John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
Hi,

Ciao T

Thorsten,

I have to apologize; I assumed things just keep improving. After evaluating several of the 24/96 small media servers I realize their performance, to say the least, is marginal. I don’t get it, my >10yr. old tech Edirol UA-5 has very clean spectra at the output. I guess I need to look into this further.
 

Attachments

  • 220px-Alles_Ist_Gut.jpg
    220px-Alles_Ist_Gut.jpg
    9.9 KB · Views: 305
It is even better if you don't smoke, that is best left to experimental electronics, and not really even then.

However I can give up smoking (and I plan to do soon), but it won't change the test procedure I prefer. If I can't imagine a live event behind that open windows no blind tests of any kinds, nor any measurements, can convince me that the equipment is better than other that allows to imagine easier as if it does not exist.
 
Scott,

I have to apologize; I assumed things just keep improving. After evaluating several of the 24/96 small media servers I realize their performance, to say the least, is marginal. I don’t get it, my >10yr. old tech Edirol UA-5 has very clean spectra at the output. I guess I need to look into this further.

Well, with respect, DAC Chip's have dis-improved mostly over the last two decades, it is only in the last 5 Years or so that we some positive development again. But at the low end it is a real cabinet of horrors.

Get an EMU 1616m External Soundcard, at least it measures well (sound at best is ordinary), not bad for the 300 or so damp squid... (from quid - pound)

Past that, almost everything cheap is now designed and made in mainland china, where the electronic arts are not even in the stone age yet. Not sure they out of the trees in fact.

Now I do not put much stock into traditional measurements so often my stuff does not measure in this sense, but I have good reasons for doing so (if people here agree with them or not) and my stuff measures different because I know "better" than those who "invented" the measurements (like Harold J. leak who promoted the importance of low (T)HD as means of advertising with zero evidence that it actually mattered).

Much of modern generica tests and measures the way it does because the chinese engineer who designed does not really understand electronics, believes real ground is the same as theoretical ground and has been working from Babelfish translated versions of Datasheets, where he would have struggled to understand a native chinese version.

The results are predictable and can be seen in much of the E-Bay kit(s) coming from that corner of the world. I think most chinese engineers are still in the first decade of the 40-50 Years KBK postulated, many even before them...

Ciao T
 
> Past that, almost everything cheap is now designed and made in mainland china, where the electronic arts are not even in the stone age yet.
> Not sure they out of the trees in fact.

I recall your high-end commercial products are also Made in China ?

http://www.ardhowenhifi.com/brands/abingdon-music-research/

"Based in London with their own manufacturing facilities in China, Abbingdon Music Research was founded in 2006 by Mr P Wayne, Vincent Luke and Thorsten Loesch."


Patrick
 
Last edited:

I am not convinced about the advantages of tying signal ground to chasis ground. EM coupling between chasis and internal electronic is mentioned as a reason why this should be done, but this would appear only to be true if this EM coupling is just between chasis and signal ground, which of course it is not. Tying chasis ground to earth ground to me seems a better option to bleed off noise current before problems of such nature can even occur. The other reason mentioned to tie signal and chasis earth together is the need to keep the signal grounds in two boxes tied, but this can be done through 1 of the signal leads when using XLR cable and connectors in an unbalanced setup.
 
Scott,



Well, with respect, DAC Chip's have dis-improved mostly over the last two decades, it is only in the last 5 Years or so that we some positive development again. But at the low end it is a real cabinet of horrors.

Get an EMU 1616m External Soundcard, at least it measures well (sound at best is ordinary), not bad for the 300 or so damp squid... (from quid - pound)

Past that, almost everything cheap is now designed and made in mainland china, where the electronic arts are not even in the stone age yet. Not sure they out of the trees in fact.

Now I do not put much stock into traditional measurements so often my stuff does not measure in this sense, but I have good reasons for doing so (if people here agree with them or not) and my stuff measures different because I know "better" than those who "invented" the measurements (like Harold J. leak who promoted the importance of low (T)HD as means of advertising with zero evidence that it actually mattered).

Much of modern generica tests and measures the way it does because the chinese engineer who designed does not really understand electronics, believes real ground is the same as theoretical ground and has been working from Babelfish translated versions of Datasheets, where he would have struggled to understand a native chinese version.

The results are predictable and can be seen in much of the E-Bay kit(s) coming from that corner of the world. I think most chinese engineers are still in the first decade of the 40-50 Years KBK postulated, many even before them...

Ciao T

I have had things happen where I'm previewing a (cough cough) $3k cd player for the purpose of equipment matching tests, to do a 'pre set-up' for a show, a dry run.... to ensure compatibility.

Curiosity makes me open the CD player.

