DIY Microphone suggestions

Thread Starter

10_2

Joined Feb 14, 2023
5
So I recently saw a video from DYI perks and he uses a "TSB-2555B Electret Capsule" and a "THAT1510".
However, building my own pre-amp seems like a daunting task and I don't have the budget to buy a kit from somewhere like mic parts.
So if anyone has suggestions as to a simplified pre-amp or a pre-amp with more instructions that would awesome.
I also am not set on the "TSB-2555B Electret Capsule" so any suggestions as to a better capsule would also be greatly appreciated.

I have some experience with soldering and basic electronics however I am a complete beginner to microphones and audio electronics in general.
My use case for the microphone would be for streaming/recording and I am planning to make it use USB-A.

Thank you for your time and have a great day :)
 

Ian0

Joined Aug 7, 2020
9,671
If you want to make a USB microphone, then you'll need a USB Codec:
https://www.ti.com/product/PCM2904
Electret capsules are much the same wherever you get them - they are not a high-quality device and quite noisy, being designed for desk telephones. You then need a preamp, which could be a simple as a transistor, three resistors and a capacitor, or you could use an op-amp.
If you need automatic gain control then there are some suitable preamp ICs.
THAT1510 is rather over-the-top for use with an condensor microphone insert that belongs in a desk telephone!
Where do you want to go with this project?
 

Thread Starter

10_2

Joined Feb 14, 2023
5
I want to have a decent-sounding microphone that is good value for money and I want to learn about how microphones work really.

If the "TSB-2555B Electret Capsule" is not a good choice then what other capsules would you recommend?

Would a pre-amp made from a transistor, three resistors and a capacitor sound any good?

How complicated would an op-amp pre-amp be? (I'm not sure if that is the correct use of pre-amp)

I am unsure of Whether I need automatic gain control. However, from my preliminary research, I probably will.

What are the advantages of using a codec over DIY perks method of taking the insides from a rca to USB converter?
Link to an example of a rca to usb converter

Thanks again for your time :)
 

Ian0

Joined Aug 7, 2020
9,671
Most of the people who make professional quality microphone capsules are in the business of selling you professional quality microphones!
I find that the MEMS microphones are better than the electret capsules, but they are tiny and surface mount. Even so, although the distortion is reasonably good, they don’t match the professional ones for noise.
This is a quick sketch of theEBF5CEE4-5C78-4404-A2BB-F3E6C22574B5.jpegthe usual transistor or op-amp preamp which you’ll find everywhere, the MEMS types don’t need a preamp.
https://www.analog.com/en/products/ssm2167.html#product-overview is a preamp with AGC.
If you take apart a phono-to-usb converter, the odds are you’ll find a PCM2904 or PCM2906 inside.
 

Thread Starter

10_2

Joined Feb 14, 2023
5
Wow ok MEMS microphones seem to be the way to go. Correct me if I'm wrong they don't need a pre amp and sound decent which is prefect for me.Do you have any specific recommendations or guide lines for picking a MEMS microphone ?

Do the MEMs microphones match the electret capsules in terms of audio quality ?

Looking into the pcm2904 it seems like a good choice if I can figure out how to wire it up. Tho i will need a gain controller right? (Assuming I go with the mems microphone)
 

Ian0

Joined Aug 7, 2020
9,671
Wow ok MEMS microphones seem to be the way to go. Correct me if I'm wrong they don't need a pre amp and sound decent which is prefect for me.Do you have any specific recommendations or guide lines for picking a MEMS microphone ?

Do the MEMs microphones match the electret capsules in terms of audio quality ?

Looking into the pcm2904 it seems like a good choice if I can figure out how to wire it up. Tho i will need a gain controller right? (Assuming I go with the mems microphone)
MEMS are better than electret capsules (in my opinion). Better control of sensitivity and more reliable. You might still need some amplification.
Most have flat frequency responses, but look at the bass. Some are designed to roll off below 100Hz to give you “telephone bandwidth”.
PCM2904 - the application circuit in the datasheet works. It doesn’t need any setting up or any software.
 

Ya’akov

Joined Jan 27, 2019
9,071
MEMS are better than electret capsules (in my opinion). Better control of sensitivity and more reliable. You might still need some amplification.
Most have flat frequency responses, but look at the bass. Some are designed to roll off below 100Hz to give you “telephone bandwidth”.
PCM2904 - the application circuit in the datasheet works. It doesn’t need any setting up or any software.
Not all condenser mics are designed for low fidelity. In particular, the TS’ capsule is designed for hifi applications, and of course there are famous examples from Neumann and AKG among others.

