I often see the recommendation to switch stock unbalanced turntable wiring to balanced. I understand how this, coupled with a preamp with good CMRR, would minimize hum.
However, in most the few turntables I have taken apart, even though the wiring is unbalanced, there is no path to ground, i.e. both wires float.
So, if you build your balanced phono input stage to terminate this unbalanced connection properly, won't you reduce hum (compared to a unbalanced input stage) since at least some of the noise in the TT wiring will be common mode? I understand that you won't get the full benefit of balanced wiring.
However, in most the few turntables I have taken apart, even though the wiring is unbalanced, there is no path to ground, i.e. both wires float.
So, if you build your balanced phono input stage to terminate this unbalanced connection properly, won't you reduce hum (compared to a unbalanced input stage) since at least some of the noise in the TT wiring will be common mode? I understand that you won't get the full benefit of balanced wiring.
There is absolutely an advantage to making that preamp input balanced if you can rewire the turntable cables.
In my setup, the cables are a pair of twisted pairs inside a single shield. The shield can be connected at either or both the turntable ground end and the phono preamp chassis. The preamp input is a DIN plug with all signal leads floating and connected to the primary of an input transformer. Despite the VERY low output of the cartridge, the setup seems to be quite insensitive to hum pickup.
In my setup, the cables are a pair of twisted pairs inside a single shield. The shield can be connected at either or both the turntable ground end and the phono preamp chassis. The preamp input is a DIN plug with all signal leads floating and connected to the primary of an input transformer. Despite the VERY low output of the cartridge, the setup seems to be quite insensitive to hum pickup.
Re: Re: Balanced input stage beneficial for unbalanced turntables?
This comment is too subtle for my understanding. Do you mean to say that if you terminate a pair of floating wires in a balanced fashion (i.e. both with a termination resistor to ground) that this is now balanced wiring?
Steve Eddy said:
This comment is too subtle for my understanding. Do you mean to say that if you terminate a pair of floating wires in a balanced fashion (i.e. both with a termination resistor to ground) that this is now balanced wiring?
Re: Re: Re: Balanced input stage beneficial for unbalanced turntables?
Basically what I'm saying is that the wiring itself doesn't know balanced or unbalanced from a hole in the ground.
For example, a simple twisted pair of wires is a simple twisted pair of wires whether they're used to connect unbalanced interfaces or balanced interfaces.
se
leadbelly said:
Basically what I'm saying is that the wiring itself doesn't know balanced or unbalanced from a hole in the ground.
For example, a simple twisted pair of wires is a simple twisted pair of wires whether they're used to connect unbalanced interfaces or balanced interfaces.
se
Re: Re: Re: Re: Balanced input stage beneficial for unbalanced turntables?
Gotta disagree with you on this one.
Having built and sold hundreds (if not thousands) of balanced phono stages, we have found that it is critical that the wires do not introduce any imbalance into the system.
The balanced circuit can only reject hum if it is common to both signals.
Most turntables only have a pair of coaxial cables, one for each channel. If you run this into a balanced input, you are going to get hum out the wazoo because the shielded conductor (no longer connected to ground, but instead to either the inverting or the non-inverting input) is going to pick up far more hum than the center conductor of the same cable.
The hum is no longer common mode and cannot be rejected.
If you want to use a balanced setup for your phono stage, you need to have balanced (ie, symmetrical) wires in place.
Yep, a simple twisted pair would be a good example of balanced (symmetrical) cables that would work well with a balanced phono stage.
I would recommend shielding the twisted pair (or twisted quad if both channels are in the same cable) just in case the CMRR of the phono stage is less-than-perfect, but I've seen it work without shielding also.
Steve Eddy said:
Gotta disagree with you on this one.
Having built and sold hundreds (if not thousands) of balanced phono stages, we have found that it is critical that the wires do not introduce any imbalance into the system.
The balanced circuit can only reject hum if it is common to both signals.
Most turntables only have a pair of coaxial cables, one for each channel. If you run this into a balanced input, you are going to get hum out the wazoo because the shielded conductor (no longer connected to ground, but instead to either the inverting or the non-inverting input) is going to pick up far more hum than the center conductor of the same cable.
