Unbalanced to Balanced Converter Circuit with 0dB Gain


I have created an unbalanced to balanced converter using a Jensen transformer - JT-11P-1. The circuit was designed by Jensen Transformers. It is the right hand circuit in the Jensen AS060 drawing, which is available online.

I assembled it into a Hammond enclosure. 

Performance is as I expected.  It reduced my already low 60, 120, 180 Hz noise peaks by an additional 10dB. These peaks now are at 90dBV, which is exemplary for my SP-6B (AR refurbed) tube preamp of early 80s vintage. My normal signal level (music) is about -10dBV, highly variable of course. Baseline RMS noise is unaffected as this noise is stochastic (random).

There is a 2.7 dB insertion loss (0.73x). The XLR output is fully balanced with pins 2 and 3 equal in magnitude but 180 deg apart.

This circuit is notable as it does not increase the signal. Commercially available converters of this type are usually -10dBV (unbal) to +4dBr (XLR), which is a "gain" of 4 (12 dB). I can't tolerate this high a signal as the SP-6B gain is already a bit high for my system at the optimum 12 o'clock position on the volume knob. One can obtain a converter box with gain. This uses an active transistor or opamp circiut for gain. The box I made is passive all the way. 

The frequency response is very flat from 20-20KHz, with only a 2dB drop between 19 and 20 KHz.

The transformers are a bit expensive - $120 each. Resistors, box and panel mounts are inexpensive. I believe I paid about $300 for everything.

Assembly is quite straightforward and not difficult if you can solder.

Finally, I should point out that my system did not suffer badly from noise problems before I added this circuit, even though I use a mix of RCA and XLR cables. I've been careful to connect all components into the same wall plug. I made sure all components are grounded properly. The box I made had a -10dB effect on the signal. I expect noisier systems will see more reduction

kevemaher