Close to Studio Quality Audio on Samsung Android devices

[AdamOutler]

So I've been working out how to get studio quality audio on Android with a powered mic for a while now and I finally have it. I'm using XLR to Android with a few adapters, some custom, some purchased. I wanted to share some pictures of how this setup works.

You can check out the previous threadhere, where I showed a resistive method. Today, we're going to use a capacitive method. The hookups are basically the same, however, this method will lead to higher ouptut.

It all starts with a great mic. You will probly already have a mic to use if you're looking at this guide. I can recommend two if you are in the market. The Blue Yeti seems to be the best quality microphone I could find, This microphone can work on USB or XLR, which means, with a proper cable, this single mic will work on the desktop as well as an Android device. The other mic, pictured below is an Audio Technica ATR-6550 http://www.amazon.com/Audio-Technica-ATR-6550-Condenser-Shotgun-Microphone/dp/B002GYPS3M . While these are both awesome mics, they serve very different purposes. The Blue Yeti is renowned for its high quality audio reproduction and the ATR-6550 is extremely directional which allows for very focused quality in a noisy environment. Both of these mics will see use from me in various conditions.








To build the Samsung mic to 1/4" adapter, you will need:
A sacrificial Samsung/Apple headphone set with built-in mic. As long as mic works on your device, this should work as well.
Ceramic Capcitor 0.10 uF with marking 104 (or similar small capcitor) Mouser.com part number: 81-RDER71E104K0K103B
A Female 1/4" inline audio jack Radio Shack Model Number 274-141

For XLR Mic:
An A3F XLR Jack to 1/4" Plug adapter/Transformer Radio Shack Model Number 274-016
A 1/4" plug-jack extension cable (make it yourself or source it from a local audio shop, Radio Shack has 25-footers, but 6-feet is more than enough for most circumstances)

For Dual XLR Mic like Yeti:
2 - A3F XLR Jack to 1/4" Plug adapter/Transformer Radio Shack Model Number 274-016
A 1/4" 2-female to 1-male Y adapter Radio Shack Model Number 42-2568
A 1/4" plug-jack extension cable (make it yourself or source it from a local audio shop, Radio Shack has 25-footers, but 2-feet is more than enough for me)




For non-XLR mics
you will just locate an appropriate connector to get mono output. For example if your microphone is a stereo mic with a 1/8" plug, you will need to get something like this

For 5V powered mics like Yeti
You can use a USB OTG adapter for many devices. For those devices which do not provide USB power, you can buy a USB OTG with power, or use this cool hack by redoano


Microphone power
Many modern mics accept 5V USB power. To allow flexibility of use between usb wall-power, desktop/laptop use and USB-OTG power from your targeted device, you can use a zip-tie to strap the USB-OTG adapter 6-8 inches from the end of the USB cable. This allows you to use the USB cable in normal mode, or apply the adapter at any time to power the mic from the device being used. This trick ensures that you never forget the stupid adapter, and the 6-8" placement from the end ensures the cable isn't bent too extraneously.







Creating the Samsung/Apple standard 4-pole adapter
This adapter will work on many other devices, however it's only a known standard on Samsung and Apple devices. I've included a pinout below, but you can check out this thread for more information.

microphone in action
Here you can find a video and a few images of the Blue Yeti which Google+ was nice enough to turn into an animated GIF for me.

conclusion and results
Due to the range of DACs(Digital to Analog Converters) in various devices, your results will vary. I find that the Galaxy Camera's headphone port provides a bit more on the high-end than the low so post-processing to remove hiss and improve bass is required. The same result has been reproduced on all four of my microphones. The Galaxy S3/4 seems a bit more level but experiences the same high-pass effect which is also correctable in post-processing.

This is a great mod if you're looking to move the microphone closer to the source or you're looking to put some hardware to use. I also hope this encourages more people to make those Android videos I love, using their Android devices. Everyone has one, now get a mic and make some cool videos.

 

[Renate NST]

When I tried a 0.1 µF capacitor, it was just because one was sitting on my desk.
I didn't put a lot of thought into the value.

The combination of series capacitance and effective input resistance makes a high pass filter.
It's hard to estimate the input resistance of a cell phone without measuring it dynamically.
Since the bias circuit can supply a milliamp or two at 2.5 V, let's go with a worst case of 1 k input resistance.

Let's find the corner frequency of our high pass filter:

F = 1 / (2 * pi * R * C)

0.1 µF and 1 k gives us a 3 dB frequency of 1592 Hz :crying:
That's why there is no bass.

I tried with a 470 µF capacitor (also just randomly on my desk)
The signal level and the bass were much better than with the 0.1 µF
That's not surprising, since the 3 dB frequency is now 0.3 Hz!

Ok, 470 µF is a bit overkill.
How about 10 µF and a 3 dB frequency of 16 Hz
A tantalum capacitior is small and will work fine.
Just make sure to have the positive end going to the cell phone input, the negative end going to the mic.

 

[AdamOutler]

How about 10 µF and a 3 dB frequency of 16 Hz

Where are you getting R? Also, this circuit should be calcualted where Frms is run against an LC-pi circuit where L is of unknown value. Note the transformer on this image.

I suppose the calculation shoud be roughly the same with resistance as inductance, but even still, the DAC is not the greatest in the device headset ports and the stock headsets suffer the same problems.

I will try a higher F cap though to see if the quality can improve. This was the smallestcap I had available at the time though which could fit into the housing.