The hum is no longer common mode and cannot be rejected.
If you want to use a balanced setup for your phono stage, you need to have balanced (ie, symmetrical) wires in place.
Steve Eddy said:
Yep, a simple twisted pair would be a good example of balanced (symmetrical) cables that would work well with a balanced phono stage.
I would recommend shielding the twisted pair (or twisted quad if both channels are in the same cable) just in case the CMRR of the phono stage is less-than-perfect, but I've seen it work without shielding also.
Re: Re: Re: Re: Re: Balanced input stage beneficial for unbalanced turntables?
Okie doke.
Agreed.
Ditto.
Yeah? Mmmm. You talking about cheap turntables that use the same cheap spaghetti wires for the cable that are given away with most other mass market stuff?
Most decent turntables that I'm aware of use twisted pairs or quads from the cartridge to either a DIN plug or RCAs for the outputs.
Far as I'm aware, a coaxial geometry is, as with a twisted pair, self-shielding by way of cancellation.
I think the problem isn't so much that the outer conductor picks up more hum, but rather that in a typical coax, the resistance of the center conductor isn't the same as the resistance as the outer conductor.
Bill Whitlock has demonstrated that in twisted pair cables, just the difference in wire resistance due to manufacturing tolerances is sufficient enough to significantly degrade common-mode rejection in electronically balanced inputs due to source impedance imbalances
Yes. But, in my opinion, because the conductor resistances are the same.
Well, I wouldn't recommend coaxial cables in a balanced system either, but for the reason I gave above.
se
Charles Hansen said:
Okie doke.
Agreed.
Ditto.
Yeah? Mmmm. You talking about cheap turntables that use the same cheap spaghetti wires for the cable that are given away with most other mass market stuff?
Most decent turntables that I'm aware of use twisted pairs or quads from the cartridge to either a DIN plug or RCAs for the outputs.
Far as I'm aware, a coaxial geometry is, as with a twisted pair, self-shielding by way of cancellation.
I think the problem isn't so much that the outer conductor picks up more hum, but rather that in a typical coax, the resistance of the center conductor isn't the same as the resistance as the outer conductor.
Bill Whitlock has demonstrated that in twisted pair cables, just the difference in wire resistance due to manufacturing tolerances is sufficient enough to significantly degrade common-mode rejection in electronically balanced inputs due to source impedance imbalances
Yes. But, in my opinion, because the conductor resistances are the same.
Well, I wouldn't recommend coaxial cables in a balanced system either, but for the reason I gave above.
se
KBK said:With what I'm doing these days with cables....shielding is the worst kind of signal damage I have ever come across.For most other folk, carry on - shielding is likely a good idea.
Hehehe. Not a terribly big fan of shielded cables myself.
I built some unshielded phono cables for a friend of mine and even with a low output moving coil into a 60dB phono stage and 104dB speakers, there was no sign of hum.
se
Re: Re: Re: Re: Re: Re: Balanced input stage beneficial for unbalanced turntables?
The key phrase here is "as far as I'm aware". And you don't have experience here. I do. You are wrong on this.
If you have a tonearm that terminates with an RCA box and then use coaxial cables that connect to a balanced phono stage with adapters, you will have hum out the wazoo, just as I said in my previous post.
Trust me on this and stop arguing. You are wrong.
Steve Eddy said:
The key phrase here is "as far as I'm aware". And you don't have experience here. I do. You are wrong on this.
If you have a tonearm that terminates with an RCA box and then use coaxial cables that connect to a balanced phono stage with adapters, you will have hum out the wazoo, just as I said in my previous post.
