Hi,
Thought I'd open this new thread to serve as a container for all to use.
The purpose is to provide some ideas on the development of a phono preamp capable of taking in MC carts in the lowish region of output types and carry the balanced topology through as far as you care to take it.
As our domestic environment gets more and more polluted by RFI and EMC I feel this approach has tremendous potential.
The thread is by no means restricted to the die-hard vinyl users and is open to CD based systems as well...they may benefit more than they realise anyways...
This new thread is a spin-off of this one:
Phono Stage only....
As far as I'm concerned all types of valves get the green light as long as they don't belong to the unobtainium superexpensive department.
Transformer coupling is accepted provided they show impeccable pedigree and bandwidth beyond reproach.
Have fun,
Thought I'd open this new thread to serve as a container for all to use.
The purpose is to provide some ideas on the development of a phono preamp capable of taking in MC carts in the lowish region of output types and carry the balanced topology through as far as you care to take it.
As our domestic environment gets more and more polluted by RFI and EMC I feel this approach has tremendous potential.
The thread is by no means restricted to the die-hard vinyl users and is open to CD based systems as well...they may benefit more than they realise anyways...
This new thread is a spin-off of this one:
Phono Stage only....
As far as I'm concerned all types of valves get the green light as long as they don't belong to the unobtainium superexpensive department.
Transformer coupling is accepted provided they show impeccable pedigree and bandwidth beyond reproach.
Have fun,
Transformers
Hello
Since I work on/with commercial audio systems everyday I feel that its onlt fair that I throw my 2 cents in. Transformer technology has come a long way since first coming into heavy play in the early 50's. It used to be that if a transformer was used one could count on reduced frequency response. Todays modern transformers are ruler or near ruler flat in response. I use transformers between pieces of equipment whether they are 2 feet apart or 1000' apart.
My opinion is that its the smart way to go especially when dealing with a delicate circuit or a circuit needing the extra isolation. Nearly any impedance and ratio are available now days at a cost that won't have you eating cereal for the next month.
Joe
Hello
Since I work on/with commercial audio systems everyday I feel that its onlt fair that I throw my 2 cents in. Transformer technology has come a long way since first coming into heavy play in the early 50's. It used to be that if a transformer was used one could count on reduced frequency response. Todays modern transformers are ruler or near ruler flat in response. I use transformers between pieces of equipment whether they are 2 feet apart or 1000' apart.
My opinion is that its the smart way to go especially when dealing with a delicate circuit or a circuit needing the extra isolation. Nearly any impedance and ratio are available now days at a cost that won't have you eating cereal for the next month.
Joe
Re: Transformers
Same here.
In the original thread that this was split off from, some recommended against using transformers due to Barkhausen noise. Noise due to the "stickiness" of the magnetic domains in the core material causing them to move (and hence polarize) not in a smooth continuous fashion, but in little jumps.
I've been using Jensen's transformers for about 20 years now and haven't noticed any deliterious effects that I could lay at the feet of Barkhausen noise. Even when I've used them at the inputs of headphone amplifiers.
Barkhausen noise is a very low frequency phenomenon (often you'll see it demonstrated at frequencies below 1Hz) and it drops like a rock as frequency increases. The domains simply can't move fast enough to keep up with the signal (this is also why hysteresis distortion drops like a rock with increasing frequency).
If we're going to dismiss transformers due to Barkhausen noise, I guess we should also dismiss tubes due to shot noise.
se
burnedfingers said:
Same here.
In the original thread that this was split off from, some recommended against using transformers due to Barkhausen noise. Noise due to the "stickiness" of the magnetic domains in the core material causing them to move (and hence polarize) not in a smooth continuous fashion, but in little jumps.
I've been using Jensen's transformers for about 20 years now and haven't noticed any deliterious effects that I could lay at the feet of Barkhausen noise. Even when I've used them at the inputs of headphone amplifiers.
Barkhausen noise is a very low frequency phenomenon (often you'll see it demonstrated at frequencies below 1Hz) and it drops like a rock as frequency increases. The domains simply can't move fast enough to keep up with the signal (this is also why hysteresis distortion drops like a rock with increasing frequency).
If we're going to dismiss transformers due to Barkhausen noise, I guess we should also dismiss tubes due to shot noise.
se
Peter Baxandall (he of the tone control) commented on Barkhausen noise in the "Microphone Engineering Handbook". He was concerned with microphone levels (which are pretty similar to cartridge levels), and concluded that at the low flux densities in microphone transformers, Barkhausen noise was not a problem.
EC8010 said:
Thanks. All I know is that the only bark in my hausen seems only to come from critters covered with fur, not MuMetal.
Certainly not everyone cares for transformers. And I don't expect that everyone should and don't really care if they don't.
I just like to encourage people to try things for themselves and come to their own conclusions. When we express a subjective dislike for something and then try and reinforce that subjective dislike by pointing to various objective causes, I think it tends to discourage people from trying things for themselves.
