Phantom Blocking Caps

I see diiferent values being used on phantom blocking caps. Looking at the INA217 datasheet the blocking caps are shown as 47uf electros. But when I built the SSL9K, it used 2.2uf caps as phantom blockers. I have some 2.2ufs left over from that project and wanted to know if I could use them for blocking caps in the 217 circuit. What I am really asking is how important is the value of the phantom blocking caps?

Correct me if I'm wrong, but do the phantom-blocking caps not form a high-pass filter?

I think you select the capacitance value to give a suitable -3dB point (eg. around 20Hz) for a given impedance.

yes they do. you'll frequently see phantom blockers as large as 470uf to counter some of the effects. It's all based on the circuit. a quick and dirty way to do this is to have a large electrolytic bypassed with a small film cap. most companies do this instead of using a better circuit.

One reason to have them higher value than low freq cutoff considerations indicate: better common-mode rejection when the caps are mismatched.

Also electrolytics can show significant distortion when there's actually signal across them, as there is near the highpass cutoff frequency. If you expect 10Hz to be the lowest possible signal frequency then size the caps for a curoff of 1/20 of that, or 0.5Hz. How big the caps should be depends on the resistance on the amplifier side of the cap:

C = 1,000,000 / (2 x pi x R x f)

where C is in microfarads, R is in ohms f is in Hz, and pi is in the oven. If the resistance is 10k (each leg) then 31.8uF is the minimum cap size; use 33uF, or bigger. Personally, I assume there will be occasional subsonic breath blasts, so I'd go bigger.

Peace,
Paul

Interesting thread indeed... i hope to be not too OT if i ask the following:
I wonder if it's an option to switch the phantom power AND hard-bypassing the input caps with relais to get the 'lytics out of the signal path of mic pre's when using dynamic or tube or ribbon mics? Are there any existing designs using this idea?
:?:
Kind regards

Martin

[quote author="smallbutfine"]
I wonder if it's an option to switch the phantom power AND hard-bypassing the input caps with relais to get the 'lytics out of the signal path of mic pre's when using dynamic or tube or ribbon mics? [/quote]

Nice idea Martin!

Surely we can just use a single pole / triple throw switch to cut phantom an short both capacitors with one switch? That would be simple...

Ooops, yes, no need for a relais anyway...a simple switch will do the trick for sure!
(I don't think that someone needs a pop-free switching with this as your mic never turns out to change its architecture suddenly:grin: ...but maybe the length of the hookup-wire in the signal path can be reduced to clean up everything as this is an idea for very clean pre's at first...)

Kind regards

Martin