High noise, low signal on NTG4+ - would a mixer help?

[johnr10]

Hi,

I have an NTG4+ which I use either with a TD70 for camera work, or a Roland Rubix 24 interface to go into the computer. The problem I am having is I have to put the input gain almost max to get a decent signal - even if I am talking into it like an SM58... (which actually, I'm tempted to get instead - this is for podcasting use and I want that close-up sound).

Obviously, when the gain is so high, there is a lot of background noise and the mix is even picking up noises from downstairs like my wife and child laughing etc (not at me, I hope!)

Just for the record as I know this will be asked. Yes, 48V is on, no, the -10db pad isn't on Rubix 24 is powered by a 5v power supply, not via USB. I do have a computer in the room, but there's not too much I can do about that at the moment, and it seems odd that I'd have to crank the input gain in the 24 more or less to max to get a decent signal in.

Would a small mixing desk with decent pre-amps fix this? I want something with a USB interface - and ideally more than 2 channels. Money is tight at the moment so I'm considering the Zedi-10 (not the FX) which has 4 in/4 out USB and I can get for 205 euros locally. Would the mic pre-amps on there be any better than the 'low-noise' pre-amps on the Rubix 24? I don't want to waste my money on the wrong solution (and really I want a desk with faders on each channel but ...)

Thanks
John

[Hugh Robjohns]

From your descriptions I suspect what you are hearing is acoustic noise rather than electronic noise.

The NTG4+ has a pretty high sensitivity at 25mV/Pa, so you really shouldn't need a massive amount of preamp gain if you're using the mic up very close to your mouth.

I know you're using phantom, and I see the Rubix specs say it is 48V (not all are!) but that it is restricted to 6mA which is below the full spec level of 10mA. I doubt the NG4+ needs more than that, but some mics do... so it might be worthwhile switching phantom off and running the mic using its internal battery option just to see if it makes any difference.

I can't find any specs for the preamp EIN figure, so I don't know ho noisy it really is.

[johnr10]

From your descriptions I suspect what you are hearing is acoustic noise rather than electronic noise.

The NTG4+ has a pretty high sensitivity at 25mV/Pa, so you really shouldn't need a massive amount of preamp gain if you're using the mic up very close to your mouth.

I know you're using phantom, and I see the Rubix specs say it is 48V (not all are!) but that it is restricted to 6mA which is below the full spec level of 10mA. I doubt the NG4+ needs more than that, but some mics do... so it might be worthwhile switching phantom off and running the mic using its internal battery option just to see if it makes any difference.

I can't find any specs for the preamp EIN figure, so I don't know ho noisy it really is.

Yes, I think it is acoustic noise, I've had digital noise on the rubix 24 before but using external power and the ground lift fixed that. It seems to be because I have to have the input gain SO high to get a decent level.

I didn't know that about the current restriction on the Rubix 24. I can't find the current specs on the Zedi-10 preamps. As they boast they are gs preamps, I'd like to think they'd supply 10mA but without specs, who knows.

Good idea trying it from the battery. It's on charge now (I don't normally use it charged).

I have just found a very very very old Behringer Tube Ultragain Mic100 in my drawers of crap - don't laugh - and wondering if it might be worth testing that to see if it makes any difference although I suspect it will just add more noise

Thanks

ps. the RIGHT answer was 'Yes, the mixer will fix your problems, go buy it now, in fact, buy the one you really want no matter how much it costs' ...

[Sam Spoons]

I'm wondering how the mic can be picking up excessive noise from people downstairs when you are using it close up (with an SM58 1-2" from your lips is it's natural habitat but you would not want to be quite that close with the Rode)? If you speak at normal levels the wanted audio (i.e. your voice) should be a lot louder than the background. Do you have a recording you could post to illustrate the problem you are having?

[johnr10]

Actually, just read the NTG4+ specs and it only needs 4.8 mA so unlikely to be that. I'll still try of course.

Thanks

[Hugh Robjohns]

Actually, just read the NTG4+ specs and it only needs 4.8 mA so unlikely to be that.

Didn't expect it was the problem, but you have to rule things out.... Another possibility is that the 48V phantom isn't actually 48V. That happens surprisingly often. Again, running on internal battery will rule that possibility out too...

With a mic having a 25mV/Pa sensitivity and located only a few inches from the mouth, with normal speech you really shouldn't need more than about 40dB of gain. Having to run the preamp flat out doesn't sound right at all.

Have you checked with a different XLR cable in case the original cable is one-legged (one of the signal wires broken)?

[johnr10]

Actually, just read the NTG4+ specs and it only needs 4.8 mA so unlikely to be that.

Didn't expect it was the problem, but you have to rule things out.... Another possibility is that the 48V phantom isn't actually 48V. That happens surprisingly often. Again, running on internal battery will rule that possibility out too...

With a mic having a 25mV/Pa sensitivity and located only a few inches from the mouth, with normal speech you really shouldn't need more than about 40dB of gain. Having to run the preamp flat out doesn't sound right at all.

Have you checked with a different XLR cable in case the original cable is one-legged (one of the signal wires broken)?

I've tried two cables, and also tested them with a multimeter and they are fine.

Running with the battery has made no real difference. There's possibly a bit of electrical hum with the '48v' from the Rubix, which is measuring at 44.3v

[johnr10]

I'm wondering how the mic can be picking up excessive noise from people downstairs when you are using it close up (with an SM58 1-2" from your lips is it's natural habitat but you would not want to be quite that close with the Rode)? If you speak at normal levels the wanted audio (i.e. your voice) should be a lot louder than the background. Do you have a recording you could post to illustrate the problem you are having?

I've just made some test recordings. I have to rush out to view a house. Can I message you with a link to them when I get back, rather than post them? Thanks

[Sam Spoons]

Better to post them so the 'hive mind' can get to work on the problem

[johnr10]

Better to post them so the 'hive mind' can get to work on the problem

Fair enough!

Ok, so I'm recording this on the ngt4+, connected to a Rubix 24. That is connected to the computer via USB. There are no options on the computer to change any settings for the input or output of the Rubix.

It's going in to Logic Pro X, recorded on a new audio track, no settings changed except to change the track input from 1&2 to only 1.

First file - https://workdrive.zohopublic.eu/externa ... F4l-lU5Xxz

It starts very quietly (on 50% input gain). Mic is powered by the '48v' from the Rubix, and I'm using this XLR cable - https://www.amazon.co.uk/dp/B016D50A1Y/ ... UTF8&psc=1

The waveform is basically flat at 50% input gain, and ok at 100% - with the obvious issues that brings.

I mentioned before I had tested it with two cables, but they are both the same brand - bought a year apart. Both are 3m (which is longer than I need but not too long to cause this problem!)

The mic is on a cheap Neweer arm which is attached to a shelf on the (solid) wall, isolated from my desk. The mic is sitting in a Rode SM3-R shock mount (https://www.amazon.co.uk/R%C3%98DE-SM3- ... 01LVZUCJM/)

The second file - https://workdrive.zohopublic.eu/externa ... F4k-lU5Xxz

In this file I say the door is closed, it wasn't. It was open, sorry. This file also tests it with 48v from the Rubix and then the internal battery of the mic.

Really hoping I am doing something very stupid (or that I can use this as a good excuse to buy a mixer )

Thanks in advance. Be warned, again, they start low at 50% gain, but do go loud when I go to 100% gain so be prepared for it to go loud and for noise.

Thanks
John

[James Perrett]

The first file makes me think that most of your noise is acoustic. Our ears are very good at ignoring background noise but, when listening to a recording, the noise becomes much more apparent.

The noise on the very last segment (on battery power) sounds different and much more like preamp noise. Was the vacuum cleaner running at the time?

Do you have a dynamic mic that you can try at the same settings to see if the noise level is the same?

[johnr10]

The first file makes me think that most of your noise is acoustic. Our ears are very good at ignoring background noise but, when listening to a recording, the noise becomes much more apparent.

The noise on the very last segment (on battery power) sounds different and much more like preamp noise. Was the vacuum cleaner running at the time?

Do you have a dynamic mic that you can try at the same settings to see if the noise level is the same?

No, unless I have something really crap somewhere from years ago, I don't think I have another mic. I was considering getting an SM58 which I guess would do the job of testing it, but if I then have to spend more money on a mixer with better pre-amps because it's something else...

I'm sorry, I can't remember if the vac was running then. Very possibly. But it's a silly robot thing and I didn't turn it on and it has a mind of its own when it decides to stop and start sometimes But even if it was, it was downstairs.

But what really concerns me is the levels. Having to have the gain so high to register a strong signal that I can hear things in other rooms for a condenser seems wrong (I'm fair from an expert as I'm sure is apparent), but the ntg4+ shouldn't pick up much from behind, and the gain is so loud that when i lightly touched soft tissue behind the mic, it was crystal clear?!

Thanks

[Sam Spoons]

As far as I can tell you issue is simply the ratio of background noise to wanted audio. The second clip reveals what appears to be preamp noise (unless, as James suggests the vacuum cleaner was running in the room) so that discounts the battery option as a cure.

The only way to improve things is to get more voice or less background (or both) into the mic. So to improve the first, get closer to the mic and speak louder, to reduce the second, close the door, make door seals so the door doesn't leak noise and/or record only when your wife and kids are not in the house.

I'd be inclined to do both, make sure external sound leaking into you room is minimised by sealing the doo and ensuring it is closed when recording and get closer to the mic. Then think about some acoustic treatment around your recording position, a close mic will help and your test recordings are not too bad but a couple of duvets hung around you when you record will help absorb any reflections off the walls (duvets are tricky to set up to remove ceiling reflections but it's just a matter of achieving the best compromise you can.

But what really concerns me is the levels. Having to have the gain so high to register a strong signal that I can hear things in other rooms for a condenser seems wrong (I'm fair from an expert as I'm sure is apparent), but the ntg4+ shouldn't pick up much from behind, and the gain is so loud that when i lightly touched soft tissue behind the mic, it was crystal clear?!

See above, it's not the high gain that's the problem it's the fact that you voice is generating -35dB (at your 50% gain setting) and the background noise is at -72, when you turn the gain up to 100% the voice is generating -3dB and the background -27dB. Turning the gain up or down will turn the voice and the background equally.

Finally, to add if you double the distance between the source and the mic you will see a 6dB decrease in level so going from 2" away to 8" away will decrease the voice level by 12dB, when you then turn up the gain to get a similar level into the recorder the background noise (which hasn't got further from the mic) also comes up by 12dB.

[johnr10]

As far as I can tell you issue is simply the ratio of background noise to wanted audio. The second clip reveals what appears to be preamp noise (unless, as James suggests the vacuum cleaner was running in the room) so that discounts the battery option as a cure.

The only way to improve things is to get more voice or less background (or both) into the mic. So to improve the first, get closer to the mic and speak louder, to reduce the second, close the door, make door seals so the door doesn't leak noise and/or record only when your wife and kids are not in the house.

I'd be inclined to do both, make sure external sound leaking into you room is minimised by sealing the doo and ensuring it is closed when recording and get closer to the mic. Then think about some acoustic treatment around your recording position, a close mic will help and your test recordings are not too bad but a couple of duvets hung around you when you record will help absorb any reflections off the walls (duvets are tricky to set up to remove ceiling reflections but it's just a matter of achieving the best compromise you can.

Thanks. I do already have some acoustic treatment in the room (the difference between when it was bare and now is *stark*). I also have two thick panels to cover two lwindows but they are not hanging up at the moment but they make a big difference too.

The vac was downstairs, not in the room btw.

I wouldn't normally record with the door open, but what concerns me is the signal is so low. It needs to be at 100% gain to get a decent waveform. At 50% it's basically flat, at 75%, it's still low. And it's at 75% the noise really starts showing. Yes, I could go closer - and I have done during some of this testing in the last few days - but it's still noisy. Maybe I'm just being overly fussy?

I've had this mic about 18 months or so, done quite a few videos with it - with the mic at least 12 inches away, and not really had a problem so I'm really confused as to what is going on now (although in the past I have recorded the audio on my 70D, not straight into the computer via the Rubix)). Perhaps the noise has always been there but I've just not noticed, although that seems hard to believe.

Now I'm going more for a podcasting setup (with video, but I want the audio to be more 'radio like' rather than just clean). I did try it in to the 70D and back out to the Rubix at the weekend, but no matter what I tried I couldn't get an acceptable signal and as this is for podcasting rather than making videos, I need it to go into the computer live, not onto the sd card.

