[MASantos]

Quote:

Originally Posted by Sleestack Hey, when the #1 head-fi poster in all of Portugal speaks, I take it seriously.

I would go with black! Maybe that black anodised as the AE-1 from the picture.



Hey I just had a brilliant idea: Why not do it piano white and ask Mikhail to incorporate an Ipod Dock?

 

[ken36]

I love the black, but the mirrored finish that Mikhail uses, while beautiful is susceptible to scratching and that bothers me. I put white gloves on when I change tubes on the ES-1. I've discussed this with Mikhail. He is thinking about it.

 

[Sleestack]

Quote:

Originally Posted by ken36 I love the black, but the mirrored finish that Mikhail uses, while beautiful is susceptible to scratching and that bothers me. I put white gloves on when I change tubes on the ES-1. I've discussed this with Mikhail. He is thinking about it.

All mirrorred finishes are susceptible to showing scratches very easily. You can use car polish and wax to clean it up. I doubt I'll go with an all mirrorred finish as I like the contrast of the posts and body on my SDS-XLR. I'm waiting for some gun metal sample finish samples from him as I do want to do something different than he has done in the past. Gunmetal does add significant cost, but I think I'm out of the realm of worrying about cost on this one.

 

[Sleestack]

Quote:

Originally Posted by MASantos Hey I just had a brilliant idea: Why not do it piano white and ask Mikhail to incorporate an Ipod Dock?

That's a disturbing image.

 

[Sleestack]

Seriously, though, nobody has any thoughts on what they would features they would add to this build? I am looking for input:

Reference Section includes:

12 Mosfet per channel output running high dissipation class A
High efficiency cooling system for output devices and differential circuits.
All Mosfets matched in phase sections and in all four channels
Teflon milled boards for each output device.
Tanatulm Nitride resistors.
All associated transistors matched in phase and in all four sections
All resistors matched to within .01% using Rhoderstien resistors
Balanced configuration in four separate monoblock units
NX top grade, Black Gate capacitors in audio section
Shiltech wiring on signal paths and connection paths
Teflon milled dividers and insolators for siganl path connections
One absoute direct path input XLR inputs for balanced, Two XLR inputs selectable with a disconnect switch for direct path.
Two sets of preamp outputs balanced. One set for reference output, one for tunable circuit. Second set is switchable between reference section, and tunable circuit.
Gold Neutrik XLR input and output connectors
Two sets of S/E input jacks, can be bridged for S/E high power, or run as 4 channels S/E.
Cardas Rhodium GFRA S/E input jacks for SE operation, one set for reference section and one set for tunable section.
One single ended headphone jack.
Audio Note silver preamp section coupling capacitors.
Singlepower designed stepped attenuators for volume control.

Tuneable Circuit (i.e. the "colorable" section) includes:

18 Mosfet per channel output running high dissipation class A, three seperate banks for adjusting power output and sonics, selectable.
High efficiency cooling system for output devices.
All Mosfets matched in phase sections and in all four channels
Teflon milled boards for each output device, and differential circuits.
Tanatulm Nitride resistors.
All associated transistors matched in phase and in all four sections
All resistors matched to within .01% using Rhoderstien resistors
Balanced configuration in four separate monoblock units
NX top grade, Black Gate capacitors in audio section
Shiltech wiring on signal paths and connection paths
Audio Note silver preamp section coupling capacitors, three value selections for adjusting sonics.
Selectable biasing switch for operating output sections and differential devices at 5 selecatable levels.
Input section freqeuncy response adjustment selector switch with 6 selecatable levels.
Two XLR inputs selectable as listed above with direct disconnet as above in reference section, one set for both sections.
Two sets of preamp outputs balanced. One set for reference output, one for tunable circuit. Second set is switchable between reference section, and tunable circuit.
Gold Neutrik XLR input and output connectors
Two sets of S/E input jacks, can be bridged for S/E high power, or run as 4 channels S/E
Cardas Rhodium GFRA S/E input jacks for SE operation, one set for reference section and one set for tunable section.
One single ended headphone jack.
Digital stepped attenuator volume control, with digital display.