 

[Renate NST]

The worst case figure for R of 1 k is based on one scenario for generating the bias current to the electret mic FET.
In very old audio systems the supply voltage would be 6 to 12 volts with a fairly high resistor value.
In most middle-of-the-road PC sound cards, there is a filtered 2.5 volt source fed through a resistor of 1 or 2 k.
In most cell phones, the bias supply is controlled by whether you are in talking/recording mode or idle.

Some cell phones could have a constant current supply which is controlled to keep the average value of the input at the bias point of 2.5 volts.
In this case, the effective resistance would be a function of how much bias current you are drawing.
With a capacitively coupled input, you would be drawing no bias current and the input could have a very high effective resistance.

I guess I'll have to make a few more measurements.

 

[AdamOutler]

@Renate NST I did some experimentation. Theres a definate difference between the .1uF to 50uF, but 50uF to 100uF did not produce much of a difference. I think 100uF is sufficient. What say you?

Here is a video which starts with .1uF, moves to stock galaxy camera, 50uF, and finally 100uF.

These parts were collected from spare circuit boards and low voltage power supplies.

 

[Renate NST]

I think it sounds better with the larger caps.
A 10 µF should really be plenty though.

The difference between 0.1 µF and 10 µF is a factor of 100
Going bigger will not make a big difference.

I think that your cell phone must have a pretty high input resistance already.
It didn't sound horrible with the 0.1 µF

 

[1512tootsie]

USB microphone /will it work with USB input and 3.5 mm output cable?

Hi I am interested in using the blue yeti to screen record on a rootedins 5.This tutorial was on the yeti xlr but the original has a USB output.If i get a 3.5mm Male AUX Audio Plug Jack to USB 2.0 Female Adapter Cable will it work with screen recording?

 

[Renate NST]

If i get a 3.5mm Male AUX Audio Plug Jack to USB 2.0 Female Adapter Cable...

If I understand you correctly, I know of no such adapter, it would be a giant step backwards.

If you have a USB mic what you want is a device that has USB host mode (or OTG).
You'll need the correct driver for the USB part and also the driver for the USB audio device.

I don't have a USB integrated mic, but I do have a Lexicon Alpha adapter.
It takes a regular mic as input and presents as a USB peripheral.
My Nook has the drivers in the kernel.

 

[1512tootsie]

Here's the link

If I understand you correctly, I know of no such adapter, it would be a giant step backwards.

If you have a USB mic what you want is a device that has USB host mode (or OTG).
You'll need the correct driver for the USB part and also the driver for the USB audio device.

I don't have a USB integrated mic, but I do have a Lexicon Alpha adapter.
It takes a regular mic as input and presents as a USB peripheral.
My Nook has the drivers in the kernel.

Cable :3.5mm Male AUX Audio Plug Jack to USB 2.0 Female Adapter Cable by VIMVIP

The cable has a USB input and a 3.5 mm output

I have a nexus 5 that's rooted.I also have a otg cable that works with a mouse USB flash drive and keyboard.Is there any pro condenser microphone that works well with android?

 

[Renate NST]

I looked at that adapter and as far as I can tell from the non-English description it's a purely mechanical adapter for car stereos that use a 4 pole 3.5 mm jack on the front panel as a USB connection.

 

[nelsonw]

Is any way to re-route via software the Mic to right pins? SImpler: That the phone "ears" the right output?

 

[RandomClear]

Perhaps, I'm blind, but I do not see any schematics on how/where connect the capcitor...

 

[chris10241]

Can I just use a normal 1/4" female to 1/8" male adapter ?

 

[rin67630]

Hi Adam, Hi Renate,
you will probably get a better adapter from an TRRS video cable. These have better shielded wires than the headphones ones.

I need imperatively an audio input that has no level compression.
Did you care on that? don't more or less all phone mic circuitry compress the signal anyway?

 

[Pankz Jyoti]

How is it possible?

 

[d-signer]

Thank you for this! It was just what I was looking for!
I'm posting my experience doing this as it may help others who are also trying to do the same (and not know much about electronics, like me ).

I have an condenser microphone that has a USB plug (for power input) and 3.5mm Tip-Ring-Sleeve (TRS) jack (for signal output) on the end of the cable. When I plug the microphone into the computer to record and power the microphone, (or use a wall socket (via a 5v phone charging adapter) for power) there is noticeable hum in the recording. So I wanted to record and power the mic using my Samsung Galaxy Note 4 android smartphone because 1) powering it by the phone would remove A/C power hum, and 2) the recording process is much easier, as I use the phone to record video too. (Note: you can record by plugging the mic into a laptop running on battery to remove the hum, but using my phone was what I really wanted to record with as it is so much easier)

I opened up an old CD ROM drive and took out the audio headphone socket. I salvaged a few capacitors from an old circuit board lying around, and cut the 3.5mm jack plug off some broken headphones.
I then wired up the circuit as best I could from following 1) this post, 2) the two posts referred to in this post:a) https://forum.xda-developers.com/showthread.php?t=2072612 and b) http://www.doityourselfgadgets.com/2014/09/how-to-connect-external-microphone-to.html
I did some testing on a breadboard (I tried adding a 1000 Ohm resistor, and moved the capacitor around the circuit, I used a 0.1 micro Farad capacitor, then changed to a 50micro Farad capacitor and noticed a benefit to the recording volume.)
What I called "testing" someone with electronics knowledge should read as "rolling dice" ...but this picture shows what I ended up with that worked for me (with the finished item after soldering and applying some heatshrink tubing to cover and insulate the adapter).