Trust me on this and stop arguing. You are wrong.
i agree, this is very interesting. hopefully, charles will drop a pearl or two.
since i have gained a little unexpected free time these days, i was thinking about this too. i recently took my turntable out of mothballs (it's been several years) and am really enjoying reconnecting with my vinyl. it's single ended right now and nothing fancy.
re: balanced phono stage ideas
there has been some interesting discussion in a number of places on jfet input discrete gain stages that can be used as transconductance amps.
how about three of those in series with impedances containing the appropriate RIAA time constants at the outputs of the first and second transconductors?
to start the process (i.e. see if the concept even has a chance) , the transconductors simply be jfet (cascoded, of course) diff amps, with resistive loads (current sources/sinks later?).
stimulated from a comment i remember from scott wurcer, i might try:
1st stage - 2SK146 (or 2SK369 *2) diff amp
2nd stage - 2SJ74 * 2 diff amp
3rd stage - 2SK170 * 2 diff amp
with 3 very low noise gain stages, you don't have to get so much gain from each stage.
gotta choose the bias points to meet the low noise needs as well as drive the frequency shaping networks. also, the last transconductor needs to be "beefy" enough to drive the load (for example, cables to the line stage,or a volume control, etc).
also have to think about the dc balance unless you add coupling caps (yuck?).
this seems feasible and sounds fun to try. if nothing else, it will definitely be educational. LTSpice can help with initial tweaks for the RIAA networks. h@!!, i even have the jfets.
what do you think, kbk? is this crazy or what?
oh yeah, charles: no global feedback!
this has got to work at least as well that pacific thing that crops up here from time to time
mlloyd1
since i have gained a little unexpected free time these days, i was thinking about this too. i recently took my turntable out of mothballs (it's been several years) and am really enjoying reconnecting with my vinyl. it's single ended right now and nothing fancy.
re: balanced phono stage ideas
there has been some interesting discussion in a number of places on jfet input discrete gain stages that can be used as transconductance amps.
how about three of those in series with impedances containing the appropriate RIAA time constants at the outputs of the first and second transconductors?
to start the process (i.e. see if the concept even has a chance) , the transconductors simply be jfet (cascoded, of course) diff amps, with resistive loads (current sources/sinks later?).
stimulated from a comment i remember from scott wurcer, i might try:
1st stage - 2SK146 (or 2SK369 *2) diff amp
2nd stage - 2SJ74 * 2 diff amp
3rd stage - 2SK170 * 2 diff amp
with 3 very low noise gain stages, you don't have to get so much gain from each stage.
gotta choose the bias points to meet the low noise needs as well as drive the frequency shaping networks. also, the last transconductor needs to be "beefy" enough to drive the load (for example, cables to the line stage,or a volume control, etc).
also have to think about the dc balance unless you add coupling caps (yuck?).
this seems feasible and sounds fun to try. if nothing else, it will definitely be educational. LTSpice can help with initial tweaks for the RIAA networks. h@!!, i even have the jfets.
what do you think, kbk? is this crazy or what?
oh yeah, charles: no global feedback!
this has got to work at least as well that pacific thing that crops up here from time to time
mlloyd1
KBK said:
Re: Re: Re: Re: Re: Re: Re: Balanced input stage beneficial for unbalanced turntables
That's all well and good except for the fact that I never said you wouldn't have hum.
What I was tentatively disagreeing with was the reason why.
You had said that it was because the outer conductor was picking up far more hum than the center conductor.
But the loop area of an ideal coaxial cable is zero. So to that end I suspect that the reason for the hum is due to the differences in resistance between the center conductor and the outer conductor causing a source impedance imbalance which degrades the CMR of the balanced input stage.
se
Charles Hansen said:
That's all well and good except for the fact that I never said you wouldn't have hum.
What I was tentatively disagreeing with was the reason why.
You had said that it was because the outer conductor was picking up far more hum than the center conductor.
But the loop area of an ideal coaxial cable is zero. So to that end I suspect that the reason for the hum is due to the differences in resistance between the center conductor and the outer conductor causing a source impedance imbalance which degrades the CMR of the balanced input stage.
se
Thanks for your thoughtful replies, Steve and Charles!
mlloyd1, you seem to be throwing around the term transconductance like you can call anything that and it'll sound better
Anyway, balanced phonostages are nothing new. If you want a balanced phonostage with JFETs, just build a Pacific preamp with mirrored sections.
If I were building a minimalist balanced phonostage today I would build one into this: http://jensentransformers.com/an/ingenaes.pdf
mlloyd1, you seem to be throwing around the term transconductance like you can call anything that and it'll sound better
Anyway, balanced phonostages are nothing new. If you want a balanced phonostage with JFETs, just build a Pacific preamp with mirrored sections.