Perhaps I'm best getting a dynamic mic instead (well, not instead, I'll keep the ntg4+ for other uses of course).

I'd get the SM7B if I could afford it but I can't justify it at the moment. The SM58 would be fine probably but it simply looks crap - and this is video too, so sadly that is a factor. Maybe something like a Podmic or an AT2020 xlr?

Thank you for your time (everyone), it really is appreciated.

[johnr10]

But what really concerns me is the levels. Having to have the gain so high to register a strong signal that I can hear things in other rooms for a condenser seems wrong (I'm fair from an expert as I'm sure is apparent), but the ntg4+ shouldn't pick up much from behind, and the gain is so loud that when i lightly touched soft tissue behind the mic, it was crystal clear?!

See above, it's not the high gain that's the problem it's the fact that you voice is generating -35dB (at your 50% gain setting) and the background noise is at -72, when you turn the gain up to 100% the voice is generating -3dB and the background -27dB. Turning the gain up or down will turn the voice and the background equally.

Finally, to add if you double the distance between the source and the mic you will see a 6dB decrease in level so going from 2" away to 8" away will decrease the voice level by 12dB, when you then turn up the gain to get a similar level into the recorder the background noise (which hasn't got further from the mic) also comes up by 12dB.[/quote]

Ah, did you add this after you posted and as I was replying?

Yes, I'm aware the background noise will rise with the signal, I just don't understand why there is so much background noise, and why the signal is so weak in the first place?

As I said in my last reply, i've done a lot of videos with this mic at 12 inches or so away, and never noticed a problem before. It's just I want that close 'radio' sound now and trying to get that has brought up these problems. When I recorded in the past on the 70D (sorry, I have a feeling I typed DR70 or something before?!) and imported from the sd card, I never noticed a weak waveform or lots of noise..?!

[Sam Spoons]

Edited to add :-

Weak waveform is not an issue, if you record at 24 bit you can just increase the gain digitally in post.

The signal is 'weak' because you are 12" from the mic and not speaking loudly. It's not a problem in itself only because there is noise in the house.

To get a 'close' radio sound you need to be close to the mic, in a quiet, well treated studio.

[/edit]

TBH, changing the mic isn't going to help,* you will still have the same ratio of wanted to unwanted sound unless you can get closer to the mic. The SM58 and SH7B are designed to be used 'up close and personal' i.e. a couple of inches away or less, if you can do that you'll see a big improvement (24dB between 12" and 1½"). But I'm afraid the only other option is to shut the wife and kids up when recording.

* A 'shotgun' mic might improve things but I have little experience of them, what I have is in a live context and none of it was good.

[zenguitar]

but what concerns me is the signal is so low. It needs to be at 100% gain to get a decent waveform. At 50% it's basically flat

Others might be able to help more, but I think this is a possible key to the problem. It is not uncommon for people to think that they have a poor recording because they can't view the waveform clearly on the DAW screen. And instead of listening properly and managing the gain staging through the DAW, they rely on what they see and start increasing the gain.

Andy

[James Perrett]

With modern budget mic preamps it is perfectly normal to have to have the gain at three or four o'clock on the dial. This isn't a problem. The reason is that the manufacturers try to make things as cheaply as possible so they avoid the use of any additional controls that could bump the price up. In the old days a mic preamp would have had a pad button but these days even a pad button is too expensive so the gain control has to cover the whole range of possible inputs. One by-product of this is that a potentiometer with a special characteristic should be used for this gain control but these are more expensive than a standard pot so many manufacturers just use a standard pot which means that all the gain is bunched up at one end.

If you really aren't happy with this then I'd suggest buying a different interface rather than using a new mixer with the one you have. The reason I say this is that the mic input specification is very vague and I suspect that Roland haven't paid much attention to the mic inputs. If you want decent mic inputs then I'd suggest the Audient iD range or, if you can afford it, something from RME.

Edit to add: As others have said, a healthy digital level is averaging around -18dBFS with peaks around -12dBFS. If you are seeing anything over -6dBFS you are probably recording too high.

