Douglas Self Line Input compared to Mackie line input

[k brown]

One advantage is the availabliity of an inverted signal, if one should need it.

 

[Newmarket]

thank you for the clarification,
so now my question is, what are the advantages of using "active" impedance balanced output with an additional opamp (like the one I showed above) versus the simpler approach of using just the series build out resistor + its matched resistor on the undriven side?

Not quite following. The output stage you show has only one op amp. As does a simple Impedance Balanced output ? There is no additional op amp , just a different circuit topology.
Or am I misunderstanding ?

 

[k brown]

Yes, "the one above" is too vague; there are several schems 'above' that post; OP should give post numbers when referencing a schem. I was going by the text referring to 'active balanced'.

 

[Dimitree]

I apologize for the unclearness of my questions,
what I wanted to ask is, considering those two approaches below, that can be used to add a "direct output" after a balanced receiver (for example the one proposed by Thor on post #24),
I was wondering what is the advantage of the first approach (the additional opamp in ground-sensing configuration) compared to the second approach with only two resistors. I assume the opamp approach introduces a noise penalty compared to the simple build out resistor, so there must be other advantages that I still haven't understand

 

[john12ax7]

One advantage with two opamps is you also have signal on the -out, so 6dB more differential output voltage if you need it. A common technique for a system is to have -6dB for the input, then +6dB on the output, overall unity gain but you now have 6dB extra internal headroom.

 

[Brian Roth]

One thing to be aware of are the "brain dead" line output designs that simply use a pair of opamps (one inverting) to create a "balanced output". OK if driving a balanced destination, and likely a train wreck driving an unbalanced destination. Hence, the "Tascam Problem"....or the "Avid interface Problem".

Bri

 

[Newmarket]

I apologize for the unclearness of my questions,
what I wanted to ask is, considering those two approaches below, that can be used to add a "direct output" after a balanced receiver (for example the one proposed by Thor on post #24),
I was wondering what is the advantage of the first approach (the additional opamp in ground-sensing configuration) compared to the second approach with only two resistors. I assume the opamp approach introduces a noise penalty compared to the simple build out resistor, so there must be other advantages that I still haven't understand

View attachment 120556

As I pointed out, the first circuit actually provides CMMR when feeding an unbalanced input. If you know you are feeding a balanced input then you don't need it.

 

[Dimitree]

As I pointed out, the first circuit actually provides CMMR when feeding an unbalanced input. If you know you are feeding a balanced input then you don't need it.

thank you, I missed this point before, now it is clear. I'm sorry if I take too long to understand those topics, it's all quite new to me.

I have seen some outputs which appeared as if the designer only partially understood the principle, such as having a matching resistor, but not including the output capacitor on the undriven side.

is this perhaps one of those designs? There is no output capacitor on the undriven side.

That output stage is a "Ground Sensing" configuration. It provides CMRR into an unbalanced input by adding the remote "Ground" into the signal transmitted. Well covered in Self's "Small Signal..." book. That version is also configured as a "Zero Impedance" output.

what puzzled me was that on the book, when talking about "zero impedance" outputs, he says: "The impedance-balancing resistor on the cold pin has been replaced by a link to match the near-zero output impedance at the hot pin."
Instead here I see a 75R on the cold pin. Is that resistor needed again because of the "ground sensing" configuration?
Unfortunately on the book there are no examples of "zero impedance" output and "ground sensing" output combined, hence my doubts.

 

[drtechno]

thank you, I missed this point before, now it is clear. I'm sorry if I take too long to understand those topics, it's all quite new to me.


is this perhaps one of those designs? There is no output capacitor on the undriven side.

View attachment 120569


what puzzled me was that on the book, when talking about "zero impedance" outputs, he says: "The impedance-balancing resistor on the cold pin has been replaced by a link to match the near-zero output impedance at the hot pin."
Instead here I see a 75R on the cold pin. Is that resistor needed again because of the "ground sensing" configuration?
Unfortunately on the book there are no examples of "zero impedance" output and "ground sensing" output combined, hence my doubts.

the 75 ohm resistors are the set points of voltage reference the zero crossing of the signal sees. The coupling capacitor provides a near zero output impedance while isolating the two signal reference voltages. When the output is shorted (tip to sleeve) or zero impedance load is applied, the 75 ohm resistors provide the critical loading for the op amp.

 

[Newmarket]

the 75 ohm resistors are the set points of voltage reference the zero crossing of the signal sees. The coupling capacitor provides a near zero output impedance while isolating the two signal reference voltages. When the output is shorted (tip to sleeve) or zero impedance load is applied, the 75 ohm resistors provide the critical loading for the op amp.

