Scott,
You are soooo right about that. Often, audible differences between amplifiers are the result of one or both of them misbehaving in a different way. Burst parasitic oscillations are a good example. Indeed, sometimes those parasitic oscillations will not even make huge measured THD; I had a case where an amplifier was making perhaps 0.02% THD when by all accounts it should have been much lower. It had a burst oscillation. When I fixed the problem, distortion dropped by more than a factor of 5. Many people would have just brushed that first THD measurement off, saying an amplifier with 0.02% THD is just fine.
BTW, just because an amplifier has a price tag in the many thousands of dollars does not mean that the designer knows how to compensate it. Nor does it mean he has a scope that will catch such HF oscillations.
Cheers,
Bob
I hope I didn't cause anyone too much trouble with my seemingly still "exotic" file format. OpenOffice/LibreOffice doesn't seem to be that widespread in audio circles yet.
Not Found
The requested URL /wordpress/wp-content/uploads/2010/10/Opamp-User-Guide-V1.01.pdf was not found on this server.
Additionally, a 404 Not Found error was encountered while trying to use an ErrorDocument to handle the request.
V1.0 downloads just fine though.
Yet, you can still approach an ideal playback system that reproduces records as well as possible. In that sense, fidelity absolutely is well-defined.
Assuming I wanted to get the best playback possible, I'd just stick with known-good sources and amps and invest the largest part of my efforts into room treatment and (the right kind of) speakers, ideally with some fancy room correction thrown in. (At this point, I'd probably have a record collection in the 1000s.)
But you know how I do most of my critical listening these days? In bed at night, with my trusty Sennheiser HD590s connected to a Clip+ (running Rockbox with Meier crossfeed enabled). And LAME -V 6 to -V 4 quality MP3s. Silence, darkness and concentration do a lot more for musical enjoyment than having even better equipment. Besides, realistically speaking, even like that I'm pretty far into diminishing returns territory already (these are good headphones, the Clip+ - being quite the technological marvel - makes a decent source and doesn't break into sweat driving them, Rockbox has a good processing chain, the software Meier crossfeed was the first that I really liked, and I repeatedly found myself unable to ABX my MP3s from their lossless sources).
Oops:
Not Found
The requested URL /wordpress/wp-content/uploads/2010/10/Opamp-User-Guide-V1.01.pdf was not found on this server.
Additionally, a 404 Not Found error was encountered while trying to use an ErrorDocument to handle the request.
V1.0 downloads just fine though.
Thankfully some classes of equipment very much can be made audibly transparent. (The demands of human hearing are, after all, finite.) It's a little hard with transducers (plus even the same loudspeakers rarely sound exactly the same in another location) but absolutely doable for electronic components like sources and amplifiers.
Canned sound has some obvious limitations, that's true. Stereo recording and playback in particular can never recreate the original sound field, it can only create an illusion that might come close. It tends to get better for a larger number of recorded channels.
Yet, you can still approach an ideal playback system that reproduces records as well as possible. In that sense, fidelity absolutely is well-defined.
Assuming I wanted to get the best playback possible, I'd just stick with known-good sources and amps and invest the largest part of my efforts into room treatment and (the right kind of) speakers, ideally with some fancy room correction thrown in. (At this point, I'd probably have a record collection in the 1000s.)
But you know how I do most of my critical listening these days? In bed at night, with my trusty Sennheiser HD590s connected to a Clip+ (running Rockbox with Meier crossfeed enabled). And LAME -V 6 to -V 4 quality MP3s. Silence, darkness and concentration do a lot more for musical enjoyment than having even better equipment. Besides, realistically speaking, even like that I'm pretty far into diminishing returns territory already (these are good headphones, the Clip+ - being quite the technological marvel - makes a decent source and doesn't break into sweat driving them, Rockbox has a good processing chain, the software Meier crossfeed was the first that I really liked, and I repeatedly found myself unable to ABX my MP3s from their lossless sources).
My experience was almost the opposite. I guess I was the last in my neighbourhood to buy a CD player, Arcam delta 70.2 @~1990 .
Last edited:
I didn't own one until 1994....a heavily modified Philips 850 MK11...but I didn't live in the Scottish Borders then; Andrew T may have done so his statement above is correct!
I still prefer my recordings to be on brown tape or black discs...and I am not interested in conversion, so evangelists, please don't even try!
The article SW is talking about (Post #327...), can be found here:
http://www.picovolt.com/win/elec/articles/Lepaisant_preamp-xfmrs_RSI.pdf
http://www.picovolt.com/win/elec/articles/Lepaisant_preamp-xfmrs_RSI.pdf
Agreed, just added a quad 33 pre to my 4396 dac and it seems a perfect match,
brilliant sounds have been added to my system.
Quad power supply mod has done wonders.
A system is the best system (for you) in the world if it always exceeds expectations. I arrived at this musical reproduction point as well but may have from a different approach but that is not important.
