KeithL
Administrator 
Posts: 10,273
|
Post by KeithL on Dec 12, 2022 10:56:26 GMT -5
I agree... Yes, any competent buffer should have a high input impedance, and a low output impedance, which might determine where you can put it. However, with a tube buffer, you have a device with a monotonic distortion curve and probably a specific overload point. This means that the amount and sort of effect it has on the sound will depend on the input level and output level. So, for example, it may be "very clean" at low input levels, "add just some coloration" at middle input levels, and "clip or get really muddy" at very high input levels. And, in that case, you're going to be using the input level to "tune how it sounds". In which case you're going to want to put it between two devices with volume controls... The volume control on the device feeding it will "tune the way it sounds"... And the volume control on the output side will be used to match its output to the rest of your system... (So probably between the DAC and the preamp.) The INPUT of the tube buffer should be easy to drive - even for the passive Volume control... And its output should likewise have no trouble driving the passive Volume control... (Again assuming a high input impedance and a low output impedance.) Where should a tube buffer go? I am using a Topping D-10s DAC and not attenuating volume in Roon. The layout choices include: Group 1 - Emotiva BasX preamp and tube buffer or Group 2 - passive volume attenuator with remote control and tube buffer With the preamp, I’m not sure that it matters whether the tube buffer is upstream or downstream of the preamp. I’m assuming that the tube buffer has a high-impedance input (maybe more than 100,000 Ohms) and a lower output impedance. But with the passive volume control, I don’t know whether the tube buffer should be inserted upstream or downstream of the (probably 10k Ohm) passive attenuator. What say you (and why)? Thanks - Glenn I would say the best method is to try it yourself both ways and see what sounds best to you. Because no matter what anyone says, you'll probably end up trying both anyway. And you should. It's just like asking for a 6 ohm (or even 4 or 8 ohm) speaker, which tap should you use from a tube amp, the 4 or 8 ohm one. You should use whatever sounds best.  |
|
|
|
Post by Boomzilla on Dec 12, 2022 10:58:37 GMT -5
I've a friend who takes much the same approach with his music. He says "I find that I like but two or three tracks from any specific album. So, I delete all the files I don't want and move the keepers to a file titled by artist."
His approach works, of course, and (at least usually) the player software will find the track being searched for.
But when the original rip was never properly identified, and the files are named "No. 1 - No. 2 - etc. then the search engine (more often than not) can't "see" the track.
Yes, I could hunt down and tag all those tracks, but IMHO it isn't worth my time. There's at least a 50% chance the the track will be available via streaming, so why confuse my player's search engine. As I mentioned - I've deleted all unlabeled albums & tracks and now my player searches at warp speed!
Glenn
|
|
|
|
Post by Boomzilla on Dec 12, 2022 11:17:05 GMT -5
...it may be "very clean" at low input levels, "add just some coloration" at middle input levels, and "clip or get really muddy" at very high input levels. And, in that case, you're going to be using the input level to "tune how it sounds". In which case you're going to want to put it between two devices with volume controls... The volume control on the device feeding it will "tune the way it sounds"... And the volume control on the output side will be used to match its output to the rest of your system... (So probably between the DAC and the preamp.) I understand your justification, KeithL, but I question your last sentence... My DAC has no "volume control." It has 100% analog output volume ("line level") at all times. That being the case, if I put the tube buffer between the DAC and the preamp (as you suggest), then I'm more likely to overdrive the tube buffer into distortion. To control the level of the tube buffer input, I'd need to locate the buffer between the preamp output and the power amplifier. Since I never turn up the preamp output to full volume, or even half, the tube buffer input will always be at a (much) lower voltage than the DAC output. Yes, the input voltage to the tube buffer will vary slightly with changes in volume output from the preamplifier, but will still never get anywhere as high in volume as the raw output from the DAC. And the power-amplifier input input impedance (being a relative constant) will be no challenge for the tube buffer output. That said... There IS a way that I could control volume at the DAC output. I can configure Roon to use its own digital volume control. But that option substitutes a (theoretically inferior) digital attenuation for the analog volume of the preamplifier. All the above discussion has been for the DAC + preamplifier + tube buffer + power amplifier chain. Things may differ (potentially significantly I think) if we instead discuss the DAC + passive volume control + tube buffer + power amplifier chain. In this second circumstance, the passive volume pot (unlike the preamplifier) has no input OR output buffer amplifiers. Instead, as the passive volume box is turned down, the impedance on its input jacks approaches a dead short to ground. What will this do to the DAC's output amps? As the passive volume box is turned up, the DAC's output buffers will rise to approximately 10k Ohms (the impedance of the passive volume pots in the passive box). Presumably, the DAC won't care? The 100k Ohm impedance of the tube buffer won't have much effect (assuming the buffer is downstream of the passive volume box) because the point of least resistance to ground will be the passive volume box - not the tube buffer input. Are these conjectures correct? |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Dec 12, 2022 11:41:26 GMT -5