Power Supply:

Four isolated separate power supplies in a separate single power section chassis
Large NX top grade Black Gate power supply capacitor sections in power supply
4 each Plitron transformers for power supply sections
Very tight regulation for each monoblock power supply section
Audio Note shunt capacitors for regualtion circuit
Eight regulated output sections for each circuit.
V-Cap Teflon capacitors for final regulation stage shunt devices.
High efficiency power supply heat transfer coupling material
Custom milled heat sinks for output devices, these are part of the milled side panels of the chassis.
Hard Gold power connector contacts.
Shiltech wiring for all power supply paths
High current, low resistance Shiltech umbilical cables.

 

[tom hankins]

Quote:

Originally Posted by Sleestack Seriously, though, nobody has any thoughts on what they would features they would add to this build? I am looking for input: Reference Section includes: 12 Mosfet per channel output running high dissipation class A High efficiency cooling system for output devices and differential circuits. All Mosfets matched in phase sections and in all four channels Teflon milled boards for each output device. Tanatulm Nitride resistors. All associated transistors matched in phase and in all four sections All resistors matched to within .01% using Rhoderstien resistors Balanced configuration in four separate monoblock units NX top grade, Black Gate capacitors in audio section Shiltech wiring on signal paths and connection paths Teflon milled dividers and insolators for siganl path connections One absoute direct path input XLR inputs for balanced, Two XLR inputs selectable with a disconnect switch for direct path. Two sets of preamp outputs balanced. One set for reference output, one for tunable circuit. Second set is switchable between reference section, and tunable circuit. Gold Neutrik XLR input and output connectors Two sets of S/E input jacks, can be bridged for S/E high power, or run as 4 channels S/E. Cardas Rhodium GFRA S/E input jacks for SE operation, one set for reference section and one set for tunable section. One single ended headphone jack. Audio Note silver preamp section coupling capacitors. Singlepower designed stepped attenuators for volume control. Tuneable Circuit (i.e. the "colorable" section) includes: 18 Mosfet per channel output running high dissipation class A, three seperate banks for adjusting power output and sonics, selectable. High efficiency cooling system for output devices. All Mosfets matched in phase sections and in all four channels Teflon milled boards for each output device, and differential circuits. Tanatulm Nitride resistors. All associated transistors matched in phase and in all four sections All resistors matched to within .01% using Rhoderstien resistors Balanced configuration in four separate monoblock units NX top grade, Black Gate capacitors in audio section Shiltech wiring on signal paths and connection paths Audio Note silver preamp section coupling capacitors, three value selections for adjusting sonics. Selectable biasing switch for operating output sections and differential devices at 5 selecatable levels. Input section freqeuncy response adjustment selector switch with 6 selecatable levels. Two XLR inputs selectable as listed above with direct disconnet as above in reference section, one set for both sections. Two sets of preamp outputs balanced. One set for reference output, one for tunable circuit. Second set is switchable between reference section, and tunable circuit. Gold Neutrik XLR input and output connectors Two sets of S/E input jacks, can be bridged for S/E high power, or run as 4 channels S/E Cardas Rhodium GFRA S/E input jacks for SE operation, one set for reference section and one set for tunable section. One single ended headphone jack. Digital stepped attenuator volume control, with digital display. Power Supply: Four isolated separate power supplies in a separate single power section chassis Large NX top grade Black Gate power supply capacitor sections in power supply 4 each Plitron transformers for power supply sections Very tight regulation for each monoblock power supply section Audio Note shunt capacitors for regualtion circuit Eight regulated output sections for each circuit. V-Cap Teflon capacitors for final regulation stage shunt devices. High efficiency power supply heat transfer coupling material Custom milled heat sinks for output devices, these are part of the milled side panels of the chassis. Hard Gold power connector contacts. Shiltech wiring for all power supply paths High current, low resistance Shiltech umbilical cables.

I think you have everything covered.

I have read this post at least ten times trying to take everything in all at once and I just cant do it. I cant believe I'm saying this but......This (even though SS) will probably be Mikhails most over the top (sonicly, build, flexible,options) amp he has built yet. I've had the chance to own, and or, listen to several of his top amps. I have no doubt this will be at the very least one of the top two or three(if not the best) amps he will have built.

