I was refereeing to the 0.25s silence I inserted between your burst in my wav file.
What were your measurement conditions? (distance, level, software, etc.)
I can do much better than that locally, but at $US120 + ~$50 shipping each from the US, why bother.
I still think a few CSDs are a better visual determinant of performance as I found yours awkward to compare.
Some details on how you actually measured them would help too.
Yes, as I mentioned before, this is just another set of tool. CSD while useful does not tell how waveforms are reproduced. The Tone Burst is actually even more useful if very high level is applied (as in 30vpp like SL). Unfortunately my microphone needs to be modified to enable such measurement.
The measurement conditions are:
Distance: On-axis, at the horn mouth plane. This is to sufficiently get horn's influence while avoiding diffraction effects from cabinet.
Level: I don't have SPL meter, so can't tell. But it's "normal listening" level. Ideally it should be at very high level/nearing tweeters' max spl.
Software: ARTA. SL has a very nice pic below how it's done (external excitation)
Spatial distortion
The measurement conditions are:
Distance: On-axis, at the horn mouth plane. This is to sufficiently get horn's influence while avoiding diffraction effects from cabinet.
Level: I don't have SPL meter, so can't tell. But it's "normal listening" level. Ideally it should be at very high level/nearing tweeters' max spl.
Software: ARTA. SL has a very nice pic below how it's done (external excitation)
Spatial distortion
Not sure what you mean by this so please expand.
Also with the DE250 and the Ewave copies you are using, are there any discontinuities between the adaptor and the WG. The Selenium adaptors I have need quite a bit of work to get them smooth and uniform in the same application (DE250 + JBL Ewaves) and the discontinuities in even the best of them are more obvious than the couple of cheap screw on drivers I have. This could account for the difference in the 6k5-8k band and as I find cheap Ti phrams much more prone to break up, I'd also be dubious about the highest test without an FR or CSD. I seem to recall a measurement on the Selenium in one of the Ewave threads and it had a narrow tall peak up high. Extra 'detail' is often extra 3H and a simple THD would have let that be seen easily.
Glancing back at the SL article (I'll read it again fully tomorrow) the Spectrum magnitudes would have made it easier to see differences than the time recording.
One last point is that tone testing is much more uncomfortable than noise or actual program, so the levels may be lower than you think.
Yes, if you read SL's page there is a very good example where burst test is more revealing than FR or CSD (in his example he compared 25TFF and Neo3). We all know Neo3 has excellent CSD, yet when subjected to burst test they clearly unable to reproduce the signals at the given condition. FR and CSD can mask all kinds of problems.
I am not an opponent of CSD, but I think it's an incomplete picture and the prettiness can be misleading.
Adapter for DE250 is fine, I had to modify it to make it smooth but not something that's too difficult to do.
I am not an opponent of CSD, but I think it's an incomplete picture and the prettiness can be misleading.
Adapter for DE250 is fine, I had to modify it to make it smooth but not something that's too difficult to do.
here are the CSD you posted in another thread, for reference:
http://www.diyaudio.com/forums/multi-way/178806-s16-constant-directivity-dipoles-8.html#post2582002
Have you think about doing the same burst test for a with/without foam comparison?
http://www.diyaudio.com/forums/multi-way/178806-s16-constant-directivity-dipoles-8.html#post2582002
Have you think about doing the same burst test for a with/without foam comparison?
I'm available also. Pro Sound Service has the DE250 for 120$US each. I'm on the East Coast US though so shipping may be higher, but not sure of that.
http://www.prosoundservice.com/m9/BC-DE250--b-c-de250-1-horn-hf-driver-8-ohm.html
Last edited:
When I get stuff from Parts Express it comes in a big box with lots of packing material.
When I ship parts to people in Europe/Asia I often use "Flat Rate Boxes" and pack them much tighter than those companies do. So, a few times I've saved people on the forums a lot of money on the shipping. I also take PayPal, which Parts Express does not do (they only accept PayPal from US customers). Those are some reasons why I bothered to make that offer...
PSS will take your CC direct and ship packages that are not over packed. I have actually purchased several items from them and found them much easier and cheaper to deal with than PE for B&C drivers. Their service is excellent compared to PE which for purchasers down here is a PITA. Why involve paypus at all?
Last edited:
Was there a horn attached? How was the system setup? The use of single frequencies could easily be misleading since, for example, one driver might have a slight resonance at 7000 Hz and another at 7100 Hz. I am a little concerned that you are placing too much meaning into your results. I admit I can "see" the differences, but relating that to your subjective comments is another thing altogether. You admitted to having drawn your conclusions prior to even doing these tests.
gainphile, could you tell us how you tested this sound quality? In my case, with simple high pass filter, the tonal difference among the drivers is quite significant, but once you equalize their responses to be the same, they sound fairly similar, no matter what material the diaphragm is made of.
I do very much appreciate gainphile's openmindness, curiosity and enthusiasm! It reflects the real nature of DIY AUDIO!
We are not pros (most of us here) so, please take all ideas and measurement results as just pieces of information (data). The discussion often opens up the whole picture and set achieved data in the frame. That's why it is published here!
It is likely that his results show the horn's effect or just deviation in mic position. A driver and horn (and baffle/box) are a system, he measured that system. Conclusions were drawn for only drivers - that was a flaw. But he was open and told us what he was doing. Others didn't publish results, so the topic was not finished.
SL's original result were revealing but it is very likely to get a bad result from NeoCD2 at 1,5kHz. But it was relevant info for his purposes.
We are not pros (most of us here) so, please take all ideas and measurement results as just pieces of information (data). The discussion often opens up the whole picture and set achieved data in the frame. That's why it is published here!
It is likely that his results show the horn's effect or just deviation in mic position. A driver and horn (and baffle/box) are a system, he measured that system. Conclusions were drawn for only drivers - that was a flaw. But he was open and told us what he was doing. Others didn't publish results, so the topic was not finished.
SL's original result were revealing but it is very likely to get a bad result from NeoCD2 at 1,5kHz. But it was relevant info for his purposes.
Last edited:
\
Horn not specified. Probably the same for both. Gainphile can verify.