Assuming that the buffer won't overload due to the input or output level the normally recommended spot would be between the preamp and power amp. And that would probably work just fine (assuming you like the way it sounds there). If the input impedance of the passive volume control REALLY approaches a short then that could be a problem. However, if we're really talking about 5k or 10k, then that isn't at all low by solid state device standards. Most solid state outputs will cheerfully drive a 5k or 10k load and some, including ours, will generally be fine down to even 1k. However, since some DACs are a bit odd, it might be worth checking with the specs on the DAC. (With most passive volume controls I've seen the input impedance will approach the resistance of the potentiometer in parallel with the load... So, with a 10k potentiometer, that's still only going to be, at worst, about half of that, so 5k, which should be OK for most solid state outputs.) Digital volume controls are an interesting subject... When used as an output control, where the digital control is near 100% at full output, but gets turned down at lower output levels... it's really not a big issue. What you don't want to do would be to turn the digital volume control in the software way down... then boost the level back up with a preamp. So, for example, you wouldn't want to leave the software output level set to 10%, and use the gain control on the preamp to control the playback level. Because, if you do that, you end up with a high signal level being represented by relatively few bits of resolution. If you take a picture and shrink it down by lowering the pixel count it will still look fine when viewed at its new small size... (And this is a fair analogy of what happens with a digital volume control.) But you wouldn't want to make it full size again by blowing up that low resolution picture... Because, at full size, the lack of resolution would be a problem... ...it may be "very clean" at low input levels, "add just some coloration" at middle input levels, and "clip or get really muddy" at very high input levels. And, in that case, you're going to be using the input level to "tune how it sounds". In which case you're going to want to put it between two devices with volume controls... The volume control on the device feeding it will "tune the way it sounds"... And the volume control on the output side will be used to match its output to the rest of your system... (So probably between the DAC and the preamp.) I understand your justification, KeithL , but I question your last sentence... My DAC has no "volume control." It has 100% analog output volume ("line level") at all times. That being the case, if I put the tube buffer between the DAC and the preamp (as you suggest), then I'm more likely to overdrive the tube buffer into distortion. To control the level of the tube buffer input, I'd need to locate the buffer between the preamp output and the power amplifier. Since I never turn up the preamp output to full volume, or even half, the tube buffer input will always be at a (much) lower voltage than the DAC output. Yes, the input voltage to the tube buffer will vary slightly with changes in volume output from the preamplifier, but will still never get anywhere as high in volume as the raw output from the DAC. And the power-amplifier input input impedance (being a relative constant) will be no challenge for the tube buffer output. That said... There IS a way that I could control volume at the DAC output. I can configure Roon to use its own digital volume control. But that option substitutes a (theoretically inferior) digital attenuation for the analog volume of the preamplifier. All the above discussion has been for the DAC + preamplifier + tube buffer + power amplifier chain. Things may differ (potentially significantly I think) if we instead discuss the DAC + passive volume control + tube buffer + power amplifier chain. In this second circumstance, the passive volume pot (unlike the preamplifier) has no input OR output buffer amplifiers. Instead, as the passive volume box is turned down, the impedance on its input jacks approaches a dead short to ground. What will this do to the DAC's output amps? As the passive volume box is turned up, the DAC's output buffers will rise to approximately 10k Ohms (the impedance of the passive volume pots in the passive box). Presumably, the DAC won't care? The 100k Ohm impedance of the tube buffer won't have much effect (assuming the buffer is downstream of the passive volume box) because the point of least resistance to ground will be the passive volume box - not the tube buffer input. Are these conjectures correct? |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Dec 12, 2022 12:00:40 GMT -5
One other thing that might be worth considering is the noise floor on the tube buffer.
Things like that vary widely in several ways - including noise floor - and some tube devices can actually be quite noisy.
You might want to confirm that it doesn't raise the noise floor noticeably if you put it between the preamp and the amp.
