallel reverse bias leakage current. Then the Q varies as the inverse of the frequency. Again Capacitors have a Dissipation fact (D), only coils and tuned circuits have a Q. D = 1/Q So at 2.5 GHZ, "Q" = 2500/50 X 4000 = 50.
Frank

 

[jiripolivka]

Hello all,
I learned that M/A COM MA46 Series varactor diodes have very high Q at 50MHz(as given in datasheet), for example 4000(amazing). But what is the Q at 1000Mhz，1500MHz，2000Mhz,2500MHz? No value is given. So what is the formular to calculate it?
has anybody the experience of using the diodes? how much is the Q in reality?

Best regards,
Yun Liu

If M/A COM gives only one value of Q at 50 MHz, there is no way to "calculate" it at higher frequencies >20 times higher.
The only way is to measure variations of junction capacitance and series resistance. Check with M/A com on what test fixture they used, and design yours for 1-2.5 GHz. It is not easy, and calculations lead nowhere. You must build your test circuit and run your tests.
Certain varactors are designed for tuning, others for frequency multiplication, special ones to generate a comb of frequencies. Therefore one cannot use one approach for all.

 

[vfone]

I have doubts about Q>6000 @ 50MHz for varactor MA46504 when actually they mention that the capacitance was measured at 1MHz.
Even my old home-made varactor capacimeter is measuring the capacitance at 20MHz.
I think this huge Q for a semiconductor is just estimated.

**broken link removed**

 

[jiripolivka]

I have doubts about Q>6000 @ 50MHz for varactor MA46504 when actually they mention that the capacitance was measured at 1MHz.
Even my old home-made varactor capacimeter is measuring the capacitance at 20MHz.
I think this huge Q for a semiconductor is just estimated.

**broken link removed**

Q is calculated from the junction capacitance at a maximum voltage, and series resistance M/A com do not indicate any details of their testing (jigs), so if you need you must make and use yours.
A real Q can be measured on a resonator tuned with the varactor, or on an oscillator line width.

 

[vfone]

There are two different things that don't have to mix each other.
First is the Q-factor of a device (varactor in this case), and second is the Q-factor of a resonant circuit. Even they interact somehow one to the other, they have different causes and meanings in a circuit.
For example an LC resonator with L and C having the same Q-factor, just changing their values (same resonant frequency but different LC ratio), this would affect a lot the resonator Q-factor.

 

[biff44]

You can easily make a circuit to test the varactor Q at your desired frequency. The only trouble is to inject DC bias into the varactor without messing up the Q measurement (the bias network can De-Q things).

I would suggest making a test fixture that is similar to how you are going to use the device. Make it up with either microstrip elements or L-C chip components OF A KNOW Q factor, form a resonant or anti resonant circuit response, measure the frequency of resonance, and basndwidth of resonance, and curve match that response to the theoretical response.

You will notice the Q is lower at 0V bias, and higher at a high reverse bias voltage.