If I were building a minimalist balanced phonostage today I would build one into this: http://jensentransformers.com/an/ingenaes.pdf
Dynamic impedance change due to mass shaping by/of the line's physical/mechanical considerations will also play a part (coaxial as balanced), besides the geometry and external field exposure differentials (difference in exposure to fields between the inner and outer paths) creating havoc, ie hum.. Some very odd (signal related) harmonic distortion products will come to the fore.
I think the two of you agree on that. I suspect that Steve was merely attempting to say that the signal does not care, it will flow. But the question is..how well will it flow, and is the result useful? Not too dang useful, is the answer.
Back to the important point and topic:
BALANCED PHONO CIRCUITS.
This was found at the 'vinyl engine' site:
The balanced line receiving can be done by means of a transformer (centre tapped or not, as described later). Or it can be done by means of a differntial amplifier circuit. These usually need to be more complex than a single op-amp, because all practical op-amps present a different load impedance on their '+' and '-' inputs, which cannot be 'fixed' purely by feedback resistors. This impedance disparity completely defeats the purpose of a balanced system. Some form of buffer circuit must preceded the differencing op-amp, and the whole configuration of 3 op-amps is then called an 'instrumentation amplifier'.
It is convenient to *imagine* the cartridge winding as centre-tapped, this being a ground reference. Each half winding then generates an equal, but opposite phase signal with respect to this ground reference.
However, if the various currents are measured - or computed theoretically, you find that there is zero current flowing in the ground reference - no current flows into or out of the 'centre tap' of the cartridge. If there is no current flow, you can sever the connection with no effect.
You can infer from this that the halfway point of the cartridge winding - when driving a balanced load - always remains at ground potential.
~~~~~~~~~~~~~
thanks to blakep we got Atmasphere’s response, one of the very few manufacturers of truly balanced phono preamps: “The result is lower noise throughout the phono system. In our case this allowed us to eliminate a stage of gain. That made the preamp more transparent, as it now makes less noise and distortion with wider bandwidth. IOW the signal path is actually simpler, not more complex, quite the opposite of the usual drone of balanced circuits being more complex!” Let us take note that at least Atmasphere believes that there can be additional advantages with balanced circuits.
http://www.vinylengine.com/phpBB2/viewtopic.php?t=10074&postdays=0&postorder=asc&start=0
From what I've seen in tube circuits, the usual is to use the first section as a input buffer, possibly with amplification and the second stage for inital eq.
In the case of balanced, it would be wisest, IMHO to use a dual section tube in all cases, but for matched mirroring of the signal. Critical in my estimation. The two-matched/mirrored-in-a-single-envelope has an intrinsic value that gets it beyond separate transistors in overall transparency, as a not-so-wild-assed guess. Unless the transistors are of a similar design. meaning of the appropriate design for the work-dual in a single case-closely matched (transistor).
The tube wins out in the fact of gas design, meaning a low mass field interaction within the tube envelope allows the two fields to, at 'light speed (relatively speaking) to balance and adjust with respect to one another. (this is balanced/mirrored not separate amplification stages with differing aspects!). Like a pair of perfectly matched and embraced lovers, the field interactions of the symmetrical tube innards will create/eneact this fundamental aspect.
This, with respects to generated distortion products, is likely to subjectively (analysis of how it sounds by the listener) place it above the transistor based ones. All things are not equal, so that line may blur a bit, but I'd bet that in a perfect world of layout and design, the (dual) tube's envelope/field in this point of consideration would and will 'carry the day', so to speak.
Perfectly symmetrical/mirrored boards are in order, here.
I think the two of you agree on that. I suspect that Steve was merely attempting to say that the signal does not care, it will flow. But the question is..how well will it flow, and is the result useful? Not too dang useful, is the answer.
Back to the important point and topic:
BALANCED PHONO CIRCUITS.
