SET amp owners thread

[Atmasphere]

Ralph,

Curiously it seems counterintuitive. Why would a few tens of miliohms in series with speakers that usually have higher impedance than speakers designed for SS affect the sound of a low damping amplifier?

In fact we see many people using vintage cables of low section with SET amplifiers.

If the speaker is 16 Ohms you can totally do that- 16 Ohms is why we got away with zip cord when we were kids. But now there are 4 Ohm speakers, and with them speaker cables are critical. Even more so if the amplifier is lower power and/or higher output impedance. I found that out the hard way/first hand.

You don't have a lot of damping factor to deal with; from the speaker's perspective the DCR of the speaker cable is part of the source impedance. So even small amounts can have a noticeable and measureable effect.

 

[SVS]

The advantage of shorter speaker cables is an advantage; in addition to helping get the bass right, things like vocals will be more easily discerned. Even though the amplifier might have decent bass in itself, its high output impedance might otherwise pose a difficulty, especially if the OPT is used on a lower impedance tap! By using balanced lines, you can eliminate the colorations that so many single-ended cables impose (if you've auditioned two such cable side by side and heard a difference, that's exactly what I'm talking about), and one of those colorations can be a lack of bass; IOW I'm not concerned about adding bass rather not losing it. I play bass (since 6th grade) so I need to be convinced by the system's ability to play bass correctly.

I fully agree with the rest of your comments that I didn't quote.

I do not argue about the advantage of shorter cables. But the cable won't add bass when the amp doesn't.
Different RCA cables have different sounds and this is not surprising. So why should RCA and balanced cables sound the same?
I think that balanced cables in general do not give any advantages for SET amplifiers. But they do no harm either. They cause damage only at the input of RIAA preamps.

 

[Solypsa]

I think that balanced cables... They cause damage only at the input of RIAA preamps.

I would be interested in a little more detail behind your statement above. No challenge, just curious..

 

[SVS]

I would be interested in a little more detail behind your statement above. No challenge, just curious..

Everything is very simple. A balanced preamp practically consists of two amplifiers, one for amplifying a "hot" signal, the other for amplifying a "cold" signal. An unbalanced preamp has only one such amplifier. Let all amplifiers have a gain of K, and the noise level brought to the input Un (I am not a native English speaker, so I may be wrong in the translation of technical terms).
The noise level at the output will be KUn.
The signal from the Uk cartridge in the unbalanced preamp is fed to the input, and at the output we measure the voltage KUk.
The signal from the Uk cartridge is fed to the inputs of the balanced preamp, is divided by 2 at its input resistances, and the voltage Uk/2 is actually present at each input. After amplification, there will be a voltage of KUk/2 at each output.
The signal-to-noise ratio in an unbalanced preamp will be KUk/KUn=Uk/Un.
The signal-to-noise ratio in the balanced preamp will be KUk/2KUn=Uk/2Un, that is, twice less, or worse by 6 dB.

I apologize for taking so long, but this is simple arithmetic that no one thinks about. Unfortunately.

 

[Atmasphere]

I do not argue about the advantage of shorter cables. But the cable won't add bass when the amp doesn't.
Different RCA cables have different sounds and this is not surprising. So why should RCA and balanced cables sound the same?
I think that balanced cables in general do not give any advantages for SET amplifiers. But they do no harm either. They cause damage only at the input of RIAA preamps.

I stated that I did not want to lose bass. I did not state that a cable could add it!

An RCA cable can only sound like a balanced cable when its working perfectly with no coloration whatsoever. When you hear differences in cables, the one takeaway you always have is both cables are incorrect. This is simply because next year, the manufactuer of the 'better' cable will have a new and improved version and if he doesn't, someone else will. I knew the person that founded the high end audio cable industry (Robert Fulton)- he lived in the Twin Cities where I live. So I've seen this phenomena going on since the 1970s and its not slowed down. When you go balanced, you get off the white elephant merry go round and don't have to deal with cable differences any longer. This is assuming of course that your equipment supports the balanced line standards.

Everything is very simple. A balanced preamp practically consists of two amplifiers, one for amplifying a "hot" signal, the other for amplifying a "cold" signal. An unbalanced preamp has only one such amplifier. Let all amplifiers have a gain of K, and the noise level brought to the input Un (I am not a native English speaker, so I may be wrong in the translation of technical terms).
The noise level at the output will be KUn.
The signal from the Uk cartridge in the unbalanced preamp is fed to the input, and at the output we measure the voltage KUk.
The signal from the Uk cartridge is fed to the inputs of the balanced preamp, is divided by 2 at its input resistances, and the voltage Uk/2 is actually present at each input. After amplification, there will be a voltage of KUk/2 at each output.
The signal-to-noise ratio in an unbalanced preamp will be KUk/KUn=Uk/Un.
The signal-to-noise ratio in the balanced preamp will be KUk/2KUn=Uk/2Un, that is, twice less, or worse by 6 dB.

