hmmm... i had bought the ones of maxw here:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=84388
thought i was going to use them to bridge some opa549's, but might not gonna get around to it.
also, trying to make my own balanced driver via opamp. will see how that goes....
so...pm/email me and i'll pass them along for what i paid
http://www.diyaudio.com/forums/showthread.php?s=&threadid=84388
thought i was going to use them to bridge some opa549's, but might not gonna get around to it.
also, trying to make my own balanced driver via opamp. will see how that goes....
so...pm/email me and i'll pass them along for what i paid
Question about phase
Hi,
First post for me - please be gentle! I thought this thread would be the obvious place to post this question, as it concerns a balanced line driver using DRV134s. If anyone feels it belongs somewhere else, please let me know.
I've just built a BLD using DRV134s. (The only difference in my layout compared to digi01's is that I'm running them from +/-12V rather than +/-15V, and have included the optional 10uF NP caps).
My problem is that the +Vout and -Vout signals are in phase. Since the purpose is to drive power amps with differential balanced inputs, this isn't good - the result is silence, of course!
Does anyone here know of any circumstance that would cause the DRV134's outputs to be in phase? Or have I just misunderstood what this chip is supposed to do? (I have previously successfully built BLDs using AD SSM2142s, but they are now hard to source and the DRV134 is reputed to be better).
Thanks,
Clive
Hi,
First post for me - please be gentle! I thought this thread would be the obvious place to post this question, as it concerns a balanced line driver using DRV134s. If anyone feels it belongs somewhere else, please let me know.
I've just built a BLD using DRV134s. (The only difference in my layout compared to digi01's is that I'm running them from +/-12V rather than +/-15V, and have included the optional 10uF NP caps).
My problem is that the +Vout and -Vout signals are in phase. Since the purpose is to drive power amps with differential balanced inputs, this isn't good - the result is silence, of course!
Does anyone here know of any circumstance that would cause the DRV134's outputs to be in phase? Or have I just misunderstood what this chip is supposed to do? (I have previously successfully built BLDs using AD SSM2142s, but they are now hard to source and the DRV134 is reputed to be better).
Thanks,
Clive
Sorry - my error
Please disregard my previous post.
I've done some more testing and the signals ARE out of phase, as should be expected. I am using a software 'scope, and was also using the PC as a signal generator. It seems that the 'scope must have been picking up the generated signal instead of the output of the line driver. Once I dragged in a second PC to act as signal generator, I discovered that the output is as it should be. My apologies to everyone. First post, too: how embarrassing
Please disregard my previous post.
I've done some more testing and the signals ARE out of phase, as should be expected. I am using a software 'scope, and was also using the PC as a signal generator. It seems that the 'scope must have been picking up the generated signal instead of the output of the line driver. Once I dragged in a second PC to act as signal generator, I discovered that the output is as it should be. My apologies to everyone. First post, too: how embarrassing
not to be embarassed... challenge creates learning opportunity
it can be frustrating but then seems trivial afterwards. i often have hard time phrasing question so find myself asking simpler (seemingly stupid) questions just to trigger someone to expound in more thorough manner rather than just the simple answer with no path of deduction.have to ask something maybe trivial afterwards with struggle.
what pc probing software you using?...
it can be frustrating but then seems trivial afterwards. i often have hard time phrasing question so find myself asking simpler (seemingly stupid) questions just to trigger someone to expound in more thorough manner rather than just the simple answer with no path of deduction.have to ask something maybe trivial afterwards with struggle.
what pc probing software you using?...
SSM2142 driving input transformers
FYI- I've used the 2142/DRV134 to drive input transformers with superb
results. I simply "tried it" and it worked and sounded very good.
You can use the $9 Edcors (600:600, or 600:10000) I also tried it with a
nickle cored beauty (Sowter 3545). The chip works great.
On the receive side, you might want to try the Edcor transformer.... $18 experiment. The upside is the Edcor does not require a +/- 15V supply. In
addition, you get galvanic isolation from the source. Also, with the transformer,
you can either take the single ended output (ground one phase), or take another
differential output (ground the center tap).
-- Jim
FYI- I've used the 2142/DRV134 to drive input transformers with superb
results. I simply "tried it" and it worked and sounded very good.
You can use the $9 Edcors (600:600, or 600:10000) I also tried it with a
nickle cored beauty (Sowter 3545). The chip works great.
