I'm back + noise problem

Morshu · Post by **Morshu** » Mon Jan 27, 2014 9:33 pm

Hey guys, this is more of a general topic than specific to the R3, but it does apply to the R3.

But, how does one record audio without any noise/ filter that noise out properly

I've been getting some harsh white noise in the background (not particularly loud, but it sucks to have it) I've been filtering it out with noise removal in audacity, but sometimes when I do this strange distortion and crap happens to the sound

The same thing happens with my moog sub phatty but the signal has a whine on top of it and isnt just white noise.

Anyone got any solution?

either to using the noise removal tool more effectively/a plugin for audacity or something?

I've been using an M Audio interface, a M-Track

Would really help to know how to fix this problem.

analoguekid · Post by **analoguekid** » Mon Jan 27, 2014 9:56 pm

Are you putting your signals through a mixer or a FX unit prior to the interface, as I had similar problems and it was mixer noise and not the individual synths. I found this out by listening to a synth full vol through headphones and not playing anything

I then removed all the synths from the mixer and replaced each one, one at a time and the noise got increasingly stronger the more I added to the mixer. I have now got around this by buying a better quality audio interface so that I don't have to pre mix things before they go in. I now only get noise from FX units which unless you pay super amounts of money for them you will never get a noise free one.

Morshu · Post by **Morshu** » Tue Jan 28, 2014 12:15 am

I mean that there is no noise when I hear the synth through the headphones I have but when I record it there is noise and feedback

I don't put the synth through a mixer before I put it into the audio interface however, nor do I see why that should be necessary really.

I'm more or less trying to figure out why it is that when I record my synth into audacity it has noise and feedback behind it(its quiet but very noticable and unpleasant)

acemonvw · Post by **acemonvw** » Tue Jan 28, 2014 6:58 pm

I use my mixer as my audio interface (Onyx 1640i), which allows one to record multiple parts at the same time (say multiple outputs from different synthesizers), so... there is a reason to use a mixer, or even in the case of simply having all of your synths routed to the computer (if only a stereo output can be recorded through USB or whatever).

If you have the outputs go straight into the an audio interface, no, you wouldn't need a mixer.

Have you made sure that the volumes from your synth are loud enough? Sometimes, if the levels are too low (sent into the DAW or whatever), you have to adjust and increase the volume on your DAW, and that results in noise. I'd try to get your synth to be loud enough, but not clipping, maybe that will reduce the signal:noise ratio.

Morshu · Post by **Morshu** » Tue Jan 28, 2014 8:39 pm

I just put my synth straight into my audio interface and I get noise in the background

I always have to filter out the noise

I have my interface's recording volume really low and my synth volume maxed out so that it records at a good volume and has less noise

but, I still continually get the noise.

do you get no noise from your synth or something with your setup

is it possible that audacity is causing the noise due to settings not being good for the synth

do you use a different program to record

xmlguy · Post by **xmlguy** » Tue Jan 28, 2014 9:23 pm

I record the R3 with an Maudio Fast Track Pro. It records cleanly with no additional noise that can be heard. I certainly don't have to attempt to "remove" noise later on. That's about as effective as trying to remove excessive salt added to soup. It's generally too late to remove noise once it's been added to the signal. It's far better to eliminate the points where the noise gets added, like eliminating salt in the ingredients before they're added to the dish.

The kind of noise you describe occurs when one or more of the amplification stages are excessively noisy at certain gain levels. Some products "cheat" by not really having line level inputs at all when shared with a mic input. There should be a separate low gain opamp used for the line input that's separate from the high gain opamp used for mic level input. But instead of doing this, it's cheaper to use the mic level opamp for both and to merely put a level cut pad in the line input to keep from overloading the mic opamp. Doing that increases noise dramatically. Also, don't max the output of the synth. You need to monitor the noise level of the output to observe when you get a strong signal with low noise. Maxing an output often produces a high signal with a lot of noise, while a lower level will produce a good, strong signal with low noise.

Many cheaper mixers also do this "line level padding" cheating, particularly when a single channel has both line and mic inputs, but no switch to choose which gain stage is used. Some of the cheap Behringer and Yamaha mixers do this. while the better and more expensive Allen & Heath mixers don't. You can see this in the schematic diagrams that they publish for these products. Unfortunately, Maudio and many audio I/O vendors don't even publish the schematics, so you don't really know how they've designed the electronics until you see the negative effects of poor design to produce a cheaper product. It shouldn't be particularly surprising when a much cheaper I/O product is much noisier than a more expensive one.

Some audio I/O inputs have two or more gain modes that can be selected via the drivers. They can have a separate high-gain mode for mics and a low-gain mode for line level. So it's worth looking into the driver options. In Windows this is done under the recording options in the sound control panel. There can be a configure button enabled for inputs with settings like this.

Other common causes for excessive noise are ground loops and computer/laptop power supplies that have a noisy ground. You need a stereo direct box with ground lift to eliminate these problems when they occur. I've seen this happen with the R3. Sometimes the R3 will have an extremely noisy audio output when it's hooked to the computer by USB directly to use the software editor/librarian. Unhooking the USB cable from the R3 is an easy test. The R3 really isn't to blame for this problem. The USB ground on the computer is the cause. The R3 simply uses its ground as the reference for the unbalanced signal - and it's not at fault when the USB ground messes it up. Sometimes you have to switch from USB-midi to hardware midi to get rid of the noise. Hardware midi is optically isolated specifically to eliminate these ground loop problems. USB ground isn't optically isolated because it's used to power devices, not merely to transmit signals.

Morshu · Post by **Morshu** » Tue Jan 28, 2014 9:51 pm

your advice gave me the exact opposite of what I was looking for, more noise

the noise appears to be unelimatable

I tried messing with those settings and my previous setting was far better than trying to record at a low volume was, though the noise was quieter, the synth was a lot quieter too so it was useless

my interface allows me to switch to +48v gain but thats not what I think I want, so I stick on the lower setting

the noise is very quiet as I said, but its still a horrible noise especially if you are doing layering, I reverted to my previous setting and turned the synth down a little, didn't really change much, I currently have my mic (the audio interface) level very low so that it reduces the noise, any help?

is it audacity creating this noise?

analoguekid · Post by **analoguekid** » Tue Jan 28, 2014 10:00 pm

Morshu wrote:your advice gave me the exact opposite of what I was looking for, more noise

the noise appears to be unelimatable

I tried messing with those settings and my previous setting was far better than trying to record at a low volume was, though the noise was quieter, the synth was a lot quieter too so it was useless

my interface allows me to switch to +48v gain but thats not what I think I want, so I stick on the lower setting

the noise is very quiet as I said, but its still a horrible noise especially if you are doing layering, I reverted to my previous setting and turned the synth down a little, didn't really change much, I currently have my mic (the audio interface) level very low so that it reduces the noise, any help?

is it audacity creating this noise?

Might be motherboard noise?

Morshu · Post by **Morshu** » Tue Jan 28, 2014 10:12 pm

it sounds sorta like a ground, but I have no midi connected and when I didn't use the interface at all and just recorded through line in, the same tone occurred.

http://www.sendspace.com/file/7fagyr

heres what I'm getting

Morshu · Post by **Morshu** » Tue Jan 28, 2014 10:50 pm

ok well here is what I've discovered

when the synth is unplugged from the power cord and is not plugged in via 1/4 cable into the audio interface, the noise reverts to being pure white noise, which is easily filterable.

and that the 1/4 inch cable emits a very light hum if plugged into the unit and the synth

but that when I plug the synth into the power cord, the really bad hum tone comes back, so i guess it is grounding, but I'm still confused why I still get noise regardless, or is that normal?

xmlguy · Post by **xmlguy** » Wed Jan 29, 2014 2:45 am

THE +48V SETTING IS NOT GAIN. THAT'S PHANTOM POWER AND CAN BURN OUT THE OUTPUTS OF THE R3 OR ANY OTHER AUDIO DEVICE NOT PROTECTED FROM PHANTOM POWER. PHANTOM POWER IS ONLY FOR CONDENSER MICROPHONES.

Please describe EXACTLY how your R3 is connected to your interface. Do not leave out any details. Include the names of the jacks and the kind of cables and connectors.

How are you listening to the audio I/O box? Powered speakers, professional headphone? how are they connected.

You shouldn't need to filter out noise, in general, unless you are using a filter as an intentional audio effect. I've recorded my R3 and I get a clean result that never needs filtering to remove excess noise. There is at least one or more problems that are causing the noise you hear. The trick is to use a very clean monitoring system to easily tell the difference between the clean baseline sound from anything with a noticeable increase in noise. That noise can be identified and quantified using an oscilloscope to compare the amount of noise before and after any circuit block. Noise adds up. Each circuit component adds some noise, even wire, but some add a lot more noise than others. You shouldn't really be getting any significant noise from you I/O box if nothing is hooked up to the inputs.

Morshu · Post by **Morshu** » Wed Jan 29, 2014 3:01 am

When nothing is plugged in it just emits a very quiet white noise, but when i plug the synth power cord into an outlet that white noise gains a ground loop hum

So basically the problem appears to be the grounding and it appears to be the power cord because even if the synth is off it emits the hum only when connected to an outlet.

What this means is I need to plug it into a different outlet, one that only has 3 prong appliances in it preferably, correct?

Reason im dodging that question you asked is because nothing is wrong with the setup whatsoever and the flaw appears to be the ground of the cord

Also, i was able to fix this for the r3 by putting it into a power converter box, but for the moog, it uses a standard 3 prong cord and is still creating a hum.

I think i indirectly answered my own question.

Is it possible however, to have no white noise whatsoever in a recording of a hardware synth without filtering said noise out?

xmlguy · Post by **xmlguy** » Wed Jan 29, 2014 9:07 am

Plugging the power supply into another is worth trying but it often doesn't solve the problem for several reasons. The R3 and any other music device that uses 2 conductor DC power will use one conductor as ground and the other as the +V supply rail. The problem is when the ground floats or is not 0V with respect to the ground of another device it's connected to. The difference in ground reference causes a current flow across ground, which in turn can cause some very bad behavior of the ground by those other devices. Their ground can cause or transmit hum and noise, and since the unbalanced music device uses case and shield ground as the reference for the music output signal, the audio input treats it all as music. This unwanted interference can be addressed in several ways, but only one is very practical. The computer power supply may be damaged or defective, so sometimes it can be replaced with a working model or be upgraded to one with a different design to keep the ground clean. That's impractical for most laptops or other computers with proprietary power supply designs, since you don't have a choice other than replacing the whole computer. The most practical fix is to buy a stereo direct box with ground lift. The ground lift eliminates the ground loop by decoupling the R3 output ground/shield from the I/O box. The unbalanced output of the R3 gets converted to a balanced signal using an audio transformer balun, which electrically isolates the music signal +V/-V from the case/shield ground. Some active direct boxes can use photodiode opto-isolators instead of a transformer balun.

http://en.wikipedia.org/wiki/Balun
http://en.wikipedia.org/wiki/Opto-isolator

"Both transformers and opto-isolators are effective in breaking ground loops, common in industrial and stage equipment, caused by high or noisy return currents in ground wires"

Here's a link to an affordable Behringer DI20 stereo direct box with ground lift, battery, and phantom power.

http://www.sweetwater.com/store/detail/DI20/

It's good for most keyboardist to always use a direct box for connecting up to mixers and house PA systems. It will avoid a lot of potential problems so that you don't have to discover them the hard way. All it takes is one mistake like putting phantom power onto an unbalanced input or a defective powersupply to fry the outputs of your keyboard or possibly the main board. Transmitting power through audio inputs and outputs is rarely a good thing. I once got an unpleasant shock on stage when attempting to find a ground loop at a concert. The band's guitar amp rig with an old tube amp had shorted a hot 110V supply rail to the case/signal ground. I found it by accidentally becoming the conductor by holding the cable jack shield in one hand while touching the amp racks shield/case in the other. The idiotic guitarist knew there was a problem with the ground on the amp, since I found that he had "solved" it by using a 3 prong to 2 prong adapter on the power cord. The whole point of having a ground is to avoid this kind of shock hazard, but you can't solve "stupid" with good electrical design.

Morshu · Post by **Morshu** » Thu Jan 30, 2014 1:08 am

What i didn't mention is that I fixed the problem by pluggin it into a different outlet with all the other stuff

now the moog sub phatty still has the issue however, the r3 does not.

The moog unlike the r3 has a straight typical 3 prong double sided cord

unlike the r3 (with its strange box with adapter and 2 prongs)

My thought here is that the moog has the feedback due to being plugged into the same outlet as a ton of 2 prong appliances and being the only 3 prong.

xmlguy · Post by **xmlguy** » Thu Jan 30, 2014 1:44 am

You discovered that ground loops can be caused by differences in ground between different building electrical circuits. The length of the wire used for each circuit have different resistance. One outlet in a room may have a much longer amount of wire to the circuit breakers than another outlet. You can lift the ground on one outlet by using a 3 prong to 2 prong adapter, but be aware that this can pose a shock hazard if a hot power wire shorts to the case and there's no ground to provide a path for the short to follow. Anyone touching the case or anything else grounded to that device can be shocked by becoming the path to ground, like I described in my previous post.

Instead of merely using a 3 to 2 prong power plug adapter, it's safer to install GFCI outlets or use a GFCI extension cord in the room and be sure to use at least one outlet that is grounded to a device that connects its case to ground. That way the GFCI should trip if any current gets shorted to ground before someone gets hurt.

Sometimes you can't use the same outlet because you need to use more power than any one circuit can provide before tripping the circuit breaker. This happens a lot in music venues where the stage power, house mixer, and speaker power amps are located far from each other and can't be run off of the same power circuit. Thus the need for a direct box with ground lift becomes a necessity to stop ground loops. Running off a single power outlet in a room may stop a ground-loop, as you found.

You should be able to take away from this experience that large amounts of noise or hum can occur due to ground loops. The solution is to find the ground loop and eliminate it one way or another. "Removing" noise using audio software shouldn't normally ever need to be done, since it's a poor attempt to treat the symptom in a manner that doesn't fix the root cause at all. Also important is that phantom power can destroy keyboards and music gear that isn't designed to handle it, so direct boxes offer protection from phantom power and have a ground lift to stop ground loops.