ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 2 3 [4]   Go Down

Author Topic: Crest Pro-Lite Amplifiers, feeling confused  (Read 29880 times)

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 17176
  • Hickory, Mississippi, USA
    • Resotune
Re: Crest Pro-Lite Amplifiers, feeling confused
« Reply #30 on: July 30, 2013, 03:43:18 PM »

I'd be curious as to how well the different amps behave into a reactive load where voltage vs. current is out of phase. I've read that amps which can produce robust current when the voltage is near the zero crossing are the ones that can deliver. All others are suspect.
This is more of an issue for old school (analog) power amps. The worst case for dissipation in analog amplifier output stages is when output voltage is zero, because voltage across the devices is the full PS rail. Back in ugly old days when power devices were weak and men were strong, we used VI limiting to protect devices from too much current at too low output voltage. It was not uncommon to add some fold-back to current limiting also, so an amp would often put out a lot less current into a dead short circuit to ground, than into a larger valid signal voltage. I suspect some too-clever designs could get tripped up by speakers with too much load angle. Some unexplained audible artifacts could be caused by premature VI limiting (a form of current limiting vs voltage). These artifacts could appear random, while they really weren't.   

Class D saturated output power devices don't suffer from the same secondary breakdown  mechanisms as linear pass elements in analog amps. In class D amps they are switched hard-on or hard-off. Note: Class D is technically analog so not really digital, just saturated switch outputs.

I haven't looked into this for years, IIRC from last time I did, loudspeakers are (were) mostly resistive, with some reactive funny business at extremes. Odd, passive crossovers can add some more fun to the mix.

I recall one amp designer speculating about using class D amps to drive electrostatic loud speakers (capacitive load) as the class D would not mind the reactive load. I am not smart enough to evaluate, but the absence of resistive voice coil heat dissipation losses in electrostatic speakers may have some utility.... or not when combined with an amp that tolerates driving pure reactance. That said I don't recall electrostatics being very efficient, but they were a huge PIA to drive.   

I suspect a too-reactive load could interfere with class D output filter design that expects a nominal (resistive) load behavior, not sure exactly how this would express.

Caveat...  I am not very close and personal with the new generation of class D amps, but I suspect the physics is the same, just using faster-bigger power devices with more of the front end overhead reduced to ICs. I like QSC's realization that they can combine class D amp outputs together, after the output filters that tend to protect them from driving back into each other's output (within reason).

JR

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

Greg_Cameron

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 675
    • Cameron Pro Audio
Re: Crest Pro-Lite Amplifiers, feeling confused
« Reply #31 on: July 30, 2013, 04:58:13 PM »

I had a feeling you'd chime in JR, thanks. Makes sense about class D vs old school. Not to get too far OT but maybe slightly, the one caveat with class D that seemed to rear it's head with the original I-Tech amps was that the speaker load forms part of the output RC filter which means varying impedance appears to affect the frequency response of the amp. Somebody on this board measured it as a high frequency roll-off. It seems that could be correct via the DSP in the amp, but then you'd have to know the exact impedance behavior of every connected load or have an analyzer in the amp to figure it out and store it. It seem like any class D would inherently suffer from this, no?
Logged
"Procrastinators of the world - contemplate uniting!"

Cameron Pro Audio

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 17176
  • Hickory, Mississippi, USA
    • Resotune
Re: Crest Pro-Lite Amplifiers, feeling confused
« Reply #32 on: July 30, 2013, 06:08:03 PM »

I had a feeling you'd chime in JR, thanks. Makes sense about class D vs old school. Not to get too far OT but maybe slightly, the one caveat with class D that seemed to rear it's head with the original I-Tech amps was that the speaker load forms part of the output RC filter which means varying impedance appears to affect the frequency response of the amp.
Indeed... IMO a design trade-off, more than an outright mistake or flaw, but communication to the market was not very good surrounding that product roll-out, so every little thing was looked at harshly. 
Quote
Somebody on this board measured it as a high frequency roll-off.
Langston... and it may have been flat at one load termination but not for all.

