ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 ... 6 7 [8] 9 10   Go Down

Author Topic: Danley DNA SC48/20K4 Pro Processor/Amp  (Read 39107 times)

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 9205
  • Atlanta GA
Re: Danley DNA SC48/20K4 Pro Processor/Amp
« Reply #70 on: December 17, 2015, 07:21:35 AM »

A neat response that misses the point completely. Maybe you should read Art's original question that I was referencing.

And if you are affiliated with a manufacturer maybe you should follow the posting rules and include that fact in a sig line.

Mac
I guess another way to look at it is this.

The output power ratings would still be the same if as 12dB crest factor signal was used.

The limit is the power supply VOLTAGE rails.  You cannot exceed them.

But what the lower crest factor  numbers show is that into higher impedance loads, the amp can produce a higher "average power", (which is not shown in the simple "power" rating) due to greater current and thermal capacity.

Once again, this is not something that is easily shown with a "simple number" answer.

And just to make it even more "it depends", consider that loudspeakers are NOT resistive loads, and have peaks in the impedance-which results is a lighter load (higher impedance), so the "average impedance load" is not as hard on the amp as a resistive load-UNTIL you start to factor in the phase of the impedance-but now we are talking about a whole different can of worms.
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Peter Morris

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1375
Re: Danley DNA SC48/20K4 Pro Processor/Amp
« Reply #71 on: December 17, 2015, 08:12:14 AM »

I guess another way to look at it is this.

The output power ratings would still be the same if as 12dB crest factor signal was used.

The limit is the power supply VOLTAGE rails.  You cannot exceed them.

But what the lower crest factor  numbers show is that into higher impedance loads, the amp can produce a higher "average power", (which is not shown in the simple "power" rating) due to greater current and thermal capacity.

Once again, this is not something that is easily shown with a "simple number" answer.

And just to make it even more "it depends", consider that loudspeakers are NOT resistive loads, and have peaks in the impedance-which results is a lighter load (higher impedance), so the "average impedance load" is not as hard on the amp as a resistive load-UNTIL you start to factor in the phase of the impedance-but now we are talking about a whole different can of worms.

BUT not if you use a 3 dB crest factor!  If you do that your 2 ohm output will be more like your 8 ohm output of 1300 watts not 5000 watts.

FWIW I think 1/8 power (12d crest) is a good average of what you find in the field - there are exceptions.

This is what happens with this type of design -
« Last Edit: December 17, 2015, 08:18:01 AM by Peter Morris »
Logged

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 16828
  • Hickory, Mississippi, USA
    • Resotune
Re: Danley DNA SC48/20K4 Pro Processor/Amp
« Reply #72 on: December 17, 2015, 10:29:06 AM »

Lies, damn lies and specifications.  8) (Sorry I know Ivan is the last honest man in the sound business.  ;D ).

From observation of Peter's (similar?) power vs. time vs. impedance graph we can see that cold power is voltage limited (so low Z loads make more short term power) while long term the power is thermal/power supply limited so same for different impedances (no doubt limited by smart protection circuits).

Presenting the different crest factors for different impedances is because that actually characterizes how the amp behaves under different loading. For those who want a single worst case answer use the 8 ohm number.

I don't think this advances us beyond the old advice to ask actual users how these work for them in similar applications, with similar loads.

How many people actually know what the duty cycle or crest factor of their particular application demands? Back last century when my day job was pimping power amps I wanted to put my patented peak/VU (ave) meter display on our power amps. With a simultaneous peak (dot) and average (bar) meter display calibrated in even dB steps (I used 3 dB steps per LED) you could literally read crest factor in real time by observing the space between peak and average readings.  8)

Of course I was shot down because it was too expensive of a feature to put on a value power amp for the very price sensitive Peavey distribution. I didn't really expect my customers to understand crest factor (not sure how many here understand it).  I did expect customers to like all the blinky lights... 8)

Hopefully in the real world we operate these amps somewhere between the cold (high peak output) and hot (power limited). So the specs don't really tell us what we will actually get out in the world.

As i have offered many times before this is complicated and I will continue to advocate for smart powered speakers or DSP presets dialed in by the factory (to protect the speakers). The end users should not have to understand and deal with this minutiae IMO.


JR

PS: Welcome to the party Mr Smalley, thanks for participating on this forum. 

Logged
Cancel the "cancel culture". Do not participate in mob hatred.

Merlijn van Veen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 309
    • www.merlijnvanveen.nl
Re: Danley DNA SC48/20K4 Pro Processor/Amp
« Reply #73 on: December 17, 2015, 10:52:34 AM »


Its also difficult one for manufacturers when theres no agreed format and we all quote and are seduced by a big headline figure. I should know better but I still do it!

I think the CAF approach is admirable. Makes you wonder why it's not being embraced like the CLF.

Then again standardization in 2015 still appears to be undesirable or at least extremely difficult to achieve, think Bennett Prescott's DSP Tower of Babel or digital audio network formats.

Well at least we've agreed on polarity ;-)


Verzonden vanaf mijn iPad met Tapatalk

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 16828
  • Hickory, Mississippi, USA
    • Resotune
Re: Danley DNA SC48/20K4 Pro Processor/Amp
« Reply #74 on: December 17, 2015, 12:44:54 PM »

I think the CAF approach is admirable. Makes you wonder why it's not being embraced like the CLF.

Then again standardization in 2015 still appears to be undesirable or at least extremely difficult to achieve, think Bennett Prescott's DSP Tower of Babel or digital audio network formats.

Well at least we've agreed on polarity ;-)


Verzonden vanaf mijn iPad met Tapatalk
I submit that the problem is not (lack of) standardization for amp power specifications, but complexity (too many variables) and lack of consistency between customer applications. The elephant in the room is coming up with meaningful specs that customers can relate to their real world applications with simple quantifiable numbers, they can use for design/purchase decisions. 

I tried for years (decades) to come up with meaningful numbers and never could.

Simultaneously amplifier technology keeps advancing delivering more value for less cost than ever. The free market (and forums like this) helps consumers determine what amps work and what don't for their applications.

For those of us who understand what these metrics mean, it is wishful thinking to expect that A) all customers need is a little more education and/or B) all we need is a simpler specification***. 

This stuff is complicated and I continue to prefer to let the design engineers wrestle with this.

JR

**** as a design engineer I like numbers but as I've argued before coming up with some arbitrary duty cycle will likely turn into a design/marketing target that could take on a life of it's own (like damping factor, slew rate, etc). I prefer the current "this amp works" or "this amp doesn't work" market feedback from end users.
Logged
Cancel the "cancel culture". Do not participate in mob hatred.

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 9205
  • Atlanta GA
Re: Danley DNA SC48/20K4 Pro Processor/Amp
« Reply #75 on: December 19, 2015, 08:38:27 AM »

Something hit me this morning while lying in bed.

This is a classic case of "to much information".

If the spec sheet had said 12dB crest factor on all the rated wattages-then this discussion would not exist.

The wattage would not go up on the 8 or 4 ohm loads-because the limiting factor is the VOLTAGE rails.

But because it was shown that additional CURRENT is available (with the lower crest factor signals into those higher impedance loads) it has caused confusion and misunderstanding.

I wonder if it is because of reasons like this that some of the larger manufacturers are simply choosing to leave of basic specs in their spec sheets-because it "confuses" people.

Many people like "nice simple numbers" that they believe are true.  So why complicate it with any real facts???????

It is also interesting to me how many users never even do any basic testing to see if the specs are actually true or not.  Especially since there are no "agencies" that govern the specs.

The nice thing about standards is that there are so many to choose from----------------

But maybe that is just me.
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

Peter Morris

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1375
Re: Danley DNA SC48/20K4 Pro Processor/Amp
« Reply #76 on: December 19, 2015, 09:27:15 AM »

Something hit me this morning while lying in bed.

This is a classic case of "to much information".

If the spec sheet had said 12dB crest factor on all the rated wattages-then this discussion would not exist.

The wattage would not go up on the 8 or 4 ohm loads-because the limiting factor is the VOLTAGE rails.

But because it was shown that additional CURRENT is available (with the lower crest factor signals into those higher impedance loads) it has caused confusion and misunderstanding.

I wonder if it is because of reasons like this that some of the larger manufacturers are simply choosing to leave of basic specs in their spec sheets-because it "confuses" people.

Many people like "nice simple numbers" that they believe are true.  So why complicate it with any real facts???????

It is also interesting to me how many users never even do any basic testing to see if the specs are actually true or not.  Especially since there are no "agencies" that govern the specs.

The nice thing about standards is that there are so many to choose from----------------

But maybe that is just me.

is that so  :o
Logged

David Sturzenbecher

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1813
  • So. Dak.
    • Sturz Audio
Re: Danley DNA SC48/20K4 Pro Processor/Amp
« Reply #77 on: December 19, 2015, 09:30:38 AM »


It is also interesting to me how many users never even do any basic testing to see if the specs are actually true or not.  Especially since there are no "agencies" that govern the specs.


It is interesting..... ::)

How about actually answering Art's question? What would the results be if the same crest factor was used at each impedance? Why did you chose to change the crest factor for your measurement, when the music wouldn't be any different at different impedances?

Mac
To be honest-I have not measured the amps in that way.

