Is there a balanced to unbalanced (single ended) board that is recommended? Twisted Pear has two options but they use op amps, and then there's the Broskie Cathode Follower, and Unbalancer Two.  Those two use tubes.  What else is out there?

Amb alpha24

Or you can use simple THS4131+LM6171 if your source can work with something like 2-10 kOhm load.

my balanced/unbalanced/cast to balanced/unbalanced.  but its a pile of transistors, single resistor gain control, and would need a buffer for low impedance loads.

Transformers

Three discrete opamps?

I've used Jensen transfo for XLR-2-SE.  Not sure it's the best/cheapest method.  May be a step-down transfo's better than 1:1

 

21 hours ago, kevin gilmore said:

my balanced/unbalanced/cast to balanced/unbalanced.  but its a pile of transistors, single resistor gain control, and would need a buffer for low impedance loads.

Is that the board that you posted with mention of using it in a modular preamp of sorts?

its the latest one.

Is it this one ?

http://gilmore.chem.northwestern.edu/boards/balancedoutputcast2.zip

There's also THAT - 

http://www.neurochrome.com/product/that-receiver/

The board's a bit pricy, but should be easy to implement.

The most significant issue with the THAT 1200 series of receivers is noise, -107dBu. It is dominated by the rather large values of noise-generating resistors inside (7k typical). With a good discrete design, or clever use of opamps you can get to -120dBu, which challenges measurement (<1uV in a 20kHz bandwidth).

Edited October 18, 20169 yr by Craig Sawyers

Interesting, just checked the datasheet and the noise figure checks out. I wonder how the guy got his -140dB noise measurements. He's an EE at TI and should know better than that.

He doesn't say what the measurement bandwidth was. Which really makes it a very loose specification. The only hint is the very sharp mains harmonics in the noise plot, which are way sub-Hz in width. To get to -158dBV (-155.8dBu) from -107dBu on a 20kHz B/W you would need a measurement bandwidth of 0.25Hz,

Shockingly badly specified.

I found this post...  Did these designs go any further?

 

On 10/3/2016 at 9:03 PM, Skooby said:

I've used Jensen transfo for XLR-2-SE.  Not sure it's the best/cheapest method.  May be a step-down transfo's better than 1:1

 

I think this might be the best solution...  anyone else do this compared to an opamp based solution?

Here's the schematic on the jensen website

http://www.jensen-transformers.com/wp-content/uploads/2014/08/as089.pdf

It will go starting with a dual mono BIII to a passive 1 in 3 out xlr switch box (the transformers will be one of the three positions) to a 1 in 3 out RCA switch box.  D you guys see any issues with that?

For my tube amps I would need the dampening circuit according to jensen.  For best response it says to keep my cable run under 2 feet.  I might make a box separate for each tube amp.  That would keep the cable run under two feet.

Any audio signal level transformer has a resonant peak at some frequency, hopefully outside the audio bandwidth, caused by leakage inductance resonating with distributed winding capacitance. The RC network on the output is chosen to damp that resonant peak.

But if you put a cable after it, which has some capacitance, it screws with the damping. The catchall of using less than two feet of cable comes from that consideration.

I understand...  That was an excellent explanation, thank you

this is coming soon

http://gilmore.chem.northwestern.edu/diffinput2.pdf

a lot less parts than the other version, almost the same thd

 

The diffinput works,  files posted , and tested.

Edited January 19, 20179 yr by congo5

That's pure win right there.   

Nice

Board files were posted yesterday and you have a working prototype today? Most impressive !!!

boards at the speed of light.

updated version of the board with input resistors just posted.

by adjusting a few resistors you can get up to 6db (12db balanced) gain