Recordings aren't very clear. xfigo with a 2i2 to improve sound?

[digitaldragon]

I have been recording basic 2 track vocal karaoke covers in Reaper for a while now. I am currently using a 2i2, an sm57 with fethead, and a cheap acer laptop.

My problem is that the sound I hear coming from the sm57, even without any processing, doesn't sound as crisp and clear as it does in various YouTube mic shootouts and vocalists throwing up vids testing out the sm57 as a vocal mic. I thought initially that it's just me, or that maybe I'm just missing something and I need to tweak plugins or eq in Reaper to get the sound I want.

But even after moving the eq sliders around, and getting it to sound acceptable to my ears, it still sounds kind of "muffled". My final recordings just always seem to be missing the clearness of sound that I'm looking for. Could it be that the pre in the 2i2 isn't good enough?

I have a fethead boosting the db to 65 I believe, so volume isn't a problem, and I play with the gain and mic distance until I'm just out of clipping range. I got to thinking that maybe the laptop's soundcard may have something to do with it. Is this possible? Or does the 2i2 automatically act as the soundcard?

I have an xfigo soundcard and I'm wondering if this will be able to work in conjunction with the 2i2 to provide better audio for whatever goes into that sm57. I'm also planning on trying a condenser mic to see if it's just an sm57 issue.

From the examples I've heard, and the fact that a lot of artists have used it as a vocal mic, I assume it's just lack of knowledge on my part and I'm doing something wrong or need better equipment. Any other suggestions to improve the clarity of my recordings?

Thanks.

 

[jimmys69]

I have been recording basic 2 track vocal karaoke covers in Reaper for a while now. I am currently using a 2i2, an sm57 with fethead, and a cheap acer laptop.

My problem is that the sound I hear coming from the sm57, even without any processing, doesn't sound as crisp and clear as it does in various YouTube mic shootouts and vocalists throwing up vids testing out the sm57 as a vocal mic. I thought initially that it's just me, or that maybe I'm just missing something and I need to tweak plugins or eq in Reaper to get the sound I want.

But even after moving the eq sliders around, and getting it to sound acceptable to my ears, it still sounds kind of "muffled". My final recordings just always seem to be missing the clearness of sound that I'm looking for. Could it be that the pre in the 2i2 isn't good enough?

I have a fethead boosting the db to 65 I believe, so volume isn't a problem, and I play with the gain and mic distance until I'm just out of clipping range. I got to thinking that maybe the laptop's soundcard may have something to do with it. Is this possible? Or does the 2i2 automatically act as the soundcard?

I have an xfigo soundcard and I'm wondering if this will be able to work in conjunction with the 2i2 to provide better audio for whatever goes into that sm57. I'm also planning on trying a condenser mic to see if it's just an sm57 issue.

From the examples I've heard, and the fact that a lot of artists have used it as a vocal mic, I assume it's just lack of knowledge on my part and I'm doing something wrong or need better equipment. Any other suggestions to improve the clarity of my recordings?

Thanks.

I would suggest posting a sample. It seems odd that you would have an issue with the sm57 being different from others results. Tho, do you have an actual SM57 or a knockoff fake one? That is a possibility. Also it could be a bad cable. Just some things to check first.

The interface (2i2) takes the internal sound in your computer out of the equation. Phantom power on the interface on? Even with that on a bad XLR cable could cause things to not work properly.

Start with the first order. Test that cable and move to the next possible issue.

 

[Mack Caster]

My problem is that the sound I hear coming from the sm57, even without any processing, doesn't sound as crisp and clear as it does in various YouTube mic shootouts and vocalists throwing up vids testing out the sm57 as a vocal mic. I thought initially that it's just me, or that maybe I'm just missing something and I need to tweak plugins or eq in Reaper to get the sound I want.

You really do not need the frethead. The sm57 is a dynamic mic and your audio interface is really all you need to power it. I think you are supplying the signal way to hot. Try this.

1. Run your 57 right into your interface and into your computer without your frethead.

2. Open up Reaper and arm a track to record.

3. Turn the gain up on your audio interface till it reads between a -12 and -18dbs on the input meter in Reaper. Record track with your 57 4 to 6 inches from your mouth.

