The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Focusrite Clarett 2Pre monitor output noise?
Old 14th July 2016
  #1
Lives for gear
Focusrite Clarett 2Pre monitor output noise?

Hey guys, hope you're all well. I'm having some drama with my new Clarett interface (which I got today).

I've recorded a video to demo the problem: https://goo.gl/photos/DcmY6QMi1wdxUKkFA

At this stage, I'm thinking of just taking it back and paying more for the UA unit, but I'd love to get your feedback. Have I perhaps done something wrong? Is it simply because my monitors are too loud?

Same issue on all 4 outputs across both my monitors.

I tried my MacBook Pro output with an RCA lead to the monitors and honestly found that to be quiet, and also tested with my Chord Mojo (really nice DAC / headphone amp) and it too is silent so I'm just not sure what's going on.

Your thoughts would be greatly appreciated guys

Thanks
Fotis

Last edited by fgimian; 14th July 2016 at 10:15 PM..
Old 14th July 2016
  #2
Lives for gear
 

Hi Fotis,

is that increased noise level steady? Eg. there's no influence to level of this noise by some computer activity?
Because if this will be the case, then it might be ground loop from computer.

This sometimes happens, when you use unbalanced audio connections at some device in setup, so signal ground and chassis ground gets connected, for example by use RCA or TS cables. Unfortunately it's bit more complex, depends also on construction of particular devices and requires to do couple of things to debug.
To exclude this kind of issue, I would try balanced TRS->XLR connection to your A7s (maybe you already have that) and do one test just with minimal setup..
That means single power strip for monitors and 2Pre PS, notebook temporarily powered from battery without any other peripheral with own PS (monitors, external HDDs). Then listen to the noise.. if it will be still there, it won't be ground loop.. if it will be good, you can try to connect notebook PS and peripherals (one by one to find the culprit).

Anyway, I've heard Clarett 2pre and it was very quiet and clean to me (published spec also confirms that).. So there's some chance, you've got a lemon or there's some setup issue like I've mentioned.

Michal
Old 14th July 2016
  #3
Gear Maniac
 
ilikefruit's Avatar
It doesn't sound like a ground loop to me. If I recall they are more humming than hissing.
Why not ask your vendor? Perhaps they have another unit you could try or compare with in the store?

Also, have you contacted Focusrite support? I chatted with them several times regarding my own Clarett (that got returned eventually) and they were very nice and knowledgeable. I'm sure they can decide from your movie if it is to be expected or not.
Old 14th July 2016
  #4
Lives for gear
 

Quote:
Originally Posted by ilikefruit View Post
It doesn't sound like a ground loop to me. If I recall they are more humming than hissing.
Why not ask your vendor? Perhaps they have another unit you could try or compare with in the store?
That's pretty common mistake, ground loop doesn't necessarily sound like classic 50/60 hum or its harmonics. But also at much higher frequency without apparent relation to mains.
Reason for that is, this can be related to changing ground potential caused by varying consumption of components in complex devices like computer, which is mainly related to its operation pattern.. So say USB microframe interval is 125us, which leads to varying consumption with frequency of 8k. Similarly other component with more complex operation patterns can cause bursts.. which someone might not initially address to ground loops. But if you isolate or separate grounds, issue disappear.
That's exactly problem, which someone describes like.. when I'm moving mouse or my graphics card is under load.. i"m hearing weird noises from my speakers.

Additionally with regards to ground loops and audio interfaces, sometimes other people immediately chime in with some recommendation for external ground isolator or transformer.. Which is certainly one way how to handle the situation, it has it place in some circumstances, but in many cases is better in long term view to really solve the root of issue (like picking more suitable way for equipment interconnection, replace badly designed device, fixing the room grounds).
Sometimes it requires some helper tools like different lifting boxes or cables or measurement devices.
Fortunately when we talking simple setups (like OP's) with notebooks, it's really easy to exclude most common source of those problems (computer power supply and its internal grounding design) by simple test when computer is battery powered. This will clearly indicate, if it is ground problem or not.

