Measuring Amplifiers - Page 43 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 969Likes
Reply
 
Thread Tools
post #1261 of 1783 Old 01-23-2017, 03:00 PM - Thread Starter
AVS Forum Addicted Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 10,300
Mentioned: 324 Post(s)
Tagged: 0 Thread(s)
Quoted: 3816 Post(s)
Liked: 3771
Quote:
Originally Posted by Shreds View Post
I still don't understand how being on drugs would make you say something that wasn't true about what you do when you burst test. To my knowledge pain drugs don't make you hallucinate.
Never said it was something I do, it was just a passing comment. You can see all my measurements are DC coupled.

Quote:
Originally Posted by Shreds View Post
What you did was a shyster move and instead of admitting it and giving credit where it's due there's always an excuse for the action. I'll continue to bring this up when we discuss things and you brag about your selflessness in contributing the the community because I find this behavior extremely hypocritical.
A shyster move? What, sharing information about the availability of a product? That makes me a shyster? Rofl, you're out of your mind. Never mind that it was being discussed on prosoundweb and being sold on one of the largest websites in the world. I wasn't publishing any info that was being discussed in confidence. Get over yourself.

Even Bosso was sharing information in the thread I posted, which at some point later, he redacted.

Quote:
Originally Posted by bossobass View Post
not,

You're killin' me, man. Tell me subwoofer stories with the new amps.

Bosso
Clearly he was very very upset. Why don't you twist some things a bit more?

Quote:
Originally Posted by Shreds View Post
Yes Rob character bashing. That's what it's called for example when you tell me to "grow up" or tell me that I'm "bitter" or to "go peddle your crap elsewhere" or telling me that I don't do anything constructive. You haven't yet addressed the new data that I've brought to this thread that points out your mistakes.
If you have an issue with my measurements, that's fine, no problem discussing it, but you're obviously still bitter and bringing your baggage here where it's completely unrelated.


Quote:
Originally Posted by Shreds View Post

Umm... So which is it? Pushing the burst to where it clips both upper and lower is certainly not under 1% distortion. It seems very hard for you to remember what your testing methodology is. I like to write things like that down and stick to them so my measurements are consistent. It looks like you just change your testing procedures at a whim to try to look right all the time.
You know, you can push a signal to where it clips or compresses without a large amount of distortion, right? Do you even know what 1% or 10% thd looks like?


Quote:
Originally Posted by Shreds View Post


I remember it fine. The issue was a miscommunication between Bosso and I. Before he saw the consolidated data I shot him a quick number on power which he posted. I gave him the peak power value and he thought it was average power. After you called him out I immediately came into the thread and showed in detail all of the data broken down and easy to understand. It was a mistake, it happens and I have no problem at all admitting that.
Quote:
Originally Posted by Shreds View Post
If we say something measured a certain way, take it to the bank.
Yet it remains uncorrected after being pointed out a number of times. How transparent of you.



Quote:
Originally Posted by Shreds View Post

To my knowledge Brian doesn't post CEA-2010 burst data. He is talking about a sine wave here not an irregular waveform of a burst test. For sine wave testing I use peak values because as I have shown they are more reliably accurate than the RMS chip in our scopes. You can easily arrive at RMS derived average power by dividing peak power in half and you end up with a more accurate RMS measurement. I also show the waveform so that people can see the wave is clean, the duration of the tone and the frequency so that there is nothing that can be misunderstood about how hard the amp is being pushed. This is considerably more information than SpeakerPower submits.
If speaker power posted ratings that were more than double what they currently post, it would be like what you're doing. You're also not using sinusoidal waveforms, so dividing your power rating by 2 won't be accurate either. It's also still completely ignoring the time factor. Also, using reported peak values aren't always accurate, as often higher frequency noise at the peaks of the signal can skew the results. This either needs to be filtered, or corrected for manually.


Quote:
Originally Posted by Shreds View Post
So let me get this straight, you can use peak numbers but I can't? You use peak to peak, and I use the positive peak because a CEA-2010 burst is an irregular waveform. Either way the time is specified in both scenarios Rob. I show the waveform with the frequency given. This shows how the signal relates to time because there is a timescale grid.
The issue is you were using an instantaneous measurement instead of an interval based measurement and reporting the wattage over an interval which resulted in your figures being off by double. It still hasn't been corrected. Just my $0.02, but if you're selling something, you should be familiar enough with its power output capabilities that you don't go posting a number that's off by such a large factor without knowing what you're doing. Transparent.


Quote:
Originally Posted by Shreds View Post


Here's something for the community:

notnyt is posting innacurate burst data. His arriving at RMS values is flawed. He has admitted to not using maximum clean output for burst numbers. His burst numbers do not reflect the clipping point of the amplifiers tested.
So when you do it, using even less data, you can "take the numbers to the bank", but in this case, you lose your mind about it. Point noted.