That said, balancing the pros and cons of assembling my own USB microphone (it’s fun to build stuff; I can learn about things; and I can make it just how I want; vs. it’s not likely to save much money, so I am paying for my fun; what I will actually learn is a fairly narrow thing about cobbling together a home-brew microphone, and while I can build it just as I want that is limited by budget and my knowledge and skill—so maybe it won’t be too much more like I want if at all) I think I would probably opt for a commercial product.

What I would do in practice is monitor the B&H Photo and Video Deal Zone which not infrequently includes very good USB microphones often at startlingly deep discounts. I have purchased a large amount of it from B&H and their service is superb. An extremely reputable dealer. You can sign up to receive a daily email with the current 24 hour offerings so you won’t miss what you are looking for.

The used market can also turn up some excellent options and is worth exploring—particularly if you can tolerate some cosmetic defects which drive prices down substantially while not affecting electrical and mechanical operation. I have sometime purchased scuffed/dented items and, because I do put some value in the cosmetics, made my own cosmetic repairs for them with paint and judicious polishing or bending of plastics and metals.

The downsides of the DIY approach, I perceive, are exacerbated for a neophyte as well.

All of the foregoing said, if your goal is to have fun with building your own kit, there is nothing at all wrong with ignoring the practicalities. I just find people tend to couch this part in economy, education, and customization which, for me are not good arguments—while, “I just want to build my own, it’s will please me to have done it” is an entirely cogent and legitimate rationale for overriding the apparent “practical” considerations that are only practical if you have the wrong idea of what the goal actually is.
 

Ian0

Joined Aug 7, 2020
9,671
Not all condenser mics are designed for low fidelity. In particular, the TS’ capsule is designed for hifi applications,
Maybe - but I don’t see datasheets, frequency response graphs, noise figures, directional response plots, sensitivity tolerances, all the sort of things that reputable manufacturers would include de rigueur.

and of course there are famous examples from Neumann and AKG among others.
I own an AKG C451, but the capsules are not for sale separately.
 

LowQCab

Joined Nov 6, 2012
4,026
The problem with a "MEMS" Microphone, like most other Electret-Microphones,
is that they are "Omni-Directional".
This is not a serious problem if your ROOM has been sound-deadened adequately.

If You seriously handle absorbing ALL Reflections and Reverberations in the ROOM,
then almost any Microphone will sound at least acceptably good.

If You do NOT handle the ROOM problems, a ~$1500.oo Microphone will sound like garbage.

The best ROOM in the house for Voice-Recording,
is a large Closet that is crammed full to the ceiling with Clothing and Blankets.

You are making a Sound-Recording in a small cubical Box.
This will cause severe resonances, usually in the ~90 to ~200 Hertz range.
( Larger rooms have lower Frequency Resonances )
These resonances must be mostly absorbed before they reflect back to the Microphone.
The best material for accomplishing this is "Rock-Wool" Building-Insulation.

Do a search for DIY Rock Wool Bass Traps
Here's a good example site ........
https://bettersoundproofing.com/diy-bass-traps/

The first choice for consistent narration recording is a
very high-quality Head-set/Boom Mic combination, with a "Noise-Canceling" Microphone.

For other types of Microphones, keeping a consistent distance between the
Microphone and your Mouth is the most important factor.
The ideal distance is usually between ~6 to ~10-inches.

A "Pop-Filter", either Foam, or stretched-fabric, between You and the Mic is a good idea.

Mic positioning,
the Mic should be placed at "odd-distances" from hard, flat surfaces.
This includes the ceiling, and the floor, and your Desk,
and the wall in front of You, and the wall behind You.
The distances to all of these surfaces from the Mic
should be substantially different than each other,
the distances should be as randomized as possible.

All of the above flies out the window if You decide to
mount the Mic FLAT on the surface of your Desk.
This is called a "PZM" Microphone,
( Pressure-Zone-Microphone ),
and they can work very well.

Mic Specifications for narration ............

You don't need a super-low "Noise-Floor" for narration.
( I'm assuming you description of "Streaming" basically means YouTube-Videos or a "Pod-Cast" )
A Noise-Floor that is ~30-db down,
is perfectly adequate, and can only be detected with a good set of Ear-Phones and a silent Room.
Almost any Mic-Pre-Amp that You will find will exceed this minimum requirement,
especially if your Mic is within ~10-inches of your Mouth.

Frequency-Response,
For high-Frequencies, response up to 15khz is more than adequate,
and 10khz may actually be better, considering that most listeners are listening on
garbage speakers, or Headphones that won't reproduce anything higher than ~10khz anyway.

For Low Frequencies,
it depends on your Ego.
Some people seriously get-off on talking 1-inch from a Microphone that has a
flat frequency response down to ~10hz,
but this does absolutely nothing for "intelligibility" and clarity,
this is why You should stay at least ~6-inches away from the Mic,
this is to reduce the artificial Bass-"Boomy-ness" that occurs when talking too close to a Mic.
There is no real advantage to Vocal-Mic Bass-Frequency-Response below around ~80hz,
it just means that You will be required to have the Mic mounted in a springy-suspension device
to get rid of all the "Rumbling-Noise" that will inevitably be picked-up and amplified.