On the receive side, you might want to try the Edcor transformer.... $18 experiment. The upside is the Edcor does not require a +/- 15V supply. In
addition, you get galvanic isolation from the source. Also, with the transformer,
you can either take the single ended output (ground one phase), or take another
differential output (ground the center tap).
-- Jim
I noticed that this Design is exactly the same as the Applications information in the DRV134 datasheet...so this isn"t an original and new design but one that does work quite well...
The DRV134 is an OK Ballanced Line driver but the Chip costs allmost $5 and you can Make one that performs just as well out of a NE5532 or a TL072 (or allmost an Inverting Dual opamp)at 1/4 the cost and in Non-Inverting mode you can even make one with Variable gain as opposed to Unity gain in the DRV134....
I have several Designs for createing Ballanced Line drivers out of Low cost opamps or Transistors (NPN & PNP) and they are just as easy to use and build as the DRV134 ones and much cheaper.....
If anyone wants the schematics Let me know....
Cheers
The DRV134 is an OK Ballanced Line driver but the Chip costs allmost $5 and you can Make one that performs just as well out of a NE5532 or a TL072 (or allmost an Inverting Dual opamp)at 1/4 the cost and in Non-Inverting mode you can even make one with Variable gain as opposed to Unity gain in the DRV134....
I have several Designs for createing Ballanced Line drivers out of Low cost opamps or Transistors (NPN & PNP) and they are just as easy to use and build as the DRV134 ones and much cheaper.....
If anyone wants the schematics Let me know....
Cheers
DRV134
Actually no you cant build one as good for under $5.00, not even close.
The DRV134 is a true ballanced cross feed back circuit, laser etched resistors for less then .1% to 1% tollerance.
To make one you will need 3 HQ ultra low distortion op amps a hand full of .1% resistors bypass caps ect....... OH yeh you will half to etch a very thought out PC board.
The DRV will drive a 600ohm high capacitive load with less then .0005%THD with a pretty flat curve from DC to 100Khz, oh and 15V/us in reality even a little better.
So if you can get the op amps etch the board get the proper layout the resistors and caps for under $5.00 good luck
Actually no you cant build one as good for under $5.00, not even close.
The DRV134 is a true ballanced cross feed back circuit, laser etched resistors for less then .1% to 1% tollerance.
To make one you will need 3 HQ ultra low distortion op amps a hand full of .1% resistors bypass caps ect....... OH yeh you will half to etch a very thought out PC board.
The DRV will drive a 600ohm high capacitive load with less then .0005%THD with a pretty flat curve from DC to 100Khz, oh and 15V/us in reality even a little better.
So if you can get the op amps etch the board get the proper layout the resistors and caps for under $5.00 good luck
Ballanced Audio
This is a post in reply to a question posted by WIM back on 10-28-2004......
Question/Statement: excuse my newby question, but what exactly is a balanced line driver?? Can i see it as a pre-amp?? or is it only to connect 2 mono amps (gainclones) to 1 speaker.
Can anyone explain it to me in newbylanguage?
...
Response: Ballanced Audio lines are essentially three Wires that form two audio lines. The audio in each is 180 degrees out of phase from each other. When an audio signal passes down the line to the distant receiver it is highly immune to noise, A.C. Hum or other induced noises.
....
The Key is the receiver (yea I know the transmitter is also of some value here - but unless you are operating in an extreme noise environment , this is not all that critical). The receiver is a diffential Amplifier. This means it compares the two signals arriving at the amp input. Since the two signals are 180 degrees out of phase, the noise will not normally be in-phase with both wire signals - therefore it will be rejected and cancelled out by the difference signal that it causes (because it will either be in-phase with any one of the two - but will then be out of phase with the other audio line - which is already 180 degrees out of phase from it's counterpart).
...
This is why the Ballanced Lines are labeled (+) and (-). It a TRS (Tip/Ring/Sleeve plug the (+) and (-) are the Tip/Ring, the sleeve is the signal common.
...
Since the signals arriving at the receiver need to be 180 degrees out of phase - but equal in amplitude - they are considered ballanced (equal). Noise induced is unballanced and since it is not 180 degrees out of phase accross the two audio paths - it is unballanced and rejected (it will cancel itself out)..
...
If you are familliar with a Push-Pull Amplifier Design - you will note that the diffrential amp stage separates the audio into two signals: a Positive and a negative - which is fed to the PNP/NPN Transistors. The signal is essentially split between the transistors.
