The best way to connect multiple amps to your DAC - Nobsound XLR Switc



Many people as part of their headphone setup will have more than one amplifier, allowing them to swap between them for different flavours to suit different moods and genres.

But, most DACs only have one output, so what is the best way to connect a DAC to multiple amps?

We have a few choices:

  1. Unplug the cables from Amp A and plug them into amp B.
  2. Use an XLR splitter to connect both amps at the same time.
  3. Use an XLR switch to let you select which amp to input to.

Option 1 is not particularly convenient, no one wants to be fumbling with cables behind their desk each time they want to listen to something different, and also if you’re doing this a lot then there is the risk you may wear out the connectors of the cable or your amplifier over time.

Option 2 addresses these problems, no cable swapping needed as the DAC will just feed both amps all the time. But it does introduce a new problem: Impedance Matching.

A DAC has an output impedance, and an amplifier has an input impedance. Ideally you want the output impedance to input impedance ratio to be as high as possible, though keeping it above 10:1 at a minimum is an ok rule of thumb. (See our glossary for some info on input/output impedance!)

Lets say you have two amplifiers. A topping A90 with an input impedance of 2000 Ohms, and an Enleum AMP-23R with an input impedance of 10,000 Ohms. You’re using an SMSL SU-9 which has an output impedance just over 200 Ohms.

The output impedance to input impedance ratio for the SU9 and 23R is 50:1, plenty! And for the SU9 and A90 it’s 10:1, so a little low but still fine.

But when we use a splitter, things change.

The effective input impedance for the two when connected via a splitter is calculated using the same formula as two resistors in parallel:

This gives us a result of 1666 Ohms! Lower than either of the two individually! This means that using a splitter in this setup, our impedance matching ratio would be less than ideal, even though either amplifier connected directly to the DAC would be fine.

Option 3 will fix this. By using a switch, we keep the DAC connected to only one amp at a time, avoiding the effect of resistance (input impedance) in parallel, keeping our impedance matching ratio nice and high, and we don’t have to fumble with cables either. The only question is; will the switch degrade audio quality? It’s a passive device and in theory shouldn’t have an effect on audio quality in the way that an active device like a preamplifier might, but does that hold up in practice? Lets find out!


The Nobsound MC103 is a popular choice as an XLR switch mostly due to its affordable price.

The MC103 is a fairly simple design, with the XLR ports connected to a rotating switch, allowing the user to select between three options.

The MC103 comes in a ‘1-in, 3-out’ variant and a ‘3-in, 1-out’ variant. There is also an MC303 with a ‘3-in, 3-out’ configuration facilitated by two switches instead of one.

The switch in each of these units changes the connection for the positive and negative XLR pins as you select the different input/output options, however the ground/shield pins are permanently bridged and are not affected by the switch.

It is also worth warning that the switch is a ‘make before break’ design, meaning that as you switch between say X and Y, they will be temporarily shorted together. This is not an issue for the ‘1 in 3 out’ version, as shorting the inputs of amplifiers is no problem, but for the ‘3-in 1-out’ option, you will short the outputs of the two DACs together, which could potentially cause damage if they are playing at the time, and have insufficient short protection. It would really have been ideal to see a ‘break before make’ switch used in these products just in case.

Besides that, the build quality seems quite nice for the money, with a grey, sandblasted metal exterior, a satisfyingly firm and clicky knob, and Neutrik connectors used for the XLR ports. The only complaint I could possibly make about the build quality is just that because of the lightness of the product, the cables can sometimes pull the box back if it’s not held down with something such as blu-tac.

So, other than the slight concern in regards to the shorting issue, the construction seems to be pretty nice, with an attractive price to match, but can this switch be considered ‘transparent’?


Measurements Include:

  • Noise
  • Crosstalk (Left to Right)
  • Crosstalk (Zone to Zone)
  • LCR (Capacitance, Inductance, Resistance)

Tests conducted with standard 4V line level XLR input

Test Setup

  • Audio Precision APx555 B-Series analyzer
  • Measurement setup and device under test are running on regulated 230V power from a Furman SPR-16-Ei
  • DER EE DE-5000 LCR Meter
  • AudioQuest Mackenzie XLR Cables


First, lets quickly check that the switch does not change distortion compared to just using an XLR cable alone.

1khz Sine, 4V input, 4V output (XLR cable loopback):

1khz Sine, 4V input, 4V output (Loopback with Nobsound MC103 in signal path):

As you can see, absolutely no change in distortion when using the switch. It’s behaving like a normal XLR cable here (as expected).

But is there a chance that it might allow for some additional noise to be picked up?


For this test, I simply measured the noise collected by an XLR cable up to 1Mhz, and then compared it to the amount of noise collected with the switch also in the path

MC103 switch and XLR cables

No change in noise pickup, and it seems that the metal enclosure is doing a good job of shielding from unwanted noise, just as the XLR cable’s own shielding does.


Even though the switch doesn’t affect distortion, and doesn’t allow for any additional pickup of external noise, there is still the chance that within the unit itself, signal from the left channel could leak (via capacitive coupling) to the right channel, or from one input to another. So let’s check

Red shows crosstalk via XLR cable loopback, Blue shows crosstalk with the MC103 in the signal path

The MC103 does have a very slight effect on crosstalk, however it’s still incredibly low, and below that of most DACs and amplifiers anyway. A Topping A90D for example has crosstalk of -110dB at 1khz and -86dB at 20khz, meaning the MC103 is close to 20dB lower at any given frequency. Performance was the same regardless of which output was tested.

LCR (Capacitance, Inductance, Resistance)

Capacitance (parallel):

Positive to Negative: 27pF

Positive to Ground: 45pF

Inductance (series):

Ground: 0.2uH

Positive: 0.2uH

Negative: 0.2uH

Resistance (series):

Ground: 0.11Ω

Positive: 0.31Ω

Negative: 0.31Ω

All of these values are consistent with a typical XLR cable. Impedance is slightly higher though as would be expected due to the switch itself.


In summary, an XLR switch is the ideal way to connect multiple amplifiers to your DAC, and avoids potential issues associated with XLR splitters, or the inconvenience of moving cables yourself.

As for performance, in almost all areas, the Nobsound MC103 behaves no differently than a short XLR cable, and should be considered ‘transparent’ other than a slight impact on crosstalk which is beyond the performance of most amplifiers and beyond likely limits of audibility as well.

The MC103 is an excellent product for the money and from an audio quality standpoint is essentially perfect. The only minor concern is the ‘make before break’ switch design, however most DACs will have a high enough output impedance to protect themselves from shorts anyway.

This is a companion discussion topic for the original entry at

Great! But what about switching RCA? Nobsound/Douk Audio doesn’t make an equivalent for RCA.

1 Like

Well they do have this one that has 2 RCA in and 2 RCA out. As well as 3.5 in/out and BT in.

1 Like

+1 for the MC103. I’ve been using mine for a couple years now….it works great :+1:

1 Like

I also am using one, no issues for me either. I couldn’t detect any before/after impact on audio quality…

I have the solo input connected to my Yggy, and then switch between a Burson Soloist GT or a Flux FA-10 Limited.

This one seems decent but no idea how it sounds or measures.