4. When finished normalize your track to around a -15dbs. Upload raw, UN-altered vocal track.

 

[Bobbsy]

By all means try without the fethead but my suspicion is that the SM57 simply may not provide the type of sound you're after.

While it's a good, versatile microphone, as a dynamic it'll never have the same sort of crisp detail you'd get with most condensers. Some people love the sound on vocals; to my ear it's a bit flat and lacking detail--great on guitar cabinets or some other instruments but never my first choice for vocals. Others would argue with me.

I'll second (or is it third or fourth?) the idea of you posting a sample. You'll need some more posts before you're allowed to post links.

 

[ecc83]

+1 to the 57 being the problem. As Bob say's, great mic for many jobs but drops off well at HF.

The YTs the OP is listening to might well be done with a cheap capacitor mic. A USB mic or one of the 'BM-800' jobs that are so common now. Not U87s for sure but DO have that post 10kHz brilliance dynamics lack.

I also agree with Mack, the SM57 should be fine straight into a 2i2. The mic works well into my KA6 and I think the 2i2 has a smell more gain? Noise performance is about the same, i.e. bloody good.

Dave.

 

[Mack Caster]

I am going to go into a little more detail then my previous post to help pinpoint your questions.

I have been recording basic 2 track vocal karaoke covers in Reaper for a while now. I am currently using a 2i2, an sm57 with fethead, and a cheap acer laptop.

Remove the frethead out of your signal chain from your mic into your computer.

My problem is that the sound I hear coming from the sm57, even without any processing, doesn't sound as crisp and clear as it does in various YouTube mic shootouts and vocalists throwing up vids testing out the sm57 as a vocal mic. I thought initially that it's just me, or that maybe I'm just missing something and I need to tweak plugins or eq in Reaper to get the sound I want.

Never judge your own voice by someone else. You may not have the same ability, training or coaching they have had.

But even after moving the eq sliders around, and getting it to sound acceptable to my ears, it still sounds kind of "muffled". My final recordings just always seem to be missing the clearness of sound that I'm looking for. Could it be that the pre in the 2i2 isn't good enough?

Your pres are just fine for what you want to do unless you have a defective unit. Are you recording over carpet? If so, that is going to such the mid range frequencies right out of your recording. Yep, sucks them right out. The sound of your recording is controlled by your recording environment. This may be why the videos sound so much better then what you are trying to mimic.

I have a fethead boosting the db to 65 I believe, so volume isn't a problem, and I play with the gain and mic distance until I'm just out of clipping range.

This is the worst thing you can do. Most interfaces are designed to send the signal to be recorded into the recorder between a -12dbs and -18dbs.

I got to thinking that maybe the laptop's soundcard may have something to do with it. Is this possible? Or does the 2i2 automatically act as the soundcard?

The 2i2 will only act as the sound card if you are using the ASIO file that is normally supplied with the unit. Have you downloaded and installed any files for the 2i2?

I have an xfigo soundcard and I'm wondering if this will be able to work in conjunction with the 2i2 to provide better audio for whatever goes into that sm57. I'm also planning on trying a condenser mic to see if it's just an sm57 issue.

The sm57 and 58 are famous for their roll in building rock and roll back in the day. I doubt you will find any professional recording studio who does not have at least 6 to 12 of these mics on hand. Do you have a high pitched voice? The freq response for the sm 57 is between 40 and 15,000hrz, If your voice exceeds 15,000hrz then this will not give you the high end you will need.

From the examples I've heard, and the fact that a lot of artists have used it as a vocal mic, I assume it's just lack of knowledge on my part and I'm doing something wrong or need better equipment. Any other suggestions to improve the clarity of my recordings?

Two things. First, you are probably 100% correct on your first sentence above this answer. Second, I am 100% positive this community can get you where you need to be!

 

[keith.rogers]

I'd like to see a sticky somewhere on how to ask for help with a recording problem. I'm continually amazed by how many posts start off by stating that something doesn't sound right, and there's never a clip posted that actually demonstrates the problem!