I only say this because when I was first getting involved in this hobby, there were quite a few things that I'd dismissed completely out of hand without even trying simply because someone else didn't care for it only to come back to it later and discover that I quite liked them.
In fact, transformers were one of those things.
se
Hi,
The trouble with trying out several MC input xformers is it's rather expensive to experiment with.
Past experiences with MC stepup xformers have been all but positive.
OTOH I'm certain things have improved considerably on all fronts as far as xformers go.
Personally, and you may view it as a prejudice, I'd rather have control over what's going on by finding optimum operating points for the tubes and so.
IOW, about a year of fiddling to find the magic and than move on to the next challenge.
Cheers,
The trouble with trying out several MC input xformers is it's rather expensive to experiment with.
Past experiences with MC stepup xformers have been all but positive.
OTOH I'm certain things have improved considerably on all fronts as far as xformers go.
Personally, and you may view it as a prejudice, I'd rather have control over what's going on by finding optimum operating points for the tubes and so.
IOW, about a year of fiddling to find the magic and than move on to the next challenge.
Cheers,
fdegrove said:
Yes, they can indeed be rather expensive. But some tubes can be even more expensive.
Sorry to hear that. Haven't had any experience with MC stepups so nothing to offer on that front.
Yes. Many of the basic techniques remain the same. I think much of the improvement has come about from materials science.
Prejudice? Naaah. As you said, they can be quite expensive. So I don't see this as being prejudice as pragmatism.
Yup.
se
OK,
I learn something from this; thanks to all who spoke of the Barkhausen phenomenon.
To avoid the high cost of a MC trafo, I'm still driven to a hybrid front end.
The quietest (and I've not tried this yet so it's anecdotal) would appear to be the jfet/tube cascode, using 2SK170 (if you can get 'em). The great advantages here are:
1. A gate connection to the MC, so no chance of current flow into the cartridge, which might otherwise damage it. You can load the cartridge without interactions resistively and capacitively at this point, too.
2. Place the tube in grounded grid, the best bandwidth and the quietest configuration for the tube.
3. Operating at 8mA and using a frame grid tube, my choice would be 6GK5. This has mu of 78, rp of 5400R. It could operate with a bias around 1-V, quite acceptable for the jfet, and a plate load around 12K and a B+ of 220V. Tube is a double triode(though not an octal, ulp!!); so both sections could be operated alongside for fully balanced input. Not sure of gain, but the gain block should be good for at least 45dB.
4. A complete, passive RC RIAA could be used off the output of the 6GK5, matched on both sides for balanced. This absorbs around 20dB of available signal, giving overall gain of just 25dB (18 times, to around 10mV at 1KHz from a 0.6mV cartridge).
5. Cap couple (polystyrene or silver mica) the output from the RIAA to two cascaded, direct coupled 6SN7s, the first with unbypassed cathode resistor and the second supported from a 1:1 interstage transformer at the same B+. First operating at 8mA and 300V B+ with 18K plate load, and the second also at 8mA, same supply, with cathode resistor of 18K, with bypass a 200V 22uF Black Gate.
6. Zout of the output transformer (in single end) should be twice the rp of a single 6SN7, around 6K5, too high. A 2:1 stepdown could be used with a higher mu tube (such as the 6GK5) to give a fourfold improvement in drive. However, the optimum arrangement would be a single ended, self-biased cathode follower, in fact 6SN7 would be fine here but there are others.
There are holes all over this but it's a first pass, a quick grab bag. Let's get stuck into it!
Cheers,
Hugh
I learn something from this; thanks to all who spoke of the Barkhausen phenomenon.
To avoid the high cost of a MC trafo, I'm still driven to a hybrid front end.
The quietest (and I've not tried this yet so it's anecdotal) would appear to be the jfet/tube cascode, using 2SK170 (if you can get 'em). The great advantages here are:
1. A gate connection to the MC, so no chance of current flow into the cartridge, which might otherwise damage it. You can load the cartridge without interactions resistively and capacitively at this point, too.
2. Place the tube in grounded grid, the best bandwidth and the quietest configuration for the tube.
3. Operating at 8mA and using a frame grid tube, my choice would be 6GK5. This has mu of 78, rp of 5400R. It could operate with a bias around 1-V, quite acceptable for the jfet, and a plate load around 12K and a B+ of 220V. Tube is a double triode(though not an octal, ulp!!); so both sections could be operated alongside for fully balanced input. Not sure of gain, but the gain block should be good for at least 45dB.
4. A complete, passive RC RIAA could be used off the output of the 6GK5, matched on both sides for balanced. This absorbs around 20dB of available signal, giving overall gain of just 25dB (18 times, to around 10mV at 1KHz from a 0.6mV cartridge).
5. Cap couple (polystyrene or silver mica) the output from the RIAA to two cascaded, direct coupled 6SN7s, the first with unbypassed cathode resistor and the second supported from a 1:1 interstage transformer at the same B+. First operating at 8mA and 300V B+ with 18K plate load, and the second also at 8mA, same supply, with cathode resistor of 18K, with bypass a 200V 22uF Black Gate.
