|
|
Joined: Jan 2004
Posts: 901 Likes: 19
Pan-dimensional being
|
|
Pan-dimensional being
Joined: Jan 2004
Posts: 901 Likes: 19 |
just got an m33c. had it on unbalanced for a few days at +1.0db and now switched to balanced. was hoping it would lower the noise floor a touch, hasn't really changed anything. however, i had to calibrate it to -10db. is it right that there is that much more gain using balanced?
Chris Srces: Squeezebox Touch | OSMC Vero 4K+ Procs: HD621 > G61R > Minidsp DDRC-88D Spkrs: 7200.2 + 5KC + 5Ks | iNuke6K > USC15 | iNuke3K > 2 x Crowson T108SS Video: JVC HD950 & 1.9m React3 2.4:1 screen Room: 5.8x2.8m | Various GIK Panels
|
|
|
|
|
Joined: Jan 2005
Posts: 854 Likes: 5
Pan-dimensional being
|
|
Pan-dimensional being
Joined: Jan 2005
Posts: 854 Likes: 5 |
Nope. It should only be 3db of additional gain when going from a single-ended to a balanced connection.
Happy Trails! Vince@Freewheelcycle.com
2ch: Jay's Audio CDT2 Mk3, PS DirectStream Dac, Oracle Delphi VI/SME V/Benz LP-S, Coincident Statement Linestage, Foundation Research V5 Phono, JL Audio CR-1 xover M-ch: OppoMod 203, Anthem AVM60, 2x 557 2x or 5x ATC SCM20ASL mk2, 4x SCM12i, 4x Seaton Submersive HP, AM2.0 sub EQ Video: JVC X990, Isco 3L A-lens/sled, 105" motorized 2.37:1 AT screen
|
|
|
|
|
Joined: Sep 2004
Posts: 137
Hitchhiker
|
|
Hitchhiker
Joined: Sep 2004
Posts: 137 |
Actually moving from unbalanced to balanced gets you + 6 dB of additional gain. Since the differential signals swing in opposite directions, the voltage difference is twice the single ended swing and the calculation for dB difference would be 20log2=6.02 dB. If we were measuring a power difference it would be 10log2 which would give us the +3 dB difference.
Meridian Advocate
|
|
|
|
|
Joined: Jan 2005
Posts: 854 Likes: 5
Pan-dimensional being
|
|
Pan-dimensional being
Joined: Jan 2005
Posts: 854 Likes: 5 |
Oops, I stand corrected I guess. I was under the impression that doubling the voltage (which you are effectively doing with a balanced line)= +3db?
2ch: Jay's Audio CDT2 Mk3, PS DirectStream Dac, Oracle Delphi VI/SME V/Benz LP-S, Coincident Statement Linestage, Foundation Research V5 Phono, JL Audio CR-1 xover M-ch: OppoMod 203, Anthem AVM60, 2x 557 2x or 5x ATC SCM20ASL mk2, 4x SCM12i, 4x Seaton Submersive HP, AM2.0 sub EQ Video: JVC X990, Isco 3L A-lens/sled, 105" motorized 2.37:1 AT screen
|
|
|
|
|
Joined: Sep 2007
Posts: 1,980 Likes: 55
Knows where his towel is
|
|
Knows where his towel is
Joined: Sep 2007
Posts: 1,980 Likes: 55 |
had it on unbalanced for a few days at +1.0db and now switched to balanced. i had to calibrate it to -10db. is it right that there is that much more gain using balanced? The 6dB levels quoted above are the correct voltage difference if you simply hook a balanced output into a un-balanced input or visa versa. The difference between Pro level (balanced) and consumer level (unbalanced) is actually 14dB. Pro having 0 VU = +4 dBu and Consumer 0 VU = −10 dBV. Sounds like Meridian don't take the cheap and easy route and do balanced properly. I used one of these when I had a 501 feeding a Pro Active Crossover. I found this to be a great source of information for the correct way to interface between balanced and unbalanced equipment. Steve
|
|
|
|
|
Joined: Jul 2006
Posts: 55
Mostly harmless
|
|
Mostly harmless
Joined: Jul 2006
Posts: 55 |
can you point me at power/voltage/loundess dependence formulas? thanks
I miss my old Meridian setup.
|
|
|
|
|
Joined: Jun 2005
Posts: 501 Likes: 1
Paranoid android
|
|
Paranoid android
Joined: Jun 2005
Posts: 501 Likes: 1 |
Great info, Steve!
Ivan - Paul's given the standard elec eng formulae above, and Steve mentioned the different reference levels.
I've actually got an OA12 in my desk drawer awaiting modification into an unbal/bal converter. I've been unable to convince myself to cut and strap what amounts to an excellent four channel DAC, though. I might try one of those Rane transformers.
Chris - are the cable runs really short? I heard loads of difference with noise in standby when changing from a fully balanced 568-557 connection to unbalanced on my G68. These were 5 meter runs, though.