If you can measure it you can fix it yourself, it seems a minor problem.
Quote
It seems that could be correct via the DSP in the amp, but then you'd have to know the exact impedance behavior of every connected load or have an analyzer in the amp to figure it out and store it. It seem like any class D would inherently suffer from this, no?

I need to think back to the last time I answered this. IIRC there is one class D topology (Bruno Putzey's) where the output filter delay is part of the class D modulation scheme, so the negative feedback loop actually has the output filter inside the feedback loop and response errors related to the output filter are reduced by loop gain margin. Most class D amps take NF from before the filter (AFAIK). 

Amplifiers, even class D, are generally flatter than the speakers they are used with so not a huge issue (IMO). If the filter response error is reasonably constant, it can be measured and corrected for (just like speaker response errors).

If you want to get picky, speaker Z changes with temperature so that response error changes slightly too, but these errors are just in the top octave, so only a small issue and only for the HF passband (not to mention that the speaker response changes with heat too).

Class D may seem new but it is actually been around for several decades so most of this stuff is well understood by those skilled in the art. The only recent development is finally delivering on the cost effectiveness class D always promised.

I wish I had these inexpensive class D platforms back in the day, we tried (trust me) but the device technology just wasn't ready for prime time yet. These will be great in powered speakers and powered mixers.

JR


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

noah katz

  • Newbie
  • *
  • Offline Offline
  • Posts: 1
Re: Crest Pro-Lite Amplifiers, feeling confused
« Reply #33 on: October 30, 2013, 08:01:49 PM »

when ARE we going to see the dsp versions of the 3.0...

+1
Logged

Jeff Bankston

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2568
Re: Crest Pro-Lite Amplifiers, feeling confused
« Reply #34 on: October 31, 2013, 02:42:34 AM »

I also want to point out that the company is using homemade boxes that are loaded with all higher end Faital Pro drivers, and processing is achieved with Ashley Protea units. Is it that the drivers are that efficient? I could save a lot of money as I expand if I were to switch as well.
i'm using faitalpro 18" woofs and they are very sensitive. the fp 18 are a lot more sensative than ciare 18's. it would take two and one half-18" ciare 18" woofs to equal the spl of one faitalpro 18. to put it another way 1-18" faitalpro will almost produce the same spl as 3-18" ciare at max power. so yes you can use less speakers that produce a higher spl in place of more speakers that produce less spl and thats the reason i use the 18" fp woofer. i hope i explained this so you understand. also if i used the ciare 18" i would need to add more amps. also when you have multiple speaker cabinets side by side you get acoustic coupling that can add 6db more spl output.
« Last Edit: October 31, 2013, 02:51:41 AM by Jeff Harrell »
Logged

Ivan Beaver

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 9534
  • Atlanta GA
Re: Crest Pro-Lite Amplifiers, feeling confused
« Reply #35 on: October 31, 2013, 07:13:48 AM »

i'm using faitalpro 18" woofs and they are very sensitive. the fp 18 are a lot more sensative than ciare 18's. it would take two and one half-18" ciare 18" woofs to equal the spl of one faitalpro 18. to put it another way 1-18" faitalpro will almost produce the same spl as 3-18" ciare at max power. so yes you can use less speakers that produce a higher spl in place of more speakers that produce less spl and thats the reason i use the 18" fp woofer. i hope i explained this so you understand. also if i used the ciare 18" i would need to add more amps. also when you have multiple speaker cabinets side by side you get acoustic coupling that can add 6db more spl output.
Sensitivity is your friend.  While the individual cabinets may cost more-you need less of them-less amplifiers-less cable-less truck space-less weight and so forth.

So the overall "cost of ownership is less.

Plus if you are not driving the cabinets as hard-there is less distortion.

Of course some people LIKE distortion in a PA-whatever
Logged
A complex question is easily answered by a simple-easy to understand WRONG answer!

Ivan Beaver
Danley Sound Labs

PHYSICS- NOT FADS!

ProSoundWeb Community

Re: Crest Pro-Lite Amplifiers, feeling confused
« Reply #35 on: October 31, 2013, 07:13:48 AM »


Pages: 1 2 3 [4]   Go Up
 



Site Hosted By Ashdown Technologies, Inc.

Page created in 0.036 seconds with 22 queries.