|
|
Post by Casey Leedom on Jan 28, 2017 14:58:14 GMT -5
Negative Feedback Loops generally do have hysteresis in them in order to prevent the Control from rapidly oscillating between States. For instance, in your Home Thermostat, if you set it for {x}˚F, the Furnace will generally not trigger until the Thermostat detects ({x}-Δlow)˚F and won't turn off till it sees ({x}+Δhigh)˚F. This prevents the Furnace from rapidly cycling on and off, which would decrease its efficiency and shorten it's life span. This hysteresis in a Control System determines it's Frequency Response with respect to changes in the measured difference between the actual state of the Operational System and the Control Directed State of the Operational System (the "Negative" part in the Control System).
For something in the Audio domain, as with Switching Power Supplies, you'd want the Frequency Response of the Negative Feedback Control System to be very high — far above human hearing. For instance, in the Krell iBias amplifiers, they claim that the Bias is raised within microseconds of measuring a High Current Draw, but only lowered after a sustained period of Low Current Draw.
Now I'll go and read the paper you referenced. Thanks for the pointer!
Casey
|
|
|
|
Post by leonski on Jan 28, 2017 18:31:53 GMT -5
Yes, about temp control. The last system I worked on used in semiconductor manufacture had above and below band setpoints. If -8c a large correction was applied but decreased as temp approaced setpoint. As temp exceeded setpoint, power was again reduced reaching zero the max listed in the table. Various additional factors like RATE and RESET along with Some other stuff I forget was all factored in. When I set a furnace up, I set the UPPER limit about 3 c above setpoint to rapidly turn power OFF. I set the lower setupoint at about 4 or 5 c, depending on the temp to be controlled to. For some things, like oxidation, which could run at 1150c or so, 3c worked well, since the system lost energy so rapidly due to heat loss. But at lower setpoints, like processes that ran at much lower temps, and were sometimes EXOTHERMIC reactions, I had to make other changes. For a while I had a computer plugged in with special, dedicated software so I could do realtime changes and 'fine-tune' the system. With the furnaces I have in mind, they are not simply 'on / off' bur rather fully proportional control with SCR firing circuits to the heater elements. And transformers the size of an End Table. |
|
|
|
Post by Gary Cook on Jan 28, 2017 18:36:04 GMT -5
I'll be curious to see how many years it'll take for the SMPS myth to finally be accepted as busted. About 20 years, that's how long I keep my power amplifiers.  Cheers Gary |
|
|
|
Post by Casey Leedom on Jan 29, 2017 0:24:18 GMT -5
|
|
|
|
Post by Casey Leedom on Jan 29, 2017 0:53:58 GMT -5
|
|
|
|
Post by leonski on Jan 29, 2017 4:10:51 GMT -5
Oh! Yes! Touch the 3rd rail an fry.
To some, it is EVIL while others say that it is OK at times in certain apps, but generally NOT global in nature. That sort of thing.
Still Others won't go Near it.
Pass has good ears And is a wonderful designer and engineer. I'll reread his article later.
Bruno P. is a BIG fan of NFB and uses a large amount in his nCore 'D' amps. Go Figger.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jan 29, 2017 16:40:43 GMT -5
As someone mentioned above, power supply design is a science in itself, so true. Digital changed the world.
I had questions of Emo using the new technology could produce the power they claimed until I did some reading.
Any switching PS that can supply 800 watts RMS per channel; 20 Hz - 20 kHz; THD < 0.5%; into 2 Ohms, I wouldn't worry about it lacking.
Their specs are impressive, especially in real world applications where no one is going to drive that much power in a HT.
So going back to thxultra's question- get the Emo & don't look back.
Well, I do know my XPS-5 Gen 2 has been "on" since 7:00 am "all channel" mode and it's much cooler than the old Denon AVR5700, 2 ch, that was on for 2 hrs in my office at low volume. I have yet to notice a time when the XPA got more than warm. Heat is wasted energy.
BasX A5125 amp? hmmm, hope they use the same switching PS technology. It may end up on my shopping list for effect channels in the new house.
If reducing the poundage of equipment & reducing the chance of a hernia, I'm all for it. lol
|
|
|
|
Post by leonski on Jan 30, 2017 0:25:53 GMT -5
Texzick
You are mixing up 2 major things:
Heat and Temperature.
