Speakerplans.com Homepage
Forum Home Forum Home > General > Amp Forum
  New Posts New Posts RSS Feed - Peavey IPR-1600 Test Results
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Peavey IPR-1600 Test Results

 Post Reply Post Reply Page  <1 234
Author
Message
cilla.scope View Drop Down
Old Croc
Old Croc

Legs all the way to her bottom....

Joined: 02 October 2009
Status: Offline
Points: 2954
Post Options Post Options   Thanks (0) Thanks(0)   Quote cilla.scope Quote  Post ReplyReply Direct Link To This Post Posted: 13 July 2010 at 11:28pm
heh ,... I'm not Lewis so I can't say if he would use it or not ...

mmm .. as to whether its use would be legitimate, yes, of course it would .. there is nothing inherently wrong with the AES17 test method as such,  it serves a purpose ... it does enable you to reach those low numbers and analyze the in-band components to very low numbers ...  there is also an argument that the AES17 test was aimed at getting to the low numbers when even a small amount of residual out-of-band noise could upset the measurement,  I don;t believe the test was ever intended to mask what I perceive as gross errors in the design process that lead to significant out of band components being present.

I fully accept that you have goen to a great deal of effort to get to an AES17 distortion figure probably an order of magnitude better than shown in Lewis measurement, and that by using a full bandwidth measurement, this work does not shine through ... on the flip side, the full bandwidth measurment does show up the ones that have problems quite significantly.

I think when comparing one Class D with another, your amp will still show up very well, regardless of whether it is AES17 or full bandwidth measurement ... where you might lose out is in a comparisom with a linear Class A/B amp, which will have low out of band noise components ... so yes, in that respect I agreee with you.
My girlfriend thinks I'm a stalker.
Well, she's not exactly my girlfriend yet ...
Back to Top
nineleaves View Drop Down
Young Croc
Young Croc


Joined: 31 August 2007
Status: Offline
Points: 1327
Post Options Post Options   Thanks (0) Thanks(0)   Quote nineleaves Quote  Post ReplyReply Direct Link To This Post Posted: 13 July 2010 at 11:40pm
This amp holds the top spot in my shrine of awful product design... they somehow went from the industrious & purposeful design of their old CS range to this crap ..ive seen better aesthetics from the chinky knock off merchants.. dont like on that basis alone.

However, supprisingly well production engineered.. dispite the unessurserily big case.
Back to Top
SickNeedHelp View Drop Down
Registered User
Registered User


Joined: 07 March 2010
Location: Meridian MS
Status: Offline
Points: 15
Post Options Post Options   Thanks (0) Thanks(0)   Quote SickNeedHelp Quote  Post ReplyReply Direct Link To This Post Posted: 13 July 2010 at 11:52pm
Originally posted by cilla.scope cilla.scope wrote:

... where you might lose out is in a comparisom with a linear Class A/B amp, which will have low out of band noise components ...
 
Hope I'm not wearing out my welcome here (assuming I had one).
 
Interesting that your made the point regarding a Class A/B comparison. The Crest 5200 is a class A/B with a reasonable reputation for fidelity. The IPR1600 proof of concept prototype used a 5200 power supply and went up against the 5200 in a blind AB comparison in our auditorium (Andy knows the one). Obviously power was within a few watts of matched. Gain was also matched.  The spec sheets easily favor the 5200, yet 10 out of 12 picked the IPR proto in the listenting test at low levels and at high levels.
 
From a marketing standpoint, we have to rethink the meaning & value of the specs we publish. I don't have a solution.  Hopefully other manufacturer's will pitch in.  For me it's:
 
1) How does it sound. Nothing else matters if it sounds bad and causes listener fatigue consciously or subconsciously.
2) Is it reliable. If it's not, what's the point, especially if you are making a living with your equipment.
3) How hard will it push a cone.
3.5) What does it cost.
4) What does it weigh. We are all getting older.
5) What does the spec sheet say. #6 is my personal belief, but given we are in a commercial competitive enviroment, I have to elevate the importance.
 
 


Edited by SickNeedHelp - 13 July 2010 at 11:58pm
Back to Top
SickNeedHelp View Drop Down
Registered User
Registered User


Joined: 07 March 2010
Location: Meridian MS
Status: Offline
Points: 15
Post Options Post Options   Thanks (0) Thanks(0)   Quote SickNeedHelp Quote  Post ReplyReply Direct Link To This Post Posted: 14 July 2010 at 12:18am
Originally posted by nineleaves nineleaves wrote:

... they somehow went from the industrious & purposeful design of their old CS range to this crap ..
 
I think we still have some of the old 70's vintage 4U cast face plates around. We can fill the other 2U of the thick steel chassis with bricks.
 
How many should I put you down for?
Back to Top
cilla.scope View Drop Down
Old Croc
Old Croc

Legs all the way to her bottom....

Joined: 02 October 2009
Status: Offline
Points: 2954
Post Options Post Options   Thanks (0) Thanks(0)   Quote cilla.scope Quote  Post ReplyReply Direct Link To This Post Posted: 14 July 2010 at 12:23am
Wearing out your welcome? JD, its an absolute pleasure I assure you, always a privilege to chat with a designer.

I think I wore out my welcome ages ago, but I'm  too thick skinned to care :)

Your list of points is interesting. One point I would make is that  not everyone has the same  priorities. For some build quality and how it will survive "life on the road" is key. For others, lightweight.  For others, cost. To cover all the bases I suspect you need more than one range of amps, not just based on price but on other factors.

Sound as you say is key though. If it doesn't sound good, then there really is no point.

Regarding the spec sheet, to be honest, I'm not sure people really pay that much attention to them anymore, as I've heard some awful sounding amps with glowing spec sheets ... or power outputs that imply efficiencies of around 180% .. so people "sort of" believe them.  Too many cheap Chinese clones about with output waveforms like an arc welder and a spec sheet like a hi-fi amp.

I don't know what the answer is to the last one ... when you have gone the extra mile to get the distortion down an order of magnitude below your competitor, how do you get that across to the public? Trciky one .. magazine reviews are often biased by advertising budgets ... spec sheets are not independently vetted ... 

maybe its time for a industry standard assured test? and independent organisation with a *reasonable* fee that makes testing verifiable and accessible not only to manufacturers but also to customers who wish to verify results. Modern automated test should allow someone to test an amp in minutes with little effort and produce repeatable and verifiable results ... 





Edited by cilla.scope - 14 July 2010 at 12:26am
My girlfriend thinks I'm a stalker.
Well, she's not exactly my girlfriend yet ...
Back to Top
IanD View Drop Down
Registered User
Registered User


Joined: 17 January 2009
Location: London
Status: Offline
Points: 400
Post Options Post Options   Thanks (0) Thanks(0)   Quote IanD Quote  Post ReplyReply Direct Link To This Post Posted: 14 July 2010 at 8:59pm
Originally posted by abeltronics abeltronics wrote:

The AES17 standard talks about measuring THD in the presence of significant out of band noise. With class D at full power, the OOB noise is already about 60dB down on the wanted signal, which is the approximate attenuation of the AES17 filter anyway, so we're measuring THD in the presence of INsignificant OOB noise. I don't think it really poses too much of a problem, and I would argue AES17 does not apply to class D amplifiers when tested this way. Remember, Audio Precision want to sell accessories for their distortion analysers, so they have a vested interest in telling everyone their THD analysers are no good.

Also, to elaborate on Cilla's point, harmonic distortion is composed of harmonics, or multiples of the fundamental frequency. So if you're testing at 10kHz, the ampliifer will produce harmonics at 20kHz, 30kHz, 40kHz, 50kHz, and so on. Putting a 20kHz brick wall filter slap bang on the end of the amplifier before the distortion analyser filters out all these harmonics and completely negates the validity of 10kHz THD tesing - all of a sudden all the distortion produced by the amplifier magically disappears. I think the validity of this methodology is questionable.

The peavey measured about 0.08% THD at 1/4 power 1kHz, which is very low and exactly where i'd exect it to be - Much lower than the FFA tested in the same manner - so I don't think HF affecting my analyser is really plausible.

It seems to me that class D manufacturers have invented a new test procedure - they have invented a method to make their amplifiers measure better by filtering out what you're trying to measure. And the raison d'etre is the AES17 document which is designed to be applied to something completely different.


Since you can't hear noise and distortion above 20kHz (or less for many people), if you want a (N+D) measurement which doesn't give a spuriously high figure caused by out-of-band noise a filter is needed. It's nothing to do with making class-D look better, it's needed for the same reason as measuring oversampled ADCs which also generate inaudible out-of-band noise -- and which are the signal source for most music that you hear.

Then you ought to measure intermodulation distortion where the distortion products do fall in the audible bandwidth, unlike harmonics. This should correlate better with what we hear anyway, especially in an amp with loop gain which falls with frequency.

It's long been established that even quite high levels of harmonic distortion are inaudible anyway because the ear can't pick out low-level harmonics in the presence of a much bigger fundamental -- intermodulation is different, it generates components which weren't there in the first place and the ear can pick out at lower levels.

There are amps with good harmonic distortion which sound crap and vice versa, THD in itself isn't really so useful as a metric of how good an amp is -- it can show whether there's a problem, but even this often doesn't hold up.

One problem with either THD or intermod measurements is that they're steady-state, they don't show effects like dynamic crossover distortion caused by bias loops which don't track output transistor temperature changes when the signal level is varying rapidly like it does on music, or similar effects in class-D amps when power levels change rapidly.
Back to Top
abeltronics View Drop Down
Registered User
Registered User


Joined: 16 July 2009
Location: Norwich
Status: Offline
Points: 265
Post Options Post Options   Thanks (0) Thanks(0)   Quote abeltronics Quote  Post ReplyReply Direct Link To This Post Posted: 14 July 2010 at 9:42pm
Originally posted by IanD IanD wrote:


Since you can't hear noise and distortion above 20kHz (or less for many people), if you want a (N+D) measurement which doesn't give a spuriously high figure caused by out-of-band noise a filter is needed. It's nothing to do with making class-D look better, it's needed for the same reason as measuring oversampled ADCs which also generate inaudible out-of-band noise -- and which are the signal source for most music that you hear.

Then you ought to measure intermodulation distortion where the distortion products do fall in the audible bandwidth, unlike harmonics. This should correlate better with what we hear anyway, especially in an amp with loop gain which falls with frequency.

It's long been established that even quite high levels of harmonic distortion are inaudible anyway because the ear can't pick out low-level harmonics in the presence of a much bigger fundamental -- intermodulation is different, it generates components which weren't there in the first place and the ear can pick out at lower levels.

There are amps with good harmonic distortion which sound crap and vice versa, THD in itself isn't really so useful as a metric of how good an amp is -- it can show whether there's a problem, but even this often doesn't hold up.

One problem with either THD or intermod measurements is that they're steady-state, they don't show effects like dynamic crossover distortion caused by bias loops which don't track output transistor temperature changes when the signal level is varying rapidly like it does on music, or similar effects in class-D amps when power levels change rapidly.
 
I completely agree with almost all of that, I don't believe the last few decimals of a percent of THD make that much difference. I tested a Conrad-Johnson 6 channel valve power amp recently (about 15000 quid's worth or something) and the THD hovered about 2%. Mostly second harmonic mind. My customer loved the sound of it, although I was a bit unsure myself. I don't believe THD correlates well with people's perception of sound quality.
 
Much has been made of my THD measurement method, and FFA's main bone of contention with me was the high THD measurements of the FFA6004. But I consider them far less important than the appearance of the waveforms on the scope....
 
But the bottom line for me is this: Just because you can't hear it, doesn't mean it shoud be there.
 
DACs have an analogue filter at the output and often a digital filter implemented in the chip. When you stick a spectrum analyser on the DAC's output, you don't get artefacts upto several megahertz like on some class D amps. A lot of effort went into producing oversampling digital filters and high quality analogue back-end filtration during the development of mass-market consumer audio. I'd hate to think it's all been pointless....
http://www.abeltronics.co.uk - Electronic Engineers
Back to Top
IanD View Drop Down
Registered User
Registered User


Joined: 17 January 2009
Location: London
Status: Offline
Points: 400
Post Options Post Options   Thanks (0) Thanks(0)   Quote IanD Quote  Post ReplyReply Direct Link To This Post Posted: 14 July 2010 at 10:25pm
Originally posted by abeltronics abeltronics wrote:

Originally posted by IanD IanD wrote:


Since you can't hear noise and distortion above 20kHz (or less for many people), if you want a (N+D) measurement which doesn't give a spuriously high figure caused by out-of-band noise a filter is needed. It's nothing to do with making class-D look better, it's needed for the same reason as measuring oversampled ADCs which also generate inaudible out-of-band noise -- and which are the signal source for most music that you hear.

Then you ought to measure intermodulation distortion where the distortion products do fall in the audible bandwidth, unlike harmonics. This should correlate better with what we hear anyway, especially in an amp with loop gain which falls with frequency.

It's long been established that even quite high levels of harmonic distortion are inaudible anyway because the ear can't pick out low-level harmonics in the presence of a much bigger fundamental -- intermodulation is different, it generates components which weren't there in the first place and the ear can pick out at lower levels.

There are amps with good harmonic distortion which sound crap and vice versa, THD in itself isn't really so useful as a metric of how good an amp is -- it can show whether there's a problem, but even this often doesn't hold up.

One problem with either THD or intermod measurements is that they're steady-state, they don't show effects like dynamic crossover distortion caused by bias loops which don't track output transistor temperature changes when the signal level is varying rapidly like it does on music, or similar effects in class-D amps when power levels change rapidly.
 