[johnr10]

With modern budget mic preamps it is perfectly normal to have to have the gain at three or four o'clock on the dial. This isn't a problem. The reason is that the manufacturers try to make things as cheaply as possible so they avoid the use of any additional controls that could bump the price up. In the old days a mic preamp would have had a pad button but these days even a pad button is too expensive so the gain control has to cover the whole range of possible inputs. One by-product of this is that a potentiometer with a special characteristic should be used for this gain control but these are more expensive than a standard pot so many manufacturers just use a standard pot which means that all the gain is bunched up at one end.

If you really aren't happy with this then I'd suggest buying a different interface rather than using a new mixer with the one you have. The reason I say this is that the mic input specification is very vague and I suspect that Roland haven't paid much attention to the mic inputs. If you want decent mic inputs then I'd suggest the Audient iD range or, if you can afford it, something from RME.

Edit to add: As others have said, a healthy digital level is averaging around -18dBFS with peaks around -12dBFS. If you are seeing anything over -6dBFS you are probably recording too high.

Thanks all.

Just on the mixer, I was thinking of one like the Allen & Heaht Zedi-10 or similar, with a usb interface (that is 4x4 which is why I am drawn to it, but there are a lot more 2x2 choices). If I did get a mixer it would be n the hope the pre-amps are better, and i'd stop using the Rubix 24 for input.

[Sam Spoons]

Your problem is not to do with your preamps, sure you might be able to improve on the ones you have but more gain will not solve your problem. That is purely related to the relative levels of voice and background. That is what you need to change.

It has been said on her several time before that a better mic will be better at picking up the unwanted audio as well as the wanted audio.

[johnr10]

Yes, thanks.

I did some decibel readings using an app on my P30 Pro (so not going to be perfect but hopefully relative). The phone was in roughly the same position as where the mic is.

Also done some recordings using the Tascam 70D both with the computer on and the computer off. I only managed to listen to a few of them on the computer (imported into Logic) before being told to get my arse downstairs... But the ones without the computer were perfectly silent - even when the gain in the menu was set to high, with the input gain knob at a decent level. I've never been able to use it at high due to the background noise ... the PC! And it's still decent at mid, and not totally terrible at low if the knob is 100%.

With the computer on, they were still noisy - perhaps a bit less, but I've not actual been able to compare yet.

Background db level from the monitor app when the computer was on was between Min 31, Avg 34, Max 38, and when the fans kick in more Min 45, Avg 46, Max 49. With the computer off, it went down to Min 22, Max 35, Avg 31.

It has a beefy graphics card with it's own fans, and the cpu is overclocked because of the video work I use it for but I may revert to normal and then I can at least get the main fans in 'silent' mode (it's actually liquid cooled on the cpu but the other fans are still kicking in at the moment and can get noisy).

Bottom line is - surprise surprise - you are all correct, it is background noise, specifically from the PC (although my HS8's have a bit of noise too - connected to the Rubix).

So ideally I'd put the computer in another room if possible - I'll look in to that - if not I'll look at isolating it more (and removing the overclocking so the fans are quieter). We're house hunting at the moment so I'm not going to go crazy with moving to another room - leave that for the new house.

And I'm probably better getting a dynamic mic for this use as that will make quietening the computer less of an issue. I hate the look of the SM58 - yes, I know, it shouldn't matter but this is on video as well as just audio recording... but at least it is tried and tested... If anyone has any suggestions, please shout.

The other issue this has shown is the noise from the HS8's. It's not loud and didn't affect the decibel app readings. But I'll double check that is coming from the Rubix 24 and if so, I'll probably get the A&H Zedi-10 and hope it is better (at least it is 4x4 USB). There is also a quiet, but audible, whine from the Rubix 24 itself for some reason.

Thanks again for all your help, it's really been appreciated and, one way or another, I'll get the background noise removed but at least now I know what it is!

[Sam Spoons]

The Zed 10 is, on paper, quieter than the Rubix but the specs are not particularly clear (at least not to me at this time of night ) but a whine from the Rubix seems likely to be something earth related rather than internally generated. More investigation needed I think to track down the source of the noise (as you have already started doing). Spending money to solve a problem that is not directly related to the kit being replaced seems a bad idea to me.

