Is that what the capacitors (C7, C8) are doing in the phantom supply? Working as coupling caps? I'm learning as I go along here.

So, all I need to do is place a capacitor in series with the output ground, to eliminate its 11v offset?

(11v)ground-----capacitor-----ground connector of line-in device
preamp output-----------------signal connector of line-in device

 

So I'd just have to not connect the phantom power circuit's ground to the preamp's ground? Doing so would be linking the 11v(split) to 0v, so I'm guessing so? I'm used to using unbalanced pres, where the signal wouldn't be received unless ground is connected too, so it seems like an odd concept.

You're not connecting the phantom power circuit's ground to the ground used by the preamp but you are connecting it to a different voltage point in the preamp's circuit. The two circuits are connect together but just at a different voltage point. This has nothing to do with balanced or unbalanced, you could do the same thing with this same preamp circuit and a standard unbalanced, unpowered mic. The key is the blocking capacitor before Q1 (and Q3) in the preamp. If it's not there (and don't get me wrong it has to be there if you're using a powered mic) your mic must use the same ground in its signal so the audio signal will revolve around that midpoint, if it is there, you're free to connect the mic ground to a different voltage point such as the low point of the battery (in which case the signal will revolve around a different midpoint but the capacitor will eliminate the difference before it gets to the preamp).

 

Is that what the capacitors (C7, C8) are doing in the phantom supply? Working as coupling caps? I'm learning as I go along here.

So, all I need to do is place a capacitor in series with the output ground, to eliminate its 11v offset?

(11v)ground-----capacitor-----ground connector of line-in device
preamp output-----------------signal connector of line-in device

Almost, connect it like this:
preamp output-----capacitor-----signal connector of line-in device
-11v(22v battery -)-----------------ground connector of line-in device

 

Here is a preamp I was planning to use some time ago: http://www.mendelt.nl/greenpre_V14/

 

Thanks tubeguy, and patricktoday, I really appreciate your help.

You're not connecting the phantom power circuit's ground to the ground used by the preamp but you are connecting it to a different voltage point in the preamp's circuit. The two circuits are connect together but just at a different voltage point. This has nothing to do with balanced or unbalanced, you could do the same thing with this same preamp circuit and a standard unbalanced, unpowered mic. The key is the blocking capacitor before Q1 (and Q3) in the preamp. If it's not there (and don't get me wrong it has to be there if you're using a powered mic) your mic must use the same ground in its signal so the audio signal will revolve around that midpoint, if it is there, you're free to connect the mic ground to a different voltage point such as the low point of the battery (in which case the signal will revolve around a different midpoint but the capacitor will eliminate the difference before it gets to the preamp).

I see what you're saying, thanks a lot for the explanation!

One last thing. If 0v from the battery (-) is being used as the ground for the phantom supply, and it's connected to the preamp's virtual ground (11v split), won't that mean that ve- (also 0v on the battery) is being shorted with 11v split?

I've compiled the circuits into a diagram so that this can be easily seen:

full circuit (also added the output coupling cap that tubeguy suggested)

 

The two "GND" nodes in the red oval should not be connected together. All three "GND" nodes on the mic circuit should be connected to 0V (the battery's (-) terminal) and do not connect to the preamp circuit.

 

In case I wasn't clear in my previous post the Gnd connection for the preamp output jack is connected to the battery 0v or bottom, not to the preamp virtual Gnd.. You have the battery connections drawn correctly.

The battery negative, or 0 volt point is the Gnd for the phantom supply circuit, and the outside world.
The Gnd's on the op-amp circuit are virtual ground's internal to the preamp circuit. You could re-label those VG

 

Super advice, thanks both of you!

So, if I'm understanding correctly (at last )..:

updated full circuit


The phantom power and preamp grounds are no longer connected, and the output ground is now on the 0v battery point. Look good to you chaps?

 

This would pull the inputs down through the input diodes. Put another 22uf np capacitor between the bottom of the input diodes and the ground part of the phantom circuit on the left from them.

