Hello. I asked myself the following question about the balanced inputs and/or outputs in the preamplifiers.
What advantages do they have since I suppose that somewhere along the path of this signal it becomes unbalanced, mainly in the power stage of an amplifier, unless it is a bridged amplifier.
This question arises because I have purchased an IFI Zen Phono phono preamplifier.
The factory, among other characteristics, states that it presents balanced circuits and even apart from unbalanced rca outputs, it has balanced outputs through a special connector.
Greetings and thanks to whoever clarifies it for me.
https://ifi-audio.com/products/zen-phono/
What advantages do they have since I suppose that somewhere along the path of this signal it becomes unbalanced, mainly in the power stage of an amplifier, unless it is a bridged amplifier.
This question arises because I have purchased an IFI Zen Phono phono preamplifier.
The factory, among other characteristics, states that it presents balanced circuits and even apart from unbalanced rca outputs, it has balanced outputs through a special connector.
Greetings and thanks to whoever clarifies it for me.
https://ifi-audio.com/products/zen-phono/
Balanced can have lower hum. Also balanced inherently has higher gain (+6dB) and so 3dB less noise at same signal level.
Balanced audio networks are very common in the broadcast business where it is used to prevent ground loops.
Various parts of the network in an installation may be running from different power sources,
Stray currents are possible if adjoining pieces of equipment are powered from different power outlets.
Any stray current in the common lead of parts of a system can result in an interfering signal on the program material.
A commonly used impedance in the network is 600 Ohms. Many different audio transformers are available for this application.
Google 'ground loop'. Consumer audio does not normally use balanced connexions.
A push-pull amp can easily be built to operate both balanced or unbalanced.🙂
Various parts of the network in an installation may be running from different power sources,
Stray currents are possible if adjoining pieces of equipment are powered from different power outlets.
Any stray current in the common lead of parts of a system can result in an interfering signal on the program material.
A commonly used impedance in the network is 600 Ohms. Many different audio transformers are available for this application.
Google 'ground loop'. Consumer audio does not normally use balanced connexions.
A push-pull amp can easily be built to operate both balanced or unbalanced.🙂
Balanced generally sounds different and better to my ears. Yes, i am aware that some, often sophisticated, listeners hate it, but it's still my opinion.
Often "balanced" is pursued just as an additional feature with little regards to sound. Manufacturers sometimes are like that 🙂
If a source is intrinsically balanced, such as a dac based on a balanced chip, having a balanced output makes a lot of sense. Same as inputs, if an amp is based on a balanced/bridged topology, or if even a differential pair is used as a balanced input it makes good sense to use a balanced connection.
If, otoh, the topology is intrinsically single ended, but marketing requires balanced inputs/outputs and these are achieved by introducing additional circuitry such as discrete inverters, opamps, or transformers, the end result usually sounds worse using the balanced interface. Non-deaf reviewers sometimes encounter this phenomenon.
Often "balanced" is pursued just as an additional feature with little regards to sound. Manufacturers sometimes are like that 🙂
If a source is intrinsically balanced, such as a dac based on a balanced chip, having a balanced output makes a lot of sense. Same as inputs, if an amp is based on a balanced/bridged topology, or if even a differential pair is used as a balanced input it makes good sense to use a balanced connection.
If, otoh, the topology is intrinsically single ended, but marketing requires balanced inputs/outputs and these are achieved by introducing additional circuitry such as discrete inverters, opamps, or transformers, the end result usually sounds worse using the balanced interface. Non-deaf reviewers sometimes encounter this phenomenon.
Firstly you have to define what you mean by 'balanced'. It has a specific meaning but is often used to describe a differential input or output...
Can a phono preamplifier with rca inputs have balanced circuitry and have a balanced output? No xlr inputs needed?.
balanced means the impedance in the 2 signal wires is the same (in balance). This doesn't have to mean it's driven on both legs. The Ifi user manual is pretty limp on explaining exactly what they mean by 'balanced' so I'd take everything they say with a pinch of salt.
The phono input could be balanced, as MC cartridges are basically floating. MMs generally have one leg tied to the case as a screen so are less so. It's not truly balanced but some measurements on here have shown a significant reduction in pickup using a balanced input. Not really recommended with RCAs though as they usually have a coax cable which defeats the purpose. My phono stages are balanced in and out, but I'm a wierdo...