(And, if it does, you're going to want a volume control between it and the amp, one way or the other.)
|
|
|
|
Post by garbulky on Dec 12, 2022 12:05:52 GMT -5
The way I see it is if a tube buffer can't handle a line level signal it is not worth being a buffer.
|
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Dec 12, 2022 12:41:26 GMT -5
I'm absolutely inclined to agree... But then, as you already know, I don't see much point in tube buffers at all.  A really well designed buffer should reduce the source impedance of the signal while adding no measurable or audible distortion or coloration. (And, yes, you can design a tube buffer whose operation in a circuit would be totally inaudible... it's just easier to do it with solid state.) Therefore, by definition, if a tube buffer alters the sound, then it is not a very well designed buffer. And, taking that as a given, or even a design goal, I would not assume that it would perform well in other ways either. Specifically, in this case, some designers believe that tube gear should make audible distortion at pretty much any level... I've seen little tube buffers that were deliberately "plate starved" because it made them sound "different". For example, some of the little low-cost tube buffers you see on eBay run a 12AU7 at very low plate voltage, and they sound quite nasty, and overload quite easily. (And, yes, you CAN run a 12AU7 with a 12V plate voltage - you just wouldn't normally want to.) The way I see it is if a tube buffer can't handle a line level signal it is not worth being a buffer. |
|
|
|
Post by Boomzilla on Dec 12, 2022 19:03:03 GMT -5
To minimize background noise from the tube buffer, higher signal voltage? If this answer is “yes,” and I think it is, then the tube-buffer location that will minimize thermionic noise the most will be between the (line-level) DAC outputs and the preamp inputs.
Glenn
|
|
|
|
Post by Boomzilla on Dec 20, 2022 9:20:45 GMT -5
OK - the decision is made. Since the NUC is working as a Roon source, I'll keep it for now. But I'm not really happy with the Apple TV3 I'm using as a streamer. It has only TOSLINK optical as an audio output and although it doesn't sound too bad, it doesn't have the transparency that I get with USB+the Topping DAC. So... Assuming that I'm about to spend more $$$, the immediate question is: Should I buy a combination streamer/dac or just a bare streamer (Ethernet input - I don't care about wireless) with a USB output? In favor of a streamer/dac: Economy - In theory bundling the two should cost less than buying separates. Compatibility - There will be no question of the streamer working seamlessly with the DAC. Simplicity - The fewer the number of interconnects in my system, the better In favor of a separate streamer and dac: Futureproofing - If one or the other becomes obsolete, I won't have to replace both. Features - Being able to select the exact mix of features I want on both the streamer and dac will allow me to customize the equipment to my needs. Sound quality - Being able to concentrate the available funding into a higher quality dac, for example, may provide better sound quality than having to take the dac that was included in the streamer. I think I'm (reluctantly) favoring option two, but if I could find a high-quality integrated streamer/dac, I'd surely consider it. Another decision influencer is the question of whether I want my system to use all unbalanced (RCA) connections or balanced (XLR) ones. The advantage of XLR, of course, is its superior noise rejection. But getting to XLR throughout the system can be costly. Further, if one wants to integrate vacuum tubes anywhere in the system, then XLR becomes exponentially more costly. The logical place to incorporate tubes, if you want them, is in the preamplifier. There are not only a number of balanced vacuum tube preamps on the market but also the tubes in preamplifiers have longer service life and less expensive replacement than do the output tubes of tube power amplifiers. If one wants tone controls in a balanced tube preamplifier, the choices are severely limited (McIntosh comes to mind). "Bare" balanced tube preamps are available from the premium (BAT) to Chinese economy units. Of course, a tube buffer can suffice instead of a tube preamplifier, but all such devices that I know of are unbalanced (RCA) only. Mr. KeithL opines (and I DO respect his opinion) that vacuum tubes are generally less accurate than transistor equipment in that they (more often than not) add second-order harmonic distortion that "sweetens" the sound. But it's been my experience that the more accurate and revealing the loudspeakers, the more potentially beneficial the tube effect on the presentation. As I've mentioned before, two of the best systems I've ever owned had vacuum tubes in the signal path, and even today, an audio amigo whose system I consider a paragon of perfection uses a DAC with a vacuum tube output stage. So, ultimately, I think I do want to stick with tubes somewhere in my system. The least expensive option for accomplishing that is with a tube buffer. Since tube buffers are all unbalanced circuitry, I might as well not concern myself with balanced circuitry in the remainder of the system... So perhaps I should be looking at an inexpensive Ethernet streamer with a USB output, a good USB DAC with RCA outputs, the tube buffer, and power amplifiers. This assumes that the DAC can be used with a good oversampling internal volume control (with a remote control function). Otherwise, I'll have to add either a remote-controlled passive volume box or a preamplifier between the tube buffer and the power amp. Although I have both available, I'm back to my experience that the system with the fewest components will provide the most transparent sound. One could argue that having a passive attenuator (no electronic components) meets tha "fewer components" criteria, its highly variable effect on system impedances can sometimes cause more harm than its simplicity adds to the sound quality. The other way around the vacuum tube dilemma would be to use a DAC with a vacuum tube output stage. Black Ice Audio (nee Jolida) make such DACs, and they sound excellent. But since there's no volume control function on them, you need a preamp function somewhere downstream. Audio amigo whose system I admire uses exactly this. A Black Ice tube DAC with a passive volume device downstream. He gets amazing results from the combo too. Hmmm.... Merry Christmas, y'all. |
|
|
|
Post by garbulky on Dec 20, 2022 9:26:27 GMT -5
I was thinking about what you said the other day. I have limited experience with a tube buffer and tube dacs. But the two I've heard (the Jolida and a yaqin tube buffer) though their impressions were positive did not quite sound like a tube amp. I am thinking that a tube amp interacts with speaker load in a more complex manner to produce its characteristic sound.