ALERT: nearly all of this is nonsensical. "Zero crossing" doesn't come into it. "Coupling capacitor" does not contribute to a (small signal) "Zero Impedance" output.
Is this comment using ANI in part ?
Variation of opinion is one thing and often healthy. But pure disinformation is wholly negative.

 

[Newmarket]

thank you, I missed this point before, now it is clear. I'm sorry if I take too long to understand those topics, it's all quite new to me.


is this perhaps one of those designs? There is no output capacitor on the undriven side.

View attachment 120569

The "Ground Sensing" topology referenced does not rely on equal impedances in the same way as a simple impedance balanced output.
In your particular case going into a balanced input IIRC an impedance balanced output should work fine.
The shown circuit works well but real world performance can be well below textbook.
Bare in mind that in studio work the major problem is often "Ground Loop" hum etc and knocking even, say, 50dB off this is an effective solution.
tbh first time I encountered this type of output (last century ) I just didn't get it as didn't get the concept that it "reads" the remote "Ground" and uses that to compensate the output to suit.

what puzzled me was that on the book, when talking about "zero impedance" outputs, he says: "The impedance-balancing resistor on the cold pin has been replaced by a link to match the near-zero output impedance at the hot pin."
Instead here I see a 75R on the cold pin. Is that resistor needed again because of the "ground sensing" configuration?
Unfortunately on the book there are no examples of "zero impedance" output and "ground sensing" output combined, hence my doubts.

 

[thor.zmt]

while browsing some old schematics, I noticed that the Soundcraft 8000 uses the same "double inverted" configuration (at least to my novice eyes) that we were discussing before, although with a slight more complex gain control network (well, more complex for me)


View attachment 120503

This has fairly high impedance's which translates into excess noise. Using modern OPA's you can easily use all resistor values reduced by a factor 3 or possibly even ten (OPA1656)

I also have another question..what would it be the best approach (noise-wise) to add a direct output after this receiver? the goal is to send this direct output to an external audio interface for recording.

Key question is what levels are needed by the audio interface. If the level at the output is "correct" for 0dBFS on peaks, just a simple impedance balanced out is fine. If attenuation or gain is needed, that means extra circuitry is needed.

I think impedance balanced output with a pair of 75R resistors after the opamp would be enough (I don't think I need fully balanced output, the connection between the console and the audio interface are extremely short, so I think I can sacrifice CMRR here) but looking at other consoles I often see this approach:


View attachment 120508

This is as others remarked "ground sensing", realistically for your application the main advantage over using just a pair of resistors is that you get more noise and distortion.

Again, it all depends on how important CMRR is to you. As said, lowest noise happens when gains and levels are correctly matched and circuits have low overall impedance and self noise from active parts. This also gives least CMRR and flexibility.

If you design mass market, mass produced gear, CMRR and compatibility is king, a few dB extra self noise are usually not seen as an issue.

If you design for your own use, your own studio, you can absolutely tailor-make everything "just so". So I would say different design targets and different performance emphasis is applicable.

Thor

 

[Brian Roth]

Go screw yourself if your dumb British mind can't handle it. Alert Bitch

WHAT??????

Bri

 

[ccaudle]

what are the advantages of using "active" impedance balanced output with an additional opamp

Double the output amplitude (+6dB) with the same power supply voltages. For example if you want or need to have +24dBu output, that is over 12V RMS, so over 17V peak. You just can't do that with 15V bipolar supplies, and probably not even with acceptable distortion with 18V bipolar supplies.
Drive both legs and now the output drivers just have to supply under 9V peak, no problem even with 12V bipolar supplies.

Just for clarity this is the circuit I am referring to:

is this [ground sensing output] perhaps one of those designs?

No, a ground sensing output circuit is not a balanced output, it is an unbalanced output which also has an input which is used to partially cancel common mode difference between the driver and receiver.

This is what I mean, a partial (or bad) impedance balanced output:



The cold leg has a resistor to match the hot leg, but no capacitor. The correct way to match the output impedance of the hot leg would be like this:



Note an essential difference between those and the circuit in your post #68: the feedback point is taken before the output resistor, so the output resistor directly increases the output impedance.
The circuit in your post #68 takes the feedback after the resistor, so that the feedback action of the op-amp will counteract the voltage drop across the 75 Ohm resistor, which makes the effective output impedance just the impedance of the capacitor in the lower frequencies (but rising with frequency since the open loop gain is decreasing with frequency, reducing the effect of the feedback).

Instead here I see a 75R on the cold pin.

In parallel with the output, not in series. When referring to output impedance it would refer to impedances in series with the output, between the output driver and receiver input.

Is that resistor needed again because of the "ground sensing" configuration?