Michal
Old 14th July 2016
  #5
Lives for gear
Quote:
Originally Posted by msmucr View Post
Hi Fotis,

is that increased noise level steady? Eg. there's no influence to level of this noise by some computer activity?
Because if this will be the case, then it might be ground loop from computer.

This sometimes happens, when you use unbalanced audio connections at some device in setup, so signal ground and chassis ground gets connected, for example by use RCA or TS cables. Unfortunately it's bit more complex, depends also on construction of particular devices and requires to do couple of things to debug.
To exclude this kind of issue, I would try balanced TRS->XLR connection to your A7s (maybe you already have that) and do one test just with minimal setup..
That means single power strip for monitors and 2Pre PS, notebook temporarily powered from battery without any other peripheral with own PS (monitors, external HDDs). Then listen to the noise.. if it will be still there, it won't be ground loop.. if it will be good, you can try to connect notebook PS and peripherals (one by one to find the culprit).

Anyway, I've heard Clarett 2pre and it was very quiet and clean to me (published spec also confirms that).. So there's some chance, you've got a lemon or there's some setup issue like I've mentioned.

Michal
Hey Michal, thanks so much for your help. Yeah, the noise is actually steady mate, I don't believe it's a grounding issue as I am using balanced heavy duty cables (TRS -> XLR). I will definitely try to isolate the power though, great idea

Quote:
Originally Posted by ilikefruit View Post
It doesn't sound like a ground loop to me. If I recall they are more humming than hissing.
Why not ask your vendor? Perhaps they have another unit you could try or compare with in the store?

Also, have you contacted Focusrite support? I chatted with them several times regarding my own Clarett (that got returned eventually) and they were very nice and knowledgeable. I'm sure they can decide from your movie if it is to be expected or not.
Thanks a lot. Yeah I emailed Focusrite yesterday as well and just got this reply today:

Quote:
With regards to the tweeter hiss that you refer too. This is something that occurs with my Clarett through my Tannoy monitors too and is common behaviour within spec. Obviously the Clarett 2Pre, the same as any interface, has a professionally acceptable noise floor, but when that is outputted to speakers which then of course amplify this, it stands to reason that it will sound increased from the idling speaker tweeter hiss.
As mentioned, from what you've demonstrated in this video, this sounds like it is within spec.
I honestly really like the unit and Focusrite Control is simple and elegant so I'm in two minds about this.

If I turn down my monitors, the noise is not noticeable, but I'm just disappointed that the noise level is worse than my MacBoox Pro when the published spec should clearly provide better results. This leads me to believe that Focusrite have published false information on their website about real-world performance.

I will try to isolate the device and speakers right now using dedicated power and see if the problem persists.

Any further thoughts are welcome
Fotis
Old 14th July 2016
  #6
Lives for gear
P.S.: Here's another little demo showing that the Clarett is noisier than the MacBook Pro audio output

https://goo.gl/photos/q7WjqCkZJDWpf1BJA
Old 14th July 2016
  #7
Lives for gear
I just tried isolating just one monitor and the Clarett to 2 separate power points. All other power was removed completely. But the issue remains identical sadly.

The headphone output also has a channel imbalance at low levels which is really noticeable with sensitive headphones, but I was willing to let that slide.

Now to decide what I will do ... The 1.1 GB download from UA just to install drivers for their unit doesn't appeal to me, nor their lack of statement on Windows compatibility for their Apollo Twin Thunderbolt interface. I already own native plugins for everything I need and am not interested in being locked into any ecosystems (e.g. UAD).

So I'm a bit torn about what to do. I really need low latency which the Clarett provides, but I don't know if I trust the audio quality of the unit after seeing this.

Happy to take any further ideas / suggestions
Fotis
Old 14th July 2016
  #8
Lives for gear
 
Lumbergh's Avatar
 

Yeah this is not a valid test: MBPs output 6dBu at best, Claretts are specced at 18dBu.

You could test it again but this time set the A7s to 12dB louder for the MBP than the Clarett. And test while playing a silent sound file (the MBP's interface probably sleeps for power management.)