This was found at the 'vinyl engine' site:
The balanced line receiving can be done by means of a transformer (centre tapped or not, as described later). Or it can be done by means of a differntial amplifier circuit. These usually need to be more complex than a single op-amp, because all practical op-amps present a different load impedance on their '+' and '-' inputs, which cannot be 'fixed' purely by feedback resistors. This impedance disparity completely defeats the purpose of a balanced system. Some form of buffer circuit must preceded the differencing op-amp, and the whole configuration of 3 op-amps is then called an 'instrumentation amplifier'.
It is convenient to *imagine* the cartridge winding as centre-tapped, this being a ground reference. Each half winding then generates an equal, but opposite phase signal with respect to this ground reference.
However, if the various currents are measured - or computed theoretically, you find that there is zero current flowing in the ground reference - no current flows into or out of the 'centre tap' of the cartridge. If there is no current flow, you can sever the connection with no effect.
You can infer from this that the halfway point of the cartridge winding - when driving a balanced load - always remains at ground potential.
~~~~~~~~~~~~~
thanks to blakep we got Atmasphere’s response, one of the very few manufacturers of truly balanced phono preamps: “The result is lower noise throughout the phono system. In our case this allowed us to eliminate a stage of gain. That made the preamp more transparent, as it now makes less noise and distortion with wider bandwidth. IOW the signal path is actually simpler, not more complex, quite the opposite of the usual drone of balanced circuits being more complex!” Let us take note that at least Atmasphere believes that there can be additional advantages with balanced circuits.
http://www.vinylengine.com/phpBB2/viewtopic.php?t=10074&postdays=0&postorder=asc&start=0
From what I've seen in tube circuits, the usual is to use the first section as a input buffer, possibly with amplification and the second stage for inital eq.
In the case of balanced, it would be wisest, IMHO to use a dual section tube in all cases, but for matched mirroring of the signal. Critical in my estimation. The two-matched/mirrored-in-a-single-envelope has an intrinsic value that gets it beyond separate transistors in overall transparency, as a not-so-wild-assed guess. Unless the transistors are of a similar design. meaning of the appropriate design for the work-dual in a single case-closely matched (transistor).
The tube wins out in the fact of gas design, meaning a low mass field interaction within the tube envelope allows the two fields to, at 'light speed (relatively speaking) to balance and adjust with respect to one another. (this is balanced/mirrored not separate amplification stages with differing aspects!). Like a pair of perfectly matched and embraced lovers, the field interactions of the symmetrical tube innards will create/eneact this fundamental aspect.
This, with respects to generated distortion products, is likely to subjectively (analysis of how it sounds by the listener) place it above the transistor based ones. All things are not equal, so that line may blur a bit, but I'd bet that in a perfect world of layout and design, the (dual) tube's envelope/field in this point of consideration would and will 'carry the day', so to speak.
Perfectly symmetrical/mirrored boards are in order, here.
leadbelly said:
Not familiar with the Pacific preamp, but if what you're saying is to just mirror two single-ended preamps to create a balanced preamp, while you will indeed end up with a balanced preamp, you won't have any common-mode rejection. For that, the input also needs to be differential, and not just balanced.
That could work well provided one isn't allergic to integrated circuits.
se
Re: Re: Re: Re: Re: Re: Re: Re: Balanced input stage beneficial for unbalanced turnta
Interesting hypothesis....
But very easy to check. Please put in some ballpark numbers and demonstrate how much the CMRR degrades. You don't have to get fancy. If you want, I'll suggest some numbers:
Rgen = 10 ohms
Rcenter = 1 ohm
Rshield = 0.1 ohm
Rin = 23.5 k per phase (47 k total)
Let us know how much the CMRR degrades. Or if you think my numbers suck, please feel free to select new ones and give us some justification.
I'll be interested to see how your suspicions play out in the real world.
Steve Eddy said:
Interesting hypothesis....
But very easy to check. Please put in some ballpark numbers and demonstrate how much the CMRR degrades. You don't have to get fancy. If you want, I'll suggest some numbers:
Rgen = 10 ohms
Rcenter = 1 ohm
Rshield = 0.1 ohm
Rin = 23.5 k per phase (47 k total)
Let us know how much the CMRR degrades. Or if you think my numbers suck, please feel free to select new ones and give us some justification.
I'll be interested to see how your suspicions play out in the real world.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.