Quick noob question for my favorite forum in the world!

[AUDIO FREQ]

In a FET Microphone, what constitutes the self noise. Is it the capsule, or the circuit? And if it is both, which one is more responsible?

So in a low noise circuit like the OPA Alice, would adding a K103 neumann capsule (which comes from the TLM103), have an even lower self noise in the final mic build?

Im trying to get the lowest self noise and highest sensitivity mic build for voiceover/soft singing recordings.

 

[mrgrooves666]

I would personally say both contribute. But you dont need a neumann capsule. There's many cheaper nice quality capsules around. The Opa is already very quiet, high sensitivity design . I have a pair of opa mics, non ePV capable, with electrets (tsb 2552 and King Korg's ali cheap find), and they are very quiet mics. Im actually surprised by their performance. I mean their noise floor is probably lower than regular ambience...

 

[AUDIO FREQ]

I would personally say both contribute. But you dont need a neumann capsule. There's many cheaper nice quality capsules around. The Opa is already very quiet, high sensitivity design . I have a pair of opa mics, non ePV capable, with electrets (tsb 2552 and King Korg's ali cheap find), and they are very quiet mics. Im actually surprised by their performance. I mean their noise floor is probably lower than regular ambience...

What is king korgs ali cheap find? Is it a large dia? I tried the TSC-2 and the TSC-103, couple of ALiexpress finds, and they all seem to generate alot of noise in my opa alice boards from jli electronics. can't really seem to find a good match.

 

[Khron]

noise in my opa alice boards from jli electronics

Well, at least some of that is down to the board design itself (and/or its execution)...

 

[rogs]

Im trying to get the lowest self noise and highest sensitivity mic build for voiceover/soft singing recordings.

If you're looking at lowest noise, you might like to take a look at my OPIC project ? ... It's a similar concept to the OPA Alice, but it uses a single sided audio output .
The line is still balanced - via passive impedance matching - but there is no second op-amp.
(That provides a differential audio output with the OPA Alice project.)
Leaving out the second op-amp will stop the extra noise generatad by the amp and it's 2 resistors from entering into the signal path.
Those components will add an extra 2 or 3dB extra to the mic noise figure.
It will require an extra 6dB of preamp gain to obtain the same signal output level, but a decent mic preamp - or gain added in a DAW - shouldn't add any significant extra noise into the system.
Don't get me wrong, the OPA Alice system does offer an pretty low noise figure, but you did mention 'lowest' self noise....

It may be possible to obtain an even lower figure from a carefully designed discrete FET input stage, but we're entering into the realms of the 'law of diminishing retuns' here, and it's probably not worth the effort?
For most real world recording situations, the noise figure from both OPA designs will likely be 'swamped' by ambient noise anyway!
You'd probably be hard pushed to detect any tiny extra improvement in noise level from a discrete FET impedance converter ....

 

[mhelin]

I'm trying to get the lowest self noise and highest sensitivity mic build for voiceover/soft singing recordings.

What is king korgs ali cheap find? Is it a large dia? I tried the TSC-2 and the TSC-103, couple of ALiexpress finds, and they all seem to generate alot of noise in my opa alice boards from jli electronics. can't really seem to find a good match.

They shouldn't generate a lot of noise, how is your interface and the audio system overall performing? Have you any other mics you could use as a reference?

 

[AUDIO FREQ]

They shouldn't generate a lot of noise, how is your interface and the audio system overall performing? Have you any other mics you could use as a reference?

I am using an Apollo Twin X and Apollo X6 rackmount interface with no plugins on the input. Just stock preamps.
I get lower noise from the Microphone Parts MP-V57 circuit board build I have than the JLI Opa Alice build.

If you're looking at lowest noise, you might like to take a look at my OPIC project ? ... It's a similar concept to the OPA Alice, but it uses a single sided audio output .
The line is still balanced - via passive impedance matching - but there is no second op-amp.
(That provides a differential audio output with the OPA Alice project.)
Leaving out the second op-amp will stop the extra noise generatad by the amp and it's 2 resistors from entering into the signal path.
Those components will add an extra 2 or 3dB extra to the mic noise figure.
It will require an extra 6dB of preamp gain to obtain the same signal output level, but a decent mic preamp - or gain added in a DAW - shouldn't add any significant extra noise into the system.
Don't get me wrong, the OPA Alice system does offer an pretty low noise figure, but you did mention 'lowest' self noise....

