Electronic – DC Blocking capacitors – What value to choose

Common tolerance codes for capacitors: J = ± 5%; K = ± 10%; M = ± 20%.

Common value code for capacitors: Two numbers, and a third number c, where c tells you the number of zeroes behind the first and second number. Usually, the result is to be read in pF.

Sometimes, there's also a value expressed in pF or µF, and you have to guess which is right. Some examples:

• A ceramic capacitor with the number "470" on it likely has 470 pF, because ceramic caps are still mostly used for small-ish values.

• "0.47" doesn't make sense in pF, because 0.47 pF would be too small for almost any practical use, so pretty much all capacitors labeled "0.47" will have a value of 0.47 µF = 470 nF.

• "470" on a large-ish film or electrolytic capacitor will likely mean the cap's value is 470 µF.

(And even more strange markings do exist...)

Now, let's use your capacitors' markings as examples for this - here's what I guess:

1. capacitor #1:

   103M Z5U 2-3KV ARC GAP KAP CHINA

   10 * 10³ pF = 10 000 pF = 10 nF. M: ± 20%

   Z5U is the type of dielectric. This is a pretty creepy type of ceramic with huge tolerances over voltage and temperature.

2. capacitor #2:

   NPO 7.5D IKV

   7.5 is a fairly uncommon value for a capacitor. Mostly, you find values from the E6 or E12 series, hardly anything else. However, 7.5 is part of the E24 series, so it is not entirely alien, and according to this source, D would mean you have a tolerance of ± 0.5 pF. NP0 is a very good type of ceramic mostly used for values below 10...100 pF (that 0 in NP0 is a zero; I remember to have read that NP0 means negative-positive-zero, i.e. nearly zero tolerance over temperature and voltage changes). I guess your cap has 7.5 pF. That I is likely a 1, meaning the maximum voltage for this cap is 1 kV.

3. Capacitor #3:

   CM 1000M 125L

   Maybe 1000 pF = 1 nF, with a tolerance of ± 20 % (M).

4. Capacitor #4:

   271 2KV

   27 * 10¹ pF = 270 pF. Maximum Voltage: 2 kV.

5. Capacitor #5:

   Z5U 4700M IKV

   Another one with a cheap type of ceramic (Z5U), probably 4 700 pF = 4.7 nF. Tolerance: ± 20 % (M). Max. Voltage: 1 kV.

Again, this is guess-work. Unfortunately, there is no standard that all manufacturers adhere to, so to be exactly sure, you would have to measure your devices and find the original data sheets with the device marking specifications, which can be very, very annoying.

Even more examples from similar questions: Identifying Capacitors, https://electronics.stackexchange.com/questions/10474/what-kind-of-capacitor-is-this

For 1.5GHz the exact type of capacitor is important as well the the capacitance value. Many types of capacitor will self resonate well below 1.5GHz. The capacitor does not act like a perfect capacitor, instead it acts like a capacitor with an inductor in series.

If your capacitor was perfect and your antenna impedance at 1.5GHz is entirely resistive then you would only need to consider the impedance of the capacitor and make it low enough that it has no effect on the rf. A perfect 100uF capacitor would have an impedance of a tiny fraction of an ohm at 1.5GHz. In reality a 100uF electrolytic capacitor dosn't work at 1.5GHz due to inductance of the leads and the coiled plates.

You have to use a tiny surface mount capacitor for 1.5GHZ.

If you have a typically antenna system characteristic impedance of 50-75ohms then making the capacitor impedance under 1ohm means it has no practical effect. Maybe you want to try for 0.1ohm if you want to get the last 0.1dB of performance if you are designing expensive satellite equipment.

TO figure out the minimum capacitance value to get below 1 ohm at 1.5GHz you can use the 1/2.pi.F.C formula from any basic electronics textbook. It works out that you need a capacitor of 100pF or larger. The practical problem then becomes picking a capacitor that has low enough parasitic inductance and dielectric loss to actually work.

The behavior of common components at microwave frequencies is often not specified by the manufacturer. This is when you realise why RF is seen as a black art. You sometimes need to spend some time reading, looking at stuff other people have deigned and testing things in your lab full of RF gear to find out what works.

If you antenna impedance is significantly inductive or capacitive then you can make it more complicated by using the block cap as part of a matching circuit.