Convert unbalanced line level signal (p10 connector) to balanced XLR

[batata004]

Hi,

I have a digital piano which provides a P10 (line level output). I want to connect it to a XLR cable because I need to run this XLR cable for at least 30 meters so I can connect it to the mixer. I cant create a cable with 30 meters using the P10 (line level output) because it is unbalanced and it will pick up a lot of noise. I understand that XLR cables have 3 wires in such a way that the + and - wires provide the signal level but inverted so the mixer can reject noise.

So, the line level output of my digital piano provides the + (SIGNAL) and the GND. I can connect the + to the pin 1 of XLR connector. The GND of my line level output I can connect to pin 2 of XLR connector. Then, there is the pin 3. I heard I can connect it to GND too. If I do this, am I still creating a balanced XLR connector? I dont think so... in this case, I will have the same problem as if I enlarge the P10 cable for 30 meters, because connecting pin 3 of the XLR to GND will not create an inverted phase of the signal in such a way the mixer can reject noise.

So how do I connect the line level output (which has only SIGNAL and GND) to a XLR connector in a way that I can run it on a 30 meter cable without acquiring much noise. How do I invert the signal? Or do I really need to invert the signal and instead connect the pin 3 to GND and still benefit from the noise rejection in the mixer?

Thanks a lot!

 

[Sokel]

If you know your piano's output impedance that's the best way to do it without transformers:

(link)

You just replicate the R-C of the output.

 

[somebodyelse]

And if you don't know the output impedance you can still get some CMRR benefit using a twisted pair to the XLR. At the P10 end signal goes to one wire of the pair. The other wire of the pair goes to ground, as does the shield. At the XLR end pin 1 is shield, pin 2 signal from the pair, and pin 3 the other of the pair. NOTE: the pin numbers are not as you suggested.

 

[Wayne A. Pflughaupt]

NOTE: the pin numbers are not as you suggested.

This^^^

@batata004, your connection scheme is all wrong. The mixer input will be looking for:
Pin 1 = Shield
Pin 2 = Signal (+)
Pin 3 = Signal (-)

To run an unbalanced signal on an XLR cable, the connection scheme is:
Signal (+) -> Pin 2
Shield (-) -> Pin 3
(Some people like to jump or tie Pin 3 to Pin 1)

However, this does not make the signal balanced. You are merely running an unbalanced signal on an XLR cable. It has no functional advantage over simply running a long unbalanced cable.

A long unbalanced cable doesn’t necessarily pick up noise.

That said, the best solution in your situation is to use a direct box. It will convert the unbalanced keyboard output to a balanced signal for the long run to your mixer.

https://www.sweetwater.com/store/detail/IMP2--whirlwind-imp-2-1-channel-passive-instrument-direct-box

BTW, I’ve always been mystified why keyboards don’t have onboard balanced XLR outputs...

Regards,
Wayne A. Pflughaupt

 

[somebodyelse]

To run an unbalanced signal on an XLR cable, the connection scheme is:
Signal (+) -> Pin 2
Shield (-) -> Pin 3
(Some people like to jump or tie Pin 3 to Pin 1)

However, this does not make the signal balanced. You are merely running an unbalanced signal on an XLR cable. It has no functional advantage over simply running a long unbalanced cable.

Whitlock says otherwise - the shielded twisted pair connection I suggested has 20-30dB more ground noise rejection than your coax. See section 3.7 "UNBALANCED TO BALANCED INTERFACES" in Jensen Transformers' AN007.

 

[batata004]

Guys you are awesome! I learned a lot in this thread! So, after reading everything you said, in my case, despite using a XLR connector, I am ending up unbalanced because I soldered the pin 3 to the pin 1 (- to shield). In this cause, I ask you: is there any advantage of using this XLR connection I have over making the P10 cable longer? Does soldering the pin 3 to pin 1 provide some kind of noise reducer in the mixer?