EQ and Compression .........

There is only 1 device that I would recommend to a new-comer ......
Go buy a used Lap-Top, and install some Software called "Breakaway Audio Enhancer"
It's "free to try", but costs about ~$30.oo if I remember correctly.
It is hands-down the best Audio-Processor anywhere, period.
( Claesson-Edwards-Software also produces extreme-high-end Professional-Broadcast-Software as well ).

This used-Laptop will be your new Microphone-Pre-Amp.
It will have ......
an automatic "Noise-Gate",
3- Band Parametric-EQ, plus an additional Low-Bass Parametric-EQ on the front interface,
up-to a 7-Band Parametric-Compressor, with Automatic-Peak-Limiting,
all the standard Controls used on most Compressors,
but with substantial automatic-adjustment via selectable "characterizations" or "Compression-Models",
which You can just click-on to try-out.

You need it even if You never record anything with a Microphone,
it makes every online recording sound much better, and more consistent, than it actually is.
.
.
.
 

MrSalts

Joined Apr 2, 2020
2,767
Instead of sound deadening a whole room, you can put a MEMs microphone in a 6" long tube (2" diameter PVC pipe). Place the mike on a 2" piece dense flexible foam and insert it into the bottom of the tube. From the top of the tube, insert some 1/4" to 1/2" slices of foam on the inside walls of the tube. Again, high density flexible foam. You'll be amazed how studio-like it sounds.
 

Ian0

Joined Aug 7, 2020
9,671
Instead of sound deadening a whole room, you can put a MEMs microphone in a 6" long tube (2" diameter PVC pipe). Place the mike on a 2" piece dense flexible foam and insert it into the bottom of the tube. From the top of the tube, insert some 1/4" to 1/2" slices of foam on the inside walls of the tube. Again, high density flexible foam. You'll be amazed how studio-like it sounds.
I‘ll try that. I had an idea rattling about in my mind that it might be possible to make a cardioid response by using two MEMS mics back to back, with suitable holes drilled in a tube, and add or subtract the signals.
 

LowQCab

Joined Nov 6, 2012
4,026
"" You'll be amazed how studio-like it sounds ""
As long as "Studio-like" involves having huge peaks at 47hz, 94hz, and 188hz.
( although probably damped to some degree, it's not possible to completely eliminate them )
The 47hz Peak is probably where the "Studio" comparison comes from.
.
.
.
 

LowQCab

Joined Nov 6, 2012
4,026
I‘ll try that. I had an idea rattling about in my mind that it might be possible to make a cardioid response by using two MEMS mics back to back, with suitable holes drilled in a tube, and add or subtract the signals.
.
I thought about that as well,
but it only works when You stay near the tuned Frequency.
But, it can be done with multiple Noise-Cancelling Mic-Elements and a fancy Pre-Amp-Circuit.
.
.
.
 

MrSalts

Joined Apr 2, 2020
2,767
"" You'll be amazed how studio-like it sounds ""
As long as "Studio-like" involves having huge peaks at 47hz, 94hz, and 188hz.
( although probably damped to some degree, it's not possible to completely eliminate them )
The 47hz Peak is probably where the "Studio" comparison comes from.
.
.
.
Sounds like a rule of thumb or measurements from a single studio you've worked in. Could you share the cause of why, specifically 47Hz and multiples thereof?
 

Ya’akov

Joined Jan 27, 2019
9,071
Maybe - but I don’t see datasheets, frequency response graphs, noise figures, directional response plots, sensitivity tolerances, all the sort of things that reputable manufacturers would include de rigueur.


I own an AKG C451, but the capsules are not for sale separately.
Well, it’s a $20 capsule. I wouldn’t compare it to an AKG or Neumann other than to say it’s not intended for telephone bandwidth work. As far as the actually fidelity, being an audio instrument, and a hobbyist component it probably has a lot of “performance enhancement” from the fact that it is owned by someone who doesn’t have the dosh for an AKG or the like.

I am sure the frequency response graphs and the pickup pattern plots are just so much “engineering” to the enthusiast with God’s audio instrument—ears. How can mere numbers compete with AB testing?

Be sure to have an Oxygen Free Copper power lead on the amp when testing, as well as an OFC USB cable, otherwise how can you know what it really sounds like?
 

Ian0

Joined Aug 7, 2020
9,671
Well, it’s a $20 capsule. I wouldn’t compare it to an AKG or Neumann other than to say it’s not intended for telephone bandwidth work. As far as the actually fidelity, being an audio instrument, and a hobbyist component it probably has a lot of “performance enhancement” from the fact that it is owned by someone who doesn’t have the dosh for an AKG or the like.