In that circuit it is needed so that when there is no connector inserted, or if the output is connected to a capacitor coupled balanced input, the reference pin for the circuit is not completely floating. If that node floats the output circuit has no DC reference, so it could float to any value, up to the point where the circuit is limited by the power supply voltages. Op-amp inputs always need a path for the DC bias current to flow.

 

[Dimitree]

thank you everyone for those valuable explanations

 

[thor.zmt]

The shown circuit works well but real world performance can be well below textbook.

Mind you, in 2023 usually the opposite is true. Consistency of quality passive parts is exceptional (e.g. reels of thin film, 100pmm SMD metal film resistors rated at 1% tend to do much, much better in practice) and improvements in Op-Amp performance over 553X have been dramatic.

So in practice, average performance tends to be much better than the "textbook" average and worst case stated for "textbook" circuits.

The same tends to be true of capacitors. Using 2.5% rated C0G SMD Capacitors in EQ Circuits with 5pcs in parallel produces on average ~ 1% Tolerance. Yet in 100% testing of a commercial low cost Phono stage 100% passed better than +/-0.1dB deviation from RIAA Standard. With 1% Metal film resistors and 2.5% rated commodity parts selling for fractions of a cent USD in volume.

We should still challenge the wisdom of these "textbook" circuits from the 1960's and 1970's in 2023. Many things have changed. Improvements are commonly possible.

One just needs to be clear what performance parameters we want to maximise, e.g. from low circuit complexity, low noise and high CMRR select any two, but not all three.

Thor

 

[JohnRoberts]

We should still challenge the wisdom of these "textbook" circuits from the 1960's and 1970's in 2023. Many things have changed. Improvements are commonly possible.

One just needs to be clear what performance parameters we want to maximise, e.g. from low circuit complexity, low noise and high CMRR select any two, but not all three.

Thor

and cost?

JR

 

[thor.zmt]

and cost?

JR

Cost for DIY is immaterial.

I charge 500 USD per day for consultancy, that's the friendly rate. So 1 hour I spend on some DIY is worth at least 62.50 USD.

Next to that almost all components cost pales.

I have, since growing up in eastern Europe never considered DIY a way to save money.

It is a way to get what you really want from what you can get your grubby mitts on and which is simply not available (or silly lala land price).

Mind you, in 2024 the mass production costs are down so much on stuff that would have been world leading 100 times over that would behave been the GOAT in 1995.

As I remarked, a recent phono pre selling under 250 USD offers a performance and consistency that, in 1995 was not possible with a 5,000 USD phono with components hand selected by certified vestal virgin's.

And I might suggest the actual realised margins on the 250 bux phono in percentages are greater than those of the 1995 5,000 USD phono, once everything is accounted for.

What is more, swapping relatively high quality passives and actives for the worst garbage barely lowers the cost, as the majority is elsewhere.

Thor

 

[JohnRoberts]

Cost for DIY is immaterial.

can you lend me a dollah?

I charge 500 USD per day for consultancy, that's the friendly rate. So 1 hour I spend on some DIY is worth at least 62.50 USD.

more than I charged when I was still accepting work... Last century. [edit: I actually charged $100/hour so maybe we were in the same ballpark. ]

Next to that almost all components cost pales.

its all relative..

I have, since growing up in eastern Europe never considered DIY a way to save money.

I used to operate a kit company back in the 70s-80s. My kit company customers were very much influenced by cost.

It is a way to get what you really want from what you can get your grubby mitts on and which is simply not available (or silly lala land price).

My dream list these days does not include analog audio paths.

Mind you, in 2024 the mass production costs are down so much on stuff that would have been world leading 100 times over that would behave been the GOAT in 1995.

Yup, it is hard to justify DIY for anything you can buy off the shelf.

As I remarked, a recent phono pre selling under 250 USD offers a performance and consistency that, in 1995 was not possible with a 5,000 USD phono with components hand selected by certified vestal virgin's.

I personally hold my own phono preamp designs in pretty high regard. I escaped the audiophile market in the 80s because I did not perceive a correlation between performance and market success. While more than 10 years before your 1995 hypothetical, I got an accidental review in Stereophile magazine comparing my $150 phono preamp favorably to a Mark Levinson ($5K ?).

After walking away from that market, I pivoted to cost effective live sound reinforcement (Peavey) because it was harder to BS an auditorium full of listeners.

And I might suggest the actual realised margins on the 250 bux phono in percentages are greater than those of the 1995 5,000 USD phono, once everything is accounted for.

What is more, swapping relatively high quality passives and actives for the worst garbage barely lowers the cost, as the majority is elsewhere.

Thor

I still have some gold plated phono jacks in my back lab , anybody interested?

JR

 

[Script]

I still have some gold plated phono jacks in my back lab , anybody interested?

Yes, please send over. I could put to really good use. Had a talk with my dentist. Will need a cap soon...