No, I use it in my music system. For measurements, I use a BUF03-based system or Pete Millett's excellent interface.
Let me rephrase my question:
Do you design your line amplifiers according to how they measure only? By measure I mean measurements relevant to audio amplifiers, like distortion, SNR, output impedance, etc.
To clarify further my question, do you evaluate the quality your line amplifiers also by listening to it, or are you satisfied by it being measured well?
If (for example) a hypothetical line amp measured well attached to my signal generator and spectrum analyzer, but played AM radio when I inserted it into my music system, I would not consider that an acceptable design.
I'm confused by your concept of building something, then not using it.
I'm confused by your concept of building something, then not using it.
Good questions, Joshua. I'm as confused as you are. I have 4 of my own preamp designs here, that I can use, yet I only use one, and the others actually have more features. I don't build products just for fun, and if they all sounded essentially the same, I would not bother with any of the newer designs. My Levinson JC-2 had virtually everything and more than I use today, yet it sits on a shelf. This is because I became sensitive to its limitations, as I detected them decades ago, and MEASURED them later.
We did our best, back in 1973, when it was first designed, but we made mistakes that I avoid now. For example, asymmetrical crosstalk, a real image killer.
We did our best, back in 1973, when it was first designed, but we made mistakes that I avoid now. For example, asymmetrical crosstalk, a real image killer.
Have you measured its microphony? Whay about the phone pre? This alone is an issue that no amount of engineering will overcome. Enough about pulling babes - you have a car for that
That is indeed a serious issue for tubes, especially in phono stages. One friend of mine speculates that the popularity of DHTs stems from some artificial forwardness and spatial qualities from the filaments' resonance.
I tested my line amp for microphonics in a rather crude way: tapping on it with a pencil while it was hooked up. No problems. The microphonics in my phono stage were definitely more noticeable, so it required some attention in isolation. That's the disadvantage of tubes in that application.
Obviously you avoid direct reply to my questions. My questions are clear, yet you reply to what I didn't ask. One possible reason for this is that you may be an ingenious designer who is reluctant to disclose his trade secrets, thus deterring others from copying them.
Let me tell you this.
Many people post various schematics on the forums here. I intend to build few amps, namely, a power amp, a preamp, a phono stage and a tape head amp.
For some reason, I opted for tube amps for all the above amps.
Since I have neither the time nor the money to build amps according to each and every schematic posted on the forums here, I do my own selection. This selection is very simple and straight forward: reports concerning the sound quality of those amps, based on listening. I don't choose schematics that only measure well, when they aren't supported by listening evaluation testimonies.
Now, according to some things you wrote here, I'm deluded.
May be I am, yet, according to my own experience with various pieces of gear for stereo setup, concerning the sound quality, listening evaluation is far more revealing than the published tests results.
Indeed, there is catch. I know that I can never follow blindly listening evaluations of others, since different people evaluate differently sound reproduction systems. Thus, I know in advance that I cannot take anything for granted before I'll hear it myself, on my own stereo setup. Yet, I will never spend my time and money to build an amp that all I know about it is that it measures well.
Of course, your approach may be just the opposite – and it may not. I have no way of knowing that, since you avoid replying directly me questions, as they were asked. Also, in another thread, I asked you to specify a statement of yours that measurements can predict the sound quality of an amp. You avoided that question by sending me to perform a search, a search which didn't bring about any meaningful results.
So, what I see from you, in various threads, is nullification of the value and use of listening tests, without any clear and substantiated stand on the issue of measurements only, versus combining measurements with listening tests. All I get is that you have something to hide on this issue, though I'm not sure what it is you are hiding and why you are hiding it. You know, attacking and criticizing a certain approach is one thing. Taking a stand and clarifying it is another thing.
As for myself, I cannot prove to others, by scientific means, the inevitable necessity of listening evaluation, on top of any measurements set. But when my limited money is at stake, I cannot and I will not risk it, based on measurements only, without any listening evaluation.
YMMV.
How 'bout this?
There are some decided engineering advantages to using tubes as the voltage amplification devices. At the top of the list are high linearity and overload immunity. My old unit was a hybrid using FETs in cascode with tubes (not unlike Allen Wright's excellent designs, though not claiming the same performance), and while it was good enough to live with for a quarter century, I still feel that its performance can be surpassed with an all-tube amplification lineup. Of course, I will still not hesitate to use semiconductors where they do best, i.e., for providing power, increasing power supply immunity, and controlling operating points- for constant voltages and constant currents, silicon is the way to go.