|
|
|
|
Post by Boomzilla on Dec 20, 2022 9:44:56 GMT -5
I was thinking about what you said the other day. I have limited experience with a tube buffer and tube dacs. But the two I've heard (the Jolida and a yaqin tube buffer) though their impressions were positive did not quite sound like a tube amp. I am thinking that a tube amp interacts with speaker load in a more complex manner to produce its characteristic sound. I'd agree - but not all tube amplifiers create the effect you're discussing. What tube amplifiers will react favorably with specific speakers is a profoundly difficult thing to predict. I've owned some tube amps (including models from Audio Research, McIntosh, Rogue Audio, etc.) that were not in any way superior sounding to solid state amplifiers. I've owned others (HealthKits, VTL compact 100s) that stunned regardless of the speakers connected to them. And I've found a few combinations (inexpensive Chinese tube amplifier + Klipsch LaScala speakers) that just happened to work together amazingly well. So, what is it that makes some tube amplifiers sound so spectacular while others fail to do so? I don't know. And I've never seen anyone else provide a cogent and rational explanation either. Note that the comments in the above paragraph are not meant to disparage vacuum tube amplifiers from Audio Research, McIntosh, or Rogue Audio - I'm sure that some of their amplifiers sound really spectacular with some speakers. But the amplifiers that I just happened to own didn't particularly impress with the speakers that I'd tried them with. So there's quite a range of potential outcomes with tube power amplifiers and various loudspeakers. You might happen to find a synergistic combination - but you might not. That's why I generally prefer NOT to use vacuum tube power amplifiers. I find that solid state amps are far less picky about what speakers they'll sound good with. That's why I generally prefer to include tube stages in the amplification chain earlier in the process and use solid state power amplifiers to drive the loudspeakers. I'd also argue that it is far, far less expensive to do what I suggest rather than trying to use tube power amplifiers. Output tubes are NOT cheap, and to keep your tube power amplifier operating at peak performance, you'll need to change those output tubes on a regular basis. Balance that with the fact that you can buy really great-sounding solid state amplifiers for prices that are a fraction of what you'd have to pay for a vacuum tube power amplifier of the same wattage. Glenn |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Dec 20, 2022 10:31:43 GMT -5
Yes... Most simple buffer circuits, tube or otherwise, have a fixed gain, and a fixed noise floor... So the higher the signal level you can successfully use the better their S/N will be... And having the buffer before the Volume control on the preamp will help to minimize noise at lower levels. (When you lower the Volume you lower both the signal and the noise floor coming from the buffer.) There are two things to keep in mind... 1. Most tube circuits have a relatively high overload margins - but not all - so make sure you don't overload the buffer at full signal level. 2. Most tube circuits have "a monotonic distortion curve" - which is a fancy way of saying that their distortion rises more or less linearly with level. This means that, unlike most solid state circuitry, a tube buffer may actually sound substantially different when operated at different signal levels. So... assuming that your tube buffer actually adds a significant amount of "tube sound" (coloration).... If you put the buffer between the DAC and the preamp - the amount of coloration will vary depending on the level of the content but NOT depending on the setting of the Volume control on the preamp. (A loud passage in the recording will have more coloration than a quiet one - regardless of how loudly you're playing it.) If you put the buffer between the preamp and the amp - the amount of coloration will vary depending on the level of the content AND on the setting of the Volume control. (The amount of coloration will depend on how loudly you're PLAYING the music rather than whether it is a loud or quiet passage at the source.) (This might seem "more natural" and "will tend to make your power amp have more the characteristics of a tube amp" - which is to have higher distortion at higher output levels - if that is your goal.) Depending on the distortion curve, and overall distortion, of the tube buffer, this may or may not make any audible difference. But, from the point of view of noise floor, putting the buffer between the DAC and the preamp is absolutely the best choice. To minimize background noise from the tube buffer, higher signal voltage? If this answer is “yes,” and I think it is, then the tube-buffer location that will minimize thermionic noise the most will be between the (line-level) DAC outputs and the preamp inputs. Glenn |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Dec 20, 2022 11:50:45 GMT -5
