SNR in Mics

[Emmathom]

still looking for a very good SDC omni microphone to capture classical musicians or acoustiques instruments, I looked at some datasheets to note the signal to noise ratio...

I thought I understood from reading your various posts that the omnis are always noisier than the cardios and that it is on this aspect (and the quality of the sound of course) that one must concentrate

so what do you think is the minimum acceptable level? trying to imagine recording a violin + cello duo playing a piece with very calm passages in a church or a theater (for example)

when I look at the specs of the DPA 4006 known to be one of the best omni microphones, I read 15dB(A) and 18dB(A) for the DPA 2015 (wide cardio)
BTW the 2015 caughts all my attention... because it could combine the qualities of both world (?)

in my research I also discovered the Rode NT 55 MP delivered with a cardio capsule + an omni which claims to offer an SNR of 15dB(A) that is a good performance (if reel) for its price ($740 a pair)

finally other manufacturers do not indicate a specific inner noise but a signal to noise ratio like Primo with the EM23 which indicates (more than) 70dB but without specify its inner noise level...

 

[k brown]

Virtually no name brand mid-to-high priced mics have audible self noise in actual use recording music.

Only really a concern for recording very quiet nature sounds, etc. Mic self noise has been a bit of an obsession in the marketplace of late.

All else being equal, omnis usually have less self noise than cardioids.

 

[rogs]

still looking for a very good SDC omni microphone to capture classical musicians or acoustiques instruments, I looked at some datasheets to note the signal to noise ratio......
........ in my research I also discovered the Rode NT 55 MP delivered with a cardio capsule + an omni which claims to offer an SNR of 15dB(A) that is a good performance (if reel) for its price ($740 a pair)

If you're looking for a very low noise solution, then SDC mics are maybe not the way to go?

Taking the SDC Rode NT55 you mentioned, and comparing it to the LDC Rode NT1 (4th generation) and you will see there are significant differences in the noise figures.

The NT55 has a published noise figure of 15dBA....... The NT1 is 4.5dBA - so over 10dB better.

But you also need to consider the sensitivities as well....
• The NT55 has a sensitivity of-38dB, which means it generates a signal of 12.6mV when a reference tone of 94dBSPL is applied
• The NT1 has a sensitivity of -29dB, which means it generates a signal of 35mV when a reference tone of 94dBSPL is applied

So you are going to need around 9dB more preamp gain for the NT55 to generate the same signal ouput as the NT1, from the same source.
Add that to the aready higher noise figure of the NT55, and you will have a significantly higher noise level in the recording.

Now how significant that will be will depend on the source being recorded of course?

Neumann consider a noise figure of 15dBA near the top end of 'acceptable' for a low noise studio mic ... see here: https://www.neumann.com/en-en/homestudio-academy/what-is-self-noise-or-equivalent-noise-level/

When it comes to the EM23, I'm not sure how they are determining that 70dB SNR..... The capsule itself has no intenal impedance converter, so I'm assuimg they are referrring to the noise generated by the published circuit?
In any event, a 70dB SNR suggests a self noise figure of 24dB -- which is quite high....

 

[MicUlli]

I also have two Primo EM23 in use, they have a measured self noise figure of 9,9 dBA and 10,4 dBA...

 

[rogs]

I also have two Primo EM23 in use, they have a measured self noise figure of 9,9 dBA and 10,4 dBA...

Looks like Primo's data sheet listing is very conservative then: 'S/N Ratio : Above 70 dB at 1k Hz ( 1 Pa, A weighted network )'
(- although to be fair they haven't actually said how much above 70dB ...)

It makes a change from manufacturers making overly favourable claims !

 

[Emmathom]

I also have two Primo EM23 in use, they have a measured self noise figure of 9,9 dBA and 10,4 dBA...

on wich mic pre (body) do you use them ? some ref. of body & pre who accpet these Primo capsules ? Thx

 

[Emmathom]

If you're looking for a very low noise solution, then SDC mics are maybe not the way to go?

Taking the SDC Rode NT55 you mentioned, and comparing it to the LDC Rode NT1 (4th generation) and you will see there are significant differences in the noise figures.

The NT55 has a published noise figure of 15dBA....... The NT1 is 4.5dBA - so over 10dB better.

But you also need to consider the sensitivities as well....
• The NT55 has a sensitivity of-38dB, which means it generates a signal of 12.6mV when a reference tone of 94dBSPL is applied
• The NT1 has a sensitivity of -29dB, which means it generates a signal of 35mV when a reference tone of 94dBSPL is applied

So you are going to need around 9dB more preamp gain for the NT55 to generate the same signal ouput as the NT1, from the same source.
Add that to the aready higher noise figure of the NT55, and you will have a significantly higher noise level in the recording.

Now how significant that will be will depend on the source being recorded of course?

Neumann consider a noise figure of 15dBA near the top end of 'acceptable' for a low noise studio mic ... see here: https://www.neumann.com/en-en/homestudio-academy/what-is-self-noise-or-equivalent-noise-level/

When it comes to the EM23, I'm not sure how they are determining that 70dB SNR..... The capsule itself has no intenal impedance converter, so I'm assuimg they are referrring to the noise generated by the published circuit?
In any event, a 70dB SNR suggests a self noise figure of 24dB -- which is quite high....

Thank you Rogs for this feedback but I am rather looking for SDC since LDCs tend to distort, to give an unfaithful image of the soundstage and particularly in omni (as I've been told several times)
Regarding the gain needed at the preamp level I think I have what I need in reserve with the ones I build (tube / BJT / Fet / AOP)

Question: has anyone measured the SNR of a Takstar CM-63 + 3U omni 5mm capsule?
Thx

 

[Khron]

3U omni 5mm capsule

I'm just nitpicking here, but I'm pretty sure that's supposed to be ".5mm", as in 0.5mm thread pitch. Just FYI

 

[Emmathom]

I'm just nitpicking here, but I'm pretty sure that's supposed to be ".5mm", as in 0.5mm thread pitch. Just FYI

Yes you're right, I'm a bit lost with all the informations I grab on this site...

BTW : I can't remember wich body-pre fits the Primo EM21 & EM23 capsules... ? (except LEM)

 

[rock soderstrom]

I can't remember wich body-pre fits the Primo EM21 & EM23 capsules... ? (except LEM)

Check this thread:
https://groupdiy.com/threads/cad-equitek-e100-capsule.77061/

 

[ricardo]

finally other manufacturers do not indicate a specific inner noise but a signal to noise ratio like Primo with the EM23 which indicates (more than) 70dB but without specify its inner noise level...

Many makers, including Primo, quote a S/N wrt to 1 Pascal; 94dB spl. So 70dBA S/N would be 94-70 = 24dBA ie spl A wtd.

My $0.02 is that if the noise is below 20dBA spl, you really need to listen to the noise and/or do spectral analysis.

The dominant noise for capacitor mikes in an A wtd measurement is the 1G/capsule capacitance noise which is 'reddish'. But this sounds just like ambient noise in a quiet studio so usually not a problem. The most irritating noise is 'white'; electrical noise and the resistive damping in the capsule.

That's not to say the 1G/capsule noise isn't detrimental. I hear it as ambient sounds are 'thick' ... rather than noisy.

If you want a really cheapo but quiet omni, have a look at a PUI 5024 capsule in one of my SimpleP48 circuits This doesn't give away anything to more complex stuff except high spl overload level.

 

[k brown]

It's hard to do better than the Primo EM273 for the combination of low noise, sound quality and price.

Very popular with the DIY Nature recording crowd.

https://micbooster.com/microphone-capsules/209-primo-em273-microphone-capsule.html

 

[Emmathom]

Line Audio OM1 are great too but I guess they make their own © capsules ? Too small for a primo to fit inside...

 

[k brown]

Line Audio OM1 are great too but I guess they make their own © capsules ? Too small for a primo to fit inside...

Line Audio Omni1 (no longer called OM1) and CM4 are made with Primo capsules, but LA significantly modifies them.

Omni1 starts with the Primo EM273 (10mm dia.), and the CM1 the EM200 (14.6mm dia.). Both have built-in FETs, unlike the EM23/21.

 

[Emmathom]

Thank you @k brown for the info. Yes they're called Omni1 now... and they sound great to me (not only for the price)

 

[k brown]

To answer your earlier question about bodies for the EM23, all the commercial ones that were available (Nakamichi CM-300, Teac, Tascam, Primo's own branded mic) used the same rather poor circuitry; only one I know of that had much better electronics, plus a very nicely-made, solid (and much smaller) body for these capsules was the Crown CM-700 (not to be confused with the Nakamichi CM-700, which may have also used the EM21/23; I don't know what their electronics were like except that they also used hard to find batteries). The Crown CM-700 didn't use batteries - it ran on phantom power.

It was fitted with the EM21 cardioid capsule, but unscrewing the cosmetic black ring on the front, the capsule can be unscrewed and an EM23 put in it's place, making an excellent, compact (only 5" long) omni mic.

Unfortunately, I almost never see them on eBay; I guess because everyone that bought them, kept them. They probably show up on Reverb from time to time.