Small mixer is leaky - is there a fix?

[DC-Choppah]

We have this little stereo mixer (https://www.sweetwater.com/store/detail/PowerMix3).
I sent the first one back because it leaked sound even when everything was turned fully counterclockwise.

The new one came and it is exactly the same to the dismay of the technical support folks.

Does anyone know if this is just an issue with a little mixer like this, and they will all be that way, or is something wrong? I really need a fix for this. Any help is appreciated. Perhaps there is a simple modification that can be done?

The way we use this guy is that the main output goes to a pair of iLoud powered micro monitor speakers. The headphone output goes to either headphones or the main mixing board. The audio inputs to this little ART mixer come from two different computers (PCs) and a synthesizer setup. We found that to eliminate hum, we put an audio isolation transformers between the PC audio output and the mixer.

The leakage is heard when you play audio on either PC. With the ART mixer's master knob down all the way on the mixer, there is still a muffled sound coming from the iLoud monitors. Sort of like you ran through a low pass filter. If you turn down the input controls on the mixer for that channel it still does not cut out the audio.

The iLoud monitor speaker volume control is on its center position and the input level to the mixer is the nominal level you get out of a PC audio jack with the software device set at 50% volume.

The problem is that this leakage level is enough to be picked up by mics. This little setup was supposed to be a computer music /DJ station but I need it to be able to be silent out of the headphone output or main output when it needs to be.

We can certainly work around the problem but have to remember to reach behind the iLoud monitors and turn them off when tracking. I need the master control knob to cut this off itself.

[resistorman]

A mixer these days, no matter how small or inexpensive, shouldn't bleed through unless there is a defect. I guess there could be, since the replacement probably came from the same container and there might be a bad batch.

It's possible that your ground loop issue has something to do with it. Ground loops are the Devil's playground. Is everything plugged into the same AC circuit? Have you had to isolate both computers?

You mentioned a synthesizer setup as the 3rd input. Does it bleed too? You might try disconnecting the PCs and play just the synths through it. Do they bleed? And/or try something completely different without a ground... your phone, for instance, as the only thing plugged into the mixer.

[DC-Choppah]

OK. I plugged in a portable mp3 player (ipod nano) into input 3 of the ART mixer. The PCs are no longer connected, so inputs 1 and 2 have nothing connected.

I can still hear leakage into both the headphones and the main output with all knobs on 0. But, I have to turn the ipod volume all the way up to hear it. If I leave the ipod at 50% volume the leakage is not audible. But, if you put your ear right up to the iLoud monitor you can hear the leakage. But this is workable.

I just thought that zero on the knob meant really zero. Nothing. Nada. But it doesn't.

Problem comes in now when the PCs are connected. PC volumes are all over the place depending on the software. So the problem is that now I can get leakage out of the monitors when I don't expect it because we are set up thinking that when the main knob is at 0 there should be no sound from the monitors. But there is.

I have one PC connected with a transformer and the other without. It doesn't matter if I use the transformer or not. I am getting leakage with or without.

[ef37a]

No such thing as "zero" when it comes to shutting a channel up DCC!

Duggy Self coined the term "offness" for things like pots and signal switches that are supposed to turn things down to inaudibility, for even quite decent rotary and linear controls this seems rarely better than -60dB ref' OP level and you have to get into specially made faders with breaks at the track bottom or ground compensated systems to get offnesses(!) into the noise floor, say -110dB .

The pots in a cheapo Behringer/Mackie di-da are likely even worse and poor internal grounding could makes matters even worse.

My solution? A wee switch box twixt the source that is breaking through and the mixer input.

Dave.

[Hugh Robjohns]

I sent the first one back because it leaked sound even when everything was turned fully counterclockwise.

I think the technical term you seek is 'crosstalk' -- when one input signal 'leaks' into an output where it shouldn't appear!

Does anyone know if this is just an issue with a little mixer like this...

Not familiar with that particular mixer, but its design is very simple and ART stuff is generally well built. I'd be surprised if it's a real problem with the mixer, and more likely an issue with the way it's being used.