From the Requirements page of His Masters Noise.
se
Perhaps SY ought to start a thread on his design concepts??
Serious.
I posted when there were no other responses, but somehow now there are a bunch... weird... heh.
I guess everyone has their own design philosophy, mine is somewhat different than SY's or Curl's or anyone else's, but is not fixed.
Not sure I see a genuine need for gain in a line stage as things now stand with the technology... so I don't use one. That obviates 99% of what SY is claiming as an advantage? Surely no active device is more linear than any active device??
In a phono stage it is virtually impossible to get a tube to be as low in noise as a good bunch of jfets or the right bipolars... so the cascode looks good to me. Dunno, pick ur poison(s)?
Or maybe you prefer some of those new dandy ultra super maximal duper low distortion opamps for these applications? They do measure low, right? Impossibly so... how could they be bad??
_-_-bear
Serious.
I posted when there were no other responses, but somehow now there are a bunch... weird... heh.
I guess everyone has their own design philosophy, mine is somewhat different than SY's or Curl's or anyone else's, but is not fixed.
Not sure I see a genuine need for gain in a line stage as things now stand with the technology... so I don't use one. That obviates 99% of what SY is claiming as an advantage? Surely no active device is more linear than any active device??
In a phono stage it is virtually impossible to get a tube to be as low in noise as a good bunch of jfets or the right bipolars... so the cascode looks good to me. Dunno, pick ur poison(s)?
Or maybe you prefer some of those new dandy ultra super maximal duper low distortion opamps for these applications? They do measure low, right? Impossibly so... how could they be bad??
_-_-bear
A better analogy is an Audemars Piquet Offshore.
Virtually indestructable and not 10 minutes per year off like some other chick chasers, looks good and pricey as H.
I hate it when a watch is really off every couple of months, or breaks down too frequently, no Omega or Rolex hype ever again.
I got round to auditing Mr Hansen's kx-r some time ago, brought along a <<0.004% preamp, didn't sound the same as a decent opamp pre to me.
If some cash burner fool would flogg the kx-r for a go-dutch rate maybe i could score some babes after all, i'm ugly as H too.
SY has the concept, I can only guess. But this statement of his strongly parallels his remarks in the now closed cable audibility thread where his weasel phrase was 'non-mundane differences'. Requests to pin down these phrases (as might be required were we to to do real science i.e. falsifying his hypothesis) have proved fruitless
This is obviously false - I've read his 5th edition book and whilst he is excessively (IMO) strong on distortion he does cover other aspects of amplifier measurements - such as slew rate and noise just to give two examples.
I agree, this is the primary weakness of his book. But then he gives no PCB layouts either - schematics can't be listened to. The second weakness of his book is citing Randi.
I've designed and built a few amps on his principles since he first expounded on them in E&WW and have never suffered complete disappointment. So do you have any examples to offer where the disappointment can be laid at the feet of Mr. Self's principles, rather than say sheer incompetence?
No, they're not.
The idea that I have "trade secrets" is totally hilarious, since I publish my schematics, detailed design information, show my calculations, and my exact thoughts on design tradeoffs. Even more hilarious because I don't do audio professionally. Never even sold a circuit board. The most I've done is made some pizza money from my articles. My profession is polymer science. Ask me how to formulate a polypropylene without antioxidants that cause endocrine disruption and THEN it's likely I'll clam up.
It might help to actually read the stuff I've published. I'm not designing for you, but there's a slim possibility that you could learn something.
Really? You must be confusing me with someone else.
In setups where it's possible, I would 100% agree. Now if you have to drive any length of cables and have multiple inputs to feed, then a unity gain line amp is (IMO) the best solution. If you don't, a switch and a pot are absolutely the best way to go, assuming your goal is having no effect on the sound by the electronics, other than raising and lowering the volume.
You could certainly make an excellent (i.e., sonically transparent in a blind bypass test) unity gain stage using bipolars, FETs, opamps, or tubes. Rolex or Patek-Philippe? It's still 4:20.
I'm sorry SY, but what you should be comparing, is a Timex to an atomic clock. After all, accurate time is now possible, but NOT super accurate time from a Timex, unless it is calibrated to WWV. This is possible, but I suspect that you don't have that accuracy, or even bother. Jewelry does not do much either, but people often buy it. I have a Casio, myself, but I once owned an Omega and I loved it at the time.
- Status
- Not open for further replies.