I am sure the frequency response graphs and the pickup pattern plots are just so much “engineering” to the enthusiast with God’s audio instrument—ears. How can mere numbers compete with AB testing?

Be sure to have an Oxygen Free Copper power lead on the amp when testing, as well as an OFC USB cable, otherwise how can you know what it really sounds like?
Got my transformer wound with silver wire,
got my mains plug with gold-plated fuse,
got my massive cables with the signal flowing in the direction of the arrows,
rewired the house with OFC cable (but what do I do about the cable between the meter and the distribution transformer?)
but seriously, I take your point about this device being a grade above what normally goes by the name “condenser insert” and I’d be interested to see datasheets if they exist. Otherwise it’s all down to Beranek’s law.
https://leachlegacy.ece.gatech.edu/beraneklaw.html
 

LowQCab

Joined Nov 6, 2012
4,026
Sounds like a rule of thumb or measurements from a single studio you've worked in. Could you share the cause of why, specifically 47Hz and multiples thereof?
.
A 6-foot long Wave-Length = 93.8hz,
47hz is roughly half that Frequency and 188hz is roughly twice that Frequency,
these will also resonate in a ~6-foot pipe, although,
they will resonate at somewhat less than half the intensity of the "Center-Frequency".

3- "Quarter-Wavelength", capped-off, "Stubs" that are
"teed" into the center of the ~6-foot-Pipe,
will almost cancel the main offending resonances.

There are plenty of online Calculators that will tell You
the Wavelength of a particular sound Frequency.

One of the biggest Sound problems in a small House or Apartment is that
the Bedrooms are 8-feet X 12-feet, with an 8-foot ceiling,
this is almost a "worst-case" set of dimensions for recording or reproducing sound.

A "less-bad" scenario involves using the general guidelines for constructing a Speaker-Box,
one these is referred to as "The-Golden-Rule" whereby the dimensions of the Box are
based on the formula "0.618 : 1 : 1.618",
this helps to reduce the chances of setting-up resonances inside the Speaker-Cabinet by
making sure that no dimension is a multiple of any other dimension.
This same line of thinking can be applied to the dimensions of a Room to reduce
Resonances or Standing-Waves that can be "excited" by interacting with each other.
.
.
.
 

Thread Starter

10_2

Joined Feb 14, 2023
5
That said, balancing the pros and cons of assembling my own USB microphone (it’s fun to build stuff; I can learn about things; and I can make it just how I want; vs. it’s not likely to save much money, so I am paying for my fun; what I will actually learn is a fairly narrow thing about cobbling together a home-brew microphone, and while I can build it just as I want that is limited by budget and my knowledge and skill—so maybe it won’t be too much more like I want if at all) I think I would probably opt for a commercial product.

What I would do in practice is monitor the B&H Photo and Video Deal Zone which not infrequently includes very good USB microphones often at startlingly deep discounts. I have purchased a large amount of it from B&H and their service is superb. An extremely reputable dealer. You can sign up to receive a daily email with the current 24 hour offerings so you won’t miss what you are looking for.

All of the foregoing said, if your goal is to have fun with building your own kit, there is nothing at all wrong with ignoring the practicalities. I just find people tend to couch this part in economy, education, and customization which, for me are not good arguments—while, “I just want to build my own, it’s will please me to have done it” is an entirely cogent and legitimate rationale for overriding the apparent “practical” considerations that are only practical if you have the wrong idea of what the goal actually is.
If I'm being honest I want to build a microphone because I want it to look cool. (Im planning on making the same stand that DIY perks made) So I'm not opposed to buying a microphone and gutting it then stuffing it into my own case. if it would be more cost-efficient and or easier. Do you have any recommendations for microphones that I could gut?
 

Thread Starter

10_2

Joined Feb 14, 2023
5
.
A 6-foot long Wave-Length = 93.8hz,
47hz is roughly half that Frequency and 188hz is roughly twice that Frequency,
these will also resonate in a ~6-foot pipe, although,
they will resonate at somewhat less than half the intensity of the "Center-Frequency".
phpqPJNky.png

Forgive me if I'm wrong but since it's a closed pipe and not an open pipe wouldn't the resonant frequency only exist at the first harmonic, third harmonic, fifth harmonic etc? as the fundamental frequency in a closed pipe is 1/4 of a wavelength as opposed to 1/2 of a wavelength in an open pipe?

Which would lead to resonance at 93.8/4 = 23.45hz (first harmonic) (93.8/4) * 3 = 70.35hz (third harmonic) and
(93.8/4) * 3 = 117.25hz etc ?

I could be completely wrong as I'm recalling this from high school physics which was a while ago.
 
Top