Iso-transformer vs Amp with balanced inputs

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
There still seems to be a lot of confusion about balanced outputs. To be clear, a balanced signal is just two wires; the ones connected to the hot and cold of the source. The third wire, the cable shield, is just that, it is a shield. It is not a signal conductor, it does not carry signal current and it is not even needed for the signal to successfully reach the other end. The confusion arises because nearly all electronically balance outputs are in fact differential and the produce equal and opposite signals relative to analog 0V. For balanced operation, the fact the two signal are referenced to analog 0V is immaterial. The only thing that matters is the voltage between the hot and the cold.

What does this mean in the context of hum loops? Well, if you have a balanced source and an unbalanced receiver, to avoid a hum loop you want to avoid connecting the ground of the sender to the ground of the receiver. To do this you add a transformer at the receiver. Its secondary connects between signal and ground of the receiver. Its primary connects to hot and cold of the sender. There is now no need to connect the screen/ground at both ends, and hence no hum loop.

Cheers

ian
 
What are some good line level 1:1 transformers I could use? Ideally minimal coloration/transparent sounding
 
Sparkfun makes a THAT 1206-based input module that would make quick work of converting those problematic Hafler unbalanced inputs to high-performance balanced inputs. Just add a pair of appropriate resistors and zener diodes to get supplies down to about ±18 V.
See SparkFun THAT 1206 InGenius Breakout - BOB-14002 - SparkFun Electronics for full details.
 
Yes, I think it's an amazing value. It's a well executed design and, of course, it uses the IC that I patented (but now expired and public-domain). The only time a good/Jensen transformer would be better is in the case of extreme common-mode voltages over about ±15 V peak. In years past, the most offensive applications in that regard were recording trucks connected to house sound systems - where the AC power hook-ups for the recording truck were "dicey" (to be kind) to the point of being dangerous.
 
These 1206 boards seem very useful. Would it be unwise to use a separate power supply for the 1206 boards outside of the amp/piece of equipment running the signal into?

I was thinking of making a balanced to unbal box, in order to make it modular so if I wanted to use it on other unbalanced equipment
 
These 1206 boards seem very useful. Would it be unwise to use a separate power supply for the 1206 boards outside of the amp/piece of equipment running the signal into?
No problem. You just need to make sure the power supply common of the add-on is well connected to the amp's input ground.
I was thinking of making a balanced to unbal box, in order to make it modular so if I wanted to use it on other unbalanced equipment
That would be very wise. A debalancer is quite useful in a toolbox.
 
It's never a good idea piling up balanced and unbalanced gear. Just one unbalanced connection ruins the CMRR of all the rest.
This is a bit of a revelation to me.

And I have been thinking about this, how does this work ?

I've learned about "the Tascam problem" on GDIY, and I start to see what happens there, these outputs do not like to be shorted and they will become noisy because of that, as they are shorted to signal ground, this will polute your signal ground.
(on a sidenote, if I look at a "volume control" in general it's a pot, shorting some of the signal to ground, how does this not create a problem ? .. or does it ?)

I used to think that by removing a problematic connection overall noise levels would improve, but I see that is not the case, though I would not be surprised that 8 bad connections cause more noise than one ? But the problem won't be gone untill all of these are fixed. (I've always learned to not short outputs, leave them floating if you cannot connect them)

And in the case of an unbalanced line to an balanced input I short pin 3 to ground.

Never thought that this would have an effect on CMRR.

But last night I got an insight, it is the loading !

If I look at a TRS output jack, going to an unbalanced output I leave Ring floating.
But looking at the TRS jack inside the mixer, it's "balancing act" internal, only works if both legs are presented by an equal load, as one leg is now tied to ground and the other is loaded by an input so this is no longer the case and CMRR will start to creep up.
(same goes for inputs)

Is this what happens ?
 
(on a sidenote, if I look at a "volume control" in general it's a pot, shorting some of the signal to ground, how does this not create a problem ? .. or does it ?)
It does not create a problem because as the wiper moves to the "ground", the resistance between top and wiper increases, so the current is limited.
Never thought that this would have an effect on CMRR.
By definition, an unbalanced line has zero CMRR. CMRR combines as the least common denominator, so a perfect balanced connection connected to an unbalanced one results in a "perfectly" unbalanced connection.
 
It does not create a problem because as the wiper moves to the "ground", the resistance between top and wiper increases, so the current is limited.
(y)
It also came to mind that the opamp is an active part, the pot is passive
By definition, an unbalanced line has zero CMRR. CMRR combines as the least common denominator, so a perfect balanced connection connected to an unbalanced one results in a "perfectly" unbalanced connection.
Ok, so my OTO BIM delay and OTO BAM reverb units with unbalanced connections connected to balanced aux send and input channels on my mixer are not messing things up ? (I did not short output pins and did short input pins on my mixer)

Thing is, I have a bunch of Lundahl LL1527 (not Jensen 😀) sitting in a box that I got for free... they did not cost me a penny, since I have those, would you advise to use those to balance these connections ?
Maybe only balance the FX outputs going to the mixer ?
Or just leave it as is as there will be nothing to gain by adding these transformers ?
 
Ok, so my OTO BIM delay and OTO BAM reverb units with unbalanced connections connected to balanced aux send and input channels on my mixer are not messing things up ? (I did not short output pins and did short input pins on my mixer)
Depends on what you mean by "messing thing up". What I can say is that the actual connection is unbalanced. Remember a connection is balanced only if both the source and receiver are balnced.
Now, the negative effects depend on the type of balanced output the source has (check post #16).
In any case, none of the advantages of a balanced connection (interference rejection and elimination of hum loops) are retained.
Thing is, I have a bunch of Lundahl LL1527 (not Jensen 😀) sitting in a box that I got for free... they did not cost me a penny, since I have those, would you advise to use those to balance these connections ?
That depends very much of how each piece of gear is capable of driving said transformers. Data says that they can be satisfactorily driven from 800 ohms. Anything more than that will result in performance degradation. Remember that all transformers benefit from being driven by a low impedance.

Maybe only balance the FX outputs going to the mixer ?
I would think that many pedals are not suitable for driving a xfmr. Typically, most Boss pedals have a 1k resistor in the output. Some have their output taken from a 50k pot.
Or just leave it as is as there will be nothing to gain by adding these transformers ?
If the cable runs are short enough, you should not see significant improvements.
 
These FX units are not pedals, desktop units and they run on +15V power supply but have unbalanced connection only.

https://www.otomachines.com/product/bam/
Specs ? ...it has Neutrik (tm) connectors ! :rolleyes:

The reverb is great though.

But kept the paper manual and it says this;
Output; single ended 100 Ohm max +20dBu
Input; single ended 1M max input +20dBu (@1% THD+N)

So driving these transformers should not be a problem I guess ?

I have quite a bit of "mismatches" in my setup and a box full of "general purpose" audio transformers, so I would like use these in a way that makes sense.

(edit)

In addition, I don't know how my mixer outputs are balanced, it is build buy the guy that can not be mentioned on the forum 😀 and all I can find is that it is "electronically balanced" all outputs <100Ohm all inputs >10K Ohm.
 
Last edited:
if I look at a "volume control" in general it's a pot, shorting some of the signal to ground,

A short is defined as a very low impedance connection, typically just a wire, an accidental contact, solder, etc. A high value resistor making a connection is not a "short" by any conventional definition.
 
Not just to be contrary but balanced (differential) interfaces are mainly found on professional gear (with the exception of low Z mic inputs). Lots of consumer and prosumer gear pass acceptable audio with unbalanced interfaces. Digital interfaces can moot this all..

JR
 
Back
Top