But ... Important bit! ... I do not recommend testing noise floor by using monitors and turning everything up. One glitch and you can wave goodbye to your hearing.


If you want to listen to the noise floor do it after calibrating your monitoring chain. If you can't hear a difference then ... it doesn't matter!



Sidenote: If you are concerned with noise you don't really want a wireless phone base station and what looks like a wifi router within 6" of your audio cables either
Old 14th July 2016
  #9
Lives for gear
Quote:
Originally Posted by Lumbergh View Post
Yeah this is not a valid test: MBPs output 6dBu at best, Claretts are specced at 18dBu.

You could test it again but this time set the A7s to 12dB louder for the MBP than the Clarett. And test while playing an empty sound file (the MBP's interface probably sleeps for power management.)


Important bit! I do not recommend testing noise floor by using monitors and turning everything up. One glitch and you can wave goodbye to your hearing.


If you want to listen to the noise floor do it after calibrating your monitoring chain. If you can't hear a difference then ... it doesn't matter!
Thanks for the advice and concern, you're right, I'm really careful though, but it is risky.

I was just thinking about the max output level of +18 dBu and to be honest, my old RME interface was switchable between different levels (-10 dBu, +4 dBU .etc). I absolutely don't need +18 dBU, which means I would be getting +14 dB of extra noise (compared to a regular +4 dBu) and since volume is controlled post D/A conversion, this translates to +14 dB of extra noise.

I have brought this up with Focusrite. I presume that this is fixed on the circuit at +18 dBu and the best solution is to simply turn down your monitors until the point that you can't hear the noise. But it really should be switchable.

The shortcuts taken on the build of these devices are now more apparent (no separate D/A conversion for headphone output, analog control of headphones which leads to the usual channel imbalances at low volumes, no switchable volume control for inputs and outputs).

I think I'll just go with your suggestion though, setting the volume lower on my monitors resolves the issue and is more than enough juice for me. I can't even turn it up to 3 o'clock without blowing out my ears, around 12 o'clock is about my regular listening level with this setup
Old 15th July 2016
  #10
Lives for gear
 
Lumbergh's Avatar
 

I wish I could get referral commission for this link ... Monitor Wizard | Sound On Sound

It might seem like some nerdy PITA exercise but its so worth doing
Old 15th July 2016
  #11
Lives for gear
Quote:
Originally Posted by Lumbergh View Post
I wish I could get referral commission for this link ... Monitor Wizard | Sound On Sound

It might seem like some nerdy PITA exercise but its so worth doing
Thanks, I will give this a go

Here's a little tip for the headphone output when used purely as a second source for mixing (I use a pair of open-back Audeze headphones for that):
  1. In Focusrite Control, go to Device Settings and set monitor Controls to All
  2. Calibrate the headphone output to be the same level as your monitors using the headphone volume (to make for an easier transition when you switch)
  3. Use the regular volume knob from now on to adjust both levels

This brings headphone volume control under control as the low level tolerance of the analog circuitry is overcome and the control is moved into the digital domain (similar to the monitor outputs).
Old 15th July 2016
  #12
Lives for gear
 
Lumbergh's Avatar
 

Yeah thats good, you're already calibrating relative monitoring levels

Now apply that to your mains to set an absolute loudness reference. That SOS article really is worth spending the time to understand and then apply. Basically MBP at full volume, Clarett at full volume, start at nothing and turn the A7s up to your comfortable monitoring level for the most dynamic stuff you listen to. Your ears and your music will both be grateful
Old 15th July 2016
  #13
Lives for gear
Quote:
Originally Posted by Lumbergh View Post
Yeah thats good, you're already calibrating relative monitoring levels

Now apply that to your mains to set an absolute loudness reference. That SOS article really is worth spending the time to understand and then apply. Basically MBP at full volume, Clarett at full volume, start at nothing and turn the A7s up to your comfortable monitoring level for the most dynamic stuff you listen to. Your ears and your music will both be grateful
Thanks a lot for the advice, this'll be my little project tonight