There are a variety of different mechanisms that produce non-linearities and other effects in tube amps and they each do something slightly different. Most preamps or buffers have relatively low distortion and, of course, no interaction at all with the speaker. Triodes tend to have higher distortion and also tend to have a more linear distortion product (their distortion increases linearly with output level). Pentodes tend to be more linear until they approach clipping (more like most solid state circuits). Both of those characteristics depend on the amount of gain and the amount and type of feedback employed. Note that tube preamps and buffers can be designed to have inaudibly low levels of coloration (with both triodes or pentodes). However, the other side of this is that many modern tube products are deliberately designed to sound " tubey". (That's a nice way of saying that they're deliberately designed badly, so as to increase distortion, and so emphasize the potential weaknesses of tubes to make them more audible.) It's also worth noting at this point that we humans tend to mentally equate distortion with "loudness". (This is why low powered devices with lots of distortion sound "annoyingly loud". It's also why good headphones often do not sound loud - even at high SPL levels.) This produces an interesting "effect" with tube devices "especially triodes" that have a level of distortion that increased more or less linearly with output level. As an example let's compare a "typical" 10 watt solid state amp to a 10 watt single-ended triode amplifier. At 0.1 watts both probably have relatively low distortion... At 1 watt, the solid state amplifier is at 0.01% THD, and the triode is at 1%... At 5 watts, the solid state amp is still at 0.02%, and the triode is at 10%... The result of this is that, at 5 watts, the triode amplifier experiences "a psychoacoustic boost in loudness" (because we hear the increased distortion as extra loudness). So, since both amps sound equally loud at 1 watt, but the triode amplifier sounds louder at 5 watts, we have the psychoacoustic equivalent of a dynamic range expansion. (The apparent difference in loudness between 1 watt and 5 watts is greater with the triode amp than with the solid state amp - so it "sounds more dynamic".) Note that this effect is greater with triodes than with pentodes, and MUCH greater with single-ended triodes than with push-pull amps of either type. (In push-pull tube amps a lot of the distortion is cancelled out - resulting in a distortion curve closer to a typical solid state amp - distortion stays relatively low until the amp "runs out".) Tube POWER AMPS have a few other mechanisms that change the way they sound. One such mechanism is output transformers. Transformers, especially audio transformers operating at relatively high power levels, tend to have all sorts of non-linearities (types of distortion). Small low cost output transformers tend to produce a lot of distortion... and even large, heavy, expensive transformers produce a significant amount of distortion. (You have low-level non-linearity, saturation effects, and hysteresis effects... just to name a few of the limitations of transformers.) Another mechanism is the sort of circuitry involved... Single-ended triodes (which are a throwback to really outdated technology) tend to have lots of distortion... Their design provides no cancellation for any of the distortion produced by a power tube... They also place a lot of demand on both the power supply and the output transformer... They tend to have "a monotonic distortion curve"... meaning that their distortion, which is mostly second harmonic, tends to increase linearly with frequency. The next step up is a push-pull triode amplifier... The push-pull arrangement reduces distortion by enabling the distortion in the two output tubes to largely cancel out. This cancellation also serves to reduce power supply noise reaching the output. And, finally, this distortion allows the magnetic fields caused by the plate current of the output tubes in the output transformer to cancel out. So... more power, lower distortion, and LESS demands on both the power supply and the output transformers.) Stepping up to a push-pull pentode output stage offers some significant improvements in linearity. As a resault, a pentode amplifier can generally deliver more power, and less distortion, until it gets close to its power limit. The next step up is push-pull pentodes in an ultralinear configuration... In an "ultralinear configuration" a separate winding in the output transformer is connected back to a grid in the output tubes... This provides a sort of local feedback which improves linearity, reduces distortion, and delivers other such improvements. Incidentally, in case you were wondering, there is a reason why tube amps ALWAYS have limited feedback and a low damping factor. In solid state amplifiers the high damping factor is largely a result of feedback (feedback can "multiply the natural damping factor of the output circuitry"). (That's not a proper technical description but it actually gets the idea across pretty well.) That lovely output transformer used in almost all tube amps produces a lot of phase shift and other non-linearities at high frequencies. And those, in turn, limit the amount of feedback that can be safely used before encountering "stability issues". (That's a nice way of saying that, if you try to apply too much feedback, your amplifier either becomes unstable and sounds really odd, or oscillates and turns into a giant doorbell.) All of these things taken together determine how a particular tube amplifier is going to sound. Also note that some companies, like Audio Research, actually design their tube amps NOT to add coloration, while many Chinese manufacturers... have the opposite goal in mind. (But, if you're going to choose a tube amp that doesn't add coloration, then you might as well buy a solid state amp, and save a lot of money...) I was thinking about what you said the other day. I have limited experience with a tube buffer and tube dacs. But the two I've heard (the Jolida and a yaqin tube buffer) though their impressions were positive did not quite sound like