Noisy Monitor Output on Focusrite Scarlett 18i8

[Angus Kerr[_3_]]

Hi

I finally started messing around with the Scarlett after upgrading my Linux kernel to 4.0, which gives the full mixer capability.

After I had got it going and was listening to a track I was working on, I decided to crank up the gain and hear the song at a mid pressure level, not ear splitting but so i could feel it.

After playback had ended, I was sitting and just fiddling with computer related stuff, and then I thought, what's that noise?

2 noises,

1: A high frequency whine at about 10k similar to that made by old CRT TV sets.

2: A low-mid 'chatty' noise that sounds like digital electronics working.

Both were independent of the volume know, i.e. same at 0 and full crank.

Noise was definitely from the Scarlett, all other channels on the monitor mixer were muted, and muting the Scarlett removed the noise.

@ Mike, I see you did a nice review of this unit. Did you notice any problems with the unit you tested.

Second question, if there is noise on the output D/A, how clean is the input A/D? will I get this on the analog lines in?

Thanks

-A.

[geoff]

On 23/09/2015 11:09 p.m., Angus Kerr wrote:

>
> Second question, if there is noise on the output D/A, how clean is
> the input A/D? will I get this on the analog lines in?
>
> Thanks
>
> -A.
>


On a laptop ? They can be particularly noisy powering things by USB.
My Scarlett 2i2 (?) - dead quiet (laptop or desktop).

Th whine is possibly something to do with Linux. It appears to ellict a
lot of whining ;-)

geoffR

[Mike Rivers[_2_]]

On 9/23/2015 7:09 AM, Angus Kerr wrote:

> I finally started messing around with the Scarlett after upgrading my
> Linux kernel to 4.0, which gives the full mixer capability.
> After playback had ended, I was sitting and just fiddling with
> computer related stuff, and then I thought, what's that noise?

> 1: A high frequency whine at about 10k similar to that made by old
> CRT TV sets.

> 2: A low-mid 'chatty' noise that sounds like digital electronics
> working.

> Both were independent of the volume know, i.e. same at 0 and full
> crank.
> @ Mike, I see you did a nice review of this unit. Did you notice any
> problems with the unit you tested.

Those noises that you're describing sound like they're the result of
electrical noise from the computer getting onto the ground of the
Scarlett. The fact that you can't turn off the noise with the volume
control means that it's not likely that it's coming from inside the
Scarlett. The simplest thing to try first is a different USB cable
connecting the Scarlett to the computer. Sometimes a plug won't make a
good ground connection or the cable isn't well shielded. Or, since it's
not powered from the computer through the USB port, just unplug the USB
cable and see if the noise goes away.

Do you have a balanced connection between the Scarlett output and your
monitor chain? If so, you might try lifting the ground connection to see
if that has an effect. Do you hear the same noise when you listen on
headphones connected to the Scarlett's headphone jack?

When I'm tracking down a noise problem like this, I try to isolate the
suspicious unit as much as possible and still have it working.

I probably have a mic-in-to-monitor-out noise measurement in my review,
but unless I mentioned a noise problem, I didn't have one.

> Second question, if there is noise on the output D/A, how clean is
> the input A/D? will I get this on the analog lines in?

Try it and see. Connect a 150 ohm resistor between pins 2 and 3 of an
XLR plug, plug that into a mic input, turn the input gain up to a
reasonable setting and make a recording. Then play the recording and
look at the DAW meter. If everything's OK, you will see some noise with
peaks around -75 dBFS. That's not the quietest I've ever measured but
it's OK for the price.


--
For a good time, visit http://mikeriversaudio.wordpress.com

[Frank Stearns]

Angus Kerr > writes:

-snips-

>1: A high frequency whine at about 10k similar to that made by old CRT TV sets.

>2: A low-mid 'chatty' noise that sounds like digital electronics working.

The crappy wall-wart or line-lump switching power supplies shipped with most gear
can inject a LOT of crap into the ground plane. What you describe sounds exactly
like that -- at least the HF whine. The chatter is probably digital "data or
processing" hash getting into the system due to a feeble or broken ground somewhere.

If you haven't got a uniformly perfect ground (and I do mean perfect -- from nose to
tail of your system) this kind of junk will readily show up.

If the switcher PS is really cheesy (likely nowhere close to meeting USA FCC Part 15
rules on RFI) those damn things will also broadcast noise. Less than well-designed
gear elsewhere in the system can then pick it up.

Are there any 1/8" TRS jacks in use for any of your analog path? Another reason to
hate this connector is that they seem many times more prone to poor grounding (and
thus noise getting in) than even a mediocre 1/4" TRS.


>Second question, if there is noise on the output D/A, how clean is the input A/D?
>will I get this on the analog lines in?

Record some test tracks with no signal. Then look at the DAW wavefrom at high
magnification or use a software spectrum analyzer. You might well be horrified at
what you see.