We found that to eliminate hum, we put an audio isolation transformers between the PC audio output and the mixer.

I suspect this is the heart of the problem. As far as I can see the ART box works with unbalanced inputs and outputs, so ground loops are likely to be an issue... However, the box itself looks like it is powered from a double-insulated wall-wart and so relies on the connected equipment for its own grounding.

In placing isolation transformers on both inputs and outputs I wonder if you have inadvertantly managed to remove all grounding from the unit completely. If it is not properly grounded, crosstalk will almost inevitably be a significant problem, because the audio reference ground for the electronics won't be a reference at all, but will float up and down with the incoming audio signals.

I suspect if you set the mixer up in a simpler system, with the output going to something grounded and the inputs coming from battery-powered sources like an iPod and laptop, it will work perfectly with no crosstalk... Worth a test, anyway!

So... I think it would help enormously if you draw out the complete audio system connected to this mixer, paying careful attention to which source and destination equipment are properly grounded to the mains safety earth, and which are floating (because they use battery power or double-insulated supplies).

From that you'll be able to see if there are any real ground loop potentials, how best to resolve them, and how to set up a sensible grounding path so that everything gets the earthing it requires!

H

[Mike Stranks]

Good response from Hugh - of course!

Just last week I had all sorts of funnies when interconnecting two bits of kit with unbalanced connections. My ART DTI actually made things worse! Then I started to think about earths and if there was one at all... things then became somewhat clearer and Plan B resulted in no unwanted noise. I swapped-out one piece of double-isolated kit for something that most definitely does have an earth/ground.

[Hugh Robjohns]

You can always add a solid earth to a system like that simply by running a wire from a known good earth to the ground connection on a connector on one device...

For convenience (and safety) I often use a Groundology plug wired to a jack plug...

http://www.soundonsound.com/reviews/gro ... ction-plug

[DC-Choppah]

I suspect if you set the mixer up in a simpler system, with the output going to something grounded and the inputs coming from battery-powered sources like an iPod and laptop, it will work perfectly with no crosstalk... Worth a test, anyway!
H

Test :
ipod shuffle (battery powered) into channel 3 of ART mixer. (No other inputs connected). iPod volume at max.

Main output to iLoud powered Monitors plugged into same power strip as ART Mixer. Monitors set at the nominal mid gain position.

All knobs on ART Mixer fully counter-clockwise.

Output at 1" from the speaker measured by SPL meter is 72 dBA!

Man, that's a lot of sound coming out of this thing with everything set to zero!

This is not going to work the way I had intended. The work around is I have to set the iLoud monitor gain at the '1' position. With that, I can work with this rig and when the master volume on the mixer is is off there is no leakage coming out that is picked up by the mics, but it is still there if you listen close. But the mixer has to be cranked way up and adds some noise when set like this. I need a kill switch for the monitors right on the mixer I guess. The knob is not doing enough and is leaky it seems.

[DC-Choppah]

The Tech at Sweetwater is now duplicating my test (he uses his cell phone as an input) on multiple ART units. He thinks it may be a bad batch?!

Hmm, if Mr Dave is right and these units really will just leak no matter what, then I just sent these folks on a wild goose chase.

[Hugh Robjohns]

I suspect if you set the mixer up in a simpler system, with the output going to something grounded and the inputs coming from battery-powered sources like an iPod and laptop, it will work perfectly with no crosstalk... Worth a test, anyway!
H

Test :
ipod shuffle (battery powered) into channel 3 of ART mixer. (No other inputs connected). iPod volume at max. Main output to iLoud powered Monitors plugged into same power strip as ART Mixer. Monitors set at the nominal mid gain position.

From what I can see, the micro monitors are not grounded -- they appear to be powered from a double-insulated wall wart supply. So if nothing in the test system above has a real ground reference and we are no further forward...

Try replacing the iPod with an output from a desktop computer -- not a laptop. (desktop computers are usually grounded via the mains supply)

The knob is not doing enough and is leaky it seems.

Maybe the pot is cheap and naff. Maybe the mixer is badly designed and has terrible crosstalk. But I find it hard to believe given the quality of everything else I've used from ART. I'm still of the opinion that the problems you describe are due to the system having no earth. Although I'm not entirely comfortable about the gain structuring either...

H

[DC-Choppah]

Try replacing the iPod with an output from a desktop computer -- not a laptop. (desktop computers are usually grounded via the mains supply)

The PC we use is a desktop computer and I am now using that as a music source for this new test. This is a normal motherboard audio output (no audio card) coming out of the stereo mini jack on the motherboard. Audio level set to to 50% on the software device.

The desktop is plugged into the same power strip as the ART mixer. Nothing else plugged into the mixer inputs just the PC.

No isolation transformer. Just a splitter to take the stereo mini TRS plug and split into Left and Right 1/4" TS inputs on the mixer.

All gain knobs set to zero position on the ART mixer.

iLoud monitors connected to main output of ART. Monitors at nominal mid gain level.

SPL level 1" from speaker is about 65 dBA. Music leaking out is very bass heavy.

It actually sounds like the bass is the same level regardless of the main gain setting at low levels. When you turn the gain up a bit, the higher frequencies come in but the bass is the same level between about 7 oclock and 9 oclock on the knob.

Leakage sounds similar in headphones.

Sweetwater tech says he can duplicate this with a signal generator on his bench.

[Hugh Robjohns]

Interesting.

So the 'leakage' is around 6dB lower when the mixer is grounded via the desktop computer. I would have expected more, but it's a step in the right direction and confirms the importance of proper grounding.

From your earlier comments I've been assuming that the level knobs are simple rotary faders, but I'm wondering now if they are actually gain controls instead. The block diagram on the ART website suggests that -- And in that case they wouldn't mute when fully counterclockwise. I subsequently found a user review on Amazon.co.uk that says:

Non-zero Volume ControlsThis was mentioned by other reviewers here. No channel can be completely reduced in volume (such that it is inaudible) including the master channel. This means that when you have any channel set to a moderate volume, even if you minimize the master volume you will still hear audio. It is low in volume but certainly audible.

That seems to be the answer -- the mixer has no 'faders'. It is designed to mix things together with a set balance achieved by adjustingnthe input stage gains, not to fade channels in and out. In other words, it isn't designed to do what you want it to do.

The constant level of bass is also rather intriguing. Can't think of a reason for that yet...

H

[resistorman]

For convenience (and safety) I often use a Groundology plug wired to a jack plug...

http://www.soundonsound.com/reviews/gro ... ction-plug

Very clever! I'll have to keep this solution in mind...

[resistorman]

That seems to be the answer -- the mixer has no 'faders'. It is designed to mix things together with a set balance achieved by adjustingnthe input stage gains, not to fade channels in and out. In other words, it isn't designed to do what you want it to do.
H

How about that. I can see why the input channels might be designed that way, but wouldn't the master and headphone controls be conventional faders? Maybe they are, and the OP is treating the input gain controls like faders, cranking them up, and driving the circuit really hard. You'd think there'd be distortion, though. Hmmm.

[DC-Choppah]

Thanks for that. Ok so it has gain controls but not faders. Faders go to - infinity.

So now I need to look again at these mini-mixers and see if there is one out there that actually has faders. That's a distinction now that I can appreciate. I like the form factor and we built a custom shelf below the table for this guy.

3 stereo channels
1u half rack size
Faders (not gain control) so that audio can be cut-off completely
2 outputs (one headphone, one main speaker)

If I can't find one I'll have to add a cutoff switch to the main output so we don't get any bleeding when tracking.

[DC-Choppah]

How about that. I can see why the input channels might be designed that way, but wouldn't the master and headphone controls be conventional faders? Maybe they are, and the OP is treating the input gain controls like faders, cranking them up, and driving the circuit really hard. You'd think there'd be distortion, though. Hmmm.

