# Selecting DC Blocking Capacitors for GPS, LoRa

I am designing a transmitter and receiver part of LoRa and GPS. It requires DC Blocking capacitors. I was looking for capacitor over digikey and I am amazed most of the manufacturer haven't specified impedance vs frequency curve.

I am interested in RF capacitor which can be used as DC blocking capacitor and pass signal of 900 MHz (LoRa) and 1.5 GHz (GPS).

Even if very few manufacturer have specified the graph but it looks like they have SRF at around 200 MHz and I think they can't be used in my application. Should I ignore SRF and check impedance at my frequency of interest? I am clueless now how to select the capacitor for RF.

The ESR is also specified at 100 kHz or some other low frequency but I want to interpolate it to my frequency of interest. I came across one application note which states that

If, for example, you are designing for a 900 MHz wireless application, and the ESR is specified at 150 MHz, the ESR at 900 MHz may be calculated by multiplying the specified ESR at 150 MHz by √ 900/150.

Is it correct? Can someone comment on what other parameters I should consider while selecting capacitor for RF application? I have mentioned few of the parameters below. Please someone validate my understanding.

1. Tolerance- I am looking for capacitor having less than 5% tolerance.
2. Voltage Rating: I am selecting caps having voltage double of my max RF signal voltage.
3. Quality Factor: I am looking for high Q.
4. ESR: I am interpolating ESR to my frequency of interest and trying to select the lowest ESR caps available on the digikey.
• What impedance do you need? and does 1nH/mm sound familiar to you? – Tony Stewart Sunnyskyguy EE75 Jun 10 '18 at 16:00
• As less as possible. I know if I can go on ESR,igh then cost may increase. Right now, I can't find any impedance specified at my frequency of interest. – abhiarora Jun 10 '18 at 16:02
• You might not find the Z(f) curve on digikey, but many will have it on the manufacturer web site. TDK, Taiyo Yuden, Murata, Kemet, for example, all have these curves available. – The Photon Jun 10 '18 at 16:07
• @TonyStewartolderthandirt Are you talking about Inductance per unit length of transmission line? – abhiarora Jun 10 '18 at 16:11
• I have edited my question. – abhiarora Jun 10 '18 at 16:11

I am amazed most of the manufacturer haven't specified impedance vs frequency curve.

Most of the manufacturers I use do have this data available (because if they don't, I won't use that manufacturer). It will be a "typical performance" curve and not a specification, but it will be published. It might not be available on the distributor website, but it will be on the manufacturer website.

The ESR is also specified at 100 kHz or some other low frequency but I want to interpolate it to my frequency of interest.

ESR is not usually a critical parameter for a signal DC blocking application.

I usually take the value at the bottom of the dip in the |Z(f)| curve as the ESR. That will be good enough for most DC-blocking applications. If you really need to know the phase shift at some specific frequency, you might need to use a more complete model (but also, variation from part to part might make your careful modelling irrelevant).

Voltage Rating: I am selecting caps having voltage double of my max RF signal voltage.

It is even more important to consider the DC difference that will be across the capacitor in your circuit.

Quality Factor: I am looking for high Q...ESR: I am interpolating ESR to my frequency of interest and trying to select the lowest ESR caps available

High Q and low ESR are just two ways to say the same thing.

Ultra-low ESR is not typically critical for a dc-blocking application. As long as you're using ceramic (NPO/C0G) parts, I wouldn't spend time sorting by ESR.

what other parameters I should consider while selecting capacitor for RF application?

You want to look at the capacitance stability with bias voltage and temperature. For your application you should probably be using NPO/C0G parts.

• Great answer. I have one doubt. Is Z(f) reactive only? I used to think it will only introduce phase shift or delay in the RF signal. ESR would rather attenuate my RF signal and dissipate it as heat. Can you elaborate on this point a bit more? – abhiarora Jun 10 '18 at 16:23
• Of course ESR will contribute a (tiny) real component to Z(f), but it will be dominated by the reactive components everywhere except at the series resonance frequency. Typical ESR of a ceramic SMT MLCC is less than 1 ohm. This will cause a small power loss, but it probably won't be near the biggest inefficiency in a realistic system. – The Photon Jun 10 '18 at 16:28
• @abhiarora I missed where you said you're designing a GPS transmitter. If you're passing 10's of W signals through these capacitors, you may need to be more careful of ESR loss than I've ever had to be. – The Photon Jun 10 '18 at 16:32
• What about SRF? It should be above my frequency of interest or I should consider a caps only with it's impedance at my frequency of interest. If impedance is acceptable, then I should be selecting the caps irrespective if it is inductive in that region? Regarding GPS, transmission is only for LoRa. – abhiarora Jun 10 '18 at 16:33
• @abhiarora, sorry I don't know what that means in terms of transmit power. – The Photon Jun 10 '18 at 16:34