My dad actually has a dB meter at home in his garage too so I might pinch that and confirm my reference levels don't exceed 85 dB.
Old 15th July 2016
  #14
Lives for gear
 

Hi Fotis,

if it's completely steady, it's not a ground loop, which is good.
I agree with Lumbergh with regards to comparison of noisefloor between different devices, it should be level matched first to get some meaningful results. Max output level of most line-outs at notebooks is about 2Vrms, while 2pre has much higher output - 18dBu is about 6Vrms.
General rule of thumb for noisefloor level optimization is source high, input low. Anyway, it looks like you've already started to explore possibilities for optimization using input trim at your monitors.
With regards to low level channel imbalance or response of headphone amp pot, this is unfortunately inevitable thing for audio taper pots and only typical cures are either stepped attenuators or digital encoders controlling another gain element. But personally I wouldn't take that as critically, because as it affect all devices, unless beforementioned construction is used. Some parts have tighter tolerances, but it towards lowest levels at the divider, still it can be easily 10-15%. As you find out, it can be optimized by digital volume adjustment, which will move analog pot to the range with better channel tracking, without negative side effects.. when digital volume is properly dithered. I've used and recommended that many times also for other interfaces with DSP mixer, where this was possible (like higher Scarletts for example).

And generally to 2pre audio performance, as mentioned, I've heard it at friend's place together with other DAC and interface and it definitely doesn't sound noisy to to me. Also I don't really think, they can make up its specification (it will be very easily revealed with great shame for them). Possibly if you're interested, maybe you can do at least line loopback (TRS-TRS) measurement of it. Typically ADC noise floor and DR at most interfaces is bit lower than its DAC, so it shouldn't be masked by that. If you calibrate levels using some test tones (eg. with bypassed volume control -14dBFs sine at output is +4dBu, so if you set input sensitivity to match -14dBFs.. then you have approx unity gain.), then you can measure RMS noise floor of that.
Unfortunately I can't give you exact tip for OS X software procedures, because I'm using PC for that kind of stuff.. However I've seen, SignalScope or Futz Measure is used by Mac based friends.. and should have trial. Possibly also free REW might work for that. There are also some analyzer plugins to DAWs, but I'm not sure how each works with regards to correct RMS power calculation or if there's possible to use A-weighting curves for example. One last note.. during noisefloor measurement, broadband noise level is directly proportional to the bandwidth, so use 44.1k rate, because higher rates will naturally leads to higher readouts and this is closest to usual 22k bandwidth used in specs.

All the best,

Michal
Old 15th July 2016
  #15
Gear Maniac
 
ilikefruit's Avatar
Quote:
Originally Posted by soundpalace View Post
So I'm a bit torn about what to do. I really need low latency which the Clarett provides, but I don't know if I trust the audio quality of the unit after seeing this.
Myself, I had a Claret 8pre which I returned since there was an issue with my particular model of computer. Instead I got the RME 802, which is a breath of fresh air after the issue I was having. It's also a solid upgrade from my previous Apogee interface.

If you are reluctant to go to UA why not stay with RME? I read that you already own a RME interface, why the switch?
Wouldn't the UC or UCX be a good upgrade for you?
My Logic install reports 3,4 ms roundtrip latency and when tracking I havn't felt any latency like before, so I am really happy with the latency and I connect to the 802 via USB. I did not try Firewire yet. Maybe better?
Old 15th July 2016
  #16
Lives for gear
Quote:
Originally Posted by msmucr View Post
Hi Fotis,

if it's completely steady, it's not a ground loop, which is good.
I agree with Lumbergh with regards to comparison of noisefloor between different devices, it should be level matched first to get some meaningful results. Max output level of most line-outs at notebooks is about 2Vrms, while 2pre has much higher output - 18dBu is about 6Vrms.
General rule of thumb for noisefloor level optimization is source high, input low. Anyway, it looks like you've already started to explore possibilities for optimization using input trim at your monitors.
With regards to low level channel imbalance or response of headphone amp pot, this is unfortunately inevitable thing for audio taper pots and only typical cures are either stepped attenuators or digital encoders controlling another gain element. But personally I wouldn't take that as critically, because as it affect all devices, unless beforementioned construction is used. Some parts have tighter tolerances, but it towards lowest levels at the divider, still it can be easily 10-15%. As you find out, it can be optimized by digital volume adjustment, which will move analog pot to the range with better channel tracking, without negative side effects.. when digital volume is properly dithered. I've used and recommended that many times also for other interfaces with DSP mixer, where this was possible (like higher Scarletts for example).

