Actually, the peak currents are present even if the amplifier is drawing a constant current from the supply, as current is drawn only during the period when the filter capacitors are being charged.
Why do you think an additional 8% of sag in the supply rails would be inaudible? At any rate, it is at least a phenomenon that could (in theory) be measured, which is I think what you asked for before
Phil, would it be possible to measure peak and average currents in a simple linear supply for a few different current draws from the supply?
JohnR
Why do you think an additional 8% of sag in the supply rails would be inaudible? At any rate, it is at least a phenomenon that could (in theory) be measured, which is I think what you asked for before
Phil, would it be possible to measure peak and average currents in a simple linear supply for a few different current draws from the supply?
JohnR
Just in passing, I'd like to note that the reason that some of the giga-buck power cables are so flippin' stiff is that they're made with stuff like 10ga. single strand like you'd find in the walls. Wrap a braid around that for RF shielding, and some fancy jacket around that for cosmetic pruposes, and you've got a commercial product.
I'd think that it might be a useful idea to design a cable so that it functioned as a low pass filter (even if only a first order slope), rather than acting as a "flat response" wire. That way it would knock down incoming digital hash, RF, what-have-you. It would also perform the same function on "outgoing" waveforms, thereby reducing the effect that one piece would have on another.
Grey
I'd think that it might be a useful idea to design a cable so that it functioned as a low pass filter (even if only a first order slope), rather than acting as a "flat response" wire. That way it would knock down incoming digital hash, RF, what-have-you. It would also perform the same function on "outgoing" waveforms, thereby reducing the effect that one piece would have on another.
Grey
JohnR said:Actually, the peak currents are present even if the amplifier is drawing a constant current from the supply, as current is drawn only during the period when the filter capacitors are being charged.
Why do you think an additional 8% of sag in the supply rails would be inaudible? At any rate, it is at least a phenomenon that could (in theory) be measured, which is I think what you asked for before
Phil, would it be possible to measure peak and average currents in a simple linear supply for a few different current draws from the supply?
JohnR
That's correct John. I expect that voltage sag in the line voltage will definitely limit the amps headroom and could lead to clipping at high levels. I was planning on testing some amps I have in my possesion at the moment:
- Ancient 20 Watt Marantz receiver.
- 5 Channel, 100 Watt Sony HT amp.
- 2 Channel 400 Watt Crown PA amp.
- 2 Channel 800 Watt QSC PA amp (switch mode supply).
- 100 Watt Fender tube quitar amp (only tube amp I currently have sad to say).
Will post results after I collect data.
It will be interesting to see how much the line voltage saqs as I increase the load on the amps. I have some monster power resistors I have been dying to find a use for (two 400 W, 8 Ohm and one 750 Watt, 4 ohm baby that could warm a room).
Any objections to using a sine wave for testing. This makes it a lot easier to take repeatable, consistant measurements.
Phil
Sorry brain fart on my part, I missed John's earlier post.
I'm not sure I agree with the the peak current draw of 20 amps. Assume you have a resistive load drawing 240 watts from a 120 VAC, 60 Hz line voltage. This will draw 2 amps RMS which results in a peak current draw of 2.83 amp (at the top of the waveform).
The amplifier John referred to will only output it's rated 240 Watts during a short transient, otherwise our speakers would burst into flames and we would all go instantly deaf. Most of the time our amps aren't putting out much more than a watt or two (unless you work in a disco).
If your amp was putting out 240 Watts continously then you would expect to see significantly more than the 2.83 amps peak current dependant on the cap charging duty cycle. If you WAG (wild assed guess) a 30% charging duty cycle then this would result in about 8.5 amp peak current. We will need to actually measure some real amps to see how good that figure turns out to be in real life.
I have also decided to amend by earlier post. I will measure current draw two ways, first with a steady sine wave and then with a complex (music) input waveform. Any suggestions on the music choice? I was thinking about some trance music because it's so repetative.
Phil
Hi Phil, agreed, if the current waveform is a sinewave. But if I understand correctly the power supply does not draw current from the line in a sinewave, but only when the capacitors are charging.
Here's a shot from the duncanamps PSUD program. The red line is the load current, the green line is the current in the transformer secondary. It would be interesting to know how the waveform looks by the time it gets back to the line cord.
JohnR
Attachments
JohnR said:Why do you think an additional 8% of sag in the supply rails would be inaudible? At any rate, it is at least a phenomenon that could (in theory) be measured, which is I think what you asked for before
It can be measured, sure, but i don't quite know what this has to do with the "audiophile" mains cable issue... any cable has resistance and will cause a voltage drop of some degree; so, you'll loose output power with an unregulated supply. For that sake, turning a heater in your house will reduce power too, but the volume/power relation is logarhitmic; that's what i mean with it being quite inaudible.
current pulses, etc.
If you guys are that worried about high current pulses and what that does to your line, amplifier, etc (and these are real concerns) then you ought to be running choke input power supplies. To further guild the lily, use a power transformer withan electrostatic shield between the primary and the secondary. And then use a dedicated 60 hertz line at least back to the circuit breaker box. A properly designed choke input power supply looks like a resistor to the 60 hz line. (Also see the SOZ choke input filter question in the Pass labs forum.) Running bear: My soldering iron is getting hot. Gotta go build something. See you guys later.
If you guys are that worried about high current pulses and what that does to your line, amplifier, etc (and these are real concerns) then you ought to be running choke input power supplies. To further guild the lily, use a power transformer withan electrostatic shield between the primary and the secondary. And then use a dedicated 60 hertz line at least back to the circuit breaker box. A properly designed choke input power supply looks like a resistor to the 60 hz line. (Also see the SOZ choke input filter question in the Pass labs forum.) Running bear: My soldering iron is getting hot. Gotta go build something. See you guys later.
I don't see any mention of quantum physics yet... sounds like a good sales pitch (it is, isn't it?). Electron drift velocity is usually much slower than a few meters per second with a properly chosen cable, and stray capacitance has nothing to do with signal or current -- it is a function of the area between the two conductors and the dielectric constant of the insulation separating them.
Joseph Cohen said:Here's something I wrote for a line of cables I was at one time importing:
"Correctly designed cable is the primary and most critical means of insuring the proper delivery of the audio signal from source to loudspeakers. The general misconception is that audio cables are like hoses with the signal traveling inside them. In fact the flow of electrons within conductors is measured at a few meters per second, while the signal itself travels on the energy of the electromagnetic force surrounding the conductors at the speed of light. For this signal flow to be perfect and unimpeded it is necessary that the electromagnetic fields surrounding the conductors retain their symmetry, which is distorted by EMI, RFI and high frequency digital pulses and especially by stray capacitance which arises as a result of signal and current that runs on the shield and throughout all the cable's materials. Stray capacitance is the single greatest cause of coloration and distortion in audio cables."
So from my point of view we are already into quantum physics when we attempt to describe the forces at work in the behavior of signal transmission. This goes way beyond capacitance, inductance and resistance and into areas that are mostly beyond the capabilities of our most sophisticated measuring instruments.
Reality bites
If one is just talking about the 2 meters of powercord, this is an insane debate. Only at the extreme edge (powersupply for your powersuppy, direct connect to the streetmains, filterbanks, etc.) is this theory every going to come into play. One would be much better off chuncking all of that money into the best damn powersupply one could build.
BUT to take it to the absurd. (he he )... Many power companys are trying to put broadband down our electric wires. In the case of the switching powersupplies, will this little amount of high(er) freq. noise in the mHz. interact with those devices? It has to be at a higher level than the current RF noise that might be in the wire...
If one is just talking about the 2 meters of powercord, this is an insane debate. Only at the extreme edge (powersupply for your powersuppy, direct connect to the streetmains, filterbanks, etc.) is this theory every going to come into play. One would be much better off chuncking all of that money into the best damn powersupply one could build.
BUT to take it to the absurd. (he he
)... Many power companys are trying to put broadband down our electric wires. In the case of the switching powersupplies, will this little amount of high(er) freq. noise in the mHz. interact with those devices? It has to be at a higher level than the current RF noise that might be in the wire...- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.