a tube amp. I am thinking that a tube amp interacts with speaker load in a more complex manner to produce its characteristic sound. I'd agree - but not all tube amplifiers create the effect you're discussing. What tube amplifiers will react favorably with specific speakers is a profoundly difficult thing to predict. I've owned some tube amps (including models from Audio Research, McIntosh, Rogue Audio, etc.) that were not in any way superior sounding to solid state amplifiers. I've owned others (HealthKits, VTL compact 100s) that stunned regardless of the speakers connected to them. And I've found a few combinations (inexpensive Chinese tube amplifier + Klipsch LaScala speakers) that just happened to work together amazingly well. So, what is it that makes some tube amplifiers sound so spectacular while others fail to do so? I don't know. And I've never seen anyone else provide a cogent and rational explanation either. Note that the comments in the above paragraph are not meant to disparage vacuum tube amplifiers from Audio Research, McIntosh, or Rogue Audio - I'm sure that some of their amplifiers sound really spectacular with some speakers. But the amplifiers that I just happened to own didn't particularly impress with the speakers that I'd tried them with. So there's quite a range of potential outcomes with tube power amplifiers and various loudspeakers. You might happen to find a synergistic combination - but you might not. That's why I generally prefer NOT to use vacuum tube power amplifiers. I find that solid state amps are far less picky about what speakers they'll sound good with. That's why I generally prefer to include tube stages in the amplification chain earlier in the process and use solid state power amplifiers to drive the loudspeakers. I'd also argue that it is far, far less expensive to do what I suggest rather than trying to use tube power amplifiers. Output tubes are NOT cheap, and to keep your tube power amplifier operating at peak performance, you'll need to change those output tubes on a regular basis. Balance that with the fact that you can buy really great-sounding solid state amplifiers for prices that are a fraction of what you'd have to pay for a vacuum tube power amplifier of the same wattage. Glenn |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Dec 20, 2022 12:02:56 GMT -5
It's also worth mentioning that speakers vary in how they interact with the amplifier.
Most modern speakers are designed with the expectation that they will be connected to an amplifier with a high damping factor.
They are designed with relatively little internal damping because they expect the amplifier to provide lots of electrical damping.
Therefore, if you connect one of these to a tube amplifier, which has very little damping, the bass will often sound "loose, sloppy, or boomy".
In contrast, most vintage speakers, and some few modern ones, are designed with the expectation that they will be connected to an amplifier with a LOW damping factor.
They are designed with an adequate amount of internal damping because they expect the amplifier to provide very little electrical damping.
And, if you connect one of those to an amplifier that also provides a lot of electrical damping, it may sound "dry and overdamped".
And, of course, some speakers are simply less sensitive to this than others.
It's also worth mentioning that you can simulate JUST THIS ASPECT of a tube amplifier quite easily with a solid state amp.
If you simply connect a big fat 1 Ohm power resistor in series with each speaker the result will be a DF of 4 (with a 4 Ohm speaker) or a DF of 8 (with an 8 Ohm speaker).
(This won't simulate the transformer and other non-linearities but it will let you hear part of what the speaker would sound like with a tube amp.)
|
|
|
|
Post by Boomzilla on Dec 20, 2022 17:06:51 GMT -5
Recent research indicates that Intel's design goal for the NUC was to make it even smaller & cheaper than a laptop. Therefore, connectors like the USB sockets are not mounted to the case or to any reinforced plate, but just wave-soldered onto the 4x4-inch motherboard. Since the sockets are vertically stacked, one above the other on the motherboard, any attempt to insert a USB-A plug in backwards, for example, on the top one of the two USB sockets gives enough bending force to break the solder joints holding the pair of USB sockets to the motherboard. It WAS a very poor design, but I guess Intel didn't care.
|
|
|
|
Post by leonski on Dec 21, 2022 0:28:31 GMT -5
Any reason INTEL Should care? I did a computer repair for a buddy. I was NOT qualified for this but got lucky.
A power connector had worked loose and he had a replacement. I successfully desoldered and REsoldered in the new part.
It was a 3 oe 4 layer board, too boot.
I've done headphone repairs with wire I could hardly SEE .....and it was nylon reinforced so you had to deal with THAT, too.......
Over the years I've been pretty lucky.
My CA DAC Magic +? Requires some REAL long #1 philips Screwdriver and the back panel is held on with maybe a Dozen or more screws.
I give it about a 90% chance of being able to fix it IF I could get at it! I'm betting a cold soldier / cracked from using that new brittle crap....
As usual, Keith goes on about some thing very simple.
Old school 'big Box' speakers were very good with low damping tube gear. The 'invention' of the 'small box'......say.....the Early AR 3 type speaker? VERY low sensitivity compared to
what went before, but also benefitted from high damping factor. About the only 'exception'? Magnepan which seems to work equallly well over a wide range of damping factors.