 

Except for the rail splitter circuit. That's awfully primitive. The ground path has to travel through a 1k resistor on either side to get back to the battery. I'm skeptical about that part, especially compared to the "recommended" power supply he mentions in the text. The rest looks correct

 

Like this, kubeek?

Patricktoday, do you reckon the primitive rail splitter will cause any issues? I suppose it's time to build it now, and I can find out... need to plan my strip board now. Thanks a load for the help!

 

I reckon it will affect the sound quality but I'm not sure. Perhaps you can put the power supply on a separate piece of stripboard for later improvement.

 

Yes, like that. As for the rail splitter, you would be better off using a tl072 and use the second half to make the virtual ground, you connect +input to the center of the R1-R2, -input to the output, and the output will be the virtual ground for the rest of the circuit.

 

I reckon it will affect the sound quality but I'm not sure. Perhaps you can put the power supply on a separate piece of stripboard for later improvement.

That's a good idea. Rod Elliott himself actually suggested that I used that particular splitter, so I expect it must be reasonably suitable. I'll have build the whole thing and find out! Strip board time... I will update the thread once I'm finished, and hopefully have some good things to report.

Yes, like that. As for the rail splitter, you would be better off using a tl072 and use the second half to make the virtual ground, you connect +input to the center of the R1-R2, -input to the output, and the output will be the virtual ground for the rest of the circuit.

Okay, I'll try that if the cave-man-version doesn't work very well. Cheers.

 

The updated circuit in post 28 is on the right track.
The circuit in post 31 - not so much.

Why? Because it removes the microphone ground reference for the phantom power to the microphone.

You are using batteries, so why not stack 2 11-12 volt batteries and get your virtual ground from the mid point ?

The batteries are the rail splitter, and very good at it.

 

Why? Because it removes the microphone ground reference for the phantom power to the microphone.

I fail to see your point, the connection in post 31 puts the phantom voltage right to the microphone, without connecting it to the amp. Now that I look at it again, the diodes should be connected to the virtual ground and the newly added cap could be left out, but that doesn´t change anything in the mic phantom supply.

 

I fail to see your point, the connection in post 31 puts the phantom voltage right to the microphone, without connecting it to the amp. Now that I look at it again, the diodes should be connected to the virtual ground and the newly added cap could be left out, but that doesn´t change anything in the mic phantom supply.

Yes, Agreed ! I apologize, I did not look carefully at where the battery negative was connected.

 

Now that I look at it again, the diodes should be connected to the virtual ground and the newly added cap could be left out, but that doesn´t change anything in the mic phantom supply.

Haha, okay, another revision. Hopefully the last one: might-be-final full circuit.

As I understand it, the diodes are there to prevent any voltage spikes, so it makes sense that they should be able to short out with something (would the cap have isolated them?).

You are using batteries, so why not stack 2 11-12 volt batteries and get your virtual ground from the mid point ?

The batteries are the rail splitter, and very good at it.

Ideally I'm trying to go with using just one battery. If I were to use two 12v batteries, would the load be equally spread between the two so that they'd be drained at the same rate? I've not ordered my battery yet, but I'm planning on using a 22v lipo, and it has balancing connectors. It might be possible to use one of the balance connectors as the mid point, but I'm not 100% sure how they work so I don't know if it would be suitable yet. I'll look into it!

Edit: It looks like the balance wires of a lipo are indeed just directly wired to the cells, meaning that I can use them to power the preamp. No splitter required. I've thus made another edit to the circuit: full circuit, another update.

 

That looks right to me, and good news on the batteries. You really ought to build this thing on a breadboard first or at least the bulk of it.

 

That looks right to me, and good news on the batteries. You really ought to build this thing on a breadboard first or at least the bulk of it.

Circuit look's good to me also.
Being able to get the VG directly from the batteries is great because it provides a much more solid VG than a resistor splitter, and doesn't waste current.