First Project: Stereo Line-To-Mic Adaptor

[Helmanfrow]

I'd like to interface my Mackie 802VLZ4 mixer's line inputs and outputs with my phone and my ThinkPad by way of their TRRS jacks.

Doing this with TRRS-to-dual-TRS adapter cables has some problems:

• Horrific electrical crackling and chattering noise at the mic input when feeding the mic input from the aux send or the main outputs. My guess is that the mic bias voltage is interfering with the signal.
• Distortion. The signal level at the aux/main outputs of the mixer is far too hot for the mic input and requires running the output fader/knob at the lowest end of its range, which is less than ideal.
• Mono/Stereo. The mic input on a TRRS connector is in mono, which is fine (the first two problems notwithstanding) if I want to feed it from a mono aux, but problematic if I want to use the stereo bus, which will have to be summed to mono first.

For the record I've tried the adapter cable only with the ThinkPad and not my phone, yet.

I decided to build a little adapter box to solve the three problems.
I took a little inspiration from some videos on YouTube

and came up with this:

https://docs.google.com/drawings/d/1ovtH-qWSbTdVd8ZfaA4Itoj1b5tdtmYiYpHW9GAHVBk/edit?usp=sharing

Mind you, while I have decades of experience making my own audio cables and occasionally troubleshooting and fixing PCBs, and while I have a pretty good basic understanding of electricity and electrical components... I have never before built an electronic circuit from scratch, so this is my first project... ever. I'm in my mid 40's. 

Therefore while I have a basic grasp of what the above circuit does, I have no idea why those specific component values were chosen, and whether or not they'll work in my application. I've got a resistor kit and some capacitors coming in so I'll be able to tweak the circuit.

What I'm hoping to find at this forum is some guidance, and advice should the device not operate as expected at first go.

I've already received the project box, connectors and other components from AliExpress and have started assembly. Image attached.

Thanks for reading.

[Helmanfrow]

The schematic leaves out a switch for quickly selecting between CTIA and OMTP pinouts (swapping Sleeve and Ring 2).

[Helmanfrow]

I built the circuit as it appears in my diagram, with the following exceptions:
R3 - 10KΩ
R4 - 270Ω
I'm still getting the same noise on the mic input when the cable is plugged into my ThinkPad's TRRS port. In fact, it sounds exactly as it did with just the adapter sinking a line level load with no circuit at all.
I've attached a sample recording of the mic input. First with the cable plugged in and then unplugged.

The ThinkPad's sound card is putting 3.85 volts on the mic pin.

The line level outputs to the mixer work fine.

There's a hole in my life right now and it's shaped like a breadboard. (Although the wire used in these resistors I bought is so thin and spindly I doubt it could be pressed into a breadboard without buckling)

[Renate]

... selecting between CTIA and OMTP pinouts ...

I'm not familiar with those terms at all.
The schematic with the mic going to sleeve is Android style.

Doesn't your ThinkPad have regular line inputs?
You'd be a lot better off that way.

I'd use a regular USB interface plugged into your phone through an OTG adapter.

[Helmanfrow]

I'm not familiar with those terms at all.
The schematic with the mic going to sleeve is Android style.

It's CTIA, the pinout which is used by basically everyone these days.

Doesn't your ThinkPad have regular line inputs?

Nope.

You'd be a lot better off that way.

Maybe, but then I wouldn't have an excuse to order capacitors on AliExpress.

I'd use a regular USB interface plugged into your phone through an OTG adapter.

I'm an audio professional (well, more like a serious hobbyist these days).
I have USB audio interfaces. That's entirely beside the point.

My goal—however silly or impractical it may seem—is to have the mic input on the TRRS jack of my ThinkPad P50 accept a line level signal from my mixer without distortion or too much noise.
My suspicion is that this is possible, and if I discover that it is not I will grudgingly put away my Chinese resistor pack and buy yet another audio interface.

[themadhippy]

My goal—however silly or impractical it may seem—is to have the mic input on the TRRS jack of my ThinkPad P50 accept a line level signal from my mixer without distortion or too much noise.

Use  a di box

[Renate]

Your box is being fed from the unbalanced outputs of the Mackie?