I completely agree with almost all of that, I don't believe the last few decimals of a percent of THD make that much difference. I tested a Conrad-Johnson 6 channel valve power amp recently (about 15000 quid's worth or something) and the THD hovered about 2%. Mostly second harmonic mind. My customer loved the sound of it, although I was a bit unsure myself. I don't believe THD correlates well with people's perception of sound quality.
 
Much has been made of my THD measurement method, and FFA's main bone of contention with me was the high THD measurements of the FFA6004. But I consider them far less important than the appearance of the waveforms on the scope....
 
But the bottom line for me is this: Just because you can't hear it, doesn't mean it shoud be there.
 
DACs have an analogue filter at the output and often a digital filter implemented in the chip. When you stick a spectrum analyser on the DAC's output, you don't get artefacts upto several megahertz like on some class D amps. A lot of effort went into producing oversampling digital filters and high quality analogue back-end filtration during the development of mass-market consumer audio. I'd hate to think it's all been pointless....


Speaking as a DAC designer who also tests them, many oversampled DACs don't bother with high attenuation analogue filters at the output because they're not needed from the sound quality point of view, in fact more complex filters can degrade the quality -- as a consequence they can have significant levels of out-of-band noise, certainly enough to cause problems with measurement equipment unless this includes sufficient internal filtering.

I agree that because you can't hear it doesn't mean it should be there, that's what I meant when I said poor distortion *can* be a symptom of a real problem -- but also it doesn't *have* to be, as in the valve PA you mentioned.

What drives me up the wall is people who say "the distortion isn't very good so the amp will sound crap" because this may or may not be true -- like so many other bold statements in audio ;-)

Of course an amp which outputs 50V of ultrasonic crud *has* got a serious problem, and one which could merrily toast HF drivers, which is definitely A Bad Thing -- which I think is where this started...

Ian
Back to Top
cilla.scope View Drop Down
Old Croc
Old Croc

Legs all the way to her bottom....

Joined: 02 October 2009
Status: Offline
Points: 2954
Post Options Post Options   Thanks (0) Thanks(0)   Quote cilla.scope Quote  Post ReplyReply Direct Link To This Post Posted: 14 July 2010 at 10:48pm
I suspect it could just as easily toast LF and mid drivers as HF drivers ... significant amounts of HF current flowing in a voice coil that has no way to convert them into sound will produce heat, its the only thing it can do.

Agreed on the distortion though .. its like loud exhausts on motorbikes, it makes you think you are going faster ... a little distortion makes it sound louder. Ultimately though, it becomes tiring on the ears and muddles the soundstage. Personally, I would rather have it crisp and clean, but thats just me, others may have other opinions.

That said, the IPR looked very clean .. I have every confidence it would meet its published spec on an aes 17 filter


Edited by cilla.scope - 14 July 2010 at 10:50pm
My girlfriend thinks I'm a stalker.
Well, she's not exactly my girlfriend yet ...
Back to Top
abeltronics View Drop Down
Registered User
Registered User


Joined: 16 July 2009
Location: Norwich
Status: Offline
Points: 265
Post Options Post Options   Thanks (0) Thanks(0)   Quote abeltronics Quote  Post ReplyReply Direct Link To This Post Posted: 15 July 2010 at 1:16am
Originally posted by SickNeedHelp SickNeedHelp wrote:

Originally posted by nineleaves nineleaves wrote:

... they somehow went from the industrious & purposeful design of their old CS range to this crap ..
 
I think we still have some of the old 70's vintage 4U cast face plates around. We can fill the other 2U of the thick steel chassis with bricks.
 
How many should I put you down for?


Was that sarcasm? You can't be in America surely.... Wink
http://www.abeltronics.co.uk - Electronic Engineers
Back to Top
SickNeedHelp View Drop Down
Registered User
Registered User


Joined: 07 March 2010
Location: Meridian MS
Status: Offline
Points: 15
Post Options Post Options   Thanks (0) Thanks(0)   Quote SickNeedHelp Quote  Post ReplyReply Direct Link To This Post Posted: 15 July 2010 at 7:25pm
Originally posted by abeltronics abeltronics wrote:


Was that sarcasm? You can't be in America surely.... Wink
 
Sarcastic? That's as bad as accusing an engineer of having an ego or being defensive when someone comments on his design failures. 
 
Now my feelings are hurt ...... Cry


Edited by SickNeedHelp - 15 July 2010 at 11:09pm
Back to Top
 Post Reply Post Reply Page  <1 234

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.06
Copyright ©2001-2023 Web Wiz Ltd.

This page was generated in 0.172 seconds.