Please help fix my error!

[sharks739]

That does help!

I'll try change the resistors R3, R4 first, and then I'll venture into R1, R2. I'd have to test them out to see how much more I need, but 9dB should be a pretty good start.

At least I now know what to do to get more gain out of an op-amp! Considering I haven't ever designed anything with one, I call that a win on my part.

For the sake of common knowledge, the mic doesn't hit the preamp before this headphone box, so that first op-amp is the only time that mic signal gets amplified on that part of the circuit.

Microphone -> PM50 mic input -> PM50 mic output -> mixer mic input -> mixer aux output -> talkback box -> PM50 headphone input

Does the fact that there's no other amplification allow me to go a little higher with the resistors?

From what I've read, PLEASE correct me if I'm wrong, the op-amp will hit saturation and clipping will occur at any more than 80% of the voltage supply for the amp, which in this case is (I assume) 12V? So, I can increase the output up to 9.6V without hitting saturation, correct? I can measure the voltage by placing my multimeter leads across the VCC and VBIAS terminals, correct? Maybe it's not a full 12V then now that I realize that it's a single supply and it's floating (is that thinking right?)

I'm getting there... slowly, but I feel like I'm making progress!

[Lotus 7]

Isn't the mic connected to the main pre amp after the pass through on the Rolls?

You said: "Microphone -> PM50 mic input -> PM50 mic output -> mixer mic input -> mixer aux output -> talkback box -> PM50 headphone input"

As such (which would be the usual connection) the Rolls input impedance is in parallel with the mic pre input impedance. The mic "sees" both input impedances all the time. I wouldn't lower R1 and R2 if you're using the feed through. It doesn't matter which order they're connected. The Rolls feed-through output is not buffered, it's just the plain, old mic signal. If you lower the input impedance by reducing R1,R2, you will also have to increase the values of the input coupling caps (C1, C2) appropriately to avoid having a low frequency roll-off.

Also, since U1A and U2 are always running at a fixed gain (after U1B), they will clip before U1B does.

Because of the op amp's biased "common" reference at approximately half the power supply voltage,(which is not actually even always centered on the power supply, the peak signal out of any of the op amps in that circuit will not be over about 4 volts. That's still a lot of signal however and way more than would normally ever be fed into any low-Z headphones. If you're using high-Z (600 ohm) phones, it won't be as loud, and the full swing may be needed.

In the Rolls circuit, the op amps are actually effectively running at +/- 6 volts when a 12 v external supply is used.

[sharks739]

Gotcha...

So even though the doesn't go to the mixer first, the mic still sees all of that impedance.

What purpose do those coupling caps serve?

Off to another topic, which I'm not sure if you are familiar with, but you seem to be a genius, so I wouldn't put it past you if you do know what I'm talking about...

In the TV and stage world, they use ClearCom gear to communicate with the TV truck, the booth, etc. They often have 2 bidirectional channels of audio flowing down a single 3 Pin XLR. Not only that, but the main unit is supplying voltage. What makes these boxes capable of accepting 2 channels and supply voltage as well? Each channel has their own volume control on the box, and you can push to talk to each channel. How does that work?

Thanks!

[Lotus 7]

Originally Posted by sharks739

Gotcha...

So even though the doesn't go to the mixer first, the mic still sees all of that impedance.

What purpose do those coupling caps serve?

Because the Rolls amplifier uses a single voltage power supply, the "signal common" has to be referenced to a "bias" voltage that's floating at approximately one half of the single power supply voltage. The capacitors couple the AC audio signal from the mic inputs (which are ground referenced to the op amp inputs which are "bias-voltage referenced". Using bipolar (+ and -) power supplys (as the op amps were intended to be used with) eliminates the need for the coupling caps and all the potential issues they create (like phase shift and low frequency response drop off).

Originally Posted by sharks739

Off to another topic, which I'm not sure if you are familiar with, but you seem to be a genius, so I wouldn't put it past you if you do know what I'm talking about...

In the TV and stage world, they use ClearCom gear to communicate with the TV truck, the booth, etc. They often have 2 bidirectional channels of audio flowing down a single 3 Pin XLR. Not only that, but the main unit is supplying voltage. What makes these boxes capable of accepting 2 channels and supply voltage as well? Each channel has their own volume control on the box, and you can push to talk to each channel. How does that work?

Thanks!

I'm not at all familiar with the ClearCom system, but with the two signal conductors, the wired common conductor and the shield (4 independent conductors total) there are several possible ways of accomplishing the functions you describe. They can run independent single-ended signals between the signal conductors and the common, between the signal conductors and the shield, and an independent differential (balanced) signal between the two signal conductors. DC Power can be provided over either conductor or both, as is conventional microphone phantom power, and can be separated and isolated from the audio with coupling capacitors or transformers.

[sharks739]

Ok, so because they didn't use a bipolar supply, they had to create a different voltage source to operate the op-amp. In the box I'm creating, I don't need that because I'm using the +- 15 V supplies to give the op-amp it's full range.

From what I've read, when using coupling caps, they can, as you mentioned, cause low frequency drop off. The way to reduce that is by using a larger capacitor. What are the drawbacks of using a higher cap in that situation?

The ClearCom stuff is strange to me, because they are using standard XLR cables to pass all of that stuff!

What are the advantages of using an analog switch over a relay? In the talent boxes, I'm using 2 relays, one to open the insert tip/ring to mute the mic audio, and then a second to route the tip to the talkback output and open the tip/ring on the main out. Does it make more sense to use an analog switch in that situation? Why or why not?

Do you have any experience with Arduino?

Thanks again Lotus. I appreciate every bit of advice you've given me, and I'm so glad there are people like you in these forums!

[sharks739]

It works!!!

I've been able to breadboard two channels so far, and I didn't have any issues!

There is no cross talk at all and the trimmer for the ducking level works perfectly. Gives me just enough and I like that I have it available for each channel.

My only oddity is something I might just have to live with, and that is the signal is still getting through to the Rolls headphone amp (and coming out of the headphones at a decent level, though distorted) when the op-amp supplies are not plugged in. Is there anyway I can prevent that from happening without adding a switch to shut the signal off?

Again, thanks for your help!