Balanced Input / Output IC's vs Discreet - real world performance

Jump to Latest
M

mp006ltk

Member
Joined 2004
#1
Hi, I've been reading through Self's Active Crossover book and I was looking over the various reference designs for balanced input and output circuits using opamps and other passives.

Then I stumbled upon an article about the THAT 1200 IC. I also noticed ICs built for balanced output.

Has anyone used these?

In real world performance do they outperform discrete circuits?

It would seem much simpler to just use a chip... This also seems to be a common enough application that TI should have a single chip for this...
 
#2
In ballanced inputs a small source+cable impedance mismatch between +and- reduces the CMRR. That chip uses a circuit that fixes this problem. These problems are usually found in larger installations where you get long cable runs and patch bays. At home where cables are short ballanced lines are usually overkill.
 
#4
Thank you. Trying not to turn this into a debate on balanced connections.

Just trying to figure out if I'm better to build a reference design or buy a chip. It seems that often an IC wins, but nobody talks about chips like these, so it makes you wonder.
 
#5
Yes these chips are available. Beware balanced comes in 2 flavors, pro and consumer with the pro version about 10db hotter. So some se/ballanced converters will change the level. Dont mismatch these levels. Ie driving a -10dbu (consumer) input with +4vu (pro) out will distort. Other way round will increase the noise.
 
#6
mp006ltk said:
Thank you. Trying not to turn this into a debate on balanced connections.

Just trying to figure out if I'm better to build a reference design or buy a chip. It seems that often an IC wins, but nobody talks about chips like these, so it makes you wonder.
Click to expand...

Unless this is about learning electronics, Buy the chip it will be easier and you will get better results.
 
#8
There is a circuit that shares feedback between the two op-amps which creates some output common mode rejection behavior but my experience results in the conclusion that using that topology is a bad idea. What happens is that the common mode rejection behavior causes the op-amps to become very sensitive to any common mode voltages and impedance on the output and become unstable and/or both clip in an effort to cancel each other. So my conclusion is that common mode rejection is a good idea for inputs but a bad idea for outputs. You can add an inverted output but the best plan is to terminate the negative input at the source ground (not the destination ground), ie provide a signal return connection that is separate from the shield-ground.
 
Last edited:
#11
I use these chips on my commercial gear - the pre uses the balanced output chip and the power amps all use the balanced input chips (the integrated is all single-ended so does not use them).

They are fuss free, cheap and easy to use. Just make sure you decouple them well. I used 1uF SMD MLCC 0805 located right next to the devices.

I guess if you are designing mass market audio like Self does, every penny counts so it makes sense then to go with opamps and discrete resistors but I’d avoid that if you can.
 
Last edited:
You must log in or register to reply here.
Top Bottom