For all of this it's time to check/upgrade cabling and certainly check/upgrade your
overall grounding scheme.

And be sure to unseat/reseat ALL your connectors, including power connectors going
into your devices.

As a last resort, audio transformers strategically placed will often solve the
issue.

Frank
Mobile Audio
--

[Angus Kerr]

-snip-
> The crappy wall-wart or line-lump switching power supplies shipped with most gear
> can inject a LOT of crap into the ground plane. What you describe sounds exactly
> like that -- at least the HF whine. The chatter is probably digital "data or
> processing" hash getting into the system due to a feeble or broken ground somewhere.

I must admit, I did not try another USB cable - I'm using the one that is supplied with the unit, it does seem rather thin to me. I also did not try unplugging the USB or just listening to the Headphone outputs. I just switched back to my old sound card which does not have these problems..

>
> If you haven't got a uniformly perfect ground (and I do mean perfect -- from nose to
> tail of your system) this kind of junk will readily show up.

I didn't suspect grounding. Will check.
>
> If the switcher PS is really cheesy (likely nowhere close to meeting USA FCC Part 15
> rules on RFI) those damn things will also broadcast noise. Less than well-designed
> gear elsewhere in the system can then pick it up.
>
> Are there any 1/8" TRS jacks in use for any of your analog path? Another reason to
> hate this connector is that they seem many times more prone to poor grounding (and
> thus noise getting in) than even a mediocre 1/4" TRS.

Only balanced TRS 1/4". I hate the 1/8th because it's impossible to solder nicely.


>
> >Second question, if there is noise on the output D/A, how clean is the input A/D?
> >will I get this on the analog lines in?
>
> Record some test tracks with no signal. Then look at the DAW wavefrom at high
> magnification or use a software spectrum analyzer. You might well be horrified at
> what you see.
>
> For all of this it's time to check/upgrade cabling and certainly check/upgrade your
> overall grounding scheme.

>
> And be sure to unseat/reseat ALL your connectors, including power connectors going
> into your devices.
>
> As a last resort, audio transformers strategically placed will often solve the
> issue.

Thanks,

-A,

[Angus Kerr]

On Wednesday, September 23, 2015 at 1:52:34 PM UTC+2, Mike Rivers wrote:
-snip->

Those noises that you're describing sound like they're the result of
> electrical noise from the computer getting onto the ground of the
> Scarlett. The fact that you can't turn off the noise with the volume
> control means that it's not likely that it's coming from inside the
> Scarlett. The simplest thing to try first is a different USB cable
> connecting the Scarlett to the computer. Sometimes a plug won't make a
> good ground connection or the cable isn't well shielded. Or, since it's
> not powered from the computer through the USB port, just unplug the USB
> cable and see if the noise goes away.

Will try this.

> Do you have a balanced connection between the Scarlett output and your
> monitor chain? If so, you might try lifting the ground connection to see
> if that has an effect. Do you hear the same noise when you listen on
> headphones connected to the Scarlett's headphone jack?

Didn't listen to the headphone jack, but thanks for the suggestion.

> > Second question, if there is noise on the output D/A, how clean is
> > the input A/D? will I get this on the analog lines in?
>
> Try it and see. Connect a 150 ohm resistor between pins 2 and 3 of an
> XLR plug, plug that into a mic input, turn the input gain up to a
> reasonable setting and make a recording. Then play the recording and
> look at the DAW meter. If everything's OK, you will see some noise with
> peaks around -75 dBFS. That's not the quietest I've ever measured but
> it's OK for the price.

Thanks I'll do that also.

Surprising that the USB device is noisier than my soundcard, sitting on the PCI bus for goodness sake. Oh, wait, I forgot this EMU 1212-M has 'mastering grade' D/A and A/D converters on it, according to the marketing blurb. So it must be good! I'm sure that I'll find these cards in all good mastering houses......

-A.

[JackA]

On Wednesday, September 23, 2015 at 7:09:31 AM UTC-4, Angus Kerr wrote:
> Hi
>
> I finally started messing around with the Scarlett after upgrading my Linux kernel to 4.0, which gives the full mixer capability.
>
> After I had got it going and was listening to a track I was working on, I decided to crank up the gain and hear the song at a mid pressure level, not ear splitting but so i could feel it.
>
> After playback had ended, I was sitting and just fiddling with computer related stuff, and then I thought, what's that noise?
>
> 2 noises,
>
> 1: A high frequency whine at about 10k similar to that made by old CRT TV sets.

I think that was more in the range of 15.75 kHz. Let's see, was it..
525 lines x 60 Hz horizontal oscillator/ 2 (passes)?