I apologize for taking so long, but this is simple arithmetic that no one thinks about. Unfortunately.

Your second sentence above is incorrect. Most balanced line preamps (like ours, which has always had a balanced differential phono section) are fully differential. Differential amplifiers are found in ALL amplifiers, including SETs.

Don't believe me? The first commercial opamps were made by George Philbrick Research and were simply a single tube; one input (inverting) was the grid of the tube and the other input was the cathode. Tubes amplify when there is a difference between the state of the grid and the cathode (usually the cathode is referenced to ground, but that is still a difference compared to the input signal). I have used this techique to drive SETs in the differential domain, obviating the need for an input transformer. To do this pin 2 of the XLR is the same as the center pin of the RCA and pin 3 is tied (via a coupling capacitor) to the cathode of the same tube. I've been doing this for over 20 years. Because the cathode is an input to the tube, this technique does not alter the character of the amplifier.

So a fully balanced differential amplifier does not consist of two amplifiers, there is simply one amplifier with both inputs used. Google is your friend... For any given stage of gain that is ideally executed differentially as opposed to single-ended, you get up to 6dB less noise. This is one reason why solid state circuits can be so much quieter; its why we can have only two stages of gain in our phono section, have passive EQ and still work with LOMC cartridges. So I think you have this backwards, based on your initial assumption that a balanced phono section (or any other balanced circuit) will be dual amplifiers (in which case you would be correct; but of course no-one does balanced that way).

 

[SVS]

Your second sentence above is incorrect. Most balanced line preamps (like ours, which has always had a balanced differential phono section) are fully differential. Differential amplifiers are found in ALL amplifiers, including SETs.

Don't believe me? The first commercial opamps were made by George Philbrick Research and were simply a single tube; one input (inverting) was the grid of the tube and the other input was the cathode. Tubes amplify when there is a difference between the state of the grid and the cathode (usually the cathode is referenced to ground, but that is still a difference compared to the input signal). I have used this techique to drive SETs in the differential domain, obviating the need for an input transformer. To do this pin 2 of the XLR is the same as the center pin of the RCA and pin 3 is tied (via a coupling capacitor) to the cathode of the same tube. I've been doing this for over 20 years. Because the cathode is an input to the tube, this technique does not alter the character of the amplifier.

So a fully balanced differential amplifier does not consist of two amplifiers, there is simply one amplifier with both inputs used. Google is your friend... For any given stage of gain that is ideally executed differentially as opposed to single-ended, you get up to 6dB less noise. This is one reason why solid state circuits can be so much quieter; its why we can have only two stages of gain in our phono section, have passive EQ and still work with LOMC cartridges. So I think you have this backwards, based on your initial assumption that a balanced phono section (or any other balanced circuit) will be dual amplifiers (in which case you would be correct; but of course no-one does balanced that way).

You tell me a history, and I told you the arithmetic. Show me my mistakes. Can a 5mV cartridge make 2x5mV specifically for a balanced preamp? How to do it?

 

[Atmasphere]

You tell me a history, and I told you the arithmetic. Show me my mistakes. Can a 5mV cartridge make 2x5mV specifically for a balanced preamp? How to do it?

I showed the history and you showed the math for something unrelated. I showed your mistake in my prior post.

Phono cartridges are balanced sources. Do you understand why??

 

[SVS]

I stated that I did not want to lose bass. I did not state that a cable could add it!

An RCA cable can only sound like a balanced cable when its working perfectly with no coloration whatsoever. When you hear differences in cables, the one takeaway you always have is both cables are incorrect. This is simply because next year, the manufactuer of the 'better' cable will have a new and improved version and if he doesn't, someone else will. I knew the person that founded the high end audio cable industry (Robert Fulton)- he lived in the Twin Cities where I live. So I've seen this phenomena going on since the 1970s and its not slowed down. When you go balanced, you get off the white elephant merry go round and don't have to deal with cable differences any longer. This is assuming of course that your equipment supports the balanced line standards.

Balanced cables also color the sound. I observe a low effect of the cable on the sound only with low output and input resistances of the power amplifiers. I, for example, never make the input resistance of the amplifier more than 10 k?.

 

[Atmasphere]

Balanced cables also color the sound. I observe a low effect of the cable on the sound only with low output and input resistances of the power amplifiers. I, for example, never make the input resistance of the amplifier more than 10 k?.

If the balanced line is driven by a source that supports balanced line standards then there will be no coloration, as long as the cable itself maintains the standards (i.e. twisted pair within a shield).

We have a legacy of balanced lines in recordings going back to the early 1950s. Prior to the War, single-ended connections where used. They were abandoned by the recording industry for a very good reason! You can't get them to be uncolored without extraordinary measures.