When you pull out a CD or LP you already know what you going to hear and how it will affect you, but hey, then the system surprises you and the outcome is better than you anticipated every time - this is the kind of system I am talking about - when you reached that point you stop tweaking because every tweak you make after this point goes backward.
When you pull out a CD or LP you already know what you going to hear and how it will affect you, but hey, then the system surprises you and the outcome is better than you anticipated every time - this is the kind of system I am talking about - when you reached that point you stop tweaking because every tweak you make after this point goes backward.
I haven't seen the circuit for this part of the design but would say the following. One pin of T0220 type regulators (or transistors) is always connected to the tab of the device package. If the print that pin goes to connects to the relevant heatsink then I would say no washer is necessary for insulation purposes. That said there are occasions where the extra stray capacitance of a heatsink can cause added noise pickup depending on circuit configuration in which case a washer might be advisable.
Are the regulators specified available in fully isolated packages ? Many 78/79 types are.
Are the regulators specified available in fully isolated packages ? Many 78/79 types are.
My bad. Now after re-looking at the PCB the heatsinks are just soldered onto the board with no electrical connection. No worries about insulating their package (TO-220) from the HS. But, I will use thermal paste as good and necessary practice.
Can I use the Artic Silver variety I use on CPUs, or should I stay away from potentially electrically conducting thermal paste?
Last edited:
I finished building its predecessor, the Precision Preamp '96, yesterday - see attached photo (excuse dodgy photography, and as you can see it hasn't made it into a case yet...). I tried to do the superb circuit and PCB design justice by carefully measuring and selecting every component and using the best quality components I could get my hands on at sensible prices. I used polypropylene and bipolar coupling caps everywhere, they mostly fit quite well. Whether they really make much difference I don't know, but I wanted the best quality parts throughout. They shouldn't hurt anyway. I also used Fairchild 5532s which are supposedly the best.
As a sound processor it works beautifully, those adjustable tone controls really make it much easiser to correct the balance of bad recordings.
The sound is interesting - there is a sort of "forced" and slightly compressed quality that really works on some recordings but not so much on others. By this I don't mean dynamics are reduced, but it's really hard to put this into words.
The bass is wonderfull, I've always thought 5532s have the best sounding bass of any opamp or discrete circuit I've tried, and this design really brings it out with stacks of punch and "meat".
The overall balance is slightly less bright compared with other preamps I have which is no bad thing as far as I'm concerned, however this means the sound is less "expansive" - the soundstage just doesn't seem to expand out of the speakers as much. With classical the timbre of strings is slightly odd, and the detail of hall acoustics does seem somewhat reduced.
Am I hearing the active gain control? Something about the way the amount of feedback is changing when the volume control is altered?
Or is it the sheer complexity of the design - are there just too many opamps and coupling caps that each have a small impact on the sound? I've never really wanted to believed this. Having worked in the recording and broadcast industries I absolutely concur with Mr Self that on the majority of recordings the signal has passed through hundreds of opamps and capacitors and hundreds of metres of cable, and it defies logic to think that a few more at the reproduction end would make so much difference.
In conclusion, I'm dissapointed and frustrated. As an engineer I really wanted to believe that a design that measures impeccably should sound "correct", but my ears just aren't buying it. Besides my wife doesn't like it at all - she has demanded that I put the other preamp back
Can anyone offer any insight to help ease the pain?
As a sound processor it works beautifully, those adjustable tone controls really make it much easiser to correct the balance of bad recordings.
The sound is interesting - there is a sort of "forced" and slightly compressed quality that really works on some recordings but not so much on others. By this I don't mean dynamics are reduced, but it's really hard to put this into words.
The bass is wonderfull, I've always thought 5532s have the best sounding bass of any opamp or discrete circuit I've tried, and this design really brings it out with stacks of punch and "meat".
The overall balance is slightly less bright compared with other preamps I have which is no bad thing as far as I'm concerned, however this means the sound is less "expansive" - the soundstage just doesn't seem to expand out of the speakers as much. With classical the timbre of strings is slightly odd, and the detail of hall acoustics does seem somewhat reduced.
Am I hearing the active gain control? Something about the way the amount of feedback is changing when the volume control is altered?
Or is it the sheer complexity of the design - are there just too many opamps and coupling caps that each have a small impact on the sound? I've never really wanted to believed this. Having worked in the recording and broadcast industries I absolutely concur with Mr Self that on the majority of recordings the signal has passed through hundreds of opamps and capacitors and hundreds of metres of cable, and it defies logic to think that a few more at the reproduction end would make so much difference.
In conclusion, I'm dissapointed and frustrated. As an engineer I really wanted to believe that a design that measures impeccably should sound "correct", but my ears just aren't buying it. Besides my wife doesn't like it at all - she has demanded that I put the other preamp back
Can anyone offer any insight to help ease the pain?