Logged
Audio Systems Design Engineer
Daktronics, Inc.
CTS-D, CTS-I
AES Full Member

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 16828
  • Hickory, Mississippi, USA
    • Resotune
Re: Danley DNA SC48/20K4 Pro Processor/Amp
« Reply #78 on: December 19, 2015, 09:44:57 AM »

Something hit me this morning while lying in bed.
You should have ducked.   8)
Quote
This is a classic case of "to much information".

If the spec sheet had said 12dB crest factor on all the rated wattages-then this discussion would not exist.
As already shared, the characteristic short term peak vs long term average power output varies as shown in the specification with load impedance. Normalizing to test the amp for all three loads using the same crest factor will only reveal optimal performance for one load and understate the other two.. 
Quote
The wattage would not go up on the 8 or 4 ohm loads-because the limiting factor is the VOLTAGE rails.

But because it was shown that additional CURRENT is available (with the lower crest factor signals into those higher impedance loads) it has caused confusion and misunderstanding.

I wonder if it is because of reasons like this that some of the larger manufacturers are simply choosing to leave of basic specs in their spec sheets-because it "confuses" people.

Many people like "nice simple numbers" that they believe are true.  So why complicate it with any real facts???????
To quote Jack Nicholson,,, "You can't handle the truth."

It's the marketers that like simple numbers they can use to try to differentiate their products in the marketplace. These "simple" numbers are often misused (slew rate, DF, etc).

Customers like numbers that they think they understand (more or less is better). Trying to actually educate customers is often a thankless job, left to marketers who are less interested in education than the selling opportunity. 
Quote
It is also interesting to me how many users never even do any basic testing to see if the specs are actually true or not.  Especially since there are no "agencies" that govern the specs.
Now you're scaring me a little... Performing these tests will be harder than static power measurements that customers rarely get correct. The higher the power, the harder to control sundry variables.
Quote
The nice thing about standards is that there are so many to choose from----------------

But maybe that is just me.

Sounds like a personal problem. As a design engineer I like numbers and use them all the time. For this specific characteristic I see simple single number specs adding more confusion and less clarity. Not only does any display graph require a time axis in addition to the spot power output, but even that chart only shows one aspect (max power) not how it performs dynamically when playing program. While that will generally fall somewhere between the extremes. 

JR

PS: Merry Christmas Ivan, how are the lights this year?
Logged
Cancel the "cancel culture". Do not participate in mob hatred.

radulescu_paul_mircea

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 75
Re: Danley DNA SC48/20K4 Pro Processor/Amp
« Reply #79 on: December 19, 2015, 10:28:32 AM »

I'm not advocating for "eight days a week " continuous power, although I am fond of a good DC-300. I simply felt the need to point out these are not amoebus things we're talking about, but concrete units.

My point was that rated power equivalent energery should be available for one second. If you can't apply that energy for one second (or double it for half a second- or equivalent fractions thereof- there's a game for you), then you can call your unit a W, but it is not an SI Watt. 

Same for multi-channel chassis. So, are you posing the argument that because continuous power is irrelevant, we should ignore conservation of energy?

In any given second 2400 joules are available on an American 20 amp circuit. You can borrow excess energy from previous timeframes but capacitors have finite capacities. If you can design the PSU such that energy equivalent to cumulative output ratings (divided by efficiency) is applicable for one second, then you're talking Watts, and not What's? (Whats?: My term for an arbitrary definition of energy over time)

I don't dive near piers. Too many ships full of shit from Shenzen.
Look,  Watt=J/s, but when you state a value for time ( 1000 watts for one hour= 1000J/s*3600s=1kWh or 3600000 Joules), you finally get only joules, because time cancel each other. If you have a laser, you have to say what is the power (in watts) and what it is it's average power (in Wh) and here comes the confusion. An amplifier generates musical signals (like the laser impulses) with high peak-average ratios and everything inside it is made to accept high bursts of power,up to 15000kW like M-drive, but it gets hot very quick so it has to have time to get cooler, so it has an average power of 5000 W. BUT, if you feed it only with sine wave continous power, it will limit itself to 2000W because only at that power rate it will noy get burned.
The 20K DSL can generate the stated power for the stated time with the stated crest factor. At 8 Ohms it will not get that hot so it will need less time to cool (6dB crest factor). But as you go lower impedance,  it will give you more power but it will get hotter.
So, the total amp has a PS that can sustain high power peaks, but it can only sustain a limited average power long term.
One thing we should remember is that any power amp has a thermal inertia. This plus the overall efficiency will affect how much power it can sustain in lower impedances
Logged

ProSoundWeb Community

Re: Danley DNA SC48/20K4 Pro Processor/Amp
« Reply #79 on: December 19, 2015, 10:28:32 AM »


Pages: 1 ... 6 7 [8] 9 10   Go Up
 



Site Hosted By Ashdown Technologies, Inc.

Page created in 0.029 seconds with 17 queries.