Neutrik NA2M-D2B-Tx balancing transformer adapter review

[solderdude]

Yep seems totally reversible. Of course as the 47k load is utterly pointless (at least with a 1.6k load) we see no differences in the highs but expect none with a 10k load either.
They just as easily could have left the resistor out.
0.1dB (1%) is what one should expect from 1:1 transformers which why the 2.5dB puzzled me (I did not assume a 'quick' measurement without respect for '0dB' line )
The 1dB drop from the 200ohm source was though.

The choice for RCA remains a bit weird as one can expect those outputs/inputs to be above -10dBV. For that reason alone it remains a 'weird' device as most people would buy this for consumer gear and RCA is not very common (but is used) in (semi-)pro circumstances.

 

[pma]

Yeah, 2.5dB was my mistake due to being to fast. My apologies.

 

[solderdude]

speed kills...

Your measurements do show how it performs when it is used when used as intended and it is not a horrible load and can even be used in a mic and -10dBu level without sacrificing much.
Some vindication for the device when used as intended.

 

[pma]

0.1dB (1%) is what one should expect from 1:1 transformers which why the 2.5dB puzzled me. The 1dB drop from the 200ohm source was though.

Re measured 0.056 ohm vs. 200 ohm source impedance with unchanged input level (please remember that in case of 200R just the 200R resistor is inserted before the adapter without changing generator level, the generator is "steady" as explained several times). It is like a worm inside brain , that forced me to re-measure source impedance effect.
Measured at -3dBu generator level.

Relative dBr Y axis non-scalled

Comparative plot in this post, this one https://www.audiosciencereview.com/forum/index.php?attachments/na2m_fr_0-056r_200r-png.293927/
was wrong as I unintentionally changed generator level between measurements. My apologies. Forget the old plot, please. I cannot edit the older post due to time limit.

Transfer ratio of outputs with 0.056/200 R source imp. Higher source impedance (200R) worsens FR flatness (acc. input impedance plot).
200 ohm vs. 1.66kohm load impedance (Cosmos) makes that 1 dB approx.

 

[pma]

Please let me continue and let me add some technical info that is probably not self-explanatory and not clear at first view.
Please avoid driving this adapter or another link transformer from coupling capacitor, unless it is very big like >1000uF. Please see what the insertion of the capacitor is doing with frequency response. From now on, please also let me to post simulations instead of measurements, they fit with reality very well and preparation of various test setups is highly time consuming.

So, this happens with coupling capacitors of 10uF, 100uF and 1000uF. We can see Q peaking even with the 100uF capacitor. This is for 2kohm transformer load. Rising load to 10kohm considerably worsens the response with much higher peaking - maybe another surprise. In a real world, the coupling capacitor creates highly audible bass booming.

BTW, anyone knows what is this?

 

[KSTR]

^ AP System One Generator Output with its clever positive feedback arrangement. A and B amps are huge discrete power op-amps and the transformer is huge and heavy.

 

[IVX]

^ AP System One Generator Output with its clever positive feedback arrangement. A and B amps are huge discrete power op-amps and the transformer is huge and heavy.

I have a pair of such trafo, EI 40-50mm not more, BTW, the core seems isn't permalloy, it is slightly rusty.

 

[solderdude]

A proposal to 'fix' them so they can be used directly with hifi gear without being driven in saturation.

An owner can modify them as shown above and then you can connect them to 4V XLR-out of a DAC to 0.4V RCA-in of an amp .... or the other version with 2V RCA-out DAC to 0.4V balanced XLR-in of an amp.
1k load to the source (DAC) and it is about 90 (or 200ohm) out (@1kHz) to the amplifier.
Of course you loose quite a few dB but you just turn the volume up a few notches.

 

[KSTR]

I have a pair of such trafo, EI 40-50mm not more, BTW, the core seems isn't permalloy, it is slightly rusty.

Volume ratio of core and copper vs. the NTE1 still is approx. 100:1.
They are old-school construction, paper-isolated interleaved standard transformer sheet metal E-I segments and complex interleaving of the windings

 

[voodooless]

A proposal to 'fix' them so they can be used directly with hifi gear without being driven in saturation.
View attachment 294327
An owner can modify them as shown above and then you can connect them to 4V XLR-out of a DAC to 0.4V RCA-in of an amp .... or the other version with 2V RCA-out DAC to 0.4V balanced XLR-in of an amp.
1k load to the source (DAC) and it is about 90 (or 200ohm) out (@1kHz) to the amplifier.
Of course you loose quite a few dB but you just turn the volume up a few notches.

It seems you can buy these transformers separately as well:

https://www.neutrik.com/en/neutrik/products/accessories/transformers

They also offer "studio" versions, which appear to have better performance, but the NTL1 for instance isn't exactly cheap at around € 50,- vs € 10,- for the one in the tested adapter.