Jack
>
> 2: A low-mid 'chatty' noise that sounds like digital electronics working.
>
> Both were independent of the volume know, i.e. same at 0 and full crank.
>
> Noise was definitely from the Scarlett, all other channels on the monitor mixer were muted, and muting the Scarlett removed the noise.
>
> @ Mike, I see you did a nice review of this unit. Did you notice any problems with the unit you tested.
>
> Second question, if there is noise on the output D/A, how clean is the input A/D? will I get this on the analog lines in?
>
> Thanks
>
> -A.

[Angus Kerr]

-snip-
Or, since it's
> > not powered from the computer through the USB port, just unplug the USB
> > cable and see if the noise goes away.
>
> Will try this.

Unplugging USB and both hf whine and chatty mid range noise go away.


>
> > Do you have a balanced connection between the Scarlett output and your
> > monitor chain? If so, you might try lifting the ground connection to see
> > if that has an effect. Do you hear the same noise when you listen on
> > headphones connected to the Scarlett's headphone jack?
>
> Didn't listen to the headphone jack, but thanks for the suggestion.

Headphone jack has chattiness, but not hf whine.

Disconnected from USB on my Studio Desktop, and connected to a Laptop. Nothing, dead quiet.

So the source of my pain lies in my Desktop. Maybe I need to change the Power Supply?

> > Try it and see. Connect a 150 ohm resistor between pins 2 and 3 of an
> > XLR plug, plug that into a mic input, turn the input gain up to a
> > reasonable setting and make a recording. Then play the recording and
> > look at the DAW meter. If everything's OK, you will see some noise with
> > peaks around -75 dBFS. That's not the quietest I've ever measured but
> > it's OK for the price.
>
> Thanks I'll do that also.

Will revert once I've tried to figure out what the deal is with my Desktop. Sucks that there's a problem, but better I find out before tracking vocals.

I do wonder, what is the point of this huge hulking thing? Apart from the nice display that is.

-A.

[Mike Rivers[_2_]]

On 9/23/2015 1:50 PM, Angus Kerr wrote:
> Disconnected from USB on my Studio Desktop, and connected to a
> Laptop. Nothing, dead quiet.

OK, so the interface and Linux driver is OK. That's a good start.

> So the source of my pain lies in my Desktop. Maybe I need to change
> the Power Supply?

I think it's more likely with with the USB ground. See if wiggling the
USB connector in the socket on the computer makes things change (for
better or worse). Or try running a ground wire between the computer
chassis and the Scarlett chassis.


--
For a good time, visit http://mikeriversaudio.wordpress.com

[Scott Dorsey]

In article >, Mike Rivers > wrote:
>On 9/23/2015 1:16 PM, Angus Kerr wrote:
>> I did say something similar, and I was a great deal younger the last
>> time I had dealings with a CRT monitor or TV.
>
>Just think. Kids these days can't grow up hearing the horizontal
>oscillator frequency because TVs today don't have flyback transformers
>to act as tweeters.

It used to be fairly common to look at recordings and see a little peak
at the NTSC flyback frequency, from TV sweep somewhere in the building
leaking into the audio. Might be the monitor on the SSL automation terminal.
Might be a CCTV monitor so the guys in the control room can see the loading
dock. All kinds of things leaked sweep and it's very hard to notice at low
levels even if you have good HF hearing.

But now things are more insidious... we have switching supplies running at
fairly random frequencies and although LED monitors don't leak sweep, they
leak noise from the backlight inverter and you can't predict where that is
going to be.

So we still get little peaks turning up above the noise floor on recordings,
they just aren't predictable.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[[email protected]]

Yep
A spectrum analyzer allows you to see things you can't hear.
Mark

[Les Cargill[_4_]]

Angus Kerr wrote:
> Hi
>
> I finally started messing around with the Scarlett after upgrading my
> Linux kernel to 4.0, which gives the full mixer capability.
>
> After I had got it going and was listening to a track I was working
> on, I decided to crank up the gain and hear the song at a mid
> pressure level, not ear splitting but so i could feel it.
>
> After playback had ended, I was sitting and just fiddling with
> computer related stuff, and then I thought, what's that noise?
>
> 2 noises,
>
> 1: A high frequency whine at about 10k similar to that made by old
> CRT TV sets.
>
> 2: A low-mid 'chatty' noise that sounds like digital electronics
> working.
>
> Both were independent of the volume know, i.e. same at 0 and full
> crank.
>
> Noise was definitely from the Scarlett, all other channels on the
> monitor mixer were muted, and muting the Scarlett removed the noise.
>
> @ Mike, I see you did a nice review of this unit. Did you notice any
> problems with the unit you tested.
>
> Second question, if there is noise on the output D/A, how clean is
> the input A/D? will I get this on the analog lines in?
>
> Thanks
>
> -A.
>

Check all the sampling rates from end to end. I usually
see that problem on ADAT Lightpipe stuff ( a 44.1k
interface is connected to a 48k one ) but that's
what it sounds like.