And generally to 2pre audio performance, as mentioned, I've heard it at friend's place together with other DAC and interface and it definitely doesn't sound noisy to to me. Also I don't really think, they can make up its specification (it will be very easily revealed with great shame for them). Possibly if you're interested, maybe you can do at least line loopback (TRS-TRS) measurement of it. Typically ADC noise floor and DR at most interfaces is bit lower than its DAC, so it shouldn't be masked by that. If you calibrate levels using some test tones (eg. with bypassed volume control -14dBFs sine at output is +4dBu, so if you set input sensitivity to match -14dBFs.. then you have approx unity gain.), then you can measure RMS noise floor of that.
Unfortunately I can't give you exact tip for OS X software procedures, because I'm using PC for that kind of stuff.. However I've seen, SignalScope or Futz Measure is used by Mac based friends.. and should have trial. Possibly also free REW might work for that. There are also some analyzer plugins to DAWs, but I'm not sure how each works with regards to correct RMS power calculation or if there's possible to use A-weighting curves for example. One last note.. during noisefloor measurement, broadband noise level is directly proportional to the bandwidth, so use 44.1k rate, because higher rates will naturally leads to higher readouts and this is closest to usual 22k bandwidth used in specs.

All the best,

Michal
Thank you so much for all the valuable information Michal, really really appreciate your help and support

Quote:
Originally Posted by ilikefruit View Post
Myself, I had a Claret 8pre which I returned since there was an issue with my particular model of computer. Instead I got the RME 802, which is a breath of fresh air after the issue I was having. It's also a solid upgrade from my previous Apogee interface.

If you are reluctant to go to UA why not stay with RME? I read that you already own a RME interface, why the switch?
Wouldn't the UC or UCX be a good upgrade for you?
My Logic install reports 3,4 ms roundtrip latency and when tracking I havn't felt any latency like before, so I am really happy with the latency and I connect to the 802 via USB. I did not try Firewire yet. Maybe better?
Well I owned a Fireface UC for many years and indeed RME units are very well built, sound awesome and have very good drivers. I sold the unit about a year ago as I had stepped away from music production.

But I didn't want to return to RME units for the following reasons:

* I don't like company's attitude towards their customers (see their forum)
* Their Mac drivers start 2 apps upon boot in the dock which is highly un-Mac like and annoying. RME are not interested in changing this behaviour. The Clarett does no such thing and has a mixer app you can launch when you please and then quit when you're done.
* There's little regard for monitors on the main outputs (no pop protection, no dedicated volume knob)
* I wanted to go with a Thuderbolt interface as I am out of USB ports on my Mac and feel that Thunderbolt would be more future-proof

From a driver / stability standpoint, believe it or not, this Clarett over Thunderbolt is the most stable I've seen. It reminds me of my old Echo PCI interface in how it performs since it's ultimately running at PCIe speeds.

But honestly, the RME units would be my first choice if I was on Windows, which is what I used primarily while owning the Fireface UC.
Old 15th July 2016
  #17
Gear Maniac
 
ilikefruit's Avatar
Quote:
Originally Posted by soundpalace View Post
But I didn't want to return to RME units for the following reasons:

* I don't like company's attitude towards their customers (see their forum)
* Their Mac drivers start 2 apps upon boot in the dock which is highly un-Mac like and annoying. RME are not interested in changing this behaviour. The Clarett does no such thing and has a mixer app you can launch when you please and then quit when you're done.
* There's little regard for monitors on the main outputs (no pop protection, no dedicated volume knob)
* I wanted to go with a Thuderbolt interface as I am out of USB ports on my Mac and feel that Thunderbolt would be more future-proof

From a driver / stability standpoint, believe it or not, this Clarett over Thunderbolt is the most stable I've seen. It reminds me of my old Echo PCI interface in how it performs since it's ultimately running at PCIe speeds.
I'm happy to hear the 2pre drivers are stable. I liked F-control but I never had the chance to appreciate the other stuff. I only got what i call "drop out galore" all the time on everything. With their (very nice) support we went through several re-installs with no change in behavior.