It may be possible to obtain an even lower figure from a carefully designed discrete FET input stage, but we're entering into the realms of the 'law of diminishing retuns' here, and it's probably not worth the effort?
For most real world recording situations, the noise figure from both OPA designs will likely be 'swamped' by ambient noise anyway!
You'd probably be hard pushed to detect any tiny extra improvement in noise level from a discrete FET impedance converter ....

I will give that a shot Rogs, shouldn't I just be able to modify the JLI Electronics OPA Alice build to mimic what youre doing on your build? It seems like the same parts are used, but the opamp is dual instead of single. Maybe bypassing the the 2nd stage of the op amp?

 

[Khron]

I will give that a shot Rogs, shouldn't I just be able to modify the JLI Electronics OPA Alice build to mimic what youre doing on your build? It seems like the same parts are used, but the opamp is dual instead of single. Maybe bypassing the the 2nd stage of the op amp?

Well, at least some of that is down to the board design itself (and/or its execution)...

I can understand @rogs has all the interest to promote his design, and for good reason, but i guess noone has yet picked up on what i was hinting at..?

As opposed to the (pretty clear) build instructions on @rogs website, JLI (or whoever designed and spec'd the boards) obviously didn't have much of a clue, or couldn't give a hoot about the high impedances involved there. As such, all the opamp pins, AND both ends of the 1G(?) resistor, ARE soldered directly to the PCB. That's NOT a recipe for lowest noise, FYI...

 

[AUDIO FREQ]

I can understand @rogs has all the interest to promote his design, and for good reason, but i guess noone has yet picked up on what i was hinting at..?

As opposed to the (pretty clear) build instructions on @rogs website, JLI (or whoever designed and spec'd the boards) obviously didn't have much of a clue, or couldn't give a hoot about the high impedances involved there. As such, all the opamp pins, AND both ends of the 1G(?) resistor, ARE soldered directly to the PCB. That's NOT a recipe for lowest noise, FYI...

Ohhhhhhhhh! Wow. I didn't even think of that. Maybe I can try lifting the op amp pins from the pcb. AND the 1g resistor? Air design? Have the pins floating in the air and soldered by wire to the connections maybe?

 

[Khron]

https://www.jlielectronics.com/diy-accessories/opa-alice-single-channel/

So ironic that they DID provide a hole right in the top middle there, for an isolation pin - just didn't use it.

The resistor leg you could probably lift out of the hole and bend it towards the opamp, but likely you'll need to add a piece of resistor leg as an extension. Probably a better idea to add that at the other end of the resistor though, in order to ease connection to the opamp pin and the wire coming from the capsule.

And speaking of the opamp, you'd only need to lift pin 3. Worth trying to gently pry / lever it up with a sewing needle or the tip of a tweezer, then clear off whatever solder is left there with some wick, then clean thoroughly with isopropyl alcohol.

And dear sweet lord, only NOW, when i looked at the board again, did i notice what a pathetically-autorouted turd it is... Or hey, for all i know, maybe Dany had a hand in that...
PS: At least in the photo there... Is that D1 diode pointing the wrong way? Why would its cathode be connected to ground (XLR1 & opamp pin 4)?

 

[AUDIO FREQ]

https://www.jlielectronics.com/diy-accessories/opa-alice-single-channel/

So ironic that they DID provide a hole right in the top middle there, for an isolation pin - just didn't use it.

The resistor leg you could probably lift out of the hole and bend it towards the opamp, but likely you'll need to add a piece of resistor leg as an extension. Probably a better idea to add that at the other end of the resistor though, in order to ease connection to the opamp pin and the wire coming from the capsule.

And speaking of the opamp, you'd only need to lift pin 3. Worth trying to gently pry / lever it up with a sewing needle or the tip of a tweezer, then clear off whatever solder is left there with some wick, then clean thoroughly with isopropyl alcohol.