For some reason I got it in my head that you were connecting some portable thing as source.
Check the noise level with the Mackie on, off and unplugged from the wall.

Do you have an audio transformers hanging around?
Can you add one to the box between the summing point and the cap?

[Helmanfrow]

Use  a di box

Hold on— am I in the right place? I thought this was a forum for electronics hobbyists. 

[Helmanfrow]

Your box is being fed from the unbalanced outputs of the Mackie?

Yes, from the RCA outputs.
I'm going to make a diagram.

For some reason I got it in my head that you were connecting some portable thing as source.
Check the noise level with the Mackie on, off and unplugged from the wall.

With nothing plugged into the box it sounds like this:  untitled2.mp3 (161.63 kB - downloaded 76 times.).
With the Mackie plugged in (using a generic 6-foot dual RCA-dual RCA cable) it sounds like the first clip I posted. The noise is the same whether the mixer is on or off, and whether or not the mixer is plugged into its power supply.

Do you have an audio transformers hanging around?
Can you add one to the box between the summing point and the cap?

Mmm... I'll check to see if I have any old modems or DSL filters in a box but I doubt it. A couple of years back I gave away all my loose electronics and PC parts so there's not much hanging around anymore.

[Helmanfrow]

Your box is being fed from the unbalanced outputs of the Mackie?

And also from the aux send, which is TRS balanced, but which seems to have no problem feeding unbalanced loads.

[Helmanfrow]

Perhaps I'm barking up the wrong tree, here, but my sense is that the chatty sputtering noise on the line is an effect of the 3.8VDC bias voltage which appears on the mic connector. The capacitor's job in this circuit is to keep that DC voltage out of the incoming signal.
I understand how 48v phantom power works but that requires a balanced connection. I don't understand how a single, unbalanced line can carry a bias voltage going one way and an audio signal going the other way.

[Renate]

Well, a lot of things could be worming their way in.

Some things to try (individually):
Remove that R5 sucking down the bias. (No reason to use the bias?)
Short the summing point to your ground. (As a test if the noise is forcing in on the ground.)
Put a small 0.001 µF cap from the summing point to ground. (RF coming in?)

If you had a transformer the best would be feeding the balanced signal in one side and taking the unbalanced off the other side.

Did you check on Windows if you have 20 dB added on the input in the control panel?

[Helmanfrow]

Remove that R5 sucking down the bias. (No reason to use the bias?)

Yeah, I pulled that out and it made a bit of difference but not much.
I noticed that some component of the noise is EMI/RFI because it changes as move the device or wave my hand around it.

Short the summing point to your ground. (As a test if the noise is forcing in on the ground.)

Ah, okay, I'll test that out, but I imagine that shorting my signal to ground is, in general, not recommended.

Put a small 0.001 µF cap from the summing point to ground. (RF coming in?)

I have a capacitor kit or two coming in on a slow boat from China. Could take a couple of weeks. I guess I can hit up my local hobby shop and pick up a small variety pack to tide me over. They sell the same crap as you get on AliExpress except you pay four times more for the convenience of getting it today. ¯\_(ツ)_/¯

If you had a transformer the best would be feeding the balanced signal in one side and taking the unbalanced off the other side.

Oh, interesting, so I could feed the balanced mono aux into its own input and unbalance it with a transformer. I like that.

Did you check on Windows if you have 20 dB added on the input in the control panel?

Yup. All digital levels are nominal. Nothing added.

[gf]

The
Remove that R5 sucking down the bias. (No reason to use the bias?)

The Audio Accessory Detection IC in the tablet/smartphone may need some DC current flow, though, to detect the presence of a microphone.

[Renate]

The Audio Accessory Detection IC in the tablet/smartphone may need some DC current flow, though, to detect the presence of a microphone.

Yeah, but my favorite is sending 3.3V down the mic line on a Nexus 7.
It switches the headphone jack into a 1.8V logic level root console. 
(This may work on other random Google Android devices.)

[Helmanfrow]

The
Remove that R5 sucking down the bias. (No reason to use the bias?)

The Audio Accessory Detection IC in the tablet/smartphone may need some DC current flow, though, to detect the presence of a microphone.