...
The only problem with using a Push-Pull amp diferential circuit is that it is like an audi rectifier. It splits the sinal and only passes a positive portion of the waveform when the audio signals are positive, and negative when they are negative. (OK Gang - hang with me - I know audio signals are complex, but alot of math will only confuse the subject).
...
The ballanced line transmitter mirrors the audio signals. When the input sine wave is going positive - the line driver puts out two sides at the same time - one going positive (+) and one going negative (-). So it is essentially two amps in parallel with their inputs being one common signal input and an inverting and a non-inverting amplifier to drive the line.
...
The key is the differebntial amp receiver that has both positive and negative inputs. When both are equal in signal strength and 180 degrees out of phase - then the signal (the sine wave) is reproduced in the amps output. The phase jitter or harmonic distortion is only a factor when the two (=)/(-) wires are of significantly different lengths. This causes the two out of phase signals to arrive at different time intervals and appear as noise. This phase difference is reproduced in the amplifier output because it is significantly equal in amplitude.
..
In a differential amplifier input the two signals must be equal in amplitude and 180 degrees out of phase for proper noise reduction. If they are equal in amplitude, but slightly out of phase the noise cancelling charateristics are compromised, but noise will still be reduced - but not totally illiminated. The same is true if In-Phase, but slightly different amplitudes.
...
If a high quality Transmitter and a matched receiver is used - the noise rejection properties are optimized and the afore mentioned conditions are minimalized (but are still present in the output signal at the receiver).
...
This is why in the early days of Broadcasting - Telephone pairs were "Normalled" when set-up for a broadcast signals. For Stereo (two pairs) they were normalled as quads (two signal pairs). This meant that they used the same length of crossover wire for each wire, in each pair, and each set of pairs in the quad. This effectively put the signals on a path of equal resistance and length, thus minimizing the chance that the transmitted audio would arrive at a different (delayed) time frame, and that the attenuation accross the pair (and quad) are equal - so no variance in signal amplitude.
...
Telco lines (in the analog world) al;so come in two flavors. Dry (no telco battery voltage) and wet (with 48VDC Telco Battery accross the pair). Normally, Broadcasters used "Dry" Pairs. Since the battery on a phone line is mostly for signalling sense (to sence when a line is pick uop or not) it is not necessary for a broadcaster who will be active continuously.
..
Incidentally, putting a DC voltage on the pair helps to eliminate attenuation over distance. But we are talking distabces in excess of 20,000 feet (cable feet). As the lines get longer from the central office the battery voltage keeps the signal manageable - because the lines are ballanced!!!!
...
Voltage on a ballanced line is alot like phantom voltage for a ballanced mic line. The line terminates in a center tapped 660 Ohm Transformer. The +48VDC Voltage is applied to the 330 Ohm center tapp of the 660 Ohm Transformer winding on the pair side leaving the telephone central office. The telephone at your house picked the +48VDC off of the center tap of the telephone hybrid in the phone receiver. When you take the phone off hook (you pick up the receiver) the phone draws a current. The central office has a relay in-line with the +48VDC feeding the Center Tap at the central office that senses the current draw at alerts the phone central office that you have picked up the phone. When you pick up the phone a swich in the phone onnects the center tap to the phone varistor network, and therefore turns on your phone, and draws power from the phone line. But I am digressing - except the voltage on a telephone pair is equal in both wires of the pair, the negative (-) power is in the earth ground. So the Telco Pair is both a 660 Ohm Ballanced line with a ballanced DC Bias when in use, which minimizes the loop loss in the signal. Once again - the pair is "ballanced"
...
Hope I have helped explain this to you....Take into consideration that phone lines are referred to a 600 ohm circuits (beacuse in the early days they were initially called 660's - which is a signalling frequency in dial tone and when a 660 tone is dropped the trunk lines go to busy. Since 660Hz iwas used in the dial tone referring to 660 when talking about impedance caused some confusion when the outside plant tech was talking to the Inside plant Tech. So the Outside plant guy always looked for 660 ohms and called it 600 ohms. When he was looking at the dial tone he looked for the 660hz, 340nz, and 220hz. When a ring tone of 20hz was received, a low pass R/C network passed the 20Hz throgh the capacitor to the ringer bell (and activated a relay that pulled the receiver off line to keep the noise of the ringer from being sent to the handset earpiece (an early problem with the first generation of phones). In the 50's they installed a set of relay contacts on the line sense relay that disconnected the phone pair from the ringer generator circuitry to keep the 20hz out all together. After all, you only want to ring a phone to tell someone to pick up, that they are getting a call - once they have the handset up and are talking it taint good to pop their eardrums.