 

[solderdude]

That one doesn't fit in the enclosure though as it is larger (hence the lower distortion and higher input level) and is encapsulated.
That one can certainly be used with line level conversion and would have measured just fine.
These are not used in any Neutrik converters. Would have driven up the price too much and also its size.

There are plenty that would do the trick as well but not in such a small size.

something like this:

https://www.rdlnet.com/downloads/datasheets/tx-afc1f.pdf

or this for stereo:

SC110N Passive Signal Converter | Alctron Audio/Alctron Electronics Co., Ltd

when on a budget:

Millenium DI-E

Passive DI Box In metal housing, Max. input level +10/ +50 dBu, Ground/ lift switch, With integrated switchable resistor network for extremely high input level (e.g. guitar box), XLR output, Dimensions (W x D x H): 100 x 130 x 40 mm

www.thomann.de

Millenium DATI

Hum Eliminator DATI (dual audio transformer / isolater), Frequency range: 10 - 50,000 Hz, Total harmonic distortion (THD): 0.01%, 2 Channels, Inputs: XLR balanced, jack balanced, RCA, Outputs: XLR balanced, jack balanced, RCA

www.thomann.de

just not the one that was tested
I do own the DATI but never really needed it.

 

[voodooless]

That one doesn't fit in the enclosure though as it is larger (hence the lower distortion and higher input level) and is encapsulated.

From what I see, it may actually still fit

 

[solderdude]

When using a hammer and grinding down the RCA contact on the inside.. maybe.
Perhaps Pavel is in for such a modification if you send him the NTL1.

 

[dualazmak]

Or my post here regarding active converter. I hope it would be allowed sharing here again.

If we would be really serious in Balanced-to-Unbalanced and/or Unbalanced-to-Balanced conversions, we need to go into active professional converters, like TASCAM LA-40MkIII.

 

[dualazmak]

And,,, if we would be "utmost extremely" serious about line level XLR-balanced --> RCA-unbalanced HiFi conversion, and if we would not care about budget, space, weight, power-consumption (operating-cost), we can use HiFi integrated amplifier which has XLR-in plus RCA-pre-out (and/or RCA-rec-out), such as ACCUPHASE E-XXX, E-XXXX series, YAMAHA A-S2XXX, A-S3XXX series (LOL!)

Some of the higher grade ACCUPHASE E-XXX, E-XXXX seiries have XLR-balanced pre-out, and therefore, can be used for RCA-unbalanced --> XLR-balanced HiFi conversion (LOL!).

 

[solderdude]

Or my post here regarding active converter. I hope it would be allowed sharing here again.

If we would be really serious in Balanced-to-Unbalanced and/or Unbalanced-to-Balanced conversions, we need to go into active professional converters, like TASCAM LA-40MkIII.

Well yes... but expensive, needs power and does not break the ground, would be using only 2 of the 8 present channels, is big (19") (aside from the lift which isn't the same thing) which is the whole point of the tested cheap transformer thingies.

This is a cheaper and smaller active one:

ART Cleanbox Pro

2-Channel level converter To convert two unbalanced channels into two symmetrical channels and vice versa, Noise and low-distortion circuit, XLR connectors for balanced inputs and outputs, RCA connectors and stereo mini jacks for unbalanced inputs...

www.thomann.de

and its even cheaper passive cousin/nephew:

ART DTI

Dual Transformer / Isolater Hum Eliminator, Frequency response 10 Hz to 50 kHz, THD: 0.01%, 2 Channels, Inputs and unbalanced XLR balanced, TRS, RCA, Output as balanced XLR, jack unbalanced, RCA, Dimensions 44.5 mm x 107 mm x 89 mm, Weight 0.34 kg

www.thomann.de

 

[pma]

That one doesn't fit in the enclosure though as it is larger

I have a competing product here, it is a custom-made transformer for me about 12 years ago. It accepts higher input voltage, but is is of course bigger.

I have measured it at +10dBV and -5.1dBV (-3dbu) and these are the results:

Frequency response

Note the difference in LF corner when level is high.


Distortion

+10dBV is the orange line. Lower distortion above 3kHz due to lower system noise floor at higher level.


THD, THD+N at 1kHz/10dBV

THD, THD+N at 50Hz/10dBV

 

[solderdude]

When you also have the device below you can fit it into the Neutrik enclosure or you would need a really nifty shoehorn.

 

[KSTR]

Or my post here regarding active converter. I hope it would be allowed sharing here again.

If we would be really serious in Balanced-to-Unbalanced and/or Unbalanced-to-Balanced conversions, we need to go into active professional converters, like TASCAM LA-40MkIII.

It does not help to fight ground loop effects on the RCA cable because there is no isolation.

 

[pma]

Now, as in the previous post, the frequency response of Neutrik NA2M adapter also at +10dBV and -5.1dBV, and added 0dBV. I hope that now everyone who was hesitating must understand why @amirm has measured so different frequency response than me, when he pushed the adapter above its specified maximum level and measured it at 1V (0dBV).