I do miss having a volume knob, but I was thinking about getting one of their remotes. Basic or Advanced?
I also would've preferred a TB interface, but in the end, the interface has to work for me. I don't mind the dual apps and I have not yet experienced their support. Maybe one day?
Old 15th July 2016
  #18
Lives for gear
Quote:
Originally Posted by ilikefruit View Post
I'm happy to hear the 2pre drivers are stable. I liked F-control but I never had the chance to appreciate the other stuff. I only got what i call "drop out galore" all the time on everything. With their (very nice) support we went through several re-installs with no change in behavior.

I do miss having a volume knob, but I was thinking about getting one of their remotes. Basic or Advanced?
I also would've preferred a TB interface, but in the end, the interface has to work for me. I don't mind the dual apps and I have not yet experienced their support. Maybe one day?
Yeah I had read your thread and that was indeed a bit of a shame mate. I used to use an SM Pro Nano Patch to control my volume for a while but ended up just getting used to using the knob on the RME unit. I think some folks here like the TC Pilot knob as well.

I honestly do prefer Focusrite Control to RME TotalMix but only because it's a bit simpler. RME's TotalMix was excellent though.

Will also admin that the support I have gotten from Focusrite is also excellent.

One little bit of advice, if you get a volume knob, do NOT put it on top of your RME unit, in fact, treat that surface with care. I ended up with permanent marks on top of mine due to the rubber legs of my Nano Patch. The finish they use appears to be a bit fragile. My old Echo Layla 3G was totally fine in the same scenario. I was really bummed about that on my RME unit and the folks at RME didn't really offer me a solution (e.g. to obtain a new top cover so that I could replace mine).

Anyway, I'm glad to say that with everyone's help I'm very happy with the Clarett now and have been mucking about all day with it.

FTY, the latency on the Focsurite unit is really impressive:

128 samples / 44.1 kHz gives me output latency of just 3.6ms which is what I'm running at right now. At 64 samples it is 1.4 ms.
Old 15th July 2016
  #19
Lives for gear
 
Lumbergh's Avatar
 

Old 15th July 2016
  #20
Gear Maniac
 
ilikefruit's Avatar
Thanks for the advice!
I will keep it in mind. I'm really looking for something with a volume knob.

Quote:
Originally Posted by soundpalace View Post
Anyway, I'm glad to say that with everyone's help I'm very happy with the Clarett now and have been mucking about all day with it.
Great! Did you solve the noise issue? Does sound good now?

Quote:
Originally Posted by soundpalace View Post
FTY, the latency on the Focsurite unit is really impressive:

128 samples / 44.1 kHz gives me output latency of just 3.6ms which is what I'm running at right now. At 64 samples it is 1.4 ms.
Oh, you're giving me latency envy!

Well, as long as I have a latency lower than 5 ms I am happy. I think 6 ms is like standing 2 meters from your guitar amp. I can live with that....
Old 16th July 2016
  #21
Lives for gear
Quote:
Originally Posted by ilikefruit View Post
Thanks for the advice!
I will keep it in mind. I'm really looking for something with a volume knob.



Great! Did you solve the noise issue? Does sound good now?
Yeah, it's totally fine now, thanks!

Quote:
Originally Posted by ilikefruit View Post
Oh, you're giving me latency envy!

Well, as long as I have a latency lower than 5 ms I am happy. I think 6 ms is like standing 2 meters from your guitar amp. I can live with that....
Haha, I'm the same, around 6ms is fine for me too. I get 6.5ms at 256 buffer which I intend to use as projects get heavier. It's still good enough for realtime playing of software instruments I think, but 3ms is just that little bit more responsive.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump