I agree with Frank's statement that noise in the PSU shows up not only as noise, but otherwise as well. In my view, the first victims of noisy/inadeqate power supplies will be the fine details in the program which may simply not be there.
Can you verify this for yourself? Actually, yes and no. Yes, because theoretically you could put two of the same power amps to work, where one is in full original state, and the other has a beefed up PSU. No because all too often such a second device is simply not available for borrowing. However, I got lucky and at one point I had two Marantz amps sitting on my desk. Theoretically different models, 1152 DC integrated amps, rated at 2*76W/8. Both were made in March 1978, according to factory stamps inside. Both were untouched, as they came from the factory.
In the first round, both had their switches taken apart, thoroughly cleand and put back together, to try to eliminate them from the equation. Then, I changed all the caps in the preamp of one, to try to find out how much of it was down to the preamp. There was a change, but it was not breathtaking. Then the other with te untouched preamp had its power amp section brought up to speed with new caps, its original Elna 15,000uF/56V caps were replaced with BC Components 22,000uF/63V, and the change was almost amazing. Subsequent measurements showed that the original caps were down to 7,100/7,200 uF only, i.e. less than half the nominal value.
Not ony was noise reduced and details started to come up which were not there a while ago, but the noise over the tweeters (titanium domes good for -3 dB up to around 22 kHz) was just barely audible with my ear about an inch away.
Lastly, one amp had its preamp and amp recapped and was then compared to the other which had only its power amp recapped. The difference was much smaller than with a recapped power amp only, but while audible as still more details, better timbre and better spatial info, it was not breathtaking. I believe this was so because the preamp is fed off its own regulated power supplies and thus depends less on the state of the main caps. Obviously, the best overall results were obtained with a complete recapping.
As for measuring the state of your own caps, just try a simple test: use musical material you are intimately acquainted with and listen to it at your normal volume, then crank up the volume to see whether the sound changes. If it does, if it becomes harsher and more bland than at low volume, then you should consider changing the main filter caps at leat. They may be worn out, but they also may be too modest, and in these days of "savings", that's not at all a rare occurence, Consider changing them for both bigger and better quality caps - how big will depend on how powerful (or not) are your bridge rectifiers and power transformer.
Each of the above steps will bring its benefits, and put together, they may be quite significant and well worth your time and trouble. The best part is that you don't need any measuring gear to inform you of the changes, your ears are quite enough. Be mindful of the full power noise lisening, do not let it rip all at once, go in steps, jut in case.
Can you verify this for yourself? Actually, yes and no. Yes, because theoretically you could put two of the same power amps to work, where one is in full original state, and the other has a beefed up PSU. No because all too often such a second device is simply not available for borrowing. However, I got lucky and at one point I had two Marantz amps sitting on my desk. Theoretically different models, 1152 DC integrated amps, rated at 2*76W/8. Both were made in March 1978, according to factory stamps inside. Both were untouched, as they came from the factory.
In the first round, both had their switches taken apart, thoroughly cleand and put back together, to try to eliminate them from the equation. Then, I changed all the caps in the preamp of one, to try to find out how much of it was down to the preamp. There was a change, but it was not breathtaking. Then the other with te untouched preamp had its power amp section brought up to speed with new caps, its original Elna 15,000uF/56V caps were replaced with BC Components 22,000uF/63V, and the change was almost amazing. Subsequent measurements showed that the original caps were down to 7,100/7,200 uF only, i.e. less than half the nominal value.
Not ony was noise reduced and details started to come up which were not there a while ago, but the noise over the tweeters (titanium domes good for -3 dB up to around 22 kHz) was just barely audible with my ear about an inch away.
Lastly, one amp had its preamp and amp recapped and was then compared to the other which had only its power amp recapped. The difference was much smaller than with a recapped power amp only, but while audible as still more details, better timbre and better spatial info, it was not breathtaking. I believe this was so because the preamp is fed off its own regulated power supplies and thus depends less on the state of the main caps. Obviously, the best overall results were obtained with a complete recapping.
As for measuring the state of your own caps, just try a simple test: use musical material you are intimately acquainted with and listen to it at your normal volume, then crank up the volume to see whether the sound changes. If it does, if it becomes harsher and more bland than at low volume, then you should consider changing the main filter caps at leat. They may be worn out, but they also may be too modest, and in these days of "savings", that's not at all a rare occurence, Consider changing them for both bigger and better quality caps - how big will depend on how powerful (or not) are your bridge rectifiers and power transformer.
Each of the above steps will bring its benefits, and put together, they may be quite significant and well worth your time and trouble. The best part is that you don't need any measuring gear to inform you of the changes, your ears are quite enough. Be mindful of the full power noise lisening, do not let it rip all at once, go in steps, jut in case.
Dejan, indeed one's ears are completely adequate for determining acceptable, vs. ill behaviour, IMO - some may have trouble tuning into this as a means of assessing behaviour, but once the "skill" is "learnt" the signature of misbehaviour is as obvious as, well, oxidation in a supposedly good wine,
. With a little practice one can pinpint the precise volume setting where the effect "switches on" for a particular audio clip - if an amplifier is good enough that this defect never occurs the term used in the audiophile world is "effortless" - a synonym for "lack of disortion", ...
Last edited:
This is highly questionable, I had self-proclaimed golden ears here who were unable to tell when the tweeters were disconnected. I like this sort of tests, and some are afraid to come for a visit
Quite so, Frank.
When I get my Marantz 170DC to pump out peaks of just above 70W, nothing changes in its tonaity, you simply get the same but louder. By that time, my speakers are delivering about 106-108 dB SPL at 1 m, or 100+ dB SPL at 3 meters. Too loud to listen to for any longer period than say 30 minutes, at least in my case.