 

[SVS]

I showed the history and you showed the math for something unrelated. I showed your mistake in my prior post.

Phono cartridges are balanced sources. Do you understand why??

The phono cartridge gives a SYMMETRICAL signal. It will be balanced when connected to a balanced input (the total voltage between the inputs is 5 mV), or unbalanced when connected to the PCA input (on which all 5 mV).
The balanced line has 3 (THREE) conductors.

 

[SVS]

If the balanced line is driven by a source that supports balanced line standards then there will be no coloration, as long as the cable itself maintains the standards (i.e. twisted pair within a shield).

We have a legacy of balanced lines in recordings going back to the early 1950s. Prior to the War, single-ended connections where used. They were abandoned by the recording industry for a very good reason! You can't get them to be uncolored without extraordinary measures.

You are teaching history again. I give you technical arguments.

 

[Atmasphere]

The phono cartridge gives a SYMMETRICAL signal. It will be balanced when connected to a balanced input (the total voltage between the inputs is 5 mV), or unbalanced when connected to the PCA input (on which all 5 mV).
The balanced line has 3 (THREE) conductors.

This is sounding like you don't understand how a balanced line connection works.

A cartridge is a balanced source because it floats with respect to ground. A center tap isn't needed and in fact would degrade the Common Mode Rejection Ratio performance. Pin 2 and 3 are the inputs to the phono section. Ground is ignored- its used only for shielding. This is not the case with an RCA connection. The cartridge can be connected either way- it really doesn't care.

When properly connected, a cartridge of 5mV will present pin 2 or pin 3 with a 5mV input, since the voltage is made with respect to those pins rather than ground. This is how all balanced connections work.

 

[morricab]

I stated that I did not want to lose bass. I did not state that a cable could add it!

An RCA cable can only sound like a balanced cable when its working perfectly with no coloration whatsoever. When you hear differences in cables, the one takeaway you always have is both cables are incorrect. This is simply because next year, the manufactuer of the 'better' cable will have a new and improved version and if he doesn't, someone else will. I knew the person that founded the high end audio cable industry (Robert Fulton)- he lived in the Twin Cities where I live. So I've seen this phenomena going on since the 1970s and its not slowed down. When you go balanced, you get off the white elephant merry go round and don't have to deal with cable differences any longer. This is assuming of course that your equipment supports the balanced line standards.


Your second sentence above is incorrect. Most balanced line preamps (like ours, which has always had a balanced differential phono section) are fully differential. Differential amplifiers are found in ALL amplifiers, including SETs.

Don't believe me? The first commercial opamps were made by George Philbrick Research and were simply a single tube; one input (inverting) was the grid of the tube and the other input was the cathode. Tubes amplify when there is a difference between the state of the grid and the cathode (usually the cathode is referenced to ground, but that is still a difference compared to the input signal). I have used this techique to drive SETs in the differential domain, obviating the need for an input transformer. To do this pin 2 of the XLR is the same as the center pin of the RCA and pin 3 is tied (via a coupling capacitor) to the cathode of the same tube. I've been doing this for over 20 years. Because the cathode is an input to the tube, this technique does not alter the character of the amplifier.

So a fully balanced differential amplifier does not consist of two amplifiers, there is simply one amplifier with both inputs used. Google is your friend... For any given stage of gain that is ideally executed differentially as opposed to single-ended, you get up to 6dB less noise. This is one reason why solid state circuits can be so much quieter; its why we can have only two stages of gain in our phono section, have passive EQ and still work with LOMC cartridges. So I think you have this backwards, based on your initial assumption that a balanced phono section (or any other balanced circuit) will be dual amplifiers (in which case you would be correct; but of course no-one does balanced that way).

Show me an actual, on the market SET with a differential stage. I have seen some with an SRPP stage, which is push pull but not an actual differential stage.

Also, the way fully differential preamps and amps are designed is with double the active elements as a single ended counterpart.

 

[Atmasphere]

Show me an actual, on the market SET with a differential stage. I have seen some with an SRPP stage, which is push pull but not an actual differential stage.

Also, the way fully differential preamps and amps are designed is with double the active elements as a single ended counterpart.

Again, any tube has a differential aspect of its operation. Did you not read my prior post regarding this?
https://www.whatsbestforum.com/threads/set-amp-owners-thread.27070/post-871190

 

[morricab]

Again, any tube has a differential aspect of its operation. Did you not read my prior post regarding this?
https://www.whatsbestforum.com/threads/set-amp-owners-thread.27070/post-871190

So what? That doesn't make all tube designs differential... especially not a SET. So, I can assume you don't know any SET designs with differential inputs, right?

 

[Atmasphere]

So what? That doesn't make all tube designs differential... especially not a SET. So, I can assume you don't know any SET designs with differential inputs, right?