I don't know if the OP here has a basso profundo's or countertenor's voice, is whisper quiet or screaming loud, etc. And, the actual record levels are truly unknown. What does 65 mean? Is that 65dB gain? Are you going into a Line input with that much gain or trying to stuff it into the Mic in?

For starters, if you record something and "it sounds bad" or at least not how you hope and you want some objective information, you should post the actual audio track. So, find out where the media is being stored for your specific DAW, and post a link to the actual sound file (.WAV or whatever format) is my suggestion. And then, if you have a mix with all your FX, post that as well. Just my early morning, still shy a cuppa joe or three, $.02.

 

[arcaxis]

- What are you monitoring/listening with? Headphones or monitors/speakers and what brand/model are they?
- Are you adding and effects to the audio such as reverb?
- Not recording in some kind of 'homemade' foam lined booth/box.

By all means as suggested try the SM57 without the FetHead and see if it makes a difference, but.......

My own experience with the FetHead with an SM7b and SM57, is that the FetHead due to its higher termination impedance (22kohms), gives both the SM7b and SM57 a slight boost on the high end giving the mic a bit more crispness and clarity. The FetHead gives 27dB of clean gain which allows both the SM7b and SM57 to hit the input of an interface at a level comparable to a hot condenser mic (SM57 > -56dB + 27dB = -29dB. Shure KSM44 sens. is -31dB.). Interface won't need to be cranked to nearly max either, which depending on how good the preamp is in it, might bring up some amount of noise.
FetHead - TRITONAUDIO

The SM57 is not one of my preferred vocal mics even with (or without) the FetHead. Have a few inexpensive condensers that sound much better.

Part of a Focusrite article on mic preamp impedance....

Dynamic moving coil and condenser microphones
Almost all professional dynamic and condenser microphones are designed to have a relatively low nominal output impedance of between 150Ω and 300Ω when measured at 1kHz. Microphones are designed to have such low output impedance because the following advantages result:
They are less susceptible to noise pickup They can drive long cables without high frequency roll-off due to cable capacitance The side effect of having such low output impedance is that the mic preamp input impedance has a major effect on the output level of the microphone. Low preamp impedance loads down the microphone output voltage, and emphasizes any frequency-related
variation in microphone output impedance. Matching the mic preamp resistance to the microphone output impedance (e.g. making a preamp input impedance 200Ω to match a 200Ω microphone) still reduces the microphone output and signal to noise ratio by 6dB, which is undesirable.
To minimise microphone loading, and to maximise signal to noise ratio, preamps have traditionally been designed to have an input impedance about ten times greater than the average microphone, around 1.2kΩ to 2kΩ. (The original ISA 11 0 pre-amp design followed this convention and has an input impedance of 1.4kΩ at 1kHz.) Input impedance settings greater than 2kΩ tend to make the frequency-related variations of microphone outputs less significant than at low impedance settings. Therefore high input impedance settings yield a microphone performance that is flatter in the low and mid frequency areas and boosted in the high frequency area when compared to low impedance settings.

 

[drtechno]

-
Part of a Focusrite article on mic preamp impedance....

Dynamic moving coil and condenser microphones
Almost all professional dynamic and condenser microphones are designed to have a relatively low nominal output impedance of between 150Ω and 300Ω when measured at 1kHz. Microphones are designed to have such low output impedance because the following advantages result:
They are less susceptible to noise pickup They can drive long cables without high frequency roll-off due to cable capacitance The side effect of having such low output impedance is that the mic preamp input impedance has a major effect on the output level of the microphone. Low preamp impedance loads down the microphone output voltage, and emphasizes any frequency-related
variation in microphone output impedance. Matching the mic preamp resistance to the microphone output impedance (e.g. making a preamp input impedance 200Ω to match a 200Ω microphone) still reduces the microphone output and signal to noise ratio by 6dB, which is undesirable.
To minimise microphone loading, and to maximise signal to noise ratio, preamps have traditionally been designed to have an input impedance about ten times greater than the average microphone, around 1.2kΩ to 2kΩ. (The original ISA 11 0 pre-amp design followed this convention and has an input impedance of 1.4kΩ at 1kHz.) Input impedance settings greater than 2kΩ tend to make the frequency-related variations of microphone outputs less significant than at low impedance settings. Therefore high input impedance settings yield a microphone performance that is flatter in the low and mid frequency areas and boosted in the high frequency area when compared to low impedance settings.

