| Dominique |
Dear diy coommunity :),
to help me choose an output coupling cap for my diy mic preamp, I'd like to ask in what range devices' line input impedances lie.
I know of a few devices which are around 10k, but that's not much to know :)
The background:
Mostly my mic preamp will be plugged into mobile recording devices, but I'd like to be able to plug it almost anywhere (but "line-in") without fearing a bass loss.
Until now, I used a 47uF aluminium electrolytic, but when my design changed, the offset voltage before the output coupling cap became lower and with a strong mic signal can switch over from negative to positive voltage!
So now I wonder whether to use two electrolytics back to back or if I go for those Wima MKS2 polyester caps which pack up to 10uF into a pretty small case (but which are rather expensive).
Cheers
Dominique |
|
|
| richie00boy |
The offset voltage provides the polarising voltage for the cap so do not worry that the signal makes it seem to change polarity.
To work out how much the bass will be affected:
f = 1 / (2 * pi * c * r)
where
f is in hertz
c is in farads
r is in ohms
Anything below 10Hz should be fine. |
|
|
| Dominique |
| quote: | Originally posted by richie00boy
The offset voltage provides the polarising voltage for the cap so do not worry that the signal makes it seem to change polarity.
|
Are you sure? Regarding my case, I recently read that during the part of the wave where the polarity of the polar electrolytic is reversed, the cap behaves diferent (than charged with correct polarity) and stronger distortions will very likely appear.
But that's just what I read - maybe if just a quarter of the wave only is charging the elco with the wrong polarity, it neither harms the elco nor disturbs the signal...?
Thanks for the formula too! Well, but the problem wasn't not being able to calculate the corner frequency, but to know how low "line" input impedances of audio gear can go. ( I am in lack of the "r" in the formula!)
Cheers
Dominique |
|
|
| richie00boy |
The DC will bias the cap and will remain there even when sinewaves are passed -- they ride on the bias.
Of course, an electro will give worse distortion performance than other cap types, but if you need that amount of capacitance you don't have much choice.
Ideally input impedance should be low (<10k) so as to minimise noise, but this can present problems when driven by esoteric equipment with high output impedance. 10k is seen as a reasonable standard in semi-pro audio, 100k would be a better load for esoteric kit, but go any higher and it gets noisy. |
|
|
| purplepeople |
It is fairly safe to assume that your output is driving 47K regardless of the actual measured impedance. A lot of pro gear I've seen works with 47K as a "nominal" input impedance although the "true" spec varies. I don't know why it is 47K and not some round number like 50.
Output impedances also vary widely. Although 10K is typical, I've seen it as low as 600R. This huge difference is probably just so that outputs can be Y-split a few times without too much problem. It's the not best for fidelity, but it does allow for a lot of flexibility when setting up in weird environments.
:)ensen. |
|
|
| Dominique |
ok, thanks for the info :)
I searched a bit on the web, but often, the input impedances of devices are hidden :confused:
Some devices are near to 10k, but I didn't find any below!
So it's not that safe to assume 47k!
I'll start a new thread because the title doesn't describe my further concerns well...
Best wishes!
Dominique |
|
|
|