861v8, DSP5200, DSP5200HC, DSP5000mk2, UHD722, MF10, Prime, Prime PSU
|
|
|
|
|
Joined: Jan 2004
Posts: 901 Likes: 19
Pan-dimensional being
|
|
Pan-dimensional being
Joined: Jan 2004
Posts: 901 Likes: 19 |
For clarity, I've switched from 568 unbal -> 33c unbal to 568 bal -> 33c bal The tail is just over 1m in length. I figured seeing as how I have the balanced capability on my 568 and 33c, I should use it. I have made up my own cable and followed a wikipedia article indicating that 1 is ground, 2 is hot and 3 is cold. So I made up my cable so my red core is cold and my white core is hot. I was expecting something of the order of 3db difference, but not 11db. Even allowing for a 1db error in my previous calibration, that is still 10db. So far, that figure doesn't add up with anyone's suggestions: 3, 6 or 14  I mean it works, but I was surprised.
Chris Srces: Squeezebox Touch | OSMC Vero 4K+ Procs: HD621 > G61R > Minidsp DDRC-88D Spkrs: 7200.2 + 5KC + 5Ks | iNuke6K > USC15 | iNuke3K > 2 x Crowson T108SS Video: JVC HD950 & 1.9m React3 2.4:1 screen Room: 5.8x2.8m | Various GIK Panels
|
|
|
|
|
Joined: Sep 2007
Posts: 1,980 Likes: 55
Knows where his towel is
|
|
Knows where his towel is
Joined: Sep 2007
Posts: 1,980 Likes: 55 |
I have made up my own cable and followed a wikipedia article indicating that 1 is ground, 2 is hot and 3 is cold. So I made up my cable so my red core is cold and my white core is hot. I always remember it as XLR, X=1, L=2, R=3, X=GROUND, L=LINE, R=RETURN I was expecting something of the order of 3db difference, but not 11db. Even allowing for a 1db error in my previous calibration, that is still 10db. So far, that figure doesn't add up with anyone's suggestions: 3, 6 or 14  Perhaps Meridian only did balanced correctly in the 568 and not in the 33c? If the 568's balanced output is +14dB over the unbalanced, and you've gone down 11dB, then there is only 3dB difference overall. Perhaps the speakers simply route the unbalanced input line and return to the same point as the balanced line and return? But that would still result in a 3dB error. It's hard to tell without seeing the output stages of the 568 or the input stage of the 33c. You could always make a balanced to unbalanced lead up. Depends if it's worth the trouble! Steve.
|
|
|
|
|
Joined: Jan 2004
Posts: 901 Likes: 19
Pan-dimensional being
|
|
Pan-dimensional being
Joined: Jan 2004
Posts: 901 Likes: 19 |
or measure the voltage across unbal v hot or cold?
Chris Srces: Squeezebox Touch | OSMC Vero 4K+ Procs: HD621 > G61R > Minidsp DDRC-88D Spkrs: 7200.2 + 5KC + 5Ks | iNuke6K > USC15 | iNuke3K > 2 x Crowson T108SS Video: JVC HD950 & 1.9m React3 2.4:1 screen Room: 5.8x2.8m | Various GIK Panels
|
|
|
|
|
Joined: Sep 2007
Posts: 1,980 Likes: 55
Knows where his towel is
|
|
Knows where his towel is
Joined: Sep 2007
Posts: 1,980 Likes: 55 |
I've had a look on the Meridian website looking for specs for the 568 and the M33c. The 568 handbook doesn't mention the exact output. I couldn't find any info on the m33c, but I could for the M60c.
It states that the Input for unbalanced is 0.775v and for balanced is 3v. That's a difference of 11.75dB.
So perhaps the 568 doesn't do the +14dB extra that pro stuff does!
Steve.
|
|
|
|
|
Joined: Sep 2007
Posts: 1,980 Likes: 55
Knows where his towel is
|
|
Knows where his towel is
Joined: Sep 2007
Posts: 1,980 Likes: 55 |
Ignore my previous posts about 14dB difference, I was wrong.
After some further reading regarding on the nominal levels of consumer and professional audio, I found some interesting points;
1. The nominal level is a time average of an audio signal, not the peak. 2. For audio, the signal to noise ratio is the difference between the nominal level and the noise floor. 3. The difference between the nominal level and peak signals is the headroom. 4. Consumer nominal level is generally given as 0 Vu = -10dBV (note dBV not dbu) 5. Professional nominal level is 0 Vu = 4dBu 6. dBu is a voltage relative to 0.77459 volts, derived from the voltage required to dissipate 1mW in a 600 ohm load. V = (1mW * 600ohm) ^0.5 7. dBV is a voltage relative to 1 volt.
From the above the following can be derived;
Professional Nominal Level of 4dBu = 1.2276 volts Consumer Nominal Level of -10dBV = 0.3162 volts The difference between professional and consumer nominal levels is therefore 3.882 or 11.78dB
Steve.
|
|
|
|
Forums18
Topics31,149
Posts296,930
Members5,503
| |
Most Online4,788 May 22nd, 2026
|
|
|
|