Sound Devices MixPre noise level

[Tom Morrow]

I just purchased a used Sound Devices MixPre. I'm a newbie to field mixers, asking for some help with diagnosing whether this thing is defective or whether I'm doing something wrong with this unit that I bought.

The main issue is noise. When I turn up the gains I hear noise through the headphone jack even when there is no signal going into the unit. With AA battery power, both the 1 and 2 gains turned all the way up and pan switch to center, I hear white noise through my mdrv6 headphones when the headphone dial is turned up to about 8:30.

By three 3 o clock on the headphone control the noise suddenly goes from medium level to a very loud feedbacky-screeching sound, and all the meter goes up from -30 dbu only lit to +8dbu. If I don't have the pan switches on Center but rather Left and Right, then that doesn't happen; the headphone dial goes all the way to max without screeching or any other lights going on.

Another weird thing is that sometimes when the gain/fader dials are all the way up, the -30 dbu light is on, but sometimes it's off. I haven't figured out a pattern yet but it's related to the pan switches as well as recent history; turning the gain dial from max, down a bit, then back to max sometimes turns off the -30dbu light.

Should I return this thing? Can anyone compare and tell me at what dial positions they can hear noise with nothing connected to their MixPre?

Thanks!

BTW for comparison I just tried the same experiment with my non-field mixer, a Behringer UB502, and got pretty much the same noise levels... with Main Mix maxed out and all other controls at unity, I started hearing noise with headphones around 10 o clock and it was medium level by about 4 o clock. However there was none of that screechy feedbacky stuff even with everything maxed up.

Also FYI before these tests I got out a jeweler's screwdriver and adjusted the limiting all the way up (right), and the tape return level all the way down (left), to make sure that wasn't causing the issue.

[Chad Johnson]

How does it sound when you put a live mic through it? Maybe you just have all this gain going by turning everything up. Set up a mic and listen to yourself talk. At the settings you have you wouldn't be able to listen without doing damage. Turn it down.

[Tom Morrow]

Okay, did a bit more testing... I fed in a signal at 100mvRMS from my NTI minirator MR2 signal generator, white noise, the highest level that didn't turn on the clipping indicator for the input on the MixPre. I only had to turn up the Input Fader to about 7:30 to get the signal up to 0dbu on the MixPre LEDs, and around 12 o clock on the headphone dial was a comfortable level to hear the tone; the headphone clipping light went on at 1 o clock. With those settings, I hear just a very faint amount of white noise with the input signal muted, so perhaps this thing isn't too noisy, but I'm still concerned about that feedbacky-screechy thing that I mentioned in the previous message.

[Tom Morrow]

Chad, that's a good testing idea, but the only mike I have right now is one that I know to be noisy so I'll have to wait until my better microphones are available next week.

[Tom Morrow]

That did give me the idea of trying to feed a portable CD audio signal in and listen. I put stereophile test CD2 into input 1, and dialed the volume just low enough that the input clipping light went off. Listening at normal listening levels noise was apparent but subtle. When I plugged the headphones directly into the CD player at normal levels, I didn't hear hardly any noise at all, and the sound is much more "clear" than when listened through the MixPre, sort of that "veil has been lifted" sound, much more dynamc through the CD directly than theough the MixPre.

Hmm, doing the same test with my UB502 mixer gives the same results: more noise and less dynamics through the mixer than plugging the headphones in directly to the CD player. I suppose the fact that I'm not using an impedance-balanced cable to connect the CD headphone out to the XLR Mic In might be an issue... I've got an impedance matching cable in the mail so I'll try that out.

But if anyone else has a MixPre and can give their impression of what settings cause noise and what settings don't, that would be very useful to me.

[Robert Wiejak]

A microphone input is a differential input. That means a voltage difference between pin 2 and 3 is what carries the information. When the inputs are left open, without anything connected to it, there is a good possibility that a very small voltage (noise) is induced internally and this is what you are hearing when you crank everything up. This happens because you have no load connected to the input. A simplest test you can do is this: Make sure phantom is off. Take a metal staple and make a small ‘U’ out of it. Using it, connect pin 2 and 3 of the XLR input. Your input is now loaded with 0 ohm source, meaning that the same voltage is present at pin 2 and 3 (effectively no signal). Now crank that input and output and hear how much noise you have. If you hear a lot of difference between when the staple is in vs. out, then your pre is working well. It is just amplifying the noise on the unloaded input. But, if there is very little difference or no difference at all and you hear a lot of noise, then you pre is generating it internally. The only time I would expect more noise than normal with good pre, is when the batteries are at or below minimum voltage.

PS. In your last example, you used a CD as a source. Any adapter you use (IMO) will effectively present itself as high impedance source. Not all pre will work well with HI Z source especially if they are specifically designed to work with LO Z sources like microphones (< 1K).

[Chad Johnson]

Did you have the CD player turned all the way up? That probably wouldn't be best. Try half way up. It still sounds like your trying to get everything right to the point of almost clipping. I don't work like that. I think at normal levels the unit should perform great. And the lack of dynamics you hear could be because your driving the limiters - effectively squashing the dynamic range.

It's hard to tell without being there myself...

[Rick Reineke]

Yeah, without having direct access to the equipment is difficult to diagnose. I would concur w/ Chad and the others, aside from shipping it to SD for evaluation you should check it out in real world conditions with other known quality gear.
The Mixpre has copious amounts of of gain available for HPs and will even drive a set of 600ohm phones to extream levels. Though I have not tried, If cranked to '11', I'm sure some BG noise will be heard though the phones. Under normal conditions with my 302 and 7506 phones, the headphone lever is usually set at around 20% ( if I recall w/o unpacking and looking at it) which is more than adequate for most conditions. (even 42nd and 8th). No functioning piece of gear is 100% silent.
In any case, if it is malfunctioning SD will rectify it immediately if not sooner. The customer service at SD one of the best.. like Lectro.

[Jon Fairhurst]

I'm not sure about the MixPre, but headphone amps are often much cheaper than the amps used in the signal chain. Just because you hear noise on headphones doesn't mean that you'll get that noise in the final recording. Have you compared the S/N of the headphone output to the S/N of the main output?

[Richard Crowley]

Originally Posted by Tom Morrow

When I turn up the gains I hear noise through the headphone jack even when there is no signal going into the unit. With AA battery power, both the 1 and 2 gains turned all the way up and pan switch to center, I hear white noise through my mdrv6 headphones when the headphone dial is turned up to about 8:30.

It is not valid to measure (even casually by listening) the noise of a mic preamp open-circuit (with nothing connected). The circuit was not designed for optimal performance with high (or no) source impedance because you never use it that way. At minimum, put a dead short across the input from pin 2 to pin 3. Ideally, you would put a low-noise resistor whose value was equivalent to the source-impedance of the microphone.

By three 3 o clock on the headphone control the noise suddenly goes from medium level to a very loud feedbacky-screeching sound, and all the meter goes up from -30 dbu only lit to +8dbu. If I don't have the pan switches on Center but rather Left and Right, then that doesn't happen; the headphone dial goes all the way to max without screeching or any other lights going on.

You are probably hearing circuit (electrical) feedback added to the circuit noise. Again, that is because you are leaving the input open-circuit, flapping in the breeze and very susceptible to picking up the (infinitely stronger) output signal from itself. It is unlikely that you would see this phenomenon if you had the input properly terminated.

Another weird thing is that sometimes when the gain/fader dials are all the way up, the -30 dbu light is on, but sometimes it's off. I haven't figured out a pattern yet but it's related to the pan switches as well as recent history; turning the gain dial from max, down a bit, then back to max sometimes turns off the -30dbu light.

So that is indicating that the noise sometimes reaches -30dB and sometimes it doesn't. That's why we call it "noise".

Should I return this thing? Can anyone compare and tell me at what dial positions they can hear noise with nothing connected to their MixPre?

I wouldn't. At least not based on the tests as you conducted them. Do the same tests with a proper termination and get back to us.

BTW for comparison I just tried the same experiment with my non-field mixer, a Behringer UB502, and got pretty much the same noise levels... with Main Mix maxed out and all other controls at unity, I started hearing noise with headphones around 10 o clock and it was medium level by about 4 o clock.

The Sound Devices product would be expected to have a lower noise level than the Behringer. But you have shown that when the test is done with no input termination, all bets are off.

However there was none of that screechy feedbacky stuff even with everything maxed up.

Probably because the Behringer is physically larger and the output parts of the circuit are physically farther away and less susceptible to electrical feedback.

[Tom Morrow]

Thanks folks for the advice.

I tried putting 180 ohm resistors between 2 and 3 on the inputs, and that did stop the screechy-feedbacky issue. It also reduced the noise level, but not terribly much, perhaps 3db to my ear.

I am still concerned about the overall noise level and dynamics of the unit, but I will wait until I get my good microphones and impedance-balanced XLR to 1/8" cable so I can do some real-world tests rather than these fake tests I've been doing.

[Tom Morrow]

Originally Posted by Chad Johnson

Did you have the CD player turned all the way up? That probably wouldn't be best. Try half way up. It still sounds like your trying to get everything right to the point of almost clipping. I don't work like that. I think at normal levels the unit should perform great. And the lack of dynamics you hear could be because your driving the limiters - effectively squashing the dynamic range.

It's hard to tell without being there myself...

I had the CD player turned down to the level that turned off the input clipping light on the MixPre, which was in practice turned down almost all the way.

I am getting a little concerned that the limiters might be causing the lack of dynamics even though they shouldn't be engaging at all. But again, I'll wait for real world tests when I have my equipment next week...

[Billy Smith]

Are you plugging the CD player (line) in to the MIC jacks on the mixpre? The mixpre is mic input only (unlike the SD302). It will not work (sound good, normal, etc) with a line lvl signal like from the CD player...