I find good parts selection...tied to the schematic for a 5 volt portable CD player. The kind where the digital and analog ~and~ clocking power supplies come off the same rail, and done via resistively set load splitting. I do my best to fix it so we're not collectively embarrassed by the situation.
 
I am not convinced about the advantages of tying signal ground to chasis ground. EM coupling between chasis and internal electronic is mentioned as a reason why this should be done, but this would appear only to be true if this EM coupling is just between chasis and signal ground, which of course it is not. Tying chasis ground to earth ground to me seems a better option to bleed off noise current before problems of such nature can even occur. The other reason mentioned to tie signal and chasis earth together is the need to keep the signal grounds in two boxes tied, but this can be done through 1 of the signal leads when using XLR cable and connectors in an unbalanced setup.

We had this, as a similar situation, rear it's head with respect to our cables being tested by a major reseller.

Their massive power amplifiers were being modulated at the ground level via poorly thought out and executed AC connectivity. The pre and source were on completely different rails.

This massive offset was trying to run down the balanced cable's signal ground, and the mess that this make of our cable's ability to do what it does, well..the cables known efficacy...was actually reversed in effect.

As soon as I heard what the sonic results were with our cables, I immediately inquired to their AC power situation. And found the high levels of separation, line lengths and elevated complex LCR of the AC.
 
diyAudio Member RIP
Joined 2005
One of the remarkable (to me, at first encounter) aspects of mains power distribution with hot, neutral, and safety conductors that was uncovered by Bill Whitlock: the difference in potential between safety grounds at different outlets is often mostly due to asymmetrical B field coupling from the hot and neutral conductors. And as one can imagine, the resulting voltages are at a very low impedance indeed, and of course dependent on current flow.

His remedy in many installations is to have the electricians remove the hot and neutral conductors from the conduit, do as tight a twist of them as feasible, and then reinstall them. He does get a lot of baffled looks :confused:

Having seen our Asian brothers slammed in this thread (I understand the frustrations that motivate a broad-brush treatment, while still striving to evaluate people as individuals) I won't even get started on some of my stories about electricians. There are at least a few good ones out there, I am sure.
 
Last edited:
diyAudio Member RIP
Joined 2005
I have had things happen where I'm previewing a (cough cough) $3k cd player for the purpose of equipment matching tests, to do a 'pre set-up' for a show, a dry run.... to ensure compatibility.

Curiosity makes me open the CD player.

I find good parts selection...tied to the schematic for a 5 volt portable CD player. The kind where the digital and analog ~and~ clocking power supplies come off the same rail, and done via resistively set load splitting. I do my best to fix it so we're not collectively embarrassed by the situation.

I had the occasion of purchasing a CD player for a friend who is in managed care. The primary requirements: not a portable, had a headphone output, and a level control for it. So I hunted down a player and got it here to inspect before bestowing it and headphones on the friend. After some listening (mediocre but acceptable) I popped it open. Eventually I downloaded a schematic.

Really not a bad set of design tradeoffs for the price (around a hundred bucks at discount). Double-bobbin power transformer placed at a fine-tuned angle to minimize hum field injection. A single-chip DAC/decoder with on-chip filtering and a rather pessimistic specification of 85dB S/N. Not much opportunity to enhance performance externally, so I left well enough alone. Bog-standard Rohm opamps after the chip and another for the headphones. Some minimalist discrete-component voltage regulators. Strategic line filtering and little RF filters in the headphone outputs.

My principal modification was to produce largish labels for the function buttons so Tupper would have a chance of reading them, instead of the absurd unreadable dark chassis markings that have long been the norm :mad:
 
Last edited:
Hey, calm down. I wonder how such stuff can escape the moderators.

I think the mistake comes if one automatically assumes that he is slagging on grounds that are outside of their given current state of affairs, with regard to evolution in audio engineering.

This aspect is not universal in the given situation (development of engineering expertize) it is a general point. Same as everywhere else. In this case, it is still near the beginning. I mean, after the war, stuff coming out of Germany had a long way to go.... Eventually it arrived. As it did in Japan, and as it did in the US, etc, etc, etc. All of this in the general sense, of course.

Regarding all of this, I think it may take place in half the time it took Japan, due to this happening in a time of a very different level and type of communication/relations capacity.
 
diyAudio Member RIP
Joined 2005
I think the mistake comes if one automatically assumes that he is slagging on grounds that are outside of their given current state of affairs, with regard to evolution in audio engineering.

This aspect is not universal in the given situation (development of engineering expertize) it is a general point. Same as everywhere else. In this case, it is still near the beginning. I mean, after the war, stuff coming out of Germany had a long way to go.... Eventually it arrived. As it did in Japan, and as it did in the US, etc, etc, etc. All of this in the general sense, of course.

Regarding all of this, I think it may take place in half the time it took Japan, due to this happening in a time of a very different level and type of communication/relations capacity.