6. Zout of the output transformer (in single end) should be twice the rp of a single 6SN7, around 6K5, too high. A 2:1 stepdown could be used with a higher mu tube (such as the 6GK5) to give a fourfold improvement in drive. However, the optimum arrangement would be a single ended, self-biased cathode follower, in fact 6SN7 would be fine here but there are others.
There are holes all over this but it's a first pass, a quick grab bag. Let's get stuck into it!
Cheers,
Hugh
THIS COULD BE FUN....
Hi,
I knew you'd not be able to resist it Mr. Hybrid...
I could do the same with a //ed 6DJ8 per signal polarity.
All we need to do here is I/V conversion.
The 6DJ8 family is linear at Vg=0, noise is neglectable when both sections are //ed and transconductance is high, so far so good.
I'm not sure a single //ed 6DJ8 would suffice in a balanced mode operation...I use 2 of those //ed in SE mode (no not you Stevie) and this setup takes in any low output MC with no apparent noise penalty so far.
Pretty soon these MC manufacturers are going to come up with carts that have no output at all anyway...
I can only hope I can get away with a single //ed twin triode in balanced mode here, I already see gain balancing pots....bummer.
He, he...I remember Steve Bench ( I think) using this tube so I looked it up this afternoon in my GE books...nice candidate indeed.
Octal it definetely isn't but Pico 7 sockets could be cool, I'll need them for my regged PSU anyway....and they're not too hard to come by.
Boy...those octals are going to stick out visually. Never mind.
Don't worry about it...if you'd listen to my train of thoughts I'd have you go bonkers within 30 secs flat...
Thanks for the input,
Hi,
I knew you'd not be able to resist it Mr. Hybrid...
I could do the same with a //ed 6DJ8 per signal polarity.
All we need to do here is I/V conversion.
The 6DJ8 family is linear at Vg=0, noise is neglectable when both sections are //ed and transconductance is high, so far so good.
I'm not sure a single //ed 6DJ8 would suffice in a balanced mode operation...I use 2 of those //ed in SE mode (no not you Stevie) and this setup takes in any low output MC with no apparent noise penalty so far.
Pretty soon these MC manufacturers are going to come up with carts that have no output at all anyway...
I can only hope I can get away with a single //ed twin triode in balanced mode here, I already see gain balancing pots....bummer.
He, he...I remember Steve Bench ( I think) using this tube so I looked it up this afternoon in my GE books...nice candidate indeed.
Octal it definetely isn't but Pico 7 sockets could be cool, I'll need them for my regged PSU anyway....and they're not too hard to come by.
Boy...those octals are going to stick out visually. Never mind.
Don't worry about it...if you'd listen to my train of thoughts I'd have you go bonkers within 30 secs flat...
Thanks for the input,
OK,
Nothing happening........
The BIG question. How best do we convert fully balanced to single ended output, without throwing away 6dB of voltage gain?
And if we continue with this topology in balanced mode, how much more voltage gain is required, and what tubes should we use?
I suspect we should have a reasonably low Zout at line level so we can separate preamp from amp and run long cables....
Cheers,
Hugh
Nothing happening........
The BIG question. How best do we convert fully balanced to single ended output, without throwing away 6dB of voltage gain?
And if we continue with this topology in balanced mode, how much more voltage gain is required, and what tubes should we use?
I suspect we should have a reasonably low Zout at line level so we can separate preamp from amp and run long cables....
Cheers,
Hugh
Hi,
According to what I read at Tubecad, there are a few possibilities.
Opamp, Anti-opamp, CF etc.
Either way, it seems the 6dB advantage would be lost by going balun, although JB doesn't seem to mention that explicitly.
I'm still in the process of digesting this though...so far it has always been the other way around; going from unbalanced to balanced.
To me at least, it seems the summing of the polarites is best done at the amps' input...or better still, go balanced up to the speaker output.
I admit, I've never dabbled with anything like it but it looks interesting enough to me.
Cheers,
According to what I read at Tubecad, there are a few possibilities.
Opamp, Anti-opamp, CF etc.
Either way, it seems the 6dB advantage would be lost by going balun, although JB doesn't seem to mention that explicitly.
I'm still in the process of digesting this though...so far it has always been the other way around; going from unbalanced to balanced.
To me at least, it seems the summing of the polarites is best done at the amps' input...or better still, go balanced up to the speaker output.
I admit, I've never dabbled with anything like it but it looks interesting enough to me.
Cheers,
fdegrove said:
Transformers safe? Yeah, right. That's what King Arthur thought about what appeared to be just a cute little bunny rabbit at the entrance to the cave of Kyre Banorg.
I suggest you heed the words of a wise man:
Follow! But! follow only if ye be men of valor, for the entrance to this cave is guarded by a creature so foul, so cruel that no man yet has fought with it and lived! Bones of four fifty men lie strewn about its lair. So, brave knights, if you do doubt your courage or your strength, come no further, for death awaits you all with nasty big pointy teeth. --Tim the Enchanter
se
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.