Smaller speakers have MOSTLY taken over and since watts are cheap? The low sensitivity is compensated by more watts......
Keith's notes about damping are spot on.
That resistor trick? Used in one of the CARVER amps.......from the SUNFIRE line....which also incidently used International Rectifier IGBTs as output devices......
Pass also has an article somewhere about adding 'networks' to SDFR speakers run on SS amps.....which help 'em perform more like they were connected to tubes...
I wonder about some NEW speakers like the KLIPSCH Forte IV. I've seen reviews with both SS and Tube gear. I'm going to have to go for a listen myself!
|
|
|
|
Post by Boomzilla on Dec 21, 2022 8:44:06 GMT -5
Any more, so far as I know, circuit boards are constructed via wave soldering. All of the surface mount components are placed in their proper locations on the motherboard, and then the entire board (with components) is microwaved just enough to melt the solder without cooking the components. Although this results in a low reject rate for the assembly line, the solder joints aren't always as robust as they might otherwise be. And traditionally, any component that must withstand physical connection/disconnection (such as USB ports) is mounted not on the motherboard itself but rather on a metal reinforcing plate with a plug/sucket connection to the motherboard. On the NUC, this practice was abandoned in lieu of economy of construction. Oh well...
|
|
|
|
Post by Boomzilla on Dec 21, 2022 12:53:29 GMT -5
...It's also worth noting at this point that we humans tend to mentally equate distortion with "loudness". (This is why low powered devices with lots of distortion sound "annoyingly loud". It's also why good headphones often do not sound loud - even at high SPL levels.) This produces an interesting "effect" with tube devices "especially triodes" that have a level of distortion that increased more or less linearly with output level. As an example let's compare a "typical" 10 watt solid state amp to a 10 watt single-ended triode amplifier. At 0.1 watts both probably have relatively low distortion... At 1 watt, the solid state amplifier is at 0.01% THD, and the triode is at 1%... At 5 watts, the solid state amp is still at 0.02%, and the triode is at 10%... The result of this is that, at 5 watts, the triode amplifier experiences "a psychoacoustic boost in loudness" (because we hear the increased distortion as extra loudness). So, since both amps sound equally loud at 1 watt, but the triode amplifier sounds louder at 5 watts, we have the psychoacoustic equivalent of a dynamic range expansion. (The apparent difference in loudness between 1 watt and 5 watts is greater with the triode amp than with the solid state amp - so it "sounds more dynamic".)... This is the first rational explanation I've heard for why my 12-Watt Heathkit tube power amplifier can sound "more dynamic" than my 250-watt solid state amplifier (and it DOES). The little Healthkits are the most dynamic power amplifiers I've ever heard - and they sound that way not only with high-sensitivity loudspeakers, but with every speaker that I've tried them with. The "increased distortion with increased output" explanation makes perfect sense and goes a long way toward explaining what, until now, has been a mystery. It also explains why the Healthkits, when pushed toward the upper limits of their output range sound LESS dynamic than they do at ordinary listening levels. That said, I've heard some 16-watt tube amplifiers (the Black Ice Audio F11s) that sound FAR LESS dynamic than the 12-watt Healthkits. The Black Ice units, however, probably run in Ultralinear Pentode mode (that the Heathkits don't). That would limit their distortion to a far lower amount until the amps approached their clipping level at the full 16 watts (which is VERY loud with just about any contemporary speaker). So at "normal" listening levels, the 12-watt amps (that clip much sooner) would sound more distorted and thus more dynamic on peaks than the 16-watt units that delay distortion until higher power levels. All that said, the distortion of the Heathkits isn't particularly audible as "distortion." Instead, the amps just sound spectacularly dynamic which makes them a LOT of fun to listen to. I think that I may keep my Heathkits so that when an audio amigo comes over who is burdened by the misconception that ONLY high powered amplifiers can produce dynamic sound, I can stun him with the little 12-watt amps that sound like a million watts! LOL Glenn |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Dec 21, 2022 14:22:55 GMT -5
Second harmonic distortion, which is the prevalent distortion with triodes, doesn't sound especially bad... In fact, in some products, second harmonic distortion is deliberately added to improve voice intelligibility and make voices sound "richer". (I've heard of this being done in some public address systems... and in a few hearing aids.) Here's a bit of trivia... Do you remember the Linda Ronstadt's signature song - Blue Bayou... And how the voice was downright penetrating on that song...? That song was enhanced using The Aphex Aural Exciter... The Aural Exciter is a hardware box which allows you to turn a knob to add precise amounts of second harmonic distortion. (It was actually quite popular back then... but they turned it up especially high on that song .) ...It's also worth noting at this point that we humans tend to mentally equate distortion with "loudness". (This is why low powered devices with lots of distortion sound "annoyingly loud". It's also why good headphones often do not sound loud - even at high