 

[DavidNY]

Quote:

Originally Posted by Sleestack Seriously, though, nobody has any thoughts on what they would features they would add to this build? I am looking for input: Reference Section includes: 12 Mosfet per channel output running high dissipation class A High efficiency cooling system for output devices and differential circuits. All Mosfets matched in phase sections and in all four channels Teflon milled boards for each output device. Tanatulm Nitride resistors. All associated transistors matched in phase and in all four sections All resistors matched to within .01% using Rhoderstien resistors Balanced configuration in four separate monoblock units NX top grade, Black Gate capacitors in audio section Shiltech wiring on signal paths and connection paths Teflon milled dividers and insolators for siganl path connections One absoute direct path input XLR inputs for balanced, Two XLR inputs selectable with a disconnect switch for direct path. Two sets of preamp outputs balanced. One set for reference output, one for tunable circuit. Second set is switchable between reference section, and tunable circuit. Gold Neutrik XLR input and output connectors Two sets of S/E input jacks, can be bridged for S/E high power, or run as 4 channels S/E. Cardas Rhodium GFRA S/E input jacks for SE operation, one set for reference section and one set for tunable section. One single ended headphone jack. Audio Note silver preamp section coupling capacitors. Singlepower designed stepped attenuators for volume control. Tuneable Circuit (i.e. the "colorable" section) includes: 18 Mosfet per channel output running high dissipation class A, three seperate banks for adjusting power output and sonics, selectable. High efficiency cooling system for output devices. All Mosfets matched in phase sections and in all four channels Teflon milled boards for each output device, and differential circuits. Tanatulm Nitride resistors. All associated transistors matched in phase and in all four sections All resistors matched to within .01% using Rhoderstien resistors Balanced configuration in four separate monoblock units NX top grade, Black Gate capacitors in audio section Shiltech wiring on signal paths and connection paths Audio Note silver preamp section coupling capacitors, three value selections for adjusting sonics. Selectable biasing switch for operating output sections and differential devices at 5 selecatable levels. Input section freqeuncy response adjustment selector switch with 6 selecatable levels. Two XLR inputs selectable as listed above with direct disconnet as above in reference section, one set for both sections. Two sets of preamp outputs balanced. One set for reference output, one for tunable circuit. Second set is switchable between reference section, and tunable circuit. Gold Neutrik XLR input and output connectors Two sets of S/E input jacks, can be bridged for S/E high power, or run as 4 channels S/E Cardas Rhodium GFRA S/E input jacks for SE operation, one set for reference section and one set for tunable section. One single ended headphone jack. Digital stepped attenuator volume control, with digital display. Power Supply: Four isolated separate power supplies in a separate single power section chassis Large NX top grade Black Gate power supply capacitor sections in power supply 4 each Plitron transformers for power supply sections Very tight regulation for each monoblock power supply section Audio Note shunt capacitors for regualtion circuit Eight regulated output sections for each circuit. V-Cap Teflon capacitors for final regulation stage shunt devices. High efficiency power supply heat transfer coupling material Custom milled heat sinks for output devices, these are part of the milled side panels of the chassis. Hard Gold power connector contacts. Shiltech wiring for all power supply paths High current, low resistance Shiltech umbilical cables.

I recall that Mikhail has built a unit with 2 tone controls. It's based on a tube design. I would ask him about that. I have found that a well executed equalizer can be most helpful when listening to older or not well recorded material.

 

[Tyll Hertsens]

One thing we've noticed is that virtually all headphones have a bass roll-off starting at about 100Hz and may be 10db down at 20Hz. So, I've often thought that there was a legitimate use for a bass boost control to bring up these lows. The danger is (and I'm assuming here) that the lack of lows is probably due to some pretty hard limits revolving around driver excursion and an inability to compress a large enough volume of air in a the rather leaky ear cup of headphones. If so, overzealously raising the very low signals may cause some clipping at what would otherwize be simply solid listening volumes. But, I think at reasonable listening levels, judicious additions of bass might be welcome and benefitial. So, I suggest considering a bass control that would only effect below 100Hz and give a max of about 10db bass boost at 20Hz.