[johnr10]

The Zed 10 is, on paper, quieter than the Rubix but the specs are not particularly clear (at least not to me at this time of night ) but a whine from the Rubix seems likely to be something earth related rather than internally generated. More investigation needed I think to track down the source of the noise (as you have already started doing). Spending money to solve a problem that is not directly related to the kit being replaced seems a bad idea to me.

yes,I agree but I would like a mixer and especially a multi usb one,even though it's only 4x4 .

[Sam Spoons]

That's fair enough, and a perfectly good reason for buying something just as long as you buy it knowing that it may not cure your background noise problem. BTW, according to the A&H website the Zed 10 does not have 4x4 I/O over USB only stereo.

And I'm probably better getting a dynamic mic for this use as that will make quietening the computer less of an issue.

No it won't, a dynamic mic will be less sensitive than your capacitor mic so, if you work at the same distance, you will need more gain which will bring up the level of background noise too. dynamic mic will be better than your NTG4+ for working close up but it can't change the ratio of wanted to unwanted audio if you don't also work much closer to the mic, only working closer will help improve that ratio*.

If you want similar polar characteristics in a 'stage vocal' type mic that is designed to be used close up consider a Shure Beta58A which has a supercardioid pickup pattern.

* The NTG4 is described as a 'super cardioid' so will have a tighter pickup pattern than an SM58/SM7B, that means it will let you work a little further away but still retain the same ratio of voice to background noise. The downside is that it also picks up more noise/sound from anything directly behind the mic than a cardioid. But, see below.

A knowledgable friend read this thread and added the following comment :-

The NTG4+ is a shotgun mic and I really wouldn't use one for the application the O/P is trying to. It's a good mic - I used to own one. However, I would never have used it for close-up voice work - especially indoors.

He could try a close-up hypercardioid, although proximity effect can be marked with some of them. For spoken word you really do need silence... a good mic/signal-chain will mercilessly expose any noises-off - however faint.

(my bold for emphasis)

https://www.shure.eu/musicians/discover/educational/polar-patterns

[johnr10]

That's fair enough, and a perfectly good reason for buying something just as long as you buy it knowing that it may not cure your background noise problem. BTW, according to the A&H website the Zed 10 does not have 4x4 I/O over USB only stereo.

And I'm probably better getting a dynamic mic for this use as that will make quietening the computer less of an issue.

No it won't, a dynamic mic will be less sensitive than your capacitor mic so, if you work at the same distance, you will need more gain which will bring up the level of background noise too. dynamic mic will be better than your NTG4+ for working close up but it can't change the ratio of wanted to unwanted audio if you don't also work much closer to the mic, only working closer will help improve that ratio*.

If you want similar polar characteristics in a 'stage vocal' type mic that is designed to be used close up consider a Shure Beta58A which has a supercardioid pickup pattern.

* The NTG4 is described as a 'super cardioid' so will have a tighter pickup pattern than an SM58/SM7B, that means it will let you work a little further away but still retain the same ratio of voice to background noise. The downside is that it also picks up more noise/sound from anything directly behind the mic than a cardioid. But, see below.

A knowledgable friend read this thread and added the following comment :-

The NTG4+ is a shotgun mic and I really wouldn't use one for the application the O/P is trying to. It's a good mic - I used to own one. However, I would never have used it for close-up voice work - especially indoors.

He could try a close-up hypercardioid, although proximity effect can be marked with some of them. For spoken word you really do need silence... a good mic/signal-chain will mercilessly expose any noises-off - however faint.

(my bold for emphasis)

https://www.shure.eu/musicians/discover/educational/polar-patterns

Thanks Sam. I'm going to get a new mic anyway - as your friend says, the NTG4+ is not right for the use I want now - and I'd imagine it will be dynamic (I've just had some good news so I can spend a little bit more than I thought on it...). And then once I have that setup through the new mixer preamps, I can set to work trying to minimise the noise in the room.

Sorry, the mixer I'm talking about is the ZEDi-10 not the ZED-10. That is 4x4, the ZED-10 is only stereo (and the smaller channel options in the ZEDi range, such as the ZEDi-8 are also only stereo).

https://www.allen-heath.com/ahproducts/zedi-10/

But, again, with this good news (it's not great news btw, I've not won the lottery ) I may get something a bit better now. I need to consider some options.