So that's where the misinformation comes from. Focusrite focuses wrong again. no, to get maximum bandwidth through the system, the impedances has to be equal. Right now, I'm waiting for transformers to come in for my DBX 286s so I can bring it up to correct specifications and operating points. Those focusrite mic preamps in the 2i2 are more suitable for the 1.2K output impedance condenser mics.

 

[rob aylestone]

I'm not sure I disagree with focusrite? The deliberate mismatch of impedance surely reduces the current flowing which must impact on the source in some way as back EMF, as resistance to change, and in a dynamic without the buffer of the mic preamp, will lump load the capsule? That's always been my understanding - is it wrong? I'd never read the focusrite text, but it appears to be as I have always understood it. The DBX preamp is a fine piece of cheaper equipment, but I really cannot see that the transformer will improve it. Colour it, modify it, etc, but not improve. Transformers are always compromise devices and the benefits of impedance conversion and isolation must be set against the detrimental effect of the transformer.

If we were talking ultimate quality electronics, then chasing the nth degree of sonic coherence might, just might be justified, but in a £150 device, there will be far more compromises than a bit of impedance mismatching.

One of the reasons I didn't get my first job at the BBC was because I was weak on the electronics side, and misunderstood bias current and impedance in audio circuits. The rule at that time was source impedance must always be lower than the destination impedance, and back then we were using mics with 25Ohm impedance, and the input circuits were over 1K, can't remember the exact figure. Did the BBC get this wrong? Most common mics will be less than 1K, so an input impedance of 1.2K makes sense. I don't think any of my condensers are above that figure.

The maximum bandwidth thing you mentioned - I don't understand. Loading dynamics has always been considered bad - is this wrong too?

Forgive me, but some modern flawed thinking appears to be creeping in somewhere - care to explain, and perhaps I can go off and read about it, but my 'audiophile warning, Will Robinson" appears to have activated.

 

[ecc83]

http://cdn.shure.com/specification_sheet/upload/81/sm57-specification-sheet-english.pdf

Shure specify the output of the 57 into an 'Open Circuit' which equates closely to the modern wisdom of pre amps having an input Z (DC resistance really) of about 2kOhms or over 10X the nominal mic impedance.

The response shown in the above .pdf is -5dB at 15kHz with a ~6dB 'hump' at 6kHz or so. A capacitor mic, especially an SDC would have a much smaller peak and extend pretty flat to 20kHz or so.

As to loading, the mics already have a transformer in them and if anything will appear as an inductive source. It is electronics 101 that if you load an inductive generator the response FALL as frequency rises, assuming a non-resonant system.
This effect is seen at its worst when you feed a guitar pickup into a 10k line input.

The fad, begun about 3 years ago AFAICT, of having variable loadings at mic inputs is SAID to give different tonalities but these are subtle from what I read and in MOST cases the input Zs are made much HIGHER than the normal 1.5-2k up to 5k and even 10k in some cases. The higher load is supposed to benefit ribbons most of all, dynamics a bit but as most capacitor mics are transformerless out, it makes buggerall difference to those. (they are essentially resistive, often the output of a chip)

Dave.

 

[arcaxis]

So that's where the misinformation comes from. Focusrite focuses wrong again. no, to get maximum bandwidth through the system, the impedances has to be equal. Right now, I'm waiting for transformers to come in for my DBX 286s so I can bring it up to correct specifications and operating points. Those focusrite mic preamps in the 2i2 are more suitable for the 1.2K output impedance condenser mics.

So a guitar PU is similar in function to a dynamic mic. A Gibson P-90 PU has an impedance of 7.47 kΩ at 6.405 kHz. By what you're stating to get 'maximum bandwidth' (widest frequency range) the P-90 should be terminated to an amplifier of equal impedance. Unfortunately doing so would make those P-90's pretty dull sounding, so we find an amp that has a high impedance of upwards towards 1MΩ and those P-90's will sparkle like they're supposed to. Matching impedance is important with some devices, such as with RF to ensure maximum transfer of signal and minimize standing waves. I'll side with Focusrite and other manufacturers that cite the same regarding the effect impedance has on dynamic microphones.