I've still been logging and providing the full data sets including the voltage numbers, which will likely be a display option on the website when I get it live.

I test until there's visible distortion of the waveform. Not all amps have the same behavior at clipping. Some will limit the signal, some will compress, some will clip, some will behave erratically. I push the amplifier to their maximum "clean" output levels. Perhaps you misunderstood when I said clipping, I was not referring to pushing them so hard the peaks flatten out, but it's more likely just something you could latch onto and use for one of your tirades. I'll generally push an amp to where the signal is more distorted than what I posted, and back off a fraction of a decibel at a time until I get an acceptable waveform.

This is the method I'm using. It will be accurate to within a fraction of a decibel. It's still more accurate than if I just used the absolute voltage peak without factoring in the rest of the wave as you do. Keep in mind environmental factors such as test duration, heat, and power conditions will cause more of a variance here than you're going on about. These are tests into heater coils in a home depot bucket, not medical research. Not that it matters really, your motives for posting here are clearly vindictive.

I'm sorry you feel so hurt, but if you're going to take the role of a vendor and advertise very misleading numbers as you have, it should be brought to light.
notnyt is offline  
Sponsored Links
Advertisement
 
post #1262 of 1783 Old 01-23-2017, 05:31 PM
Senior Member
 
Join Date: Dec 2011
Posts: 369
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 141 Post(s)
Liked: 68
A home depot bucket! That will likely add a few % of variance just because it's orange. Or is it another less impacting color?

On a more serious note: For the user of these amps recording the behavior and accuracy of any signal level / clipping indicators would be of huge value for those that are trying to push these things to the edge of clipping or trying to leave some buffer or e.q.ing capacity. Or what they think is the edge of clipping anyway.

What would be good advice on using the short term power numbers vs. the long term power when designing enclosures? The cost / benefit of different approaches to different designs would make for an interesting read?

Thanks for the testing and the entertainment. I love reading about amp and woofer tests as much as I enjoy listing to my subs....
cellarnoise is offline  
post #1263 of 1783 Old 01-23-2017, 10:22 PM
AVS Forum Special Member
 
Augerhandle's Avatar
 
Join Date: Jan 2009
Location: About 25" away from my computer screen
Posts: 5,337
Mentioned: 32 Post(s)
Tagged: 0 Thread(s)
Quoted: 1253 Post(s)
Liked: 1256
Click image for larger version

Name:	test-a-few-rsvv4b~2.jpg
Views:	900
Size:	58.1 KB
ID:	1915969

"The wise understand by themselves; fools follow the reports of others"-Tibetan Proverb
_____________________ http://www.scientificamerican.com/article/auger-handle/ ________________________

Last edited by Augerhandle; 01-23-2017 at 10:26 PM.
Augerhandle is offline  
Sponsored Links
Advertisement
 
post #1264 of 1783 Old 01-24-2017, 11:38 AM
AVS Forum Special Member
 
JohnDean's Avatar
 
Join Date: Jul 2012
Location: Franklin, TN
Posts: 1,421
Mentioned: 28 Post(s)
Tagged: 0 Thread(s)
Quoted: 546 Post(s)
Liked: 255
May I ask who @Shreds is, and what he's selling? I don't have a dog in this fight but am curious, for which company he is a vendor.

Thanks, I'll go back to letting the people smarter than I argue it out.
JohnDean is offline  
post #1265 of 1783 Old 01-24-2017, 12:30 PM - Thread Starter
AVS Forum Addicted Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 10,300
Mentioned: 324 Post(s)
Tagged: 0 Thread(s)
Quoted: 3816 Post(s)
Liked: 3771
Quote:
Originally Posted by JohnDean View Post
May I ask who Shreds is, and what he's selling? I don't have a dog in this fight but am curious, for which company he is a vendor.

Thanks, I'll go back to letting the people smarter than I argue it out.
He's bossobass's offspring. They sell clone amps and subs from China. He's bent since I made a stink about them advertising double the actual power output of the amps.

Last edited by notnyt; 01-24-2017 at 12:43 PM.
notnyt is offline  
post #1266 of 1783 Old 01-24-2017, 01:51 PM
AVS Forum Special Member
 
diy speaker guy's Avatar
 
Join Date: Jan 2011
Posts: 1,416
Mentioned: 42 Post(s)
Tagged: 0 Thread(s)
Quoted: 510 Post(s)
Liked: 525
Quote:
Originally Posted by notnyt View Post
They sell clone amps and subs from China.
Well now you gave him another reason to be mad. These guys were swearing Bosso's amps are not clones, remember?
diy speaker guy is offline  
post #1267 of 1783 Old 01-24-2017, 02:14 PM - Thread Starter
AVS Forum Addicted Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 10,300
Mentioned: 324 Post(s)
Tagged: 0 Thread(s)
Quoted: 3816 Post(s)
Liked: 3771
Quote:
Originally Posted by diy speaker guy View Post
Well now you gave him another reason to be mad. These guys were swearing Bosso's amps are not clones, remember?
notnyt is offline  
post #1268 of 1783 Old 01-24-2017, 03:09 PM
AVS Forum Special Member
 