As some of you suggested, indeed using a DIBOX will be my best alternative. I just would like to give it a try to the solution provided by @Sokel in creating a small circuit to create the inverted phase. @Sokel provided an image and a link but the circuit does not look that simple (because the dashed box is supposed to be the only circuit I should create but as I read the link he provided, I would need to make more thing). Do you guys have a link to a video or an easier source where I can learn how to invert the phase? Isnt there already a chip that I can buy that already does this inverting thing? This way I could feed my + to the chip and it would generate the - that I would connect to pin 3 of XLR.

One more thing: how can I measure the impedance of line out of my keyboard? As I read about XLR, the impedance of my keyboard cannot be more than 0,01% different than the impedance I will create in the - (pin 3) otherwise the circuit will get noisy.

 

[DVDdoug]

One more thing: how can I measure the impedance of line out of my keyboard? As I read about XLR, the impedance of my keyboard cannot be more than 0,01% different than the impedance I will create in the - (pin 3) otherwise the circuit will get noisy.

We don't want "matched" impedance.

Generally we want a lower impedance output driving a higher impedance input. The lower noise "dominates" the overall impedance and determines the noise sensitivity.

I'd try a simple "adapter" first and if you get noise I'd go with an active direct-box. An active Di has lower output impedance for better noise immunity.

BTW, I’ve always been mystified why keyboards don’t have onboard balanced XLR outputs...

I don't know why guitars don't have balance low-impedance outputs like microphones. Tradition, I suppose. And the pickup has to be high impedance and it picks-up noise so maybe it wouldn't help. Active pickups do have low-impedance (unbalanced) coming out of the guitar but the pickup windings are still high impedance.

 

[somebodyelse]

We don't want "matched" impedance.

Generally we want a lower impedance output driving a higher impedance input. The lower noise "dominates" the overall impedance and determines the noise sensitivity.

Actually in this case we do, but not in the sense you mean. We want the source impedance of the hot and cold (pins 2 & 3) to match to give the best common mode rejection ratio (CMRR). If we know the output impedance of the single ended output connected to 'hot' we can add a matching impedance to the 'cold', turning out single ended output into an impedance balanced output.

 

[Sokel]

@Sokel provided an image and a link but the circuit does not look that simple (because the dashed box is supposed to be the only circuit I should create but as I read the link he provided, I would need to make more thing). Do you guys have a link to a video or an easier source where I can learn how to invert the phase? Isnt there already a chip that I can buy that already does this inverting thing? This way I could feed my + to the chip and it would generate the - that I would connect to pin 3 of XLR.

One more thing: how can I measure the impedance of line out of my keyboard? As I read about XLR, the impedance of my keyboard cannot be more than 0,01% different than the impedance I will create in the - (pin 3) otherwise the circuit will get noisy.

Best is if you can find your piano's schematic so we can see its output.
Then it's only a matter of a resistor and a capacitor in series connected from RCA's output ground to XLR's cold (3) .

 

[somebodyelse]

Does soldering the pin 3 to pin 1 provide some kind of noise reducer in the mixer?

No. Soldering pin 3 to pin 1 makes ground noise rejection worse. See post #5.

I just would like to give it a try to the solution provided by @Sokel in creating a small circuit to create the inverted phase. @Sokel provided an image and a link but the circuit does not look that simple (because the dashed box is supposed to be the only circuit I should create but as I read the link he provided, I would need to make more thing).

The bit in the box doesn't invert the phase. It adds an impedance to pin 3 to match the output impedance of the existing unbalanced output, which is why the components shown are the same as in the existing unbalanced output circuit. Matching the impedance on the hot and cold maximises the common mode rejection ratio (CMRR) as discussed in post #8. This gives the best rejection of interference from magnetic fields , such as from the cable passing near a transformer.

Inverting the phase on pin 3 isn't necessary for CMRR, but doubles the signal level. Interference pickup is unchanged, so the ratio of signal to interference is improved slightly. This is less significant than the change from matching the impedance.

If you really want to make a board that inverts phase it's probably best to use a balanced line driver IC like the THAT 1646. SparkFun used to sell ready made modules for this as detailed in their hookup guide (which also has some explanation of how balanced outputs work). The board design is still published so you could make your own. Alternatively this audioXpress article includes designs to be made with off the shelf prototyping board.