--
Les Cargill

[Angus Kerr]

On Thursday, September 24, 2015 at 12:24:34 AM UTC+2, Les Cargill wrote:
> Angus Kerr wrote:
> > Hi
> >
> > I finally started messing around with the Scarlett after upgrading my
> > Linux kernel to 4.0, which gives the full mixer capability.
> >
> > After I had got it going and was listening to a track I was working
> > on, I decided to crank up the gain and hear the song at a mid
> > pressure level, not ear splitting but so i could feel it.
> >
> > After playback had ended, I was sitting and just fiddling with
> > computer related stuff, and then I thought, what's that noise?
> >
> > 2 noises,
> >
> > 1: A high frequency whine at about 10k similar to that made by old
> > CRT TV sets.
> >
> > 2: A low-mid 'chatty' noise that sounds like digital electronics
> > working.
> >
> > Both were independent of the volume know, i.e. same at 0 and full
> > crank.
> >
> > Noise was definitely from the Scarlett, all other channels on the
> > monitor mixer were muted, and muting the Scarlett removed the noise.
> >
> > @ Mike, I see you did a nice review of this unit. Did you notice any
> > problems with the unit you tested.
> >
> > Second question, if there is noise on the output D/A, how clean is
> > the input A/D? will I get this on the analog lines in?
> >
> > Thanks
> >
> > -A.
> >
>
> Check all the sampling rates from end to end. I usually
> see that problem on ADAT Lightpipe stuff ( a 44.1k
> interface is connected to a 48k one ) but that's
> what it sounds like.
>
> --
> Les Cargill

Unfortunately, this problem happens with just the unit connected. No other peripheral equipment. I was hoping to use this in conjunction with my Focusrite Octopre, since with the EMU 1212 I've only got 12 channels in (if you count an SPDIF coaxial as an extra chennel - god knows how I am going to sync that without a master clock. Ok for studio work, not so much for a live recording.

Cheers,

-A.

[Angus Kerr]

On Wednesday, September 23, 2015 at 9:37:38 PM UTC+2, Scott Dorsey wrote:
> In article >, Mike Rivers > wrote:
> >On 9/23/2015 1:16 PM, Angus Kerr wrote:
> >> I did say something similar, and I was a great deal younger the last
> >> time I had dealings with a CRT monitor or TV.
> >
> >Just think. Kids these days can't grow up hearing the horizontal
> >oscillator frequency because TVs today don't have flyback transformers
> >to act as tweeters.
>
I remember as a kid, that whine was LOUD. Didn't seem to bother my folks or grandparents at all. Now I know that didn't even hear it.

> It used to be fairly common to look at recordings and see a little peak
> at the NTSC flyback frequency, from TV sweep somewhere in the building
> leaking into the audio. Might be the monitor on the SSL automation terminal.
> Might be a CCTV monitor so the guys in the control room can see the loading
> dock. All kinds of things leaked sweep and it's very hard to notice at low
> levels even if you have good HF hearing.
>

Lol, a friend of mine who works on Foley audio and music (basically the whole soundtrack) for documentaries who is nearing 60, said he mixed the whole movie down when a young guy said 'what's that noise?" "What noise?" he said. Turns out there was a hf tone throughout almost the whole movie. Now he runs it through a spectrum analyser, just to be sure.

-A.

[geoff]

On 24/09/2015 10:51 p.m., Angus Kerr wrote:
> Les Cargill
> Unfortunately, this problem happens with just the unit connected. No other peripheral equipment. I was hoping to use this in conjunction with my Focusrite Octopre, since with the EMU 1212 I've only got 12 channels in (if you count an SPDIF coaxial as an extra chennel - god knows how I am going to sync that without a master clock. Ok for studio work, not so much for a live recording.
>
> Cheers,
>
> -A.
Are you getting multiple streams - maybe one 'direct monitoring' and
another via an alternate source with some latency, summing to weirdness.

geoff

[Angus Kerr]

On Thursday, September 24, 2015 at 11:24:02 PM UTC+2, geoff wrote:
> On 24/09/2015 10:51 p.m., Angus Kerr wrote:
> > Les Cargill
> > Unfortunately, this problem happens with just the unit connected. No other peripheral equipment. I was hoping to use this in conjunction with my Focusrite Octopre, since with the EMU 1212 I've only got 12 channels in (if you count an SPDIF coaxial as an extra chennel - god knows how I am going to sync that without a master clock. Ok for studio work, not so much for a live recording.
> >
> > Cheers,
> >
> > -A.
> Are you getting multiple streams - maybe one 'direct monitoring' and
> another via an alternate source with some latency, summing to weirdness.
>
> geoff

No, not in this case. I just connected the Scarlett, all by herself, all virgin like, and maybe she wasn't as clean as I would have hoped...