diy speaker guy's Avatar
 
Join Date: Jan 2011
Posts: 1,416
Mentioned: 42 Post(s)
Tagged: 0 Thread(s)
Quoted: 510 Post(s)
Liked: 525
Anyway, like most everyone else I greatly appreciate what Not is doing here. If I'm not mistaken his results line up pretty well with other 3rd party measurements of the same products. I don't even care where Not chooses to end the test - whether it's right before clipping, right at the point of clipping, or with both + and - slightly clipped, as long as the procedure is described, which it has been. If the results are within even 10 percent of what I can expect if I buy an amp Not measured that's way more than good enough for me.

Not too many people are measuring amps and the ones that do and do so well are a great service to the community. I don't see any problem with Not's procedures or results.
diy speaker guy is offline  
post #1269 of 1783 Old 01-24-2017, 07:25 PM
AVS Forum Addicted Member
 
BassThatHz's Avatar
 
Join Date: Apr 2008
Location: Northern Okan range (NW Cascades region)
Posts: 10,815
Mentioned: 235 Post(s)
Tagged: 0 Thread(s)
Quoted: 3777 Post(s)
Liked: 4297
Quote:
Originally Posted by Shreds View Post

I can't even make out half of the waveforms on this screen
Go see an optometrist then!

It is clear just by looking at it that:
Yellow is wall-socket voltage
light blue is current.
Pink is Vout
Cyan is Vin.
Indigo is V*I

Looking into the sun! = Color Blind. No fighter piloting for you!


Quote:
Originally Posted by Shreds View Post
irregular waveform
No. Actually it is a highly predictable polynomial.

Just playing around with a graphing calculator I was able to simulate a CEA-2010'ish looking polynomial. Wasn't that hard!
Attached Thumbnails
Click image for larger version

Name:	563.png
Views:	862
Size:	51.5 KB
ID:	1918065   Click image for larger version

Name:	einstein-duh.jpg
Views:	834
Size:	25.7 KB
ID:	1918081  
notnyt likes this.
BassThatHz is offline  
post #1270 of 1783 Old 01-24-2017, 08:29 PM
Senior Member
 
Join Date: Dec 2011
Posts: 369
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 141 Post(s)
Liked: 68
The world changes fast...
I remember a German test of the Inuke3000 when it first came out and it "burst" 1700 watts or so short term (1/2 second?). I might have a saved copy of that test somewhere.

I just replaced the fan in a newly purchased Inuke3000 dsp and noticed it has what must be the "hot" transistors mounted on a vertical aluminum looking cooling card under the fan cooling tunnel... The new board has a slightly different layout and color. The color I would guess adds 10% to the output / horsepower?

Note the inuke with the green pcb is my 2013ish amp that I likely wasted $2 installing heat sinks on its hot parts back in the day. It has never given me any trouble, but currently only one channel is used...
Attached Thumbnails
Click image for larger version

Name:	Old-New_inuke3000dsp-with-love-haha.jpg
Views:	96
Size:	130.5 KB
ID:	1918153  
cellarnoise is offline  
post #1271 of 1783 Old 01-24-2017, 08:41 PM - Thread Starter
AVS Forum Addicted Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 10,300
Mentioned: 324 Post(s)
Tagged: 0 Thread(s)
Quoted: 3816 Post(s)
Liked: 3771
Quote:
Originally Posted by cellarnoise View Post
The world changes fast...
I remember a German test of the Inuke3000 when it first came out and it "burst" 1700 watts or so short term (1/2 second?). I might have a saved copy of that test somewhere.

I just replaced the fan in a newly purchased Inuke3000 dsp and noticed it has what must be the "hot" transistors mounted on a vertical aluminum looking cooling card under the fan cooling tunnel... The new board has a slightly different layout and color. The color I would guess adds 10% to the output / horsepower?

Note the inuke with the green pcb is my 2013ish amp that I likely wasted $2 installing heat sinks on its hot parts back in the day. It has never given me any trouble, but currently only one channel is used...
With my tests it sustained 1600w for a couple seconds before going into protect. I have the older style iirc, with the weird little metal wings / heat dissipaters on the transistors. Still fan modded..


I need something that can sustain 1kw into 4 ohms or 2x500 into 8 ohms with dsp and fans that won't run at idle. Nothing fancy, for transducers, don't want to spend too much. Any thoughts?

Last edited by notnyt; 01-24-2017 at 08:44 PM.
notnyt is offline  
post #1272 of 1783 Old 01-24-2017, 08:59 PM
Member
 
bpgunning's Avatar
 
Join Date: Jul 2007
Location: ABQ
Posts: 103
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 48 Post(s)
Liked: 91
Quote:
Originally Posted by notnyt View Post
I need something that can sustain 1kw into 4 ohms or 2x500 into 8 ohms with dsp and fans that won't run at idle. Nothing fancy, for transducers, don't want to spend too much. Any thoughts?
crown xls?
JohnDean likes this.
bpgunning is offline  
post #1273 of 1783 Old 01-24-2017, 08:59 PM
AVS Forum Special Member
 
VisionxOrb's Avatar
 
Join Date: Sep 2003
Location: Oceanside, Ca
Posts: 1,102
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 353 Post(s)
Liked: 244
Quote:
Originally Posted by notnyt View Post


I need something that can sustain 1kw into 4 ohms or 2x500 into 8 ohms with dsp and fans that won't run at idle. Nothing fancy, for transducers, don't want to spend too much. Any thoughts?
notnyt look at the qsc pld line, they have thermal controlled fans there silent at idle, full dsp etc.

2000 watts of power and climbing

-Fred
VisionxOrb is offline  
post #1274 of 1783 Old 01-24-2017, 09:02 PM - Thread Starter
AVS Forum Addicted Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 10,300
Mentioned: 324 Post(s)
Tagged: 0 Thread(s)
Quoted: 3816 Post(s)
Liked: 3771
Quote:
Originally Posted by bpgunning View Post
crown xls?
Was thinking 2502.. do the fans spin at idle? I've been working on trying to get all my gear dead silent.
notnyt is offline  
post #1275 of 1783 Old 01-24-2017, 09:08 PM
Member
 
bpgunning's Avatar
 
Join Date: Jul 2007
Location: ABQ
Posts: 103
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 48 Post(s)
Liked: 91
Quote:
Originally Posted by notnyt View Post
Was thinking 2502.. do the fans spin at idle? I've been working on trying to get all my gear dead silent.
Also in that price range you could save a couple bucks and grab a Crest Pro Lite 3.0 DSP. I have a regular 3.0 (non DSP) to eventually run my DIY 18 sub and its fan does not run at idle. As for the XLS, I honestly don't know. IIRC @lukeamdman tested a 2500, maybe he can comment on the fans?
bpgunning is offline  
post #1276 of 1783 Old 01-24-2017, 09:18 PM - Thread Starter
AVS Forum Addicted Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 10,300
Mentioned: 324 Post(s)
Tagged: 0 Thread(s)
Quoted: 3816 Post(s)
Liked: 3771
Quote:
Originally Posted by bpgunning View Post
Also in that price range you could save a couple bucks and grab a Crest Pro Lite 3.0 DSP. I have a regular 3.0 (non DSP) to eventually run my DIY 18 sub and its fan does not run at idle. As for the XLS, I honestly don't know. IIRC @lukeamdman tested a 2500, maybe he can comment on the fans?
seems like xls2502 is a bit cheaper
notnyt is offline  
post #1277 of 1783 Old 01-24-2017, 09:22 PM
Advanced Member
 
Shreds's Avatar
 
Join Date: May 2014
Location: NC
Posts: 875
Mentioned: 66 Post(s)
Tagged: 0 Thread(s)
Quoted: 571 Post(s)
Liked: 632
Quote:
Originally Posted by notnyt View Post
peak to peak divided by 2 multiplied by 0.707 divided by the impedance will get you the RMS value of the 1.5 cycles at the center of the burst.

This is 1.5 cycles. There is more time of the negative swing than the positive which is why a true RMS value will always read less than what you state as RMS. Does everyone see on the left is notnyt's way and on the right is the true RMS reading of the tool and how much this can throw off your power measurement in only 1kW? This becomes an even more significant error in 10kW.

If you are pushing a burst to 10%THD, there is absolutely no way to accurately post RMS data of a distorted burst waveform in all the frequencies that you test by using your peak to peak method and the tools that you have. Just like there is absolutely no way to accurately post RMS data of the following waveforms:

Guitar transient:


Bass transient:


Drum kit transient:


Music transient:


Movie with bass transient:


RMS doesn't relate to the real world, peaks do. You can use the positive peak of a CEA-2010 burst at any frequency and compare it with your sub output on any movie scene with bass that you choose with full sub bandwidth content, learn what signal is more demanding and see how the amplifiers behave differently comparing the 2 signals. The fact that you guys don't find this data interesting stuns me. This is why I use peak for burst testing, not because I don't understand what I'm measuring and certainly not because I'm trying to trick anyone. Dividing the peak power I list for a burst in half will give an idea of what average power is but there is no way to accurately calculate RMS for a burst on our scopes.

Now as far as steady state sine wave testing, I use peaks to arrive at RMS because peaks don't lie whereas the cheap RMS chips in our scopes are not as reliable. I illustrate this here. I list the peak power, I state that I list the peak power and tell anyone who wants the average power derived from RMS to divide the peak value by 2. I show the time grid and spell out how long the sine wave was held for. I also zoom into the waveform as much as possible to show whether it's clipping or clean instead of showing a jumbled up and squashed mess.



notnyt has been accusing me of a lot of things that he knows I don't do to make me look as bad as possible. I didn't come into this thread because I'm hurt or vindictive or because I have baggage (whatever that means). I posted because desertdome linked a discussion that I was part of. I'm pointing out things that I believe to be errors and including data with my conclusions because I think it will help you all have more accurate information to refer to.

Quote:
Originally Posted by notnyt View Post
I generally push the cea-2010 signal enough to clip the lower and the upper when reporting max burst numbers.
I think based on this and the RMS data that I show above that this thread has no reliable clean burst data listed. Also, using notnyt's methodology using peak to peak to derive RMS on burst's cannot accurately tell you exactly where the amp clips on transients. This is very important information on both steady state sine waves and burst polynomial (thanks BTH) waveforms.
Shreds is offline  
post #1278 of 1783 Old 01-24-2017, 09:23 PM
Member
 
bpgunning's Avatar
 
Join Date: Jul 2007
Location: ABQ
Posts: 103
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 48 Post(s)
Liked: 91
Quote:
Originally Posted by notnyt View Post
xls2502 seems cheaper than the pro lite...
Ah, I was seeing $600 for the xls2502 pretty much everywhere, but Amazon has it for $479. Pro Lite 3.0 DSP $499 at Audiosavings
bpgunning is offline  
post #1279 of 1783 Old 01-24-2017, 09:29 PM
AVS Forum Special Member
 
VisionxOrb's Avatar
 
Join Date: Sep 2003
Location: Oceanside, Ca
Posts: 1,102
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 353 Post(s)
Liked: 244
Xls isnt a real dsp though its just crossovers. But it is silent at idle.

2000 watts of power and climbing

-Fred
VisionxOrb is offline  
post #1280 of 1783 Old 01-24-2017, 09:51 PM - Thread Starter
AVS Forum Addicted Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 10,300
Mentioned: 324 Post(s)
Tagged: 0 Thread(s)
Quoted: 3816 Post(s)
Liked: 3771
Quote:
Originally Posted by VisionxOrb View Post
Xls isnt a real dsp though its just crossovers. But it is silent at idle.
I could run it through my processor I guess. I have enough channels... How's the prolite dsp?
notnyt is offline  
post #1281 of 1783 Old 01-24-2017, 09:59 PM
Member
 
bpgunning's Avatar
 
Join Date: Jul 2007
Location: ABQ
Posts: 103
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 48 Post(s)
Liked: 91
Quote:
Originally Posted by notnyt View Post
I could run it through my processor I guess. I have enough channels... How's the prolite dsp?
Slightly better it sounds like, but still not exactly exciting - delays, some PEQ, limiter, bass boost, and some filters/crossovers but I think the EQ stops at 31 Hz, can't do anything below.
bpgunning is offline  
post #1282 of 1783 Old 01-24-2017, 10:27 PM - Thread Starter
AVS Forum Addicted Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 10,300
Mentioned: 324 Post(s)
Tagged: 0 Thread(s)
Quoted: 3816 Post(s)
Liked: 3771
Quote:
Originally Posted by Shreds View Post
This is 1.5 cycles. There is more time of the negative swing than the positive which is why a true RMS value will always read less than what you state as RMS. Does everyone see on the left is notnyt's way and on the right is the true RMS reading of the tool and how much this can throw off your power measurement in only 1kW? This becomes an even more significant error in 10kW.
The significance in decibels will always remain the same.

Anyway, this is an interesting discussion, with all the other bs aside. I have thought about this before, but in the end the results seem to be on point with other tests. I could easily use the RMS measurement on the scope, but it would significantly slow the process. That said, the peak in the center will be more distorted than the bottom via either compression, clipping, or other means. I don't want to just use the maximum value from the center as it doesn't represent the full signal. However, there may still be some headroom on the lower half which could be eeked out. Even so, pushing the lower half harder may also impact the upper half. I considered using a custom waveform as well, but the CEA-2010 burst is a standard, and makes duplication for anyone else wanting to test much easier.

There's lots of back and forth here, so I'm just using the peak to peak measurements for the burst numbers and going with that. This is how this test is defined, and that's how it will be.

