Unbalanced "phantom" power

Re: "If there is DC on the output of C1, the cap on the input to the next stage will block it whether R3 exists or not . If the next stage does not have an input cap, any DC on the output of C1 will still appear on the input of the next stage whether R3 exists or not. What am I missing? Or is it there simply to bias the electrolytic?"

You are missing that when plugging or un-pluging your input connection from your Mic circuit; there will be a leakage voltage appearing on your Capacitor C1 which may well exceed your Soundcard maximum input as it is not conducted to ground by the Resistor R3.
When you connect with a charge or leakage present you are applying a voltage pulse which can also exceed your input in terms of it's conduction via it's input capacitor! PC Soundcard often runs on 5V DC and your circuit is approaching 30V DC.

Hope this now clears it up - might be explained better by others though!

As to your CCS, have you investigated the noise from the IC you are looking at?
You may well find for the low current you are seeking that a discrete CCS circuit is more suitable

Mik

audiomik said:

Re:You are missing that when plugging or un-pluging your input connection from your Mic circuit; there will be a leakage voltage appearing on your Capacitor C1 which may well exceed your Soundcard maximum input as it is not conducted to ground by the Resistor R3.

Click to expand...

Yes!

Man am I happy you brought that up. R3 will not help you in that event. That is why in the original circuit Elliott used a zener voltage snubber at the output. I forgot that! And I forgot the diode to prevent kickback through the CCS. That is probably not much of a problem with batteries, but I put one in anyway.

audiomik said:

As to your CCS, have you investigated the noise from the IC you are looking at?
You may well find for the low current you are seeking that a discrete CCS circuit is more suitable

Click to expand...

I checked the data sheet for the LT1086. They do not address CCS use. But they do say that as a VR the output noise is a function of the output voltage. The spec is that noise is .003% of the output voltage. Since a CCS is a variable voltage regulator that varies voltage to maintain a given current, I suppose that spec will apply to the CCS as well. I have breadboarded the circuit and find that the output voltage of the CSS is about 10V when set up as Elliott describes in his article. If the VR spec applies to CCS use, then the CCS noise voltage would be .0003 V RMS. Unfortunately, I do not know if that is a lot or a little or if a discrete CSS produces more or less noise. Anyone here know?

Of course I could build it exactly as Elliott suggests but use batteries as the supply. However I was hoping for a really simple battery version for the intermediate DIYer who might be scared off by the original circuit.

BTW, the corrected battery circuit below uses two 9V batteries for 18V. I breadboarded the LT1086 CCS with 18V and loaded it with 1K as suggested by Elliott and I used an additional 100 ohm stopper as well. I have run it for over 5 hours and it seems stable as a brick. But noise level is a mystery for now.

Thank you for your posts.

Here is the latest version:

Besides adding noise, LT1086 might show output impedance drop at higher frequency and respective unwanted signal attenuation. I would rather go for a NJFET current source.

oneoldude

how are you progressing with this?

Will be interested to see what results you are getting given that the current source has a relatively high impedance and that the frequency response of your microphone circuit may have it's HF response affected as a result of cable capacitances

Mi