I suspect there is a religion of sorts in the SET world where things like differential operation are verboten. Its based on myth and a misunderstanding of how tubes actually work.

They operate differentially. The cathode voltage, if it swings in tandem with the grid, will result in no amplification. There must be a difference between them (other than just the bias/operating point) in order to amplify. This can come from ground, or the cathode can be actively driven.

Cathode drive (grounded grid) preamps have been around for a while, so the idea that there is cathode drive should not be mysterious; it should also not be mysterious that when the cathode is driven, the output is in phase with the input, whereas when driven by the grid, the output is out of phase. This should tell you something, but in case its not obvious at this point, it means that the tube amplifies due to differential operation...

As a result all SETs have differential inputs, even if the designer is unaware of this fact. Normally they are set up so that the difference is between the ground and the signal applied to the grid. As I described earlier, this means you can drive them differentially by driving the cathode as well as the grid. Also as I described earlier, I've been doing this for about 20 years. The reason for doing this is that with a true balanced source, neither output is referenced to ground. So if you only connect one side of it (for example pin 2 of the XLR, leaving no connection to pin 3) you'll get a buzz due to an open circuit. By connecting pin 3 to the cathode, the tube can be driven differentially (as all tubes are) and there's no longer an open circuit.

So I do know of SETs with differential inputs, which is to say: All of them. This is different from saying they all have balanced inputs BTW. I know of only a few with that.

 

[SVS]

This is sounding like you don't understand how a balanced line connection works.

A cartridge is a balanced source because it floats with respect to ground. A center tap isn't needed and in fact would degrade the Common Mode Rejection Ratio performance. Pin 2 and 3 are the inputs to the phono section. Ground is ignored- its used only for shielding. This is not the case with an RCA connection. The cartridge can be connected either way- it really doesn't care.

When properly connected, a cartridge of 5mV will present pin 2 or pin 3 with a 5mV input, since the voltage is made with respect to those pins rather than ground. This is how all balanced connections work.

I understand, but you don't seem to. You persistently write about a symmetrical line and call it balanced. You did not write about the signal relative to the ground. It is connected to the input contacts of the amplifier. The inputs are connected to the ground by input resistances that are the divider of the input signal, and the point of their connection is the ground, relative to which the signal from the cartridge floats.

 

[Atmasphere]

I understand, but you don't seem to. You persistently write about a symmetrical line and call it balanced. You did not write about the signal relative to the ground. It is connected to the input contacts of the amplifier. The inputs are connected to the ground by input resistances that are the divider of the input signal, and the point of their connection is the ground, relative to which the signal from the cartridge floats.

You make the point yourself- the cartridge floats. It does not reference ground.

The input resistors are there merely to allow the tube to operate and so can be of any impedance, preferably matched so as to minimize their effect on the CMRR. They do not divide the input signal at all. They do represent a load to the cartridge as a whole however. Since the industry standard is 47K, in our circuit there are two resistors, matched, which are half that value.

If you look at the tonearm, a properly set up arm will have 5 wires. Four are the cartridge, and the 5th is ground, to which the cartridge makes no connection. As I said before, in a balanced line situation, the signal does not reference ground. If a resistive input like ours is used, the input resistors are usually much higher impedance than the source; clearly they are not 'dividing' the signal with respect to the input.

I can give you the example of an input transformer. It has only two input wires for its primary winding. The cartridge would connect to these two wires. The ground is from the tonearm tube to the chassis on which the transformer is mounted. That works the same way: ground is ignored.

 

[SVS]

I suspect there is a religion of sorts in the SET world where things like differential operation are verboten. Its based on myth and a misunderstanding of how tubes actually work.

They operate differentially. The cathode voltage, if it swings in tandem with the grid, will result in no amplification. There must be a difference between them (other than just the bias/operating point) in order to amplify. This can come from ground, or the cathode can be actively driven.

Cathode drive (grounded grid) preamps have been around for a while, so the idea that there is cathode drive should not be mysterious; it should also not be mysterious that when the cathode is driven, the output is in phase with the input, whereas when driven by the grid, the output is out of phase. This should tell you something, but in case its not obvious at this point, it means that the tube amplifies due to differential operation...

I can see that this will be news to you, but the mode of operation of the tube with a common grid has been used since the invention of the tube. This is the highest frequency mode and is used in radio signal receivers and transmitters.
This is not a differential mode.

 

[SVS]

The input resistors are there merely to allow the tube to operate and so can be of any impedance, preferably matched so as to minimize their effect on the CMRR. They do not divide the input signal at all. The do represent a load to the cartridge as a whole however. Since the industry standard is 47K, in our circuit there are two resistors, matched, which are half that value.

This is a new word in electrical engineering and radio engineering. After it, the debate can be stopped.