Quote:
Originally Posted by Shreds View Post
If you are pushing a burst to 10%THD, there is absolutely no way to accurately post RMS data of a distorted burst waveform in all the frequencies that you test by using your peak to peak method and the tools that you have. Just like there is absolutely no way to accurately post RMS data of the following waveforms:
Never said I was pushing a burst to 10% THD. I asked if you knew what 1% and 10% THD even looked like. It's apparent that may not be the case.

For anyone curious

This is 1%



This is 10%



As I mentioned, the waveforms I posted weren't even 1%. I even included example waveforms, but you completely ignore this.

While I agree with you that it's not perfect, it's good enough. You're also looking at a perfect burst signal. When an amplifier is being pushed to its limits, the signal will not be perfect.

Here is the same example I provided earlier with markings at the same positive and negative values. Were I to just use the peak value at the center, as opposed to the Vpp values divided by 2, the power rating would be higher than actually produced. Also, as the signal is slightly flat-topped on the center peak, that will actually raise the RMS value, so it kind of evens out the deficit on the bottom swing.



I have an IPR2 on my bench. I'll actually do that comparison at some point (RMS from scope vs calculated via p-p). I'm still troubleshooting this KRK amp for my neighbor. It works fine on startup for a few seconds, then the output level starts dropping significantly, then once it gets warm it works fine again. I think there may be a bad cap on it, but there's a short window to troubleshoot and I haven't exactly been in the best mindset for it lately. Thankfully, I'm recovering now from this last stone and doing a bit better. This will be an interesting test. I hope to get to it soon.

Quote:
Originally Posted by Shreds View Post
RMS doesn't relate to the real world, peaks do. You can use the positive peak of a CEA-2010 burst at any frequency and compare it with your sub output on any movie scene with bass that you choose with full sub bandwidth content, learn what signal is more demanding and see how the amplifiers behave differently comparing the 2 signals. The fact that you guys don't find this data interesting stuns me. This is why I use peak for burst testing, not because I don't understand what I'm measuring and certainly not because I'm trying to trick anyone. Dividing the peak power I list for a burst in half will give an idea of what average power is but there is no way to accurately calculate RMS for a burst on our scopes.
You can test your way if you choose, however this way is easily repeatable and cleaner to visualize. Testing with sinusoidal data will provide enough information. If you want to use movie soundtracks to test amps, feel free, just document the process thoroughly, stay consistent, and don't do things like posting peak power values over a time interval as you've done in the past.

Quote:
Originally Posted by Shreds View Post
Now as far as steady state sine wave testing, I use peaks to arrive at RMS because peaks don't lie whereas the cheap RMS chips in our scopes are not as reliable. I illustrate this here. I list the peak power, I state that I list the peak power and tell anyone who wants the average power derived from RMS to divide the peak value by 2. I show the time grid and spell out how long the sine wave was held for. I also zoom into the waveform as much as possible to show whether it's clipping or clean instead of showing a jumbled up and squashed mess.

And if you were to use the peak-peak data here, you would arrive at incorrect RMS readings, as some peaks are higher on the top and bottom of the waveform. Look at the first positive swing. Also, your waveform is distorted enough at the end that using peak data will not be as accurate as the RMS calculation. Setting the measurement window properly and using the scopes RMS calculation on near sine output works just fine.

Quote:
Originally Posted by Shreds View Post
notnyt has been accusing me of a lot of things that he knows I don't do to make me look as bad as possible. I didn't come into this thread because I'm hurt or vindictive or because I have baggage (whatever that means). I posted because desertdome linked a discussion that I was part of. I'm pointing out things that I believe to be errors and including data with my conclusions because I think it will help you all have more accurate information to refer to.
Yet you still haven't corrected the error that I pointed out 6+ months ago that started our whole debate. Now you admit it was a miscommunication between you and Bosso. Had you done that earlier and corrected it, it could have saved significant headaches. I do actually appreciate the contribution and discussion, but you blow things way out of proportion.

Quote:
Originally Posted by Shreds View Post
I think based on this and the RMS data that I show above that this thread has no reliable clean burst data listed. Also, using notnyt's methodology using peak to peak to derive RMS on burst's cannot accurately tell you exactly where the amp clips on transients. This is very important information on both steady state sine waves and burst polynomial (thanks BTH) waveforms.

Rofl, that's a little dramatic. I think you don't know enough about the process to state that. The test method is reliable and accurate enough for these purposes. The test methods have been outlined and have remained fairly consistent along the way with some improvements being added thanks to previous feedback. The test isn't "where does the amp clip" the test is the maximum burst values using a cea-2010 signal.


On another semi related topic, many amps introduce their switching noise into the signal at clipping levels. Unless you're filtering the signal, using peak numbers anywhere will be invalid.

This is actually under 1% THD, but 1% THD+N, as it raises the noise floor more than the harmonics.