And dear sweet lord, only NOW, when i looked at the board again, did i notice what a pathetically-autorouted turd it is... Or hey, for all i know, maybe Dany had a hand in that...
PS: At least in the photo there... Is that D1 diode pointing the wrong way? Why would its cathode be connected to ground (XLR1 & opamp pin 4)?

With all this talk about teflon turrets. I think im just going to buy some. Anyone have a good link where I can get some? I checked mouser and digikey but the sizes are so overwhelmingly confusing!

 

[Khron]

Why? More expense, and one more thing to fiddle with in already cramped quarters? You'd need an extra piece of wire between the opamp pin and the teflon turret anyway, so...

 

[ElSmurf]

Anyone have a good link where I can get some?

Look for the Keystone 11218, looks like it might fit.

 

[AUDIO FREQ]

Why? More expense, and one more thing to fiddle with in already cramped quarters? You'd need an extra piece of wire between the opamp pin and the teflon turret anyway, so...

idk khron, i trust your word. some other dude in another post suggested teflon turrets so i figure that might be the way to go. but since you say no need, then hey youve helped me out so much i trust your word my finnish brother from another mother.

 

[Khron]

- Lift opamp pin 3
- Clean off remaining solder from it and from the PCB pad
- Clean with isopropanol
- Remove the through-hole resistor; clean it with alcohol too, and afterwards try to maneuver it with a pair of clean tweezers
- Solder one end to the lifted opamp pin, while resting the other end on the PCB surface (just so you don't put any undue strain on the opamp pin)
- Add whatever length of wire you need, between the board-end of the resistor and the PCB hole it came from
- Solder capsule wire to the 1G-opamp-pin joint
- Clean the area with isopropanol again, just in case
- Job done

 

[k brown]

The quietest mics I own have the high impedance points connected 'in air'.

I think the Teflon posts are primarily are about mass-production practicailties.

 

[AUDIO FREQ]

- Lift opamp pin 3
- Clean off remaining solder from it and from the PCB pad
- Clean with isopropanol
- Remove the through-hole resistor; clean it with alcohol too, and afterwards try to maneuver it with a pair of clean tweezers
- Solder one end to the lifted opamp pin, while resting the other end on the PCB surface (just so you don't put any undue strain on the opamp pin)
- Add whatever length of wire you need, between the board-end of the resistor and the PCB hole it came from
- Solder capsule wire to the 1G-opamp-pin joint
- Clean the area with isopropanol again, just in case
- Job done

That makes me wander. Ive seen people build mics without a pcb, just wires and components connected to each other in the air. I might try this in the future, but what do you think Khron? Good idea? Bad idea?

 

[Khron]

I'm not a fan, if i'm honest. Sure, if you reeeeeeeeeeeally want to cobble something together NOW to see if it works, i can sorta see the excuse for a "deadbug" execution, but... Even for small boards (think a literal thumbnail), i prefer to at least score and isolate "islands" on a copper-clad PCB and solder things onto that. It's much more practical / easy to prevent unintentional contacts on any sort of pcb, than a 3D-point-to-point assembly.

... But i'm just weird like that

 

[k brown]

I was only advocating for doing the HiZ points of a mic 'in air'; certainly not the entire microphone.

 

[rogs]

I will give that a shot Rogs, shouldn't I just be able to modify the JLI Electronics OPA Alice build to mimic what youre doing on your build? It seems like the same parts are used, but the opamp is dual instead of single. Maybe bypassing the the 2nd stage of the op amp?

You could try that - although for the reasons I mentioned above - about ambient noise being the main noise source - it's probably not worth it.

Some months back I had a problem with my Rode NT1. (See this post ) It showed me that my op-amp circuit was about 3dB worse than Rode's noise figure for the NT1 (which is specified as a pretty low 4.5dB(A) ).
Add another couple of dB on top of that extra 3dB for the OPA Alice configuration and you're still only talking about 10dB(A) or so of self noise.

According to Neumann that's still considered 'extremely low noise'..... As they describe on this page the ambient noise will still be the dominant noise source.