Neumann likes the 5x mic Z criteria as well What Is Impedance? Neumann Home Studio Academy

And Shure hints at higher Z preamps Should I match impedances of my microphone to my mixer? | Knowledge Base | Shure Americas

 

[drtechno]

I'm not sure I disagree with focusrite? The deliberate mismatch of impedance surely reduces the current flowing which must impact on the source in some way as back EMF, as resistance to change, and in a dynamic without the buffer of the mic preamp, will lump load the capsule? That's always been my understanding - is it wrong? I'd never read the focusrite text, but it appears to be as I have always understood it. The DBX preamp is a fine piece of cheaper equipment, but I really cannot see that the transformer will improve it. Colour it, modify it, etc, but not improve. Transformers are always compromise devices and the benefits of impedance conversion and isolation must be set against the detrimental effect of the transformer..

Sizing the phantom blocking caps incorrectly does effect the bandwidth of the microphone. And it could be as simple as changing them to the correct size (maybe). The mic in question (sm58) has a transformer in it. This causes a fixed point of resistance on its secondary. Now on transistorized input preamps, the requirements for proper amplification it needs the source voltage as well as its current. Isolated gate devices (moset, fet, or even tube) only require a voltage potential from the source signal.

The dbx mic pre went through major design change and it seems that they changed the preamp section without redoing the input circuit correctly. Beyond that, there are other design concerns like insert isolation and insert gain structures, and the output is not a real balance connection. The 286s mic preamp section is the modern circuit equivalent of a Neve 1073, which is really a cheap, poor quality circuit that actually got new wings because the transformers gets around most of the design flaws with that circuit. Going to transformer coupling will be a major improvement because capacitor coupling always induces a lot of response issues. Losses in a properly selected transformer will be far less compared to the losses measured using capacitive coupling. Transformers are actually the closest physical device that are ideal for ac coupling.

If we were talking ultimate quality electronics, then chasing the nth degree of sonic coherence might, just might be justified, but in a £150 device, there will be far more compromises than a bit of impedance mismatching.
.

Well, a Neve 1073 is a $25 circuit that has the distorton of 10% with a signal to noise ratio of 40db if you dare to take one apart and set it up as a capacitive coupling circuit like the mic preamps we're discussing.

One of the reasons I didn't get my first job at the BBC was because I was weak on the electronics side, and misunderstood bias current and impedance in audio circuits. The rule at that time was source impedance must always be lower than the destination impedance, and back then we were using mics with 25Ohm impedance, and the input circuits were over 1K, can't remember the exact figure. Did the BBC get this wrong? Most common mics will be less than 1K, so an input impedance of 1.2K makes sense. I don't think any of my condensers are above that figure..

In electronics, the complex signal will maintain its phase and frequency response when both the source and the next stage is equal. To complicate this, in capacitor coupling design you have to take in consideration of the devices connected, so with this design consideration oversizing the capacitor to no more than 1/10th of the input impedance. The operator's "rule of thumb" was based off of this, however, it can't be applied to all mic preamps either.

The maximum bandwidth thing you mentioned - I don't understand. Loading dynamics has always been considered bad - is this wrong too?

yes, everything gets a load regardless if it is a condenser or a dynamic. Electrically, the microphone's coupling are either reactive coil transducer (dynamic transformer-less), transformer coupled (some dynamic and condensers) and capacitive coupled (transformer-less condenser). All three type's loading requirements can vary, but all of them will operate correctly 100% of the time if the impedances are matched.

Forgive me, but some modern flawed thinking appears to be creeping in somewhere - care to explain, and perhaps I can go off and read about it, but my 'audiophile warning, Will Robinson" appears to have activated.

I love "Lost In Space" that was my favorite show back in the day.......