...
But like I said I Digress...
...
Ballanced audio is the way to go for truly clean audio (like noise figures in excess of -90 dbmr0. Especially when you have long distances, and have to contend with possible signal contamination by inductive AC voltages and spike noise from machinery (such as the smoothier blender in the break room - hey? Your Boss doesn't have one in Your Break Room? - what a shame!!!)...
...
For short home audio the 100K un-ballanced lines work just fine (after all very few people can hear down to -70dbmr0 - much less -90). Oh yea - before I get flamed for the 100K un-ballanced line statement - it is an arbitrary standard. Home audio is normally terminated with a relatibve impedance of 100K Ohms (which is why the preamps have 100K resistors accross the output and common, and 100K resistors accross the input signal and common). You design circuits without them - shame on you... You have to have at least one between an Amp and Preamp, or your load has to be a transformer - which will also work). Using the input transistor as a load is not a good deal. Remember - you do want clean laudio with as low a noise figure as possible - right?
...
But that's a whole different subject altogether, and my fingers are getting tired....
<,,,
Take Care
Dave R. Mason
High Forrest Technologies
This is a post in reply to a question posted by WIM back on 10-28-2004......
Question/Statement: excuse my newby question, but what exactly is a balanced line driver?? Can i see it as a pre-amp?? or is it only to connect 2 mono amps (gainclones) to 1 speaker.
Can anyone explain it to me in newbylanguage?
...
Response: Ballanced Audio lines are essentially three Wires that form two audio lines. The audio in each is 180 degrees out of phase from each other. When an audio signal passes down the line to the distant receiver it is highly immune to noise, A.C. Hum or other induced noises.
....
The Key is the receiver (yea I know the transmitter is also of some value here - but unless you are operating in an extreme noise environment , this is not all that critical). The receiver is a diffential Amplifier. This means it compares the two signals arriving at the amp input. Since the two signals are 180 degrees out of phase, the noise will not normally be in-phase with both wire signals - therefore it will be rejected and cancelled out by the difference signal that it causes (because it will either be in-phase with any one of the two - but will then be out of phase with the other audio line - which is already 180 degrees out of phase from it's counterpart).
...
This is why the Ballanced Lines are labeled (+) and (-). It a TRS (Tip/Ring/Sleeve plug the (+) and (-) are the Tip/Ring, the sleeve is the signal common.
...
Since the signals arriving at the receiver need to be 180 degrees out of phase - but equal in amplitude - they are considered ballanced (equal). Noise induced is unballanced and since it is not 180 degrees out of phase accross the two audio paths - it is unballanced and rejected (it will cancel itself out)..
...
If you are familliar with a Push-Pull Amplifier Design - you will note that the diffrential amp stage separates the audio into two signals: a Positive and a negative - which is fed to the PNP/NPN Transistors. The signal is essentially split between the transistors.
...
The only problem with using a Push-Pull amp diferential circuit is that it is like an audi rectifier. It splits the sinal and only passes a positive portion of the waveform when the audio signals are positive, and negative when they are negative. (OK Gang - hang with me - I know audio signals are complex, but alot of math will only confuse the subject).
...
The ballanced line transmitter mirrors the audio signals. When the input sine wave is going positive - the line driver puts out two sides at the same time - one going positive (+) and one going negative (-). So it is essentially two amps in parallel with their inputs being one common signal input and an inverting and a non-inverting amplifier to drive the line.
...
The key is the differebntial amp receiver that has both positive and negative inputs. When both are equal in signal strength and 180 degrees out of phase - then the signal (the sine wave) is reproduced in the amps output. The phase jitter or harmonic distortion is only a factor when the two (=)/(-) wires are of significantly different lengths. This causes the two out of phase signals to arrive at different time intervals and appear as noise. This phase difference is reproduced in the amplifier output because it is significantly equal in amplitude.
..
In a differential amplifier input the two signals must be equal in amplitude and 180 degrees out of phase for proper noise reduction. If they are equal in amplitude, but slightly out of phase the noise cancelling charateristics are compromised, but noise will still be reduced - but not totally illiminated. The same is true if In-Phase, but slightly different amplitudes.
...