I have a few basic filters I made for this that normalize the waveform, but need to remove them before doing HF testing as it slightly rolls off at the upper end of things.

Last edited by notnyt; 01-24-2017 at 10:36 PM.
notnyt is offline  
post #1283 of 1783 Old 01-24-2017, 10:33 PM - Thread Starter
AVS Forum Addicted Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 10,300
Mentioned: 324 Post(s)
Tagged: 0 Thread(s)
Quoted: 3816 Post(s)
Liked: 3771
Quote:
Originally Posted by bpgunning View Post
Slightly better it sounds like, but still not exactly exciting - delays, some PEQ, limiter, bass boost, and some filters/crossovers but I think the EQ stops at 31 Hz, can't do anything below.
Hm, yeah, that just won't do. I like the DSP on the behringer amps. They're quite good, especially at the price. I'm thinking I might just build my own fan controller for it. I don't care if it spins up during a movie, it won't be noticeable. I just don't want the noise when the room is quiet or while listening to music.
notnyt is offline  
post #1284 of 1783 Old 01-25-2017, 07:08 AM
AVS Forum Special Member
 
lukeamdman's Avatar
 
Join Date: May 2011
Location: Burnsville, MN
Posts: 3,790
Mentioned: 198 Post(s)
Tagged: 2 Thread(s)
Quoted: 1385 Post(s)
Liked: 2000
Quote:
Originally Posted by bpgunning View Post
Also in that price range you could save a couple bucks and grab a Crest Pro Lite 3.0 DSP. I have a regular 3.0 (non DSP) to eventually run my DIY 18 sub and its fan does not run at idle. As for the XLS, I honestly don't know. IIRC @lukeamdman tested a 2500, maybe he can comment on the fans?
I turned on the 2500 last night and after 20-25 minutes the fan was still idle.

I have seen it spin up before, but even then my ear had to be within an inch to hear it.
notnyt likes this.
lukeamdman is offline  
post #1285 of 1783 Old 01-25-2017, 07:18 AM
AVS Forum Special Member
 
JohnDean's Avatar
 
Join Date: Jul 2012
Location: Franklin, TN
Posts: 1,421
Mentioned: 28 Post(s)
Tagged: 0 Thread(s)
Quoted: 546 Post(s)
Liked: 255
Quote:
Originally Posted by notnyt View Post
Hm, yeah, that just won't do. I like the DSP on the behringer amps. They're quite good, especially at the price. I'm thinking I might just build my own fan controller for it. I don't care if it spins up during a movie, it won't be noticeable. I just don't want the noise when the room is quiet or while listening to music.
Did you test one of these amps? I thought you tested a Crown XLS gen1.

I do have extra 1502s sitting here.

FWIW I've run a XLS1500 to close to clipping in 2ohm stereo for 20-30 minutes and can't recall ever hearing the fans.
JohnDean is offline  
post #1286 of 1783 Old 01-25-2017, 08:43 AM
AVS Forum Special Member
 
dwaleke's Avatar
 
Join Date: Feb 2006
Location: Michigan
Posts: 2,721
Mentioned: 30 Post(s)
Tagged: 0 Thread(s)
Quoted: 1460 Post(s)
Liked: 751
Almost no difference between xls gen 1 and 2.

This was confirmed by crown when the gen 2 was released.
dwaleke is online now  
post #1287 of 1783 Old 01-25-2017, 09:11 AM
AVS Forum Special Member
 
VisionxOrb's Avatar
 
Join Date: Sep 2003
Location: Oceanside, Ca
Posts: 1,102
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 353 Post(s)
Liked: 244
The QSC PLD has a full fledged DSP with saved profiles etc. A friend of mine switch from XLS to them. heres some info for them

PLD Series is more than just an amplifier; it is also a capable and sophisticated loudspeaker processor. The close integration of processing and amplification allows the DSP to monitor and better respond to amplifier behavior, making dynamics processing far more accurate and effective than that typically achieved with separate components. This synergistic approach employs both RMS and Peak Limiters that allow the combination of amplifier and loudspeaker to produce more output without being pushed to distortion or destruction.
In addition to the dynamics processing, the onboard DSP offers four channels of crossover filters, parametric EQ, and alignment delay - everything needed to optimize a loudspeaker system. PLD amplifiers also offer Intrinsic Correction™, a combination of IIR and FIR Filtering and loudspeaker processing methodology first developed for our WideLine line array loudspeakers. Intrinsic Correction compensates for the non-linearities in loudspeaker array and horn design and results in notably superior performance from your QSC loudspeakers.

PLD amplifiers feature Flexible Amplifier Summing Technology™ (FAST) that actively distributes the total amplifier power in various combinations across one, two, three, or all four outputs. This flexibility allows PLD Series amplifiers to drive (for example) four full-range loudspeakers or subwoofers; a high-power subwoofer and a bi-amplified speaker; a single high-power full-range loudspeaker and subwoofer combo; or multiple power-hungry subwoofers from a single, very high output mono-block.