 

[drtechno]

So a guitar PU is similar in function to a dynamic mic. A Gibson P-90 PU has an impedance of 7.47 kΩ at 6.405 kHz. By what you're stating to get 'maximum bandwidth' (widest frequency range) the P-90 should be terminated to an amplifier of equal impedance. Unfortunately doing so would make those P-90's pretty dull sounding, so we find an amp that has a high impedance of upwards towards 1MΩ and those P-90's will sparkle like they're supposed to. Matching impedance is important with some devices, such as with RF to ensure maximum transfer of signal and minimize standing waves. I'll side with Focusrite and other manufacturers that cite the same regarding the effect impedance has on dynamic microphones.

Interesting, I measure 7.4Kohm of resistance with a DMM and when I connect the pickup to a Z-meter its 147Kohm. Its within the 10x rule for a capacitive coupled fet preamp of 1Mohm.

Neumann likes the 5x mic Z criteria as well What Is Impedance? Neumann Home Studio Academy

Well they adapted their own "5x rule" because not everyone fallows the rules and size the coupling cap to make the "10x rule" to apply

And Shure hints at higher Z preamps Should I match impedances of my microphone to my mixer? | Knowledge Base | Shure Americas

Well their paper is partially incorrect. I've never seen a degradation in signal when the impedances match. Also, capacitor coupled mic preamps can vary from 1000 ohms - 5000ohms on their input z.

any loss greater than 1db should be reviewed for design consideration. A 3db of loss is 1/2 of the source signal and all losses like this are not equal across the frequency range.

Also, here's a better definition of Electrical Impedance: Electrical impedance, the ratio of the voltage phasor to the electric current phasor, a measure of the opposition to time-varying electric current (AC) in an electric circuit.

But this definition might be more digestible for the general audience:

the effective resistance of an electric circuit or component to alternating current, arising from the combined effects of ohmic resistance and reactance.

or we can take a more mathematical definition:

the total opposition to alternating current by an electric circuit, equal to the square root of the sum of the squares of the resistance and reactance of the circuit and usually expressed in ohms.

 

[drtechno]

well, its not really changing the input impedance of the microphone preamp per se. Its only changing the termination resistance that is bridging between a high input z preamp and the low z source. The tonal changes are from the bandwidth response change that is caused by the Xc of the coupling capacitor shifting its roll off points.

Also, the design of these is with a FET input op amp or discrete FET input, that has an input z of 47K-1M all that is required for them is for the voltage to develop across this terminating resistance.

The transformer-less condenser mic fallows a slightly different rules, as its looking for a terminal resistance for its amplifier that is inside. So changing its termination resistance will only raise and lower its noise floor as it's load changes. The microphone preamp's coupling cap should be big enough value wise to be pose an insignificant change in total coupling Xc from the output coupling capacitor being in series with the preamp input caps.

 

[arcaxis]

Interesting, I measure 7.4Kohm of resistance with a DMM and when I connect the pickup to a Z-meter its 147Kohm. Its within the 10x rule for a capacitive coupled fet preamp of 1Mohm.

At what frequency does your impedance meter test at?

 

[drtechno]

but I guess you guys don't know about changing the input termination to reflect a better match of impedances. Granted it is a patch, but it does work.

The formula for this bridging would be:

the reciprocal of bridging resistance= the reciprocal of the optimal fixed bridging source load - the reciprocal of input impedance :

1/Br= (1/Rs)-(1/Rin)

for a sm57, the actual impedance is 310 ohms, the optimal bridging load is 1/2 (actual z) + actual Z. This would be (1/2(310))+310 or 465 ohms. The focusrite is 3K input. so:

1/Br=(1/465)-(1/3000) = 1/Br= (.00215)-(0.00033) = 1/Br = 0.00182 which Br= 1/0.00182 which is 549.45 ohms. at 1% it would be a 549 ohm resistor.

So if you stick this 549 ohm resistor across pins 2 and 3 in a mic cord, and use this cord with your sm57 or sm58, you will get the maximum bandwith transfer with your mic with a 2i2 mic pre. now if you don't have readily access to 1% resistors, you could use anything from 520-620 ohms to get a similar effect.

 

[drtechno]

At what frequency does your impedance meter test at?

View attachment 100952

Pink noise so I can get the average.