Windows 10—or at least the Realtek driver—knows when I've plugged a TRRS cable into the jack. It chimes, notifies me that something's been plugged in and adds "External mic" to the list of available input devices.

[Helmanfrow]

Yeah, but my favorite is sending 3.3V down the mic line on a Nexus 7.
It switches the headphone jack into a 1.8V logic level root console. 
(This may work on other random Google Android devices.)

Ha! That's awesome. Who figures this sh!t out?

[Helmanfrow]

Incidentally, whereas the P50 puts out 3.85 or so VDC on the first Ring, the Pixel 4a puts out under a volt, and it's on the sleeve. I thought most headset mics required between 3-10vdc bias.

[Helmanfrow]

I have abandoned neither this thread nor this project.
I am simply awaiting a shipment of components from AliExpress.
I ordered several capacitor kits (ceramic, Mylar film & electrolytic), as well as some breadboards and other bits and bobs.
That was on April 11 so stuff should start arriving within the next week, given my experience with AliExpress.

[Helmanfrow]

Okay, I'm back with some new wind in my sails. Maybe.

Progress report:
Received lots of cheap components from AliExpress.
Discovered commercial product that does exactly what I want but costs way too much.
It's called the ISOPODci by Whirlwind. Image attached.
It uses transformers but it seems that this is for balancing the signal to and from the mixer. There's no indication as to how they're blocking DC from the mic input.
I've written Whirlwind to ask if they'd share that information with me. We'll see how that goes.

[Helmanfrow]

Well, I never heard back from Whirlwind.
Short of a lot of trial and error, how would I determine the specs for the transformers I would need to recreate the circuit in the ISOPODCI?

[floobydust]

I wouldn't make the jump to using transformers just yet. Did you ever draw up a sketch or schematic? Using the console's unbalanced line in/aux sends?

I'd add a blocking capacitor for the mic, and likely you'll need a filter network if the headphone out is Class D which the loudspeakers are.
Headphone outs are supposed to be Class G according to the Thinkpad Conexant info out there. There is also the resistor value to spoof a headset being connected.
If the Thinkpad is mains powered, there's another source of noise.

edit: that youtube video circuit is painful, I would say you need to attenuate (from line to mic) at least 35dB

[Buriedcode]

Whilst I don't want to complicate things asking for "more", I'll second what floobydust asked - a diagram or a sketch.  Could be in sketchup, MSpaint google docs drawing aynthing, just to show what plugs into what at first.  You can then label the connections and add resistors/caps/switches later on.  Dealing with simple TRS is generally easy, but the second ring can make things difficult to visualise everything all the switching posibilities.

I've no doubt what you need can be built with just passives and probably quite simply, but formalising the requirements can help you break the problemn down into straightforward steps.  Even something as reasonable as a "passive adapter" can start to get quite complicated as more and more specifications/requirements are added later on.  Even just listing the specs of the audio source and console input will help.

A transformer is AC coupled so will eliminate the DC bias whilst also providing isolation to prevent ground loops - which can be handy if you're dealing with multiple units with earthing but night not be necessary.  It *sounds* like you just need a passive attenuator, with coupling caps, with the "ish" bit being what you do with the extra ring on the trrs connector, as well as calculating the resistor values - because there really isn't much of a standard in terms of phone/tablet audio out level.

[Helmanfrow]

Whilst I don't want to complicate things asking for "more", I'll second what floobydust asked - a diagram or a sketch.  Could be in sketchup, MSpaint google docs drawing aynthing, just to show what plugs into what at first.  You can then label the connections and add resistors/caps/switches later on.  Dealing with simple TRS is generally easy, but the second ring can make things difficult to visualise everything all the switching posibilities.

Yes! As a matter of fact, the very first thing I did before any of this started was to make a drawing (In Google Drawings 😂)!

I can't believe I neglected to share it. Attached!

This is the schematic I used when wiring up my current prototype.

[Helmanfrow]

Here's the Google Drawings file:
https://docs.google.com/drawings/d/1Z8dJF-GQFFmn9XFq55FDHRczhozJ_WphQiuHTWk5OVQ/edit?usp=sharing