SPL levels.) This produces an interesting "effect" with tube devices "especially triodes" that have a level of distortion that increased more or less linearly with output level. As an example let's compare a "typical" 10 watt solid state amp to a 10 watt single-ended triode amplifier. At 0.1 watts both probably have relatively low distortion... At 1 watt, the solid state amplifier is at 0.01% THD, and the triode is at 1%... At 5 watts, the solid state amp is still at 0.02%, and the triode is at 10%... The result of this is that, at 5 watts, the triode amplifier experiences "a psychoacoustic boost in loudness" (because we hear the increased distortion as extra loudness). So, since both amps sound equally loud at 1 watt, but the triode amplifier sounds louder at 5 watts, we have the psychoacoustic equivalent of a dynamic range expansion. (The apparent difference in loudness between 1 watt and 5 watts is greater with the triode amp than with the solid state amp - so it "sounds more dynamic".)... This is the first rational explanation I've heard for why my 12-Watt Heathkit tube power amplifier can sound "more dynamic" than my 250-watt solid state amplifier (and it DOES). The little Healthkits are the most dynamic power amplifiers I've ever heard - and they sound that way not only with high-sensitivity loudspeakers, but with every speaker that I've tried them with. The "increased distortion with increased output" explanation makes perfect sense and goes a long way toward explaining what, until now, has been a mystery. It also explains why the Healthkits, when pushed toward the upper limits of their output range sound LESS dynamic than they do at ordinary listening levels. That said, I've heard some 16-watt tube amplifiers (the Black Ice Audio F11s) that sound FAR LESS dynamic than the 12-watt Healthkits. The Black Ice units, however, probably run in Ultralinear Pentode mode (that the Heathkits don't). That would limit their distortion to a far lower amount until the amps approached their clipping level at the full 16 watts (which is VERY loud with just about any contemporary speaker). So at "normal" listening levels, the 12-watt amps (that clip much sooner) would sound more distorted and thus more dynamic on peaks than the 16-watt units that delay distortion until higher power levels. All that said, the distortion of the Heathkits isn't particularly audible as "distortion." Instead, the amps just sound spectacularly dynamic which makes them a LOT of fun to listen to. I think that I may keep my Heathkits so that when an audio amigo comes over who is burdened by the misconception that ONLY high powered amplifiers can produce dynamic sound, I can stun him with the little 12-watt amps that sound like a million watts! LOL Glenn |
|
KeithL
Administrator
Posts: 10,273
|
Post by KeithL on Dec 21, 2022 14:47:37 GMT -5
Just for the record... that is pretty much how surface mount components are attached... although I'm pretty sure they use infrared or air heating and not microwaves. (The solder, usually a low-temperature variety, and the flux, are applied to the pads as a paste... and the entire thing is heated to melt the solder.) But "wave soldering" is something else entirely... which was quite commonly used in the days of through-hole components. It involves loading all the components into the board, then passing the entire board over a "wave" of molten solder, which just touches the bottom of the board. And, after the whole board is soldered, you trim off all the leads, and you have a completed board. The picture shows the actual solder wave... with the boards being carried over it on a sort of conveyor belt (the frame the boards are clamped into rides on rails). There are fluxing and pre-heating steps before that... and cleaning and lead trimming steps after... and usually at least a few components that must be added by hand afterwards. Here's a video of the basic process... www.youtube.com/watch?v=inHzaJIE7-4 (it shows the flux wave, the heater/drier, and the solder wave... )  And, yes, surface mount components have a few issues... The biggest is simply that, since they're just mounted on the surface of the board, any excessive flexing of the board, or even expansion and contraction, can break them loose. The other big annoyance is the current ROHS compliant lead-free solder. In general lead-free solder requires higher temperatures, doesn't adhere as well, and tends to require more aggressive fluxes. Combine that with the modern requirement of using "environmentally friendly water cleanable" flux... And, overall, it becomes more difficult to get a good solder connection, and easier to fail to do so. And, yes, the NUC is intended as a "super-economy computer"... There are plenty of other "small profile" computers if you want to pay more for better quality and more options. Any more, so far as I know, circuit boards are constructed via wave soldering. All of the surface mount components are placed in their proper locations on the motherboard, and then the entire board (with components) is microwaved just enough to melt the solder without cooking the components. Although this results in a low reject rate for the assembly line, the solder joints aren't always as robust as they might otherwise be. And traditionally, any component that must withstand physical connection/disconnection (such as USB ports) is mounted not on the motherboard itself but rather on a metal reinforcing plate with a plug/sucket connection to the motherboard. On the NUC, this practice was abandoned in lieu of economy of construction. Oh well... |
|
Where should a tube buffer go?
