Decoupling Caps question

[jammybstard]

are the 10u Electrolytics used to decouple overkill?
Assuming an output z of 1M from the splitterBlend and an Imput Z of 1M for the next device. total Z would be about 500K.
F = 1/(2*Pi*R*C) = 1/(2*Pi*500k*10u) = 1/31 = 0.03 HZ

I'd rather use some of the polyprop caps I have;
I have 4u2, 2u2, 560n, 470n, 330n and on

How do I work out how low I can safely go?

[Hides-His-Eyes]

Any of those would be fine, honestly.

[merlinb]

How do I work out how low I can safely go?

It's simply a matter of noise rejection. If you want maximum immunity to mains hum then the output impedance must remain low down to 50Hz. If you want the output impedance to be no more than 1k at 50Hz, say, then you need at least:
1/(2*pi *1000 * 50) = 3uF

I prefer using 10uF to keep output Z negligible, and because of their low cost and small footprint.

(FYI, these are coupling caps, not decoupling caps  )

[jplebre]

(FYI, these are coupling caps, not decoupling caps)

so do they still fill in the same purpose? is it just a chance due to the place they take in the circuit?

[merlinb]

(FYI, these are coupling caps, not decoupling caps)

so do they still fill in the same purpose? is it just a chance due to the place they take in the circuit?

Decoupling caps are used to isolate one part of a circuit from another (e.g. prevent one circuit from influencing another in some way). Coupling caps link circuits together. It's just nomenclature.

[jplebre]

ty

[familyortiz]

As posted, coupling caps couple the signal from one section to another, in this case you are ac coupling the signal and blocking any dc from going to the next section. The high value is good to allow more signal to pass and I generally settle for values of 1 - 10uF.
Decoupling is a term usually used for the decoupling caps placed on the power supply or right on the IC's supply pins to decouple noise (filter it to ground) from the supply, generally on the order of 0.1 - 0.01uF.

[jammybstard]

(FYI, these are coupling caps, not decoupling caps  )

My bad!

I've never thought of it in that way before; and I don't quite understand. I'll have to have a think!
I always thought the value of a coupling cap was decided based on it forming an RC circuit with impedance that followed and if the C was too small you'd loose low frequencies. I've not come accross noise rejection before.

[jplebre]

based on it forming an RC circuit with impedance that followed

if I understood it correctly, it's for both reasons?

[merlinb]

based on it forming an RC circuit with impedance that followed

if I understood it correctly, it's for both reasons?

Both both reasons come into play. However, control of bass frequencies is usually done at the input to an effect, or 'inside' the effect somewhere. The output coupling cap is not usually expected to apply any bass cut in modern designs. This was less important in antiquated transistor designs from the 60s, when all the hiss, crackle and hum from the whole rig would have swamped any minor improvement elsewhere. Nowadays we can demand much higher performance.