You get leakage for any position of the input control, even with it completely counter-clockwise. Slightly less leakage actually when at mid position, then back to the full leakage when at max position. It is mostly bass that leaks through. The lower, the more leakage. Songs with sub-base especially leak through. It's basically 100% leakage at 60 Hz and below to my ear.

Do I claim this is a defect or not? Sweetwater thinks it is a defect and may be asking ART for a new batch to correct the problem.

I just want the leakage to be gone actually.

Do all these little mixers have this feature?

[Hugh Robjohns]

Slightly less leakage actually when at mid position, then back to the full leakage when at max position.

Huh? As it now appears that these input controls are actually gain controls, you seem now to be suggesting that the gain goes down in the middle of the scale... which can't be right!

Could it be that you have the same material on more than one input, but in opposite polarities (or at least phase shifted)? That kind of scenario could create the symptoms you mention of 'bass leakage' and the 'leakage' maximising when the gains are matched around the middle...

I just want the leakage to be gone actually. Do all these little mixers have this feature?

I've had a quick scan to see what's available in the way of simple half-rack line mixers. A few have proper faders on the inputs, but they are mono -- like the Behringer MX400, for example.

https://media.music-group.com/media/PLM ... 0_M_EN.pdf

I suspect a simple muting switch on the output of your existing system might be the most pragmatic way forward!

H

[ef37a]

Just did a quick test.

Setup: 2496 sends 1.12 V rms (+3.2dBu ) 1kHz then 40Hz DAW generated at -1dBFS to the line inputs of a (V cheap!) Wharfedale 16/02 mixer and thence to a pair of Tannoy 5As.

Leaving the Main Mix Out pot at 12 o'clock I set the channel pot for +6dB on the mixer LEDs. Very loud at 1kHz.
Turning the channel pot to zero gives NO sound even with Mk1 lug at the speaker. Yes, I am rather deaf at 1kHz but I CAN hear 40Hz pretty well as much as the next man.

I suspect the ART mixer has some bad internal ground paths, in a word, poor design.

This 'un..http://www.parkerblack.com.au/products/ ... harfedale/

Dave.

[Hugh Robjohns]

I suspect the ART mixer has some bad internal ground paths, in a word, poor design.

No, the ART turns out to have gain controls rather than faders... It wasn't designed to allow input signal to be muted with the controls at minimum...

H

[ef37a]

I suspect the ART mixer has some bad internal ground paths, in a word, poor design.

No, the ART turns out to have gain controls rather than faders... It wasn't designed to allow input signal to be muted with the controls at minimum...

H

Ah! Well still poor design IMHO! Should have kill switches.

Dave.

[Hugh Robjohns]

Ah! Well still poor design IMHO! Should have kill switches.

I don't think that's fair criticism. It's designed to provide a pre-mix of a number of stereo sources, which it clearly does. Adding other facilities would add to the cost.

At the end of the day, it simply wasn't designed to do what the OP wants it to do. That's clearly frustrating for the OP, but I don't really think ART can be blamed for it!

H

[ef37a]

Yeah, my fault Hugh. I should have had a gander!
I was under the illusion that it was of the Behringer Xenyx 802 stamp? I now see it is much cruder than that.

Dave.

[DC-Choppah]

I found this: http://www.ebay.com/itm/Rolls-MX28-3-Ch ... 2054891752

But now I am not sure if this also has gain controls (which don't go to zero), instead of faders.

[Hugh Robjohns]

The manual is here: http://www.rolls.com/pdf/M_MX28.pdf

It's rather vague on the detail, but does include a schematic which suggests the knobs are all gain controls, rather than faders.

Take a look at the Berhinger mx882. I think that might do what you want for very little money (it's quite a clever and versatile unit, actually).

[DC-Choppah]

Take a look at the Berhinger mx882. I think that might do what you want for very little money (it's quite a clever and versatile unit, actually).

Thanks for that. I could make it fit I think, but I am not sure this supports a headphone output if it is unpowered?

I am just going to go with a little mute button for the main monitors. Now with regards to that mute button, when in mute position, I don't think I want an open circuit on the monitors. That will hum and buzz like having the cables unconnected.