The PLD Series amplifiers use QSC fourth generation class-D power amp design in combination with a custom power stage utilizing a new, purpose-built output device. These innovative MOSFET devices provide high voltage operation without needing a full bridge output and offer superior audio quality due to co-location of the semiconductors. Additionally PLD amps benefit from the famed and road-proven PowerLight power supply, further enhanced with Power Factor Correction (PFC) that aligns the current waveform with the AC mains voltage waveform. PFC enables PLD Series amps to draw current from the wall in a more efficient and controlled manner resulting in incredible power from a single standard AC breaker. Additionally, the PLD Series amps offer multi-stage sleep modes, saving energy when possible without ever sacrificing performance. The result is an exceptionally powerful and flexible platform that offers low weight and outstanding efficiency.

http://www.qsc.com/live-sound/produc...rs/pld-series/

2000 watts of power and climbing

-Fred

Last edited by VisionxOrb; 01-25-2017 at 09:15 AM.
VisionxOrb is offline  
post #1288 of 1783 Old 01-25-2017, 07:26 PM
Senior Member
 
Join Date: Dec 2011
Posts: 369
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 141 Post(s)
Liked: 68
KnotNy,
I ran my 2013 inuke3000dsp with just heat sinks for around a month with no fan, probably longer as I forgot. Granted it was bridged into a single ftw-21 and it I was playing with it often. Lots of e.q. Into sub 20hz and I tried to stay out of the clip light always. Never noticed a difference to when I installed a cheap low flow fan (cheap 12v .09 amp Best Buy..) and then continued to play around before my first sanway clone. Did not ever shut down except when pushed to extended clipping. Others have ran the 3000 fanless without problems if used for HT and not pushed to the clip light constantly.

Though I still will stand behind my previous limted tests that on subs it sounds different from a ultra low tuned amp like the fp14000 clone. Maybe damping factor or whatever? Class d or what? Should not matter on an arse shaker. That's what I use my 2013 for now, low end booty shakers!

The new inuke is going to power cheap auras and a hand me down 2 ohm si18 ported to 10ish. Hope my fauger enjoys it all
cellarnoise is offline  
post #1289 of 1783 Old 01-26-2017, 05:28 AM
AVS Forum Special Member
 
krholmberg's Avatar
 
Join Date: Jun 2005
Posts: 1,329
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 198 Post(s)
Liked: 93
@notnyt That would be awesome if you could test an XLS 2nd series amp as well as a ProLite. I've seen an independent bench test of a XLS 2500 that shows it's quite solid to 10Hz, then drops off a cliff. Everything I've read about the ProLites suggests they not only run deep but also strong but that they get loud. I read a comment yesterday by bossobass that was not complementary of the 2502 for sub duty. It would be quite nice to see independent verification by a trusted AVS member. I'm considering a XLS2502 vs Crest CC1800/2800 vs Crest ProLite 3.0 (nonDSP). Leaning towards the 2502 since its quiet and has a dc trigger but only if it works well with my quad 18" IB sub. The same person that tested the 2500 showed the CC2800 didn't even go down to 10Hz well so its likely off the list.

Sent from my T-Mobile device using Tapatalk

Krister

Last edited by krholmberg; 01-26-2017 at 05:50 AM.
krholmberg is offline  
post #1290 of 1783 Old 01-26-2017, 05:31 AM - Thread Starter
AVS Forum Addicted Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 10,300
Mentioned: 324 Post(s)
Tagged: 0 Thread(s)
Quoted: 3816 Post(s)
Liked: 3771
Quote:
Originally Posted by krholmberg View Post
That would be awesome if you could test an XLS 2nd series amp as well as a ProLite. I've seen an independent bench test of a XLS 2500 that shows it's quite solid to 10Hz, then drops off a cliff. Everything I've read about the ProLites suggests they not only run deep but also strong but that they get loud. I read a comment yesterday by bossobass that was not complementary of the 2502 for sub duty. It would be quite nice to see independent verification by a trusted AVS member. I'm considering a XLS2502 vs Crest CC1800/2800 vs Crest ProLite 3.0 (nonDSP). Leaning towards the 2502 since its quiet and has a dc trigger but only if it works well with my quad 18" IB sub. The same person that tested the 2500 showed the CC2800 didn't even go down to 10Hz well so its likely off the list.

Sent from my T-Mobile device using Tapatalk
@lukeamdman tested a 2500 and some others...

http://data-bass.ipbhost.com/index.p...plifier-tests/
krholmberg and lukeamdman like this.
notnyt is offline  
Sponsored Links
Advertisement
 
Reply DIY Speakers and Subs

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off