You might also want to consider a phase splitter so that you could take an unbalanced signal in and convert it to a pair of balanced signals. This isn't optimal for balanced mode, but it's probably better than being limited to unbalanced operation with unbalanced inputs.

One rather crazy option would be to insert a tube stage somewhere in the middle that allows you to get some tube "lushening". If Mikhail is tricky enough (and he can be pretty tricky with tubes, I would think) he may be able to add a dial for changing the bias on this tube stage to force the tubes farther and farther into 2nd harmonic distortion; that would give you a "lushness" control.

Since youve got all the channels, you may want to conside an "invert" switch to play with polarity. In unbalanced mode it would mean switching between normal and inverted channels; and in balanced mode it would mean swapping inverted and normal channels. My feeling has always been that you can hear differences in polarity, but they're not large.

You might want to add a 1/8 inch jack for IEMs. You can allways use an adaptor, but why, if you don't have to.

We get asked for balance controlls from time-to-time. If you've got all that fancy digitally controlled attenuators, it wouldn't be too hard to have a way to change one and not the other to shift balance.

And...........crossfeed, of course.

 

[augustwest]

"I've often thought that there was a legitimate use for a bass boost control to bring up these lows. . . . . So, I suggest considering a bass control that would only effect below 100Hz and give a max of about 10db bass boost at 20Hz."


Tyll -

The "bass contour" adjustment sounds like an idea that's time has come, and sounds like a good addition to any top flight headphone amp. . . . . Sort of reminds me of the "loudness" control they used to put on amps/receivers back in the "golden age" of hi-fi during the 70's.

"One rather crazy option would be to insert a tube stage somewhere in the middle that allows you to get some tube "lushening". If Mikhail is tricky enough (and he can be pretty tricky with tubes, I would think) he may be able to add a dial for changing the bias on this tube stage to force the tubes farther and farther into 2nd harmonic distortion; that would give you a "lushness" control."

This is a cool, but out there idea! Sounds like it might take a lot of engineering to make it a practical application.


- augustwest

 

[Sleestack]

Quote:

Originally Posted by Tyll Hertsens One thing we've noticed is that virtually all headphones have a bass roll-off starting at about 100Hz and may be 10db down at 20Hz. So, I've often thought that there was a legitimate use for a bass boost control to bring up these lows. The danger is (and I'm assuming here) that the lack of lows is probably due to some pretty hard limits revolving around driver excursion and an inability to compress a large enough volume of air in a the rather leaky ear cup of headphones. If so, overzealously raising the very low signals may cause some clipping at what would otherwize be simply solid listening volumes. But, I think at reasonable listening levels, judicious additions of bass might be welcome and benefitial. So, I suggest considering a bass control that would only effect below 100Hz and give a max of about 10db bass boost at 20Hz. You might also want to consider a phase splitter so that you could take an unbalanced signal in and convert it to a pair of balanced signals. This isn't optimal for balanced mode, but it's probably better than being limited to unbalanced operation with unbalanced inputs. One rather crazy option would be to insert a tube stage somewhere in the middle that allows you to get some tube "lushening". If Mikhail is tricky enough (and he can be pretty tricky with tubes, I would think) he may be able to add a dial for changing the bias on this tube stage to force the tubes farther and farther into 2nd harmonic distortion; that would give you a "lushness" control. You might want to add a 1/8 inch jack for IEMs. You can allways use an adaptor, but why, if you don't have to. We get asked for balance controlls from time-to-time. If you've got all that fancy digitally controlled attenuators, it wouldn't be too hard to have a way to change one and not the other to shift balance. And...........crossfeed, of course.

Some great suggestions there Tyll. Thanks.

We will be incorporating some sort of bass "boost"in the "fun" section, as well as a few other variables. The "reference" section will obviously be no-nonsense.

Adding tubes .... you're killing me. I'm trying to get away from tubes. I'm dangerous around glass.

I've never actually used an amp with crossfeed. When do people generally use crossfeed? Is it used for older stereo recordings where everything is hard panned left and right?

