john k... said:
This is in stark contrast to our study, which found that levels much higher than this, approaching 20%, were inaudible.
I don't have that book, could you ellaborate more on the test.
Thinking out loud, it seems just from mathematics, a formula can be generated as a guidance for selecting driver size and frequency range to keep doopler distortion below certain desired levels.john k... said:
Using MLS to conduct tests like this seems very interesting. Seems that using a program like SoundEasy, one can just run the MLS signal first, take a measurement to generate frequency response, then continue runing MLS in Auto mode until the desired time is reached, then stop it to generate a new frequency response right over the first one, etc. and do it for as many times as you like. This method also has an advantage of taking out the effects of amp performance variation as well.
Colloms states in his book that FMD becomes significant in fullranges as soon as the xmax reaches 3mm peak. If you keep this value for multiway speakers (where it is less of an issue) you are on the safe side - given that Colloms' statement is true. It is then not that difficult to determine "optimal" driver size for each frequency band.
As far as LMS and power compression goes: The most clever proposal to measure power compression I have seen so far is indeed Dave's in post #58. This should be a very reliable and repeatable type of measurement IMO. What is open for discussion is how the signal should be shaped in terms of frequency and level rise.
Regards
Charles
So investing serious money in a PD 21" with its 150mm voice coil dont seem so crazy after all ... and the use of a 12" midwoofer
And could the good perceived sound from ribbons come from lack of thermal compression, despite their higher distortion ... this really begins to interest me
I never thought thermal compression to be an issue in normal hifi ... seems we may have to consider this
And could the good perceived sound from ribbons come from lack of thermal compression, despite their higher distortion ... this really begins to interest me
I never thought thermal compression to be an issue in normal hifi ... seems we may have to consider this
At least you fellow citizens do
See "dynamic measurements" here:
http://www.gattiweb.com/images/dynaudio/esotec-d260_data.pdf
And the power compression scope diagrams down left:
http://www.gattiweb.com/images/dynaudio/21w54_data.pdf
Regards
Charles
Earl,
Martin Collom’s book doesn't describe the tests, (at least not in detail when it does). It references the works of a number of authors and reports on the results with some additional comments. But I think more importantly is recognizing that the quoted results only provide indications of the threshold when FM is audible. Other comments in the book suggest that under normal listening conditions with normal program material (normal is considered 85dB at 2 M) these effect probably aren't significant (masked) in most cases.
However, I have done my own tests on FM, http://www.geocities.com/kreskovs/Doppler1.html, and at the levels shown in the last figure the FM distortion was clearly audible to me at -40dB, but would probably be masked with full range source material. On the other hand, with many dipole speakers the excursion of the midrange drivers is really being pushed to the limits and I think nonlinear distortion, both HD and FM, may be a more significant issue.
Thinking out loud a little more about dynamics I would think that rather simple tests would suffice. For example, the impulse response could be measured at a series various levels, making sure that the system was always at the same initial condition (VC and motor structure temperature, etc. This would be a linearity test of sorts. The max output level of the pulse could be compared to the input peak level as a measure of dynamic compression. For example if a pulse with 1 V peak input generated an output with a peak of .8 V than an input with a peak of 4 V would be expected to generate a peak output of 3.2V.
Then another series of tests could be conducted with raised VC/motor temperature. For example, the speaker would first be subjected to play back of a specific selection of source material over some well defined period of time. Then the system would be impulse tested as described above. The system would be allowed to cool and the test then repeated form scratch. This would take time, but would assure fairly uniform initial conditions for each test. This second series of tests could be conducted for different play back levels to look at the effects of different motor temperature.
Martin Collom’s book doesn't describe the tests, (at least not in detail when it does). It references the works of a number of authors and reports on the results with some additional comments. But I think more importantly is recognizing that the quoted results only provide indications of the threshold when FM is audible. Other comments in the book suggest that under normal listening conditions with normal program material (normal is considered 85dB at 2 M) these effect probably aren't significant (masked) in most cases.
However, I have done my own tests on FM, http://www.geocities.com/kreskovs/Doppler1.html, and at the levels shown in the last figure the FM distortion was clearly audible to me at -40dB, but would probably be masked with full range source material. On the other hand, with many dipole speakers the excursion of the midrange drivers is really being pushed to the limits and I think nonlinear distortion, both HD and FM, may be a more significant issue.
Thinking out loud a little more about dynamics I would think that rather simple tests would suffice. For example, the impulse response could be measured at a series various levels, making sure that the system was always at the same initial condition (VC and motor structure temperature, etc. This would be a linearity test of sorts. The max output level of the pulse could be compared to the input peak level as a measure of dynamic compression. For example if a pulse with 1 V peak input generated an output with a peak of .8 V than an input with a peak of 4 V would be expected to generate a peak output of 3.2V.
Then another series of tests could be conducted with raised VC/motor temperature. For example, the speaker would first be subjected to play back of a specific selection of source material over some well defined period of time. Then the system would be impulse tested as described above. The system would be allowed to cool and the test then repeated form scratch. This would take time, but would assure fairly uniform initial conditions for each test. This second series of tests could be conducted for different play back levels to look at the effects of different motor temperature.
About the Dynaudio tweeter shown ... it says it will take 1000watt peaks without compression, but it doesnt say how many times it will last that
I think I saw at PD site that power compression is measured at a standard of 10% of max power
I think I saw at PD site that power compression is measured at a standard of 10% of max power
JBL for instance state the value fo 10%, 50% and 100%. So how much is the compression for 10% for your PD woofer ?
Regards
Charles
Regards
Charles
john k... said:
John,
I agree with your first sentence above, but then you go on to describe something that is rather complex.
I tried the test at different power levels and not until on reached a fairly significant level was there much to look at. Thats why I just turned on the "juice" and tracked the changes as the system heated. This showed some very real effects in a very simple test. As has been stated here, it was probably more power than the system would see in Hi-fi, but let me tell you, its NOT more power than it would see in the clubs in Bangkok.
I've not been much for "clubing" in the past, but in Bangkok its a must see experience. Its cheap - like everything else in Thailand - and the scenery is not to be believed.
Sorry, I missread the information on PD drivers
http://www.precisiondevices.co.uk/showdetails.asp?id=16
1. Power compression is the reduction of sensitivity at the specified power ( 2db at 1000watt)
2. Distortion is measured at 10% of the rated power (AES standard), less than 1%
http://www.precisiondevices.co.uk/showdetails.asp?id=16
1. Power compression is the reduction of sensitivity at the specified power ( 2db at 1000watt)
2. Distortion is measured at 10% of the rated power (AES standard), less than 1%
soongsc said:
Well as I have said before, I think the difference is symantic, but yes, our primary focus is pro. The consumer market is too insane for me.
phase_accurate said:
I originally designed the Summa for my own Home Theater installations (at first my own). I was not happy with any of the loudspeakers on the market for this application. Home Theater has some really high SPL demands (Commercial Theaters hit 120 dB SPL at the seats) and Hi-Fi speakers just didn't cut it at these levels. But the Pro stuff tended to sound so bad - mostly the poor horn designs.
So I guess that you could say that Home Theater was my target market. I still think that's where the biggest opportunity lies. Once you hear a Home Theater with really spectacular sound - and I mean sound better than a commercial theater - nothing else comes close.
I don't buy the agument that different speakers are required for the different applications. There typically are different speakers for these applications, but thats not a requirement. It came about because of the serious compromises required for high output in terms of sound quality. If one can get Hi-Fi sound with Pro sound output levels then there is no compromise necessary.
I certainly think outdoor speakers are going to be different from indoor speakers.😉gedlee said:
soongsc said:
UNTRUE!! Our speakers are virtually waterproof. As long as they point downwards so that water cannot collect in the woofer cone, they can be used outdoors with no problem. Remember they are not wood. The only sensitive part is the woofer cone and this is said to be made weather resistant.
A pair of Summas went through Katrina. The woofers were ruined from floating in the water, some internal mold damage, but otherwise they were fine. They are back functioning as we speak.
I wonder (out loud);
It seems to me that since, from the transducer point of view that “what goes wrong” with a loudspeaker is strongly related to level, that there is another way one could examine what’s going on, taking advantage of a field where a lot more $ has been spent..
http://www.matrixtest.com/Literat/MTN109.htm
http://en.wikipedia.org/wiki/Third-order_intercept_point
I would bet this frame of reference would be useful for a “figure of merit” comparison if one based the comparisons on a specific sound level and frequency band. Tracking dynamic nonlinearity makes sense.
Best,
Tom Danley
Earl is correct the difference between musical sounding pro-speakers and home is that they have to produce a lot more acoustic power, the goals are the same / very similar, sizes and configurations and room effects make the design part harder.
After all if your making speakers for an auditioning room which seats 50 people, what you want to present is no different than what you would ideally present for a Theater of 500 people or two on a couch.. The distances to the audience / room size just make it perhaps exponentially harder, the larger the audience and room is..
It seems to me that since, from the transducer point of view that “what goes wrong” with a loudspeaker is strongly related to level, that there is another way one could examine what’s going on, taking advantage of a field where a lot more $ has been spent..
http://www.matrixtest.com/Literat/MTN109.htm
http://en.wikipedia.org/wiki/Third-order_intercept_point
I would bet this frame of reference would be useful for a “figure of merit” comparison if one based the comparisons on a specific sound level and frequency band. Tracking dynamic nonlinearity makes sense.
Best,
Tom Danley
Earl is correct the difference between musical sounding pro-speakers and home is that they have to produce a lot more acoustic power, the goals are the same / very similar, sizes and configurations and room effects make the design part harder.
After all if your making speakers for an auditioning room which seats 50 people, what you want to present is no different than what you would ideally present for a Theater of 500 people or two on a couch.. The distances to the audience / room size just make it perhaps exponentially harder, the larger the audience and room is..
Gee, We seem to be going backwards here. On the first page of this thread it was suggested 1) to measure the "scaling" ie, "double" the level of the impulse to see if you get "double" the output, 2) to use impulses, 3) to use impulses (or MLS) on top of a carrier (so its response could be easily subtracted). Gee the level of the carrier and the onset time of the pulse (or duration of the carrier) ould be easily manipulated and measured.gedlee said:
A reasonable and interesting question has become side-tracked.
-Tom
- Status
- Not open for further replies.