The phono input could be balanced, as MC cartridges are basically floating. MMs generally have one leg tied to the case as a screen so are less so. It's not truly balanced but some measurements on here have shown a significant reduction in pickup using a balanced input. Not really recommended with RCAs though as they usually have a coax cable which defeats the purpose. My phono stages are balanced in and out, but I'm a wierdo...
Alas converting an incoming differential signal to single-ended usually introduces more noise than you've saved as its not that simple to do well, 3 opamps are needed to do a reasonable job as you need to buffer before the differential amp stage. In practice balanced is no quieter. And even if done really well the improvement is only 3dB because you have noise from two input devices, not from one, so you lose 3dB from having two sources of noise.
Lets be clear here - a differential signal prevents ground-loop problems and is thus very useful for preventing hum - but this is not the same thing as balanced.
Balanced means the signals are equal and opposite - which is a particular form of differential signalling. There's also the meaning of balanced impedance meaning the two wires carrying the signal see the same impedance (often not the case, incidentally, if there's some naive differential->single-ended conversion circuit).
Any differential signal on a balanced impedance system ought to have the best chance of cancelling external interference, but only balanced signalling has the best chance of not creating noise that affects other cables. Alas its pretty impossible to create systems with balanced impedance upto RF frequencies, so you can't rely on balance or differential signalling to deal with RF interference.
Estoy entendiendo el concepto.
Por lo que dice Billshurv, el IFI ZEN entrando por RCA no es una entrada balanceada.
Bueno, ¿cómo puede este preamplificador tener salidas XRL si su señal de entrada no está balanceada? Se puede hacer esto?
He encontrado una página donde se hacen las medidas de este preamplificador y se comparan con los datos del fabricante.
A veces mayor resultado.
https://www.soundstagenetwork.com/i...reamplifier-measurements&catid=414&Itemid=547
Por lo que dice Billshurv, el IFI ZEN entrando por RCA no es una entrada balanceada.
Bueno, ¿cómo puede este preamplificador tener salidas XRL si su señal de entrada no está balanceada? Se puede hacer esto?
He encontrado una página donde se hacen las medidas de este preamplificador y se comparan con los datos del fabricante.
A veces mayor resultado.
https://www.soundstagenetwork.com/i...reamplifier-measurements&catid=414&Itemid=547
You've fallen into the trap of accepted interpretation vs what it actually means. You can have a single ended output and still get the CMRR benefits of a balanced link. All covered in detail in Bruno Putzey's 'G-word' article. A good read, along with the Whitlock papers on the THAT chips that get around impedance imbalances.
On the phono front I did find the measurement last night. You can get about 20dB rejection with a balanced input, after which common mode pickup from the cartridge makes any more rejection impossible (after all a cartridge coil looks like an antenna, albeit an inefficient one). If that 20dB is worth the additional complexity to you, then it's worth doing.
And these 'references' show what?
-------------------
You appear to be well informed. So it would appear you are having difficulty with
concepts that are not in agreement with what you had previously thought.
But not to worry, as long as the virus doesn't spread, we are safe.
-------------------
You appear to be well informed. So it would appear you are having difficulty with
concepts that are not in agreement with what you had previously thought.
But not to worry, as long as the virus doesn't spread, we are safe.
Hello. I have seen the preamplifier board (IFI ZEN PHONO) and I am pretty sure that the negative side of the input RCAs are not in contact with ground.
Yes they are in the output RCAs.
Can it be said that it is a balanced preamplifier despite having inputs through RCA connectors?
Yes they are in the output RCAs.
Can it be said that it is a balanced preamplifier despite having inputs through RCA connectors?
600 ohms was once THE standard, but that was when we had a power-based distribution system. 600 ohm source equipment was terminated by 600 ohm input z equipment, matching source and load z means maximum power transfer, but it all went up in heat anyway. Those days have been over for over 40 years. We moved to a voltage-based distribution system, output z is now “low”, input z is “bridging”. Signals get distributed free from noise pickup, power is saved, Life’s better.
Yeah, but be careful. The classic “repeat coil” 600/600 transformer must be very carefully applied in today’s world. Nothing will terminate that transformer properly by default, and a few outputs may not drive more than one of them well. Typical line input transformers, if used at all, are 10K or so, and the secondary can be terminated with a resistor. Outputs with transformers are pretty rare, but follow the same idea, low z line driver, transformer that can be either self terminated or deal with a bridging load.