 

[Akathriel]

No offense to Tyll but crossfeed is yuck. If you're a trained listener (which I assume you are Sleestack) your brain will crossfeed naturally. Hirsch used this analogy, "It's like giving a man a crutch, sure he can get around but hes never going to learn to walk." These were my impressions from a 2006 Balanced Max. Also the Headroom crossfeed circuit adds a large bass boost as well, which you may like, but if you don't makes you not want to use the circuit. In general if you want to add it to the fun section thats up to you (I wouldn't) but definately dont clutter up the reference signal.

 

[Tyll Hertsens]

Forum troll, huh. I guess so. How about an "In my opinion..." or an "In my experience..." I don't buy Hirsch's cruch explaination. If your eyesight is poor and you wear glasses is that a cruch? Should we forgo glasses in order to build up our brains ability to compensate for a technical problem? Why not poke an eye out? After all, it sees pretty much the same thing as the other. But in so doing your missing the comparison between eyes that give you depth perception.

Without crossfeed there is some information in a stereo signal that is only heard in one ear or the other, and the listener is denied the time domain differences that give your brain the ability to decode a stereo presentation to locate sounds in the normal way. Sure, folks have every right to prefer listening without crossfeed, and I've got no problem with that, but the reasons for crossfeed are not subjective. Crossfeed is not an attempt to creat a prefferable esthetic, but an attempt to technically model a natural listening acoustic. When I say that crossfeed makes for a more natural psychoacoustic listening environment, I don't have to add "in my opinion" because it's not my opinion---it is a scientifically developed understanding and an attempt to follow objective scientific conclusion with a technique to improve the headphone listening experience. Our approach may be different than Jan Meyer's, but were after the same thing, and were after it not because we share the same opinion, but because we share some education on the science of hearing. But we both fall short of a perfectly corrected audio image for a number of reasons: First, science tells us that audio localization is far more complicated that just inter-aural time differences (pinna reflections and cues changing with head movement play very import rolls). Second, everyones ears have different shapes, so no one set of compensations will work for everyone. And last, a two-speaker reproduction system is a significantly simplified representation of live sound that is a markedly unatural starting point for headphone corrections. So, yes, there are a number of ways in which crossfeed falls short of perfectly natural, but it is---not in my opinion, but in the claim of a significant amount of science, which virtually no one here, including myself, is qualified to rebutt---a real improvement in creatring a more natural audio image on headphones. In other words, you have every right to express your dislike of it, but you are completely incompetent to judge it's technical merit.

Sorry about the thread crap there, Sleestack. Back to the topic at hand:

Quote:

Adding tubes .... you're killing me. I'm trying to get away from tubes. I'm dangerous around glass.

Quote:

This is a cool, but out there idea! Sounds like it might take a lot of engineering to make it a practical application.

Actually, I've talked with Pete Millett at great lenght about this and it shouldn't be too hard. We tried it on the Millett Hybrid but those tubes are so insensitive to bias changes that we couldn't reliable pull it off. None the less, there are (from what Pete said) a number of tubes with characteristics that would allow significant control of the amount of 2nd harmonic distortion dialed into the audio. But, if you don't want to deal with glass, then it's out. I suspect there would be ways to do it with solid state components like FETs that might allow you to effect the load line of a stage asymetrically to get the same thing, but I haven't looked into it.

Quote:

I've never actually used an amp with crossfeed. When do people generally use crossfeed? Is it used for older stereo recordings where everything is hard panned left and right?

Crossfeed is most obviously an improvement on old recordings with hard-panned instruments, but I find it benefitial in almost every situation. However, it's completely useless for mono recordings, and classical stuff that is miked mid-side (a microphone technique used commonly in the past for its mono compatability for AM radio broadcast) gets little benefit from crossfeed. So, it is absolutely true that the benefits of the crossfeed circuit vary with the nature of the source materiel.

As Akathriel mentioned, there are tonal artifacts to the mono componant of the audio signal from the crossfeed circuit marked by a boost in the bass and a dip around 2kHz. There are arguments saying that it's actually a good thing as the crossfeed FR curve is somewhat similar to the "diffuse field" correction, but I do need to acknowledge that the artifacts exist. We have worked VERY hard to rid ourselves of the FR changes due to the crossfeed circuit and have made significant gains in this regard in our recent changes to the crossfeed circuit. If you want to persue this, I'm sure I could find a way to loan you an amp for some extended listening (because you do need to listen for a while to give you brain time to adapt and develope a preference. And I would be happy to sell Mikhail a module and give him enough instruction on how to fire it up so that it could be dropped into your amp to provide a crossfeed switch.

 

[Hirsch]

Quote:

Originally Posted by Tyll Hertsens Forum troll, huh. I guess so. How about an "In my opinion..." or an "In my experience..." I don't buy Hirsch's cruch explaination. If your eyesight is poor and you wear glasses is that a cruch? Should we forgo glasses in order to build up our brains ability to compensate for a technical problem? Why not poke an eye out? After all, it sees pretty much the same thing as the other. But in so doing your missing the comparison between eyes that give you depth perception. Without crossfeed there is some information in a stereo signal that is only heard in one ear or the other, and the listener is denied the time domain differences that give your brain the ability to decode a stereo presentation to locate sounds in the normal way. Sure, folks have every right to prefer listening without crossfeed, and I've got no problem with that, but the reasons for crossfeed are not subjective. Crossfeed is not an attempt to creat a prefferable esthetic, but an attempt to technically model a natural listening acoustic. When I say that crossfeed makes for a more natural psychoacoustic listening environment, I don't have to add "in my opinion" because it's not my opinion---it is a scientifically developed understanding and an attempt to follow objective scientific conclusion with a technique to improve the headphone listening experience. Our approach may be different than Jan Meyer's, but were after the same thing, and were after it not because we share the same opinion, but because we share some education on the science of hearing. But we both fall short of a perfectly corrected audio image for a number of reasons: First, science tells us that audio localization is far more complicated that just inter-aural time differences (pinna reflections and cues changing with head movement play very import rolls). Second, everyones ears have different shapes, so no one set of compensations will work for everyone. And last, a two-speaker reproduction system is a significantly simplified representation of live sound that is a markedly unatural starting point for headphone corrections. So, yes, there are a number of ways in which crossfeed falls short of perfectly natural, but it is---not in my opinion, but in the claim of a significant amount of science, which virtually no one here, including myself, is qualified to rebutt---a real improvement in creatring a more natural audio image on headphones. In other words, you have every right to express your dislike of it, but you are completely incompetent to judge it's technical merit. Sorry about the thread crap there, Sleestack. Back to the topic at hand:

Not so fast...

The short response: The stereo image is learned in any event. We don't normally hear sounds as coming from anywhere but their source, using the disparity in the signals to the ears as part of the cues. However, with stereo speakers, we fool ourselves into thinking that there's a continuous image in front of us that has both height and depth, and that we can hear sounds coming from very localized points in space. This is a learned perception. Now, with crossfeed, you're trying to mimic the effects of the two channel speaker rig in a headphone setup. However, the brain is perfectly capable of performing the same kind of learning that it did with the speaker rig, and creating a three-dimensional image out of the cues coming from headphones the same way it learned to process the cues from the speakers.

What crossfeed does is to try and mimic the speaker situation by bleeding information from one channel to the other. But why bother? The time domain information needed to create the stereo image is already present in the stereo signal. The actual three dimensional sound field has been bolixed up enough by trying to present it through two stereo channels. The human brain is perfectly capable of sorting this out, if given the opportunity to do so. If a person is only using crossfeed, though, that person will never learn the perceptual skillset need to integrate the information from headphones into a stereo image, simply because there's no need. The unique characteristics of the headphone listening experience have been turned into a simulation of the speaker experience. Crossfeed may make this jump easier, in going from speakers to headphones, but also prevents the learning needed to uniquely enjoy the headstage produced by a good headphone rig, IMO.



Back to your regularly scheduled thread...

 

[Sleestack]

Quote:

Originally Posted by Hirsch Back to your regularly scheduled thread...

Polished gunmetal with gunmetal titanium posts.

 

[Tyll Hertsens]

I'll answer here in order to help keep Sleestacks thread clean.