The cultural factors are important, but in a curious way more flexible of late in China than in Japan. In the latter country there was the odd custom of letting someone who had appropriately deferred to an elder engineer for long enough, finally be given her/his chance to do a new product design. But so many of the lessons learned were not passed on, so we found o.k.-sounding products being superseded by ones with problems, problems (like excessive slew-rate DAC outputs screwing up the following stages) that had been at least partially solved.

But I have been exceedingly frustrated, trying to get overseas folk to consider designs and design modifications, and rejecting things out of hand for spurious reasons. One group insisted that they would be fine tying together the outputs of a filterless class D stereo amplifier chip to make a single amp for driving a lower impedance. I pointed out that the manufacturer did not guarantee either offset voltage or gain matching, to no avail. I even got said manufacturer to say Don't do it! Nope. My client knew better, because they HAD done it and it worked fine :eek:

In another case I suggested a level control IC and sent a schematic for a touch-sensitive decoder to operate it. This was rejected because it was pointed out that the level control IC had pushbutton inputs. Yes, but by controlling those with transistors instead of mechanical switch closures... Nope. Not gonna do it. Morons, both ignorant and arrogant.

As they say however: This too shall pass.
 
One of the remarkable (to me, at first encounter) aspects of mains power distribution with hot, neutral, and safety conductors that was uncovered by Bill Whitlock: the difference in potential between safety grounds at different outlets is often mostly due to asymmetrical B field coupling from the hot and neutral conductors. And as one can imagine, the resulting voltages are at a very low impedance indeed, and of course dependent on current flow.

His remedy in many installations is to have the electricians remove the hot and neutral conductors from the conduit, do as tight a twist of them as feasible, and then reinstall them. He does get a lot of baffled looks :confused:

Having seen our Asian brothers slammed in this thread (I understand the frustrations that motivate a broad-brush treatment, while still striving to evaluate people as individuals) I won't even get started on some of my stories about electricians. There are at least a few good ones out there, I am sure.

Not really. If you actually understand how safety ground is wired, it is that the neutral carries current and is connected to the safety ground at multiple points! (Code calls for it at the building point of entry, but there are multiple transformers in a large building and multiple building entries near small buildings.)

If it were capacitive coupling then the voltage would drop with load, it rarely does. If it were magnetic as suggested then twisting the hot and neutral tightly would increase the voltage difference between the neutral and safety ground.

The neutral voltage is supposed to be at "ground" but as it is really acting as a voltage divider from the two or three hot wires, it can only be at ground if all of the loads are balanced.

The safety ground wire is connected to the conduit system which has multiple poor quality actual earth grounds. So as a result if there is any path from the neutral to the safety ground current will flow. The path can be just the primary of the power transformer windings to the case.

I have had to return equipment that was manufactured with design assistance from Bill Whitlock, it just did not handle the safety ground correctly. As I recall it had 40 db s/n at safety to neutral voltage differences that were acceptable to the NEC. The manufacturer disagreed. They measured 60 db s/n. I had to explain to them that they were measuring from clipping to noise not from standard "0" db signal reference.
 
All due to some accounting *** in the early 1900's deciding to save a nickel (5 cents) on gauge size costs. (the reason behind the USA and Canada having unbalanced AC across the nations)

I did my best, sort of..to fix this house.

I brought in a 600lb ac conditioner, and removed all the electronics, and went to a clean hardwired balanced AC from there. +450lbs of transformer. Very nice laminate steel, as well. +500KVA peak, should be enough to be relatively linear for my use....
 
Last edited:
diyAudio Member RIP
Joined 2005
Not really. If you actually understand how safety ground is wired, it is that the neutral carries current and is connected to the safety ground at multiple points! (Code calls for it at the building point of entry, but there are multiple transformers in a large building and multiple building entries near small buildings.)

If it were capacitive coupling then the voltage would drop with load, it rarely does. If it were magnetic as suggested then twisting the hot and neutral tightly would increase the voltage difference between the neutral and safety ground.

The neutral voltage is supposed to be at "ground" but as it is really acting as a voltage divider from the two or three hot wires, it can only be at ground if all of the loads are balanced.

The safety ground wire is connected to the conduit system which has multiple poor quality actual earth grounds. So as a result if there is any path from the neutral to the safety ground current will flow. The path can be just the primary of the power transformer windings to the case.

I have had to return equipment that was manufactured with design assistance from Bill Whitlock, it just did not handle the safety ground correctly. As I recall it had 40 db s/n at safety to neutral voltage differences that were acceptable to the NEC. The manufacturer disagreed. They measured 60 db s/n. I had to explain to them that they were measuring from clipping to noise not from standard "0" db signal reference.

Well I'll have to refer you to Bill. His case histories sounded compelling, but your mileage may have varied.
 
Status
Not open for further replies.