The balanced/differential I/O these days is typically one of the one-chip solutions from THAT or TI. There is a type of differential line receiver that is pre-compensated for the 6dB level pickup issue, and several output circuits that can behave essentially like a transformer output, where you can unbalance it by shoring one leg to ground, and nothing bad happens. Building one from 3 Opamps is difficult, you can do ok with a precision package of 8 or so laser trimmed resistors, but the single chip solutions work so much better with less fuss. Hardly any modern pro gear attempts a 3-opamp instrumentation amp input anymore.
No, they do not sound different from single-ended solutions, so long as the comparison is precision level matched and double-blind. If not, you just hear whatever you expect to hear.
The point of balancing/differential signals is noise immunity. Common Mode Rejection Ratio, usually only worth looking at in the audio band. RFI immunity is dealt with with filtering, toroids, ferrites, etc.
I want to point out that a balanced symmetrical signaling system can also provide the opportunity to eliminate signal currents from traveling within the ground and power supply system - balanced and symmetric signaling does not have to have only an induced noise benefit.
When a balanced and symmetric signal is sent to a balanced load, the net signal current between the sender and the receiver is zero - the positive and negative parts of the transmitted signal cancel at the load. When a signal is sent in a balanced but not symmetric signaling system, then the signal return current is not zero, since there is a net signal current transmitted from the sender to the receiver, and this signal current must travel back to the sending device through the ground connection between the sending and receiving devices.
Similarly, the signal current sent to a load must be sourced from the system of power supply circuitry and bypass capacitors within a sender, and if a symmetric signaling system is used, the positive and negative halves of the circuit can be arranged to cancel much of the required signal currents. If a single ended signal is sent to a receiving device, then that signal current must be completely sourced from the power supply and bypass components, and there is no equal and opposite current that can be summed together to cancel this current draw. Within a symmetric signaling device, there is always a "dual" to every branch current that is of the same magnitude but opposite polarity, and with a clever architecture and PCB layout, these cancellations can reduce signal current coupling to ground and the power supply by a significant factor.
Having signal currents travel within the power supply and ground system is not a guarantee of a poor quality system, but it does point to the possibility of many complex interactions between connected pieces of equipment. In my simplistic view, it seems better to remove signal currents from the ground and power supply system wherever possible, and this points to the benefit of balanced and symmetric signaling, something that is often treated as irrelevant by those who view balanced signaling solely as a noise mitigation technique.
I have a question to you guys re: balanced signals.
If I have a non-symmetrical but balanced line, with one line at ground level, the other at 1Vrms, obviously the signal has a level of 1Vrms.
If I have a symmetrical line with each line 1Vrms (in opposite phase of course) do I call that 1Vrms or 2Vrms? Is there a sort of standard?
Edit: reason I asked is I am working on equipment that needs to accept both balanced and single ended inputs. Do I need to put a 6dB gain in the SE path to maintain equal loudness in the two situations?
Jan
BTW In researching this on the 'net I came across this gem:
"the point of balanced (3 conductors) is to have a discrete path to ground (or chassis) so that noise can escape.." Can you imagine a newbee reading this and having a totally wrong idea for the rest of his life. Where do people get this sh*t?
If I have a non-symmetrical but balanced line, with one line at ground level, the other at 1Vrms, obviously the signal has a level of 1Vrms.
If I have a symmetrical line with each line 1Vrms (in opposite phase of course) do I call that 1Vrms or 2Vrms? Is there a sort of standard?
Edit: reason I asked is I am working on equipment that needs to accept both balanced and single ended inputs. Do I need to put a 6dB gain in the SE path to maintain equal loudness in the two situations?
Jan
BTW In researching this on the 'net I came across this gem:
"the point of balanced (3 conductors) is to have a discrete path to ground (or chassis) so that noise can escape.." Can you imagine a newbee reading this and having a totally wrong idea for the rest of his life. Where do people get this sh*t?
Last edited:
Mark, beg to differ. Balanced concerns itself with impedance balance in sender, receiver, cable, so any commonly introduced noise is cancelled at the receiver.
A balanced cable has in principle no reference to ground and can be anywhere within the system dynamic range (mostly determined by the power supply voltages). A balanced cable can have 1V with respect to some reference like ground, and 0V in the other, and still be balanced.
Symmetrical means both cables have the same level with respect to ground, and this will maximize headroom (with dual equal supplies). It will not help in cancelling noise, unless the cables are also balanced.
Jan
Last edited: