Did-you mean "live music" that always goes across those bad sounding PA systems ?
Even "live" records are engineered and produced in a way to be as close as possible to the "studio sound".
I believe that if some tend to prefer live recordings, it is just because the feeling and the groove are different. For various reasons, like the adrenalyn and ambiance of the audience, the fact that the musicians all play (and interact) together in the same time, etc.
A fellow is driving along when his tire hits something and goes flat. He pulls over to the side of the road, gets out the spare tire and tools. He takes off the hub cap, loosens the nuts and then jacks up the car. Next he removes the lug nuts carefully placing them in the hub cap. As he removed the flat tie it bounces off the hub cap flipping the nuts out and they fall down a drain he hadn't noticed. Looking around he sees a fenced in crowd behind him. He realizes he is parked in front of the state mental institution.
One of the inmates looks at him and says "Simple take one lug nut off of each of the other three wheels and use them to get to where you can get replacements." The driver looks amazed. The inmate seeing the surprise then says " We're crazy not stupid."
So try not to confuse the issues between clever and IQ.
One of the inmates looks at him and says "Simple take one lug nut off of each of the other three wheels and use them to get to where you can get replacements." The driver looks amazed. The inmate seeing the surprise then says " We're crazy not stupid."
So try not to confuse the issues between clever and IQ.
This is more like it. When someone says they "hear" an added .01 in thier circuitry.... some take the part and lay it on the table and test it for non-linearity. That is borderline, mental.
IMO
THx-RNMarsh
Speculating without evidence that what they hear has anything to do with linearity or DA is equally so.
Then again, if we're casting aspersions, bypassing a large electrolytic with a tiny capacitor then parking a long, inherently inductive wire after it and claiming it affects circuit function is pretty mental, too. Yes it's best to measure and Chris threw a link to Gerhard's old test of bypass, but it's also extremely simple to model in SPICE to get a feel for the behaviour.
So I built a very nice phono preamp which had provision for small polystyrene "trim" caps of a few tens of pF to fine tune the HF response, depending on the actual value of a larger EQ cap (about 10% change). I built it without the trim cap and listened for a couple of weeks. At that time I was using a Sumiko MC cartridge which I think had a somewhat hot top end. After a while I felt like on some records it was too bright so decided to add the PS cap. I soldered in the cap and listened for a while, and noted the smoother top end and felt a bit smug. Afer several records I wasn't so sure and some doubt crept in, then I had a start and wondered where I had soldered the cap. I popped the top off the unit and looked. Sure enough, I had soldered it in parallel with a much larger inter-stage couplings cap rather than the EQ cap, where it could not have had any effect at all. Of course, I "heard" the smoother HF, because that was what I expected, at least for a while.
Now that was mental!
Richard Marsh has a long experience with cap differences. Scott may not think that it matters much, but I agree with Richard Marsh on the importance of cap differences.
Our experience may go back 40-50 years, when we as circuit designers first had to select what caps we preferred to use for both coupling caps, bypass caps, and especially, EQ caps, like the RIAA equalization.
In each of these cases, we have a different set of needs, and these must be addressed.
For example, coupling caps have to be a large enough value to couple to the outside world, typically, and therefore NPO ceramic, or Teflon, just don't come in values large enough to do the job, without great cost and size. Of course, the military or a research lab might be able to afford them, but not mere mortals. Therefore we look for compromise caps that are lower cost/uF and more compact. The first thing that comes to mind is electrolytic caps, and they served a useful function especially when the caps were biased up with some fixed DC voltage. However, when bipolar power supplies became popular in audio circuits, the DC bias was reduced to almost zero. This actually puts a polar electrolytic cap into some compromise, and if it is Tantalum, will actually add distortion to the circuit. DA will also increase with 0V on Tantalum caps, compared to a DC biased one.
Now what about aluminum caps? They do not add as much distortion, but the have plenty of DA, up to 10%, the way we sometimes measure it, and that is just a single cap. What if you cap couple throughout a preamp? How much DA then? This is when, about 40 years ago, we tried to find acceptable coupling caps with polypropylene, Mylar, or polycarbonate (that were popular in the past). Now these caps are not perfect, but they sure beat electrolytic caps, in every measurable way, and we found that they sounded better in listening tests too! But first Richard, and then I found that direct coupling was even better and reduced the circuit 'footprint' as well, and even the cost, compared to using the best coupling caps available. This is why we design this way.
We are not trying to make 'effects generators' so removing high DA caps will help us make a more accurate audio stage. I believe that multiple hi DA caps can actually make an 'effects box' if one is inclined to want one. I was actually going to build one, once with a number of especially lousy electrolytic caps in series.
Now I don't have room here to go on about cap self resonance, ceramic non-linearity, etc that are equally important, but we have discussed these things before as well. We should discuss these factors again, because they seem lost to the majority of active contributors here.
Our experience may go back 40-50 years, when we as circuit designers first had to select what caps we preferred to use for both coupling caps, bypass caps, and especially, EQ caps, like the RIAA equalization.
In each of these cases, we have a different set of needs, and these must be addressed.
For example, coupling caps have to be a large enough value to couple to the outside world, typically, and therefore NPO ceramic, or Teflon, just don't come in values large enough to do the job, without great cost and size. Of course, the military or a research lab might be able to afford them, but not mere mortals. Therefore we look for compromise caps that are lower cost/uF and more compact. The first thing that comes to mind is electrolytic caps, and they served a useful function especially when the caps were biased up with some fixed DC voltage. However, when bipolar power supplies became popular in audio circuits, the DC bias was reduced to almost zero. This actually puts a polar electrolytic cap into some compromise, and if it is Tantalum, will actually add distortion to the circuit. DA will also increase with 0V on Tantalum caps, compared to a DC biased one.
Now what about aluminum caps? They do not add as much distortion, but the have plenty of DA, up to 10%, the way we sometimes measure it, and that is just a single cap. What if you cap couple throughout a preamp? How much DA then? This is when, about 40 years ago, we tried to find acceptable coupling caps with polypropylene, Mylar, or polycarbonate (that were popular in the past). Now these caps are not perfect, but they sure beat electrolytic caps, in every measurable way, and we found that they sounded better in listening tests too! But first Richard, and then I found that direct coupling was even better and reduced the circuit 'footprint' as well, and even the cost, compared to using the best coupling caps available. This is why we design this way.
We are not trying to make 'effects generators' so removing high DA caps will help us make a more accurate audio stage. I believe that multiple hi DA caps can actually make an 'effects box' if one is inclined to want one. I was actually going to build one, once with a number of especially lousy electrolytic caps in series.
Now I don't have room here to go on about cap self resonance, ceramic non-linearity, etc that are equally important, but we have discussed these things before as well. We should discuss these factors again, because they seem lost to the majority of active contributors here.
Last edited:
Certainly possible for you and maybe someone else.
My point is, measuring the part all by itself without the context of the topology and layout and circuitry values etc and without knowing the parasitic effects in that circuit, the system I/O et al...... taking one part out of that context and measuring it alone and making sweeping proclamations from that DUT, is not too cool. IMHO. As always.
THx-RNMarsh
You guys keep looking at the 0.01 uf cap as a noise shunt. Try looking at it as a very fast battery on your raw and regulated DC lines.
Which only works when that 0.01 is sitting on the power pins of the relevant load. Otherwise your very fast battery has to go through the quicksand of wire inductance.
... unless it marries the electrolytic next door and gives birth to a
nice parallel resonance, which means no decoupling at all at
some frequencies. Now, that might be audible. And, the better
the small cap, Q-wise, the more pronounced the resonance.
nice parallel resonance, which means no decoupling at all at
some frequencies. Now, that might be audible. And, the better
the small cap, Q-wise, the more pronounced the resonance.
I got interested in this phenomenon last month, when I studied Figure 16.7 in Bob Cordell's power amp book. Bob recommends (a) learning the ESL and ESR of both capacitors; (b) calculating the self resonant frequency and Q of the parallel resonant circuit {ESL of big cap // C of small cap}; (c) installing a damping resistor to lower the Q of this resonant circuit, to an acceptable value that prevents ringing. Such as Q=0.5.
Lots of inexpensive test equipment will measure ESR, but it takes a Network Analyzer to measure ESL. I didn't want to spend that kind of money (> $10K used) so I built my own custom test fixture that measures ESL for about fifty bucks. It's documented (here). I would be glad to measure the ESL of capacitors that readers mail to me; please include return postage if you want me to mail it back.
Mark Johnson
Anti-intellectualism - Wikipedia
Did an engineer shoot your dog or steal your lunch money?
It is more than a little ironic that someone who "designs" and sells electronics is so proud of their lack of knowledge of said discipline.
That Wikipedia article could use some corrections. The description of Confirmation Bias appears quite biased itself. Casts doubt on the on the quality of the rest of the article.
See what happens to phase also.... where it starts deviating from 90 degrees on either side of resonance. I own several network analyzers and precision LCR instruments as well as impedance analyzer. If you would like, I can compare your data with mine. Or maybe measure some parameter you cant.... in circuit or out.
THx-RNMarsh
Last edited:
Enough damping to prevent ringing may suffice to correct the worse problems but may result in less total attenuation of noise than originally intended or perhaps less ideal voltage source approximation of locally stored energy source. Perhaps more to consider than just damping?
- Status
- Not open for further replies.