Anyone wanting masurements and proof is welcome to come visiting and should bring his own measuring gear with him. I guarantee a great lunch, and if need be, just as great a dinner as well.
When I get my Marantz 170DC to pump out peaks of just above 70W, nothing changes in its tonaity, you simply get the same but louder. By that time, my speakers are delivering about 106-108 dB SPL at 1 m, or 100+ dB SPL at 3 meters. Too loud to listen to for any longer period than say 30 minutes, at least in my case.
Anyone wanting masurements and proof is welcome to come visiting and should bring his own measuring gear with him. I guarantee a great lunch, and if need be, just as great a dinner as well.
This is highly questionable, I had self-proclaimed golden ears here who were unable to tell when the tweeters were disconnected. I like this sort of tests, and some are afraid to come for a visit
While I can't argue with that, Pavel, for myself, I don't need any other ears than my own to tell what I'm listening to and how it sounds to me. And since this is about my own system, uin my own room, my satisfaction is all that's required.
The man who doesn't recognize whether the tweeter is working or not is, in my view, with challenged hearing. Even older folks are expected to be able to hear up to about 10-12 kHz, by which time any tweeter in any loudspeaker box must be doing its thing or the sound will be impaired, sort of like standard AM at best.
It all depends upon what one is listening for - the focus of attention of the "golden ears" may be upon something you have little interest in, and vice versa. As an example, many here would try and assess smoothness of FR; whereas I'm listening for the telltale signs of low level distortion - the absence of the latter is what leads to long term listening satisfaction, which to me is far more important than having a system which is performing well in the conventional, technical sense.
That's fine, to me it is just an information how valuable are comments on sound of some members of our local forum. You know it is extremely difficult to make a view from comments on sound from people who communicate through virtual world only. The real life may be very different.
This would be good enough to do solo piano at realistic in-room levels - which if all is well means such recordings can be played continually without discomfort, or unease, at these levels.
The Yamaha keyboard, with only 10W amplifiers on each side, does this rather nicely on MIDI classical piano pieces, demonstrating what's possible if all is done well, engineered correctly, in a single package .
This is completely irrelevant, as irrelevant as the value of R in an RC filter. The value of C and its ESR determine noise.
Last edited:
Agreed completely.
All too often, if you dig deep enough, you find that the so-called golden ears are highly impressed by some companies and price tags. In that respect, I agree with the unsighted part of the blind tests, it should indeed be hidden from the panel preferably even what's lined up, let alone seeing them.
As you say, from forums you really know very little. However, from real life auditions, you can learn so much more. In my case, I have a friend whose hearing coincides with mine almost perfectly, let's say 99%, so mcuh so that on his word alone I would buy audio gear (and in fact, did - NEVER looked back).
If only you could come around, you could hear them for yourself, I don't think you'd need more than 30 minutes to understand, especially so if you brought your onw musical matrial. THAT'S how you would quickly and reliably know how I hear things.
Last edited:
Piano yes, but what you really want to hear is Bach on church organs - that's really mind blowing. Taxes the amp heavily, to be sure, but it's most convincing.
Ah, yes .. I have my Peter Hurford at the Sydney Opera House organ CD at the ready, to test all combatants, - this instrument is a noble beast; very full, rich tone ... and has knocked most pretenders out of the ring with ease, in the past. They usually struggle with generating the depth, and intricacy of the tonal structures - the soundscape heard is quite thin and 2 dimensional ....
- this instrument is a noble beast; very full, rich tone ... and has knocked most pretenders out of the ring with ease, in the past. They usually struggle with generating the depth, and intricacy of the tonal structures - the soundscape heard is quite thin and 2 dimensional ....
I specified it in my post. Where Ed gets his numbers and what they are referred to is a mystery.
So after much huffing, puffing, random number generation, and dirt-kicking, the answer remains: there is no evidence supporting the assertion that power line noise manifests itself in any way other than noise.
Nonsense. 20W into a resistive load at 120V RMS is 167mA RMS, but we do not have a resistive load - we have a DC power supply.simon7000 said:
No. The RMS charging current is typically three times the DC output current.
My method, based on your scenario, estimates the power sent back to the source resistance by the PSU. This power is dissipated in the source resistance - only a resistance can dissipate power so this is the only place the power can go.
What is "launch energy"? How did you get from 9mW to 3.6W - apart from multiplying by an arbitrary factor of 400?
I have learnt that on DIYaudio when someone claims to have 'measured' something this may not mean what it appears to mean. A number taken from a meter or a laptop screen is raw data. To turn this into a measurement you need a correct theory of how the experimental setup works.
What is "noise source impedance"?
The key is the noise distribution. The s/n is specified, as I mentioned, over 20-20k unweighted. Noise limitation is the input tube. So, you can draw the appropriate conclusion. For comparison purposes, the JC-1 measured the same way shows -73.3dB/ref 2.83V. The Ashly amps that Ed favors only spec s/n with no reference level given (http://www.ashly.com/products/data-sheets/klr-series.pdf)- if the reference level is full output, the noise performance is pretty mediocre. But who knows.
I still haven't seen a scintilla of evidence that there's any effect other than noise for power line noise- and that's just basic engineering to eliminate. Not as much fun for story-telling, admittedly.
I still haven't seen a scintilla of evidence that there's any effect other than noise for power line noise- and that's just basic engineering to eliminate. Not as much fun for story-telling, admittedly.
- Status
- Not open for further replies.