If a high quality Transmitter and a matched receiver is used - the noise rejection properties are optimized and the afore mentioned conditions are minimalized (but are still present in the output signal at the receiver).
...
This is why in the early days of Broadcasting - Telephone pairs were "Normalled" when set-up for a broadcast signals. For Stereo (two pairs) they were normalled as quads (two signal pairs). This meant that they used the same length of crossover wire for each wire, in each pair, and each set of pairs in the quad. This effectively put the signals on a path of equal resistance and length, thus minimizing the chance that the transmitted audio would arrive at a different (delayed) time frame, and that the attenuation accross the pair (and quad) are equal - so no variance in signal amplitude.
...
Telco lines (in the analog world) al;so come in two flavors. Dry (no telco battery voltage) and wet (with 48VDC Telco Battery accross the pair). Normally, Broadcasters used "Dry" Pairs. Since the battery on a phone line is mostly for signalling sense (to sence when a line is pick uop or not) it is not necessary for a broadcaster who will be active continuously.
..
Incidentally, putting a DC voltage on the pair helps to eliminate attenuation over distance. But we are talking distabces in excess of 20,000 feet (cable feet). As the lines get longer from the central office the battery voltage keeps the signal manageable - because the lines are ballanced!!!!
...
Voltage on a ballanced line is alot like phantom voltage for a ballanced mic line. The line terminates in a center tapped 660 Ohm Transformer. The +48VDC Voltage is applied to the 330 Ohm center tapp of the 660 Ohm Transformer winding on the pair side leaving the telephone central office. The telephone at your house picked the +48VDC off of the center tap of the telephone hybrid in the phone receiver. When you take the phone off hook (you pick up the receiver) the phone draws a current. The central office has a relay in-line with the +48VDC feeding the Center Tap at the central office that senses the current draw at alerts the phone central office that you have picked up the phone. When you pick up the phone a swich in the phone onnects the center tap to the phone varistor network, and therefore turns on your phone, and draws power from the phone line. But I am digressing - except the voltage on a telephone pair is equal in both wires of the pair, the negative (-) power is in the earth ground. So the Telco Pair is both a 660 Ohm Ballanced line with a ballanced DC Bias when in use, which minimizes the loop loss in the signal. Once again - the pair is "ballanced"
...
Hope I have helped explain this to you....Take into consideration that phone lines are referred to a 600 ohm circuits (beacuse in the early days they were initially called 660's - which is a signalling frequency in dial tone and when a 660 tone is dropped the trunk lines go to busy. Since 660Hz iwas used in the dial tone referring to 660 when talking about impedance caused some confusion when the outside plant tech was talking to the Inside plant Tech. So the Outside plant guy always looked for 660 ohms and called it 600 ohms. When he was looking at the dial tone he looked for the 660hz, 340nz, and 220hz. When a ring tone of 20hz was received, a low pass R/C network passed the 20Hz throgh the capacitor to the ringer bell (and activated a relay that pulled the receiver off line to keep the noise of the ringer from being sent to the handset earpiece (an early problem with the first generation of phones). In the 50's they installed a set of relay contacts on the line sense relay that disconnected the phone pair from the ringer generator circuitry to keep the 20hz out all together. After all, you only want to ring a phone to tell someone to pick up, that they are getting a call - once they have the handset up and are talking it taint good to pop their eardrums.
...
But like I said I Digress...
...
Ballanced audio is the way to go for truly clean audio (like noise figures in excess of -90 dbmr0. Especially when you have long distances, and have to contend with possible signal contamination by inductive AC voltages and spike noise from machinery (such as the smoothier blender in the break room - hey? Your Boss doesn't have one in Your Break Room? - what a shame!!!)...
...
For short home audio the 100K un-ballanced lines work just fine (after all very few people can hear down to -70dbmr0 - much less -90). Oh yea - before I get flamed for the 100K un-ballanced line statement - it is an arbitrary standard. Home audio is normally terminated with a relatibve impedance of 100K Ohms (which is why the preamps have 100K resistors accross the output and common, and 100K resistors accross the input signal and common). You design circuits without them - shame on you... You have to have at least one between an Amp and Preamp, or your load has to be a transformer - which will also work). Using the input transistor as a load is not a good deal. Remember - you do want clean laudio with as low a noise figure as possible - right?
...
But that's a whole different subject altogether, and my fingers are getting tired....
<,,,
Take Care
Dave R. Mason
High Forrest Technologies