http://www.zaphaudio.com/tweetermishmash/csd.html
What exactly does CSD mean? How does it relate to real world speaker response, and how do I read to graphs to tell which is a good driver?
CSD ?
What I can gather is that it shows how the speaker stores energy... so basically what happens to a speaker after a sound stops. Certain tweeters can "ring" at specific frequencies, and this should be visible on the graph down the millisecond axis (look for the green tails), as it resonates beyond the note being played.
So just imagine a speaker is being given power, and then the power is cut. The speaker goes back to dead center, but it will still resonate a bit around there due to the stored energy. It's a lot like dropping a ball... each bounce (or resonance for a speaker) will decrease in motion. The stiffer a speaker's suspension, and other things, determine how quickly it will stop.
What is graphed is... at the very back (dark red) you have the frequency response of the driver. As you move up the millisecond axis, that is as time is increasing, you can see how the speaker still makes sound after the signal has stopped.
The ideal graph would be the one that extends the least, so basically in those graphs you linked, red = good, green = bad. The Seas 25TFFN looks really good in those graphs. This is funny, because it is a pretty cheap driver, which compares favorably against many higher end ones. Does that mean it will sound better? Absolutely not.
Be aware that these are differences of milliseconds, so I wouldn't consider this a determining factor for anything.
So just imagine a speaker is being given power, and then the power is cut. The speaker goes back to dead center, but it will still resonate a bit around there due to the stored energy. It's a lot like dropping a ball... each bounce (or resonance for a speaker) will decrease in motion. The stiffer a speaker's suspension, and other things, determine how quickly it will stop.
What is graphed is... at the very back (dark red) you have the frequency response of the driver. As you move up the millisecond axis, that is as time is increasing, you can see how the speaker still makes sound after the signal has stopped.
The ideal graph would be the one that extends the least, so basically in those graphs you linked, red = good, green = bad. The Seas 25TFFN looks really good in those graphs. This is funny, because it is a pretty cheap driver, which compares favorably against many higher end ones. Does that mean it will sound better? Absolutely not.
Be aware that these are differences of milliseconds, so I wouldn't consider this a determining factor for anything.
Last edited by stipud on Mon Dec 11, 2006 10:36 am, edited 2 times in total.
-
Francious70
- Half Baked
- Posts: 3533
- Joined: Wed Dec 06, 2006 2:58 pm
- Location: TN, YEEEEEEEEEEEEEEEEEEHAW!!!!
- Contact:
CSD is essentially the frequency resonance that occurs from a reflective surface or improperly damped cone.
(X axis = frequecy, Y= decibles, Z(?)= time)
If you play a fundemental note as shown in the dark red on the graphs, that you linked, each lighter shade is the a reflection or a cone fluctuation from that fundemental. A speaker which is well damped and has nothing to reflect off of should die almost completely off in a very short time as seen in the graph below which would also be my first choice of tweeter from what is graphed if all I had to go by was graphs alone.
(X axis = frequecy, Y= decibles, Z(?)= time)
If you play a fundemental note as shown in the dark red on the graphs, that you linked, each lighter shade is the a reflection or a cone fluctuation from that fundemental. A speaker which is well damped and has nothing to reflect off of should die almost completely off in a very short time as seen in the graph below which would also be my first choice of tweeter from what is graphed if all I had to go by was graphs alone.
I think we've established that "Ka Ka" and "Tukki Tukki" don't work.
Eloquently put Errin!
I agree with the choice of tweet. The Seas 25 has great CSD features, but the frequency response looks pretty messed up. The Hiquphon OW1 looks like it would be a killer above 2kHz, and it's CSD is very well constrained and similar across all frequencies. The Seas 29 and Vifa 26 look really good as well.
I agree with the choice of tweet. The Seas 25 has great CSD features, but the frequency response looks pretty messed up. The Hiquphon OW1 looks like it would be a killer above 2kHz, and it's CSD is very well constrained and similar across all frequencies. The Seas 29 and Vifa 26 look really good as well.
-
Francious70
- Half Baked
- Posts: 3533
- Joined: Wed Dec 06, 2006 2:58 pm
- Location: TN, YEEEEEEEEEEEEEEEEEEHAW!!!!
- Contact:
thanks guys...
The OW1 and OW4 are supeior to the rest of the tweets when it comes the CSD....
I am leaning towards the OW1s for my new tweet, the off axis performace of the Hiq are amazing, the frequency respose is amazingly flat, csd is the best, and the reveiws of how it sounds have been ++++. The only thing holding me back is their so so performace in HD. But this tweeter has basically knocked the expensive scan 6600 out of the running as I dont see how the extra 220 for a pair of drivers is worth.
The best over value looks like the vifa 26nc only negative from that tweeter is slight roll off @ ~ 15khz
The OW1 and OW4 are supeior to the rest of the tweets when it comes the CSD....
I am leaning towards the OW1s for my new tweet, the off axis performace of the Hiq are amazing, the frequency respose is amazingly flat, csd is the best, and the reveiws of how it sounds have been ++++. The only thing holding me back is their so so performace in HD. But this tweeter has basically knocked the expensive scan 6600 out of the running as I dont see how the extra 220 for a pair of drivers is worth.
The best over value looks like the vifa 26nc only negative from that tweeter is slight roll off @ ~ 15khz
Can one send others to war if hes not willing to go himself?