Let me ask a question? What has more Heat? An Iceberg or a Red Hot needle? Between those 2, what has the higher temperature?
The amount of energy is basically a factor of temperature AND mass. Given that a speaker is <1% efficient, the Remainder is Heat.
The Other thing to consider is that the Denon is a poor amp. Probably not made to even consider a set of 4 ohm speakers, it might even have a backpanel switch to protect the PS when running
low impedance speakers. The switch might be labeled '6ohms' or some such. It will have marginal heat sinking an No Way will it even run 1/3 of its RMS rating into 5 or 7 real speakers. Probably not
even 10% of RMS with 10 db peaks, like real audio will have.
And finally, the number you NEVER hear would be the efficiency of an amp at say 1/2 power, from wall plug to speaker. 'D' amps kill at that measure. And If you have ever seen a 'D' amp module, they
have very little in the way of heat sink.
|
|
|
|
Post by thxultra on Jan 30, 2017 9:54:23 GMT -5
As someone mentioned above, power supply design is a science in itself, so true. Digital changed the world. I had questions of Emo using the new technology could produce the power they claimed until I did some reading. Any switching PS that can supply 800 watts RMS per channel; 20 Hz - 20 kHz; THD < 0.5%; into 2 Ohms, I wouldn't worry about it lacking. Their specs are impressive, especially in real world applications where no one is going to drive that much power in a HT. So going back to thxultra's question- get the Emo & don't look back. Well, I do know my XPS-5 Gen 2 has been "on" since 7:00 am "all channel" mode and it's much cooler than the old Denon AVR5700, 2 ch, that was on for 2 hrs in my office at low volume. I have yet to notice a time when the XPA got more than warm. Heat is wasted energy. BasX A5125 amp? hmmm, hope they use the same switching PS technology. It may end up on my shopping list for effect channels in the new house. If reducing the poundage of equipment & reducing the chance of a hernia, I'm all for it. lol Ended up getting a XPA-2 and a XPA-5 second hand. So far they sound amazing. Really amazed at the amount of power they have can't see why I would ever need more. I had it cranked over the weekend and think I maybe got 3 lights to show on the meters on the XPA-2. Pretty impressive I must say. Very clean sound as well a huge improvement over my old QSC and Crown amps. One of these days I will need to replace my hight channels amps as well will be going emotiva for those as well. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jan 30, 2017 14:20:39 GMT -5
Texzick You are mixing up 2 major things: Heat and Temperature. Let me ask a question? What has more Heat? An Iceberg or a Red Hot needle? Between those 2, what has the higher temperature? The amount of energy is basically a factor of temperature AND mass. Given that a speaker is <1% efficient, the Remainder is Heat. The Other thing to consider is that the Denon is a poor amp. Probably not made to even consider a set of 4 ohm speakers, it might even have a backpanel switch to protect the PS when running low impedance speakers. The switch might be labeled '6ohms' or some such. It will have marginal heat sinking an No Way will it even run 1/3 of its RMS rating into 5 or 7 real speakers. Probably not even 10% of RMS with 10 db peaks, like real audio will have. And finally, the number you NEVER hear would be the efficiency of an amp at say 1/2 power, from wall plug to speaker. 'D' amps kill at that measure. And If you have ever seen a 'D' amp module, they have very little in the way of heat sink. What!? I really don't need an audio design education class, been there, done that 40 yrs ago. Why are you trying to correct me on what you think I said and didn't? Not impressed on your diatribe. Save it for someone who cares. |
|
|
|
Post by leonski on Jan 30, 2017 14:32:44 GMT -5
WTF? Just the reason that as a designer you should already know why your Denon runs so hot and the better built XPA runs cooler.
They might be (not likely) using the same amount of power from the wall.
I hardly think what I wrote rises to the level of 'a forceful and bitter personal attack', do you, really?
Though I must admit that diatribe IS a real cool word and difficult to find opportunity to properly use.
You Could get your XPA hot, but that would require a pretty loud afternoon. Or adjust the Bias, Mr. Designer, to a high value.
And yes, getting rid of the large power supply Toroid is of benefit. Next up? Heatsinks which are both heavy and bulky.
Shipping difficulty and costs seem to be driving this, not necessarily sheer sound quality.
If you already know the difference between